The Captor's Image: Greek Culture in Roman Ecphrasis (Classical Culture and Society) [1 ed.] 9780199735877, 0199735875

An influential view of ecphrasis--the literary description of art objects--chiefly treats it as a way for authors to wri

132 100

English Pages 304 [294] Year 2013

Report DMCA / Copyright

DOWNLOAD PDF FILE

Table of contents :
Cover
Contents
Acknowledgments
Abbreviations
Introduction. Greek Culture in Roman Ecphrasis
1. Staging Ecphrasis in Early Latin Literature: From Naevius to Plautus and Terence
2. Becoming Ariadne: Marveling at Peleus’s Coverlet with the Inconsistent Narrator of Catullus 64
3. The Challenge of Rustic Art: Ideals of Order in Vergil, Eclogues 3 and Horace, Satires 1.8
4. Describing the Divine: The Ecphrastic Temples of Vergil, Georgics 3.13–36 and Propertius, Elegies 2.31
5. Heroic Objects: Ecphrasis in the Aeneid and Metamorphoses
6. Sex, Satire, and the Hybrid Self in Petronian Ecphrasis
7. The Patron’s Image: Philhellenism, Panegyric, and Ecphrasis in Statius and Martial
Epilogue. Captives and Captors: Apuleius and Philostratus
Bibliography
Index
A
B
C
D
E
F
G
H
I
J
K
L
M
N
O
P
R
S
T
V
W
X
Z
Recommend Papers

The Captor's Image: Greek Culture in Roman Ecphrasis (Classical Culture and Society) [1 ed.]
 9780199735877, 0199735875

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

The Captor’s Image

CLASSICAL CULTURE AND SOCIETY Series Editors Joseph Farrell and Robin Osborne Emotion, Restraint, and Community in Ancient Rome Robert A . Kaster Making Mockery: The Poetics of Ancient Satire Ralph M. Rosen Readers and Reading Culture in the High Roman Empire: A Study in Elite Communities William A . Johnson Apollonius of Rhodes and the Spaces of Hellenism William G. Thalmann The Captor’s Image: Greek Culture in Roman Ecphrasis Basil Dufallo

The Captor’s Image Greek Culture in Roman Ecphrasis

B A S I L D U FA L L O

1

3 Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford New York Auckland Cape Town Dar es Salaam Hong Kong Karachi Kuala Lumpur Madrid Melbourne Mexico City Nairobi New Delhi Shanghai Taipei Toronto With offices in Argentina Austria Brazil Chile Czech Republic France Greece Guatemala Hungary Italy Japan Poland Portugal Singapore South Korea Switzerland Thailand Turkey Ukraine Vietnam Oxford is a registered trademark of Oxford University Press in the UK and certain other countries. Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016

© Oxford University Press 2013 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by license, or under terms agreed with the appropriate reproduction rights organization. Inquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above. You must not circulate this work in any other form and you must impose this same condition on any acquirer. Library of Congress Cataloging-in-Publication Data Dufallo, Basil. The captor’s image : Greek culture in Roman ecphrasis / Basil Dufallo. p. cm.—(Classical culture and society) Includes bibliographical references and index. ISBN 978–0–19–973587–7 1. Latin literature—History and criticism. 2. Ekphrasis. 3. Greek literature—Influence. 4. Authors, Latin. 5. Art, Greek—Influence. 6. Civilization, Greco-Roman. I. Title. PA6021.D84 2012 870.9´22—dc23 2012017753 ISBN 978–0–19–973587–7

1 3 5 7 9 8 6 4 2 Printed in the United States of America on acid-free paper

To Cathy, Will, and Sylvie.

This page intentionally left blank

CONTENTS

Acknowledgments ix Abbreviations xi Introduction. Greek Culture in Roman Ecphrasis 1

1. Staging Ecphrasis in Early Latin Literature: From Naevius to Plautus and Terence 14 2. Becoming Ariadne: Marveling at Peleus’s Coverlet with the Inconsistent Narrator of Catullus 64 39 3. The Challenge of Rustic Art: Ideals of Order in Vergil, Eclogues 3 and Horace, Satires 1.8 74 4. Describing the Divine: The Ecphrastic Temples of Vergil, Georgics 3.13–36 and Propertius, Elegies 2.31 108 5. Heroic Objects: Ecphrasis in the Aeneid and Metamorphoses

137

6. Sex, Satire, and the Hybrid Self in Petronian Ecphrasis 177 7. The Patron’s Image: Philhellenism, Panegyric, and Ecphrasis in Statius and Martial 206 Epilogue. Captives and Captors: Apuleius and Philostratus 244 Bibliography 253 Index 273

vii

This page intentionally left blank

ACKNOWLEDGMENTS

I have been especially fortunate to finish this book in the superb intellectual environment of the Institute for the Humanities at the University of Michigan, where I was Hunting Family Faculty Fellow from 2010 to 2011. Special thanks are due to the Director, Daniel Herwitz, and to the other members of the Institute’s seminar for that year. My warm thanks as well to my colleagues in the Classical Studies and Comparative Literature departments and in the Contexts for Classics research consortium at U-M for providing the vibrant atmosphere in which this project first took shape and grew. In particular, the help of a number of friends was especially valuable: Benjamin Acosta-Hughes, Michael Dewar, Duncan Kennedy, David Potter, and Jay Reed provided detailed suggestions and criticisms on significant portions of the manuscript as it developed; Micaela Janan and David Wray offered many helpful observations on the shape and presentation of the project as a whole. Claudia Arno provided crucial editorial assistance. My sincere thanks in addition to Alessandro Schiesaro, Jonathan Powell, Maria Wyke, and the Institute of Classical Studies Latin Seminar at the University of London, and to audiences at Bristol University, Brown University, and the APA, CAMWS, and ACLA conferences, all of whom offered welcome comments on papers drawn from this book. I am also grateful to the Dean’s Offices at U-M’s College of Literature, Science, and the Arts and Rackham Graduate School, as well as to the departments of Classical Studies and Comparative Literature, for funds that allowed for essential research travel to Italy in 2008 and 2010. Part of Chapter 2 was originally published in Cultural Critique 74 (2010) 98–113. My thanks to Jeff Moen for permission to republish this. In addition, part of Chapter 6 originally appeared as “Ecphrasis and Cultural Identification in Petronius’ Art Gallery,” Word & Image 23.3 (2007) 290–304 and is reprinted by permission of the publisher (Taylor & Francis Ltd, http://www.informaworld.com). The New York office of Oxford University Press has made the entire publication process a pleasure. Many thanks to Stefan Vranka for his early interest in ix

x

Acknowledg ments

the project and especially for his practical advice about its final form. I am also deeply grateful to the series editors, Joseph Farrell and Robin Osborne, whose remarks provided much-appreciated help at an important juncture. Abundant thanks as well to Deirdre Brady, Sarah Pirovitz, Sravanthi Sridharan, Jay Boggis, and Lisa Stallings. The two anonymous readers for the press offered many invaluable observations and will recognize my indebtedness to them. In addition, I would like to thank the following persons, each of whom has in one way or another provided valued assistance: Paolo Asso, Tim August, Robin Axelrod, Mariette Baker, Jennifer Belt, Michelle Biggs, Tom Cannavino, Beau Case, Ruth Caston, Doretha Coval, Rosanna Di Pinto, Maria Daniela Donninelli, Robert W. Dunkin, Andrew Feldherr, Paula Frank, Onorina-Fiamma Fulgenzi, Elaine Gazda, Jessica Feinstein, Robert Germany, Tom Habinek, David Halperin, Michèle Hannoosh, Stephanie Harrell, Volker Heuchert, John Dixon Hunt, Terry Janzen, Richard Janko, Kimberly Johnson, Anika Keller, Daria Lanzuolo, Eleanor Windsor Leach, Paolo Liverani, Andrew Meadows, Carole Newlands, Antonio Paolucci, Yopie Prins, Michael Putnam, Matthew Roller, Maurizio Rulli, Ruth Scodel, Sonia Schmerl, Mira Seo, Michael Slade, Chris Sutherns, and Debbie Walls. Finally, I express my profound love and thanks to my wife, Catherine Sanok, and to my children, Will and Sylvie, for countless forms of support, companionship, and patience. This book is dedicated to them.

A B B R E V I AT I O N S

For abbreviations of ancient authors and works, I have followed the lists in the Oxford Latin Dictionary, edited by P. G. W. Glare, and Liddell, Scott, Jones, and McKenzie’s A Greek-English Lexicon. Except where indicated, I have used the most recent Oxford or Teubner edition and translations are my own. Austin-Bastianini Keil Lobel and Page LSJ

OLD Pfeiffer Skutsch Strzelecki Thilo and Hagen

Wendel

Colin Austin and Guido Bastianini, eds. Posidippi Pellaei quae supersunt omnia. Milan: LED 2002. Heinrich Keil, ed. Grammatici Latini. 8 vols. Leipzig: Teubner 1855–1880. Edgar Lobel and Denys Page, eds. Poetarum Lesbiorum fragmenta. Oxford: Clarendon Press 1955. Henry George Liddell and Robert Scott, eds. A Greek-English Lexicon. Revised and augmented by Sir Henry Stuart Jones and Roderick McKenzie. Oxford: Clarendon Press 1996. P. G. W. Glare, ed. Oxford Latin Dictionary. Oxford: Clarendon Press 1982. Rudolf Pfeiffer, ed. Callimachus. 2 vols. Oxford: Clarendon Press 1949–1953. Otto Skutsch, ed. The Annals of Q. Ennius. Oxford: Clarendon Press 1985. Wladyslaw Strzelecki, ed. Cn. Naevii Belli Punici carminis quae supersunt. Leipzig: Teubner 1964. Georg Thilo and Hermann Hagen, eds. Servii grammatici qui feruntur in Vergilii carmina commentarii. 3 vols. Leipzig: Teubner 1881–1902. Carl Wendel, ed. Scholia in Theocritum vetera. Stuttgart: Teubner 1914.

xi

This page intentionally left blank

The Captor’s Image

This page intentionally left blank

Introduction Greek Culture in Roman Ecphrasis

Greek culture came to the Romans in many forms, but perhaps none so freighted with ambiguity—with pleasure and suspicion, popular appeal and elite sophistication, material value and immaterial sublimity—as the Hellenic art object. My long fascination with this Roman attitude has been critical to the genesis of this book. What follows, however, is not primarily about actual Greek art, but rather about literary descriptions of art that is somehow Greek, whether in provenance, subject matter, embodied ideals, or intertextual allusion: ecphrases in the narrow, modern sense of the term.1 Vergil’s account of sculptures by the legendary Greek artist Daedalus in Aeneid 6, with its visual narrative of deception, lust, wandering, and loss uncannily similar to Aeneas’s own, is among the most familiar examples of my subject matter in Latin literature.2 The rapt attention with which Aeneas scrutinizes these images, and Vergil’s reminiscences here of earlier poets both Greek and Latin, well illustrate my main argument: that ecphrases like this one stage a larger, ambivalent receptivity to Greek culture, a 1

Ἔκφρασις is a Greek rhetorical term originally meaning “description” in a broad sense. By late antiquity, descriptions of works of art (ἐκφράσεις ἀγαλμάτων) were recognized formally as a subcategory, if not a full-fledged genre, and the term “ecphrasis” has been used in modernity with this exclusive sense. For helpful overviews, see Becker 1995: 2n1; Zanker 2004: 6–7; Gurd 2007: 306n6; and for the development of the modern notion, Webb 1999. Recently Webb has called renewed attention to the breadth of the ancient trope (as also directed, that is, at objects other than visual art) and its particular meaning in the Progymnasmata (a series of beginning rhetorical exercises dating from the first to the fifth century CE) and other rhetorical handbooks (Webb 2009). Although I agree with Webb about the trope’s essentially rhetorical nature in the ancient context (cf. below, n25) as well as about the importance of not confusing the ancient and modern definitions of the term, I nevertheless find the modern term useful to name a recognizably distinct type of ancient description (cf. Elsner 2002: 2–3; Francis 2009: 4; Goldhill 2009; Squire 2009: 143–44). With my understanding of ancient artistic ecphrasis, at least from the Hellenistic period in Greece, as including not only description but also supplementation and interpretation, cf. Zanker 2003; 2004: 7–16, 82–86 and many of the studies of Latin authors cited in what follows. 2 A detailed discussion of this passage appears in Chapter 5 below.

1

2

introduction

set of changing social attitudes reflecting the rapidly shifting political conditions of the late Republic and early Principate.3 Latin authors use ecphrasis not simply to assert Roman dominance over Greece, as one might expect given the fact that so much Greek art arrived in Rome as plunder, but also, in a manner especially revelatory of Roman self-consciousness over the nuances of this relationship, to construct Roman identity as Greek, ambiguously, and with varying purposes, in social, cultural, and political terms.4 This book is therefore polemical, and it may assist readers approaching it in all innocence to spell out right away the view of ecphrasis I effectively reject. Simply put, this rejected view consists in treating ecphrasis primarily as a way in which authors can write about writing without appearing to do so. No sensitive reader of an ecphrasis like the Vergilian passage just mentioned could deny that it is programmatic in an important way: that Daedalus’s sculpture stands 3 With my use of the metaphor of staging, cf. Elsner 2007: 87 (on this and other ecphrases in the Aeneid): “We are effectively in the mightily complex world of a Roman visuality in which the act of spectatorship is defined as being on stage and observed as one looks.” See further Chapter 1 below. 4 The absence of a book on this subject may seem surprising given the recent explosion of work on ancient ecphrasis, but it becomes more understandable once we see how much of this has been studies of individual authors rather than focused treatments of diachronic periods. Among a very many studies of specific Roman authors, one may cite Barchiesi 1997, Putnam 1998, Hardie 2002, Henderson 2002, and Schmale 2004. A more diachronic view of ecphrasis, though only partly synthesized as such, emerges from two special issues of journals, Ramus 31.1–2 (2002) and Classical Philology 102.1 (2007), which treat various aspects of the trope in both Greek and Latin literature. Ecphrasis is a theme in Porter’s recent study, The Origins of Aesthetic Thought in Ancient Greece (2010), as well as in the collection edited by Belloni et al., Le Immagini nel testo, il testo nelle immagini (2010), Squire’s Image and Text in Greco-Roman Antiquity (2009), Elsner’s influential collection, Art and Text in Roman Culture (1996), and its companion, Goldhill and Osborne’s Art and Text in Ancient Greek Culture (1994), but these volumes, too, range widely over a series of topics germane to their titles. And this is also true of comparative studies of imagery and aesthetics in Roman wall painting and literature, such as Leach 1988 and 2004 and Grüner 2004 (cf. Barchiesi 2005 on Augustan literature and the other arts). For comprehensive surveys of Greco-Roman ecphrasis, one can go back to Friedländer 1912 and Downey 1959, while Ravenna 1974 surveys prevalent themes in Latin examples with reference to Greek precedents. Prioux 2008 treats both Greek and Latin epigrams on artistic collections. Among the only classical scholarship on ecphrasis or closely related topics to embrace a temporal scope similar to my own is recent work on Hellenistic Greek literature, such as Manakidou 1993, Männlein-Robert 2007, and Prioux 2007. An important exception is the central section of Elsner 2007 (Chapters 4–7). Elsner makes a series of extremely valuable connections between ecphrasis in Catullus, Vergil, Ovid, Petronius, and Apuleius, the visual arts, and Roman attitudes toward the gaze. His theme is “the pattern of cultural constructs and social discourses that stand between the retina and the world, a screen through which the subjects of this inquiry . . . acquired (at least in part) their sense of subjectivity” (xvii). But as this broad definition itself suggests, no single argument about ecphrasis links these chapters, and while the matter of Greek culture’s influence on Roman culture is very much present, it, too, is not an organizing concern. My study complements Elsner’s by its systematic treatment of a question implicit in his own. For more on the Roman gaze as a phenomenon represented in literature, see Bettini 1999; Fredrick 2002; Bartsch 2006.

Introduc ti on

3

for the Aeneid and that the plastic art’s pathos and detail, for instance, give a kind of concrete form to those features of the larger poem in which it appears. The attention ecphrasis draws to the processes of representation and reception can make it a useful tool for analyzing how a literary text functions artistically. And through ecphrasis, a writer may, as is often pointed out, assert the verbal medium and its powers over those of the visual artist. Modern theory and criticism on ecphrasis have done much to propagate this view, even comparing such self-assertion to the colonizer’s domination of the colonized, while appearing in general much less invested in a perspective such as my own. Indeed, what I have just sketched amounts to a “standard” view of ecphrasis still often expressed in scholarship on antiquity and other periods.5 By contrast, my claim is that in Roman literature at least, ecphrasis is also, and more centrally, about competition between cultures—Greek and Roman, literary and visual. By “competition,” however, I refer to something far more complex and subtle than simply overt, agonistic struggle or attempts at domination. Roman ecphrasis, in spite of its associations with conquest and colonization, paradoxically helps us perceive the cultural and political stakes inherent in the trope’s use when literary texts confirm, as much as they challenge, the priority of the visual image. As the verbalization of a seen, silent, and passive visual object by a seeing, speaking, knowledgeable subject, ecphrasis may sustain ideas of Roman rivalry with, even superiority to, Greek culture and society. And yet insofar as the ecphrastic image is associated, at the same time, with origins, divinity, full presence, and stability, as opposed to the viewer’s belated, partial, contingent, and biased representation of it, ecphrasis can work quite differently, pointing to a fascination with Greek culture and to its strong influence on Roman identity.6 As my title suggests, this book centers on the sentiment enshrined in Horace’s 5 On relations between the verbal and visual arts, especially in ecphrasis, as the site of texts’ programmatic self-assertion and struggle for dominance over images, see esp. Mitchell 1986, 1994: 151–81; Heffernan 1993. For the language of colonization applied to ecphrasis, see Bryson 1981: 18; Mitchell 1994: 157, 180. For the continued influence of Mitchell’s views on recent work in Classics, see many of the individual studies cited in the chapters to follow. Other texts that remain central to discussions of literary ecphrasis both in Classics and elsewhere include duBois 1982; Baxandall 1985; Krieger 1992; Becker 1995; and Hollander 1995. Becker 2003 offers a critique of the notion that a contest between the arts is a given in ancient ecphrasis. A compendious bibliography of recent work on ecphrasis is compiled by Wandhoff 2003: 1–12; see further Squire 2009: 140–46. 6 Note that neither my view nor the one I reject depends upon the actual existence of the art object in question (“actual” ecphrasis vs. “notional” in the terms of Hollander 1995: 4–5). Baxandall is right to affirm that there can be no absolute or pure description of an art object apart from earlier verbal description or specification, whether of the particular art object or of such objects in general: the way we talk about art always depends on something previously thought or said (cf. esp. Baxandall 1985: 1–5). This mitigates the force of the opposition between imaginary and real works of art, while harmonizing with a major trend in classical reception studies more generally, according to which the present meaning of any classical text (including artwork construed as a

4

introduction

famous dictum, “Captured Greece captured its uncivilized conqueror and brought the Arts to rustic Latium” (Ep. 2.1.156–57). Yet the very context of this remark in Horace’s poem, which includes a description of himself as a boy in southern Italy being schooled in a Latin translation of the Odyssey (69–71), reminds us that no simple model of dominance and submission is adequate to describe relations between Roman/Italian cultures and Greek/Hellenistic cultures over their long history.7 Latin authors, rarely limiting themselves to a “superior” and/or “inferior” position with respect to Greece, engage in imitation, rivalry, and innovation of many varieties.8 Thus alongside dominance and submission emerge paradigms of inheritance, transformation, shared purpose, and play, to name only a few. All these are important to Roman ecphrasis as I see it, and all, however muted, transient, or difficult to detect, are ways in which the Hellenic cultures to which ecphrasis alludes exert their sway over Roman subjects. Roman ecphrasis is a site of cultural competition, in other words, both in the way that various Roman and Hellenic cultures themselves can be said to compete, through ecphrasis, for influence over a Roman sense of self and in the way that Roman ecphrastic authors vie to display their receptivity to Greek culture (often in response to patrons who also wish to display such an attitude). The two phenomena are closely linked, since even the former, impersonal sort of competition occurs via the author’s receptive posture, as staged within broader cultural circumstances that favor a receptive response, in turn, from contemporary audiences: a broader Roman philhellenism expressed through visual, verbal, and other means. Thus a second aim of this book is to insist that Roman ecphrasis, for all its formal literariness as a “trope” framing a verbal image, must also

“text” available for “reading”) is always “constructed by the chain of receptions through which their continued readability has been effected,” making it impossible to recapture “any originary meaning wholly free of subsequent accretions” (Martindale 1993: 7 and, on the extended sense of “text” as “any vehicle of signification,” such as a work of visual art, 13. Cf. Martindale 2006; Hardwick and Stray 2008: 2–5). Even if I were to describe for you the Mona Lisa as we stood in front of it, this description would to some extent be that of a painting imagined and articulated by me for myself as the “real” Mona Lisa, available for construal as the image Leonardo intended us to see. The actual existence, however, of an artwork purportedly described by ecphrasis remains significant, and can carry enormous meaning at the social, cultural, and political levels, as it would, to anticipate the opening analysis of Chapter 1, in the event that the Latin poet Naevius’s ecphrastic Gigantomachy were actually the one on a Greek temple of Zeus in Sicily. 7 For an especially rich and illuminating survey of this context, see Wallace-Hadrill 2008. 8 We are speaking here in part, no doubt, of Bloom’s “anxiety of influence” and its effects upon the self-consciously belated author emulating his literary predecessors, but must take care nonetheless to avoid being bound exclusively by Bloom’s reductive psychologizing categories. For an application of psychoanalytic theory directly to the ecphrastic confrontation of object, viewer, and listener, see Elsner 2004. Extending his approach, one might posit that Roman ecphrasis in part plays out an Oedipal drama in which Rome seeks to assure itself of omnipotence through an (impossible) appropriation of a feminized, maternal Hellas in the latter’s fetishized images.

Introduc ti on

5

be understood as a cultural practice like other Roman cultural practices, deeply embedded within the changing collectivity of behaviors through which the Romans sought, year upon year and decade upon decade, to redefine Romanness with and against Greekness. Given this focus, let me hazard, at the outset, an analogy with just one feature of Rome’s wider visual culture. The dynamic I ascribe to ecphrastic images is very like the one surrounding the distressed Greek heroine figure (e.g., Ariadne, one of Vergil’s own focal points; cf. Chapter 2 on Catullus 64) in Pompeian wall paintings’ framed panels. Of these one scholar has observed that since they “remain marked as Greek, their position in the frame reenacts the process of conquest and control which brought them into the Roman sphere” but at the same time they are “a confession of interest in Greek culture, and in erotic pleasure,” othering the viewer as they confirm his (male) Roman identity.9 So, to return to my opening example, the Trojan, proto-Roman Aeneas, as he gazes intently at Daedalus’s images, reminds Vergil’s audience both of Daedalus himself and of another Greek, Theseus, insofar as Aeneas’s passion, too, has resulted in an abandonment and a distressed heroine, Dido. Through an authorial intervention, moreover, highlighting the absent image of Icarus’s fall (a figure for Aeneas’s loss of his father Anchises but also his and Theseus’s loss of beloved women), Vergil enfolds as well the perspectives of poet, narrator, and external audience within this web of viewing and desire. Seen in this light, the Aeneid passage shows how the formal properties of ecphrasis actually lend themselves to the ambivalent dynamic between Rome as receptive adaptor and Greece as received cultural model. Here the silent Greek art object, described by the Roman subject, Vergil, or his narrator, infuses this subject with voluble wonder, an emotion central to the ecphrastic tradition that the Romans inherited from the Greeks. This response, refracted through the ambiguously proto-Roman internal viewer, Aeneas, implies both the lack of speech characteristic of visual phenomena and the constitutive mystery and fascination of their silence, in this sense their aesthetic fullness. The place of the viewer in ecphrasis allows the author both to represent and to thematize reception; that is, to raise the question of receptivity within the literary work. How much of the Trojan Aeneas’s Greekness (or Theseus’s, or Ariadne’s) any particular Roman audience might assimilate to itself is indeed a question Vergil thrusts to the fore, not to answer definitively, but to pose in relation to that of Greek culture’s place in Augustan Rome, as we shall see. Like the viewer who looked at the paintings in Pompeii, audiences of the Aeneid found themselves encouraged to scrutinize what Greece meant to them, where it fit into Romanness and vice-versa, and in this case how Augustus’s visual program, hinted at through reminiscence of his Palatine temple of Apollo, presented Rome’s Greek inheritance as one in which all Romans had a stake. 9

Fredrick 1995: 279.

introduction

6

Roman culture, virtually from the moment it makes sense to speak of it as such, is always already Greek. Rome’s powerful neighbors the Etruscans were assimilating Greek culture by the eighth century bce (which is also the period of the earliest Greek colonization in southern Italy), and the Etruscan influence on the nascent Roman state provides an early conduit for Greek influence in turn. But any narrative of Latin literature in relation to Greek art takes major impetus from the story of plunder mentioned above.10 By the time Latin letters emerged in the third and second centuries BCE, the Mediterranean world had long regarded Greek goods and cultural signifiers as a lingua franca of prestige. In fact, it is no coincidence, given Rome’s development of a literary tradition adapted from the Greek, that conquering Roman generals of the period sought to bring back evidence of their success in Magna Graecia and, eventually, the Greek mainland in the form of dazzling artworks. Chapter 1 will examine a pair of early Latin plays “translated” from Greek sources, Plautus’s Menaechmi and Terence’s Eunuchus, in which characters describe paintings similar to those contemporary Romans would have seen paraded in triumphal processions. The event usually cited as opening the floodgates is Marcellus’s sack of Syracuse in 211 bce, a takeover designed to prevent the loss of this important city to Carthage in the Second Punic War. A vast amount of loot, including art objects, was brought back and displayed, making the conquest a byword among ancient authors. Livy, recounting the comparable spoils garnered two years later by Fabius Maximus after his capture of Tarentum, describes enormous quantities of silver and 3080 pounds of gold, along with numerous statues and paintings, the whole “almost” equaling what was taken from Syracuse (27.16.7). With these victories and the lavish spectacles accompanying them, a race to acquire Greek art was on. Similar display soon followed the defeat of cities on the Greek mainland, as Roman imperial dominance moved inexorably eastward. Flaminius’s victory over Eretria resulted in a stunning collection of statues and paintings as part of a triumph in 194 bce lasting three full days.11 Fulvius Nobilior’s triumph in 187 bce following his victory in Ambracia yielded artworks numbering in the many hundreds;12 L. Aemilius Paullus’s celebration in 167 bce of his conquest of King Perseus of Macedon again lasted three days, of which the first was scarcely adequate, according to Plutarch, for viewing the statues, paintings, and other images drawn in 250 wagons.13 Metellus’s decisive defeat of Macedonia in 148 bce 10

At least to go by our sources. But as Gruen 1992: 93 notes, earlier events provided substantive opportunities for the Romans to become acquainted with Greek art, and their experience of plunder may not have been as novel as the sources suggest; see contra, however, the more recent views of McDonnell 2006. For the following discussion I rely especially on Gruen 1992: 84–130 and Pollitt 1986: 153–59; see further Beard 2007: 143–86; Germany 2008: 101–7; Miles 2008: 60–75. 11 Liv. 32.16.17; 34.52.4–5. 12 Plb. 21.30.9; Liv. 39.5.15. 13 Plu. Aem. 32–33.

Introduc ti on

7

resulted in a similar haul and was especially notable for including all twenty-five bronze equestrian statues of the massive Granicus Monument, created by the Greek sculptor Lysippus for Alexander to celebrate his first significant victory over the Persians.14 In a gesture with enormous ramifications for artistic production at Rome, Metellus apparently also brought a team of Greek artists to work on the porticus Metelli, which he had set up in a specially designated area of the Campus Martius. Within this enclosure, the Granicus monument faced a pair of temples: the already existing temple to Juno Regina and a new temple to Jupiter Stator, designed by a Greek architect, Hermodorus of Salamis.15 And the tide still did not ebb: the treasures of Greek art conveyed home after Mummius’s sack of Corinth in 146 bce, often cited as the last in this series, seem to have outnumbered those of any previous triumph.16 Ancient accounts of such artworks’ effect on the Romans are tinged with dubious moralizing over the erosion of traditional Roman values caused by the new habits of luxuria this art introduced. What real worth the artworks did possess, such sources imply, came solely from the fact that they signified a Roman conquest rather than from their intrinsic artistic qualities or from the engrossing mythical narratives they depicted. Instructive, however, for our purposes is the collective emphasis on wonder at and discussion of Greek art as habits pervasive in Rome at this time. Already in the second-century bce account of the Greek historian Polybius, the extravagant display of artworks from Syracuse has profoundly changed old-time Roman character through the “zealous imitation of the conquered” (τὸν τῶν ἡττωμένων ζῆλον) it represents (Plb. 9.10.6).17 In the next century, the Roman polymath Varro belittles the habit of visiting wealthy elites such as Lucullus to view their “picture-collections” (Var. R. 1.2.10, pinacothecas, the Greek term), which he denigrates in comparison with the visual delights of agricultural bounty produced by the well-run villa. Varro’s contemporary, Cicero, writing in the Paradoxa Stoicorum of contemporary Roman reactions to artworks like those brought back by Mummius, scornfully equates the wonder they inspired with slavery (5.37, “when I see you looking on and marveling and crying out admiringly, I judge you to be the slave of every foolishness” [intuentem te, admirantem, clamores tollentem cum video, servum te esse ineptiarum omnium iudico]).18 Later, the Augustan author Livy describes the Syracusan plunder as setting off “the very beginning of wonder over works of Greek art” (Liv. 25.40.2,

14

Vell. 1.11.3–4; Plin. Nat. 34.64. Vell.1.11.2–5; Vitr. 3.2.5; Plin. Nat. 36.35. 16 Str. 8.6.23. 17 Cf. Germany 2008: 104–5. 18 On this passage, cf. esp. Barchiesi 2005: 291: “The important thing, for Cicero, is that Greek sculpture and painting should remain forms of entertaining, but not achieve mastery over the senses. Otherwise, he argues, we (Roman élite slave-owners as well as art viewers) are slaves to our own possessions.” 15

introduction

8

primum initium mirandi Graecarum artium opera) at Rome. He has the Elder Cato complain of “far too many people praising and wondering at [mirantes] the ornaments of Corinth and Athens and laughing at clay antefixes of the Roman gods” (34.4.4). The theme of wonder is expanded upon still further by Plutarch (first to second centuries CE), who says Marcellus was blamed by Rome’s elder citizens for filling the populace with “leisure and idle chatter in affecting clever views about the arts and artists, and even taking up the better part of a day with this.” Marcellus, for his part, boasts in Plutarch’s account that he “taught the Romans, who were previously ignorant, to honor and wonder at [θαυμάζειν] the beautiful and wondrous works of Greece” (Plu. Marc. 21.5). Apparently early Roman authors like those treated in Chapter 1 could already count on their audiences’ interest in Greek art and how to talk about it, a practice the authors could echo in turn through literary ecphrasis. Our very testimonia, moreover, suggest the ongoing cultural currency of this idea in later periods. The display of art taken from conquered cities not only served the self-asserting purposes of particular Roman generals, but also reflected a broader desire “to create an international identity on behalf of a city emerging as a world power.”19 The politics of art and empire are intimately connected at Rome, and Greece, together with other conquered regions such as the Near East and Africa, is a major source of the Romans’ identity-forming visual culture. In time, Roman artists, too, emerged, no doubt many more of them than attested by the historical record, which reflects elite understandings of Greek artists’ prestige as well as scorn for actually practicing art oneself.20 As the Elder Pliny’s chapters on art at Naturalis historia 34–36 attest, the Romans persisted in a view of artistic production as a fundamentally Greek activity, subject to ongoing elite Roman suspicion since the third century bce.21 The few exceptions, however, to this last sentiment demonstrate that things were not so simple at first. The aristocratic C. Fabius Pictor was a painter (hence his cognomen) who in 303 bce created a set of murals in the temple of Salus and was even so bold as to sign them (about a century later his kinsman Q. Fabius Pictor wrote, in Greek, the first work of Roman historiography). The prominent playwright M. Pacuvius (220–c. 130 bce), a dramatist adapting Greek plays for the Roman stage, painted as well, and his work appeared in the temple of Hercules in the Forum Boarium.22 And a text such as Pliny’s own, which may itself derive 19

Leach 2004: 5. The names of some Roman painters, although virtually no Roman sculptors, have come down to us. On this complex matter see Gruen 1992: 131–41. 21 Cf. Plin. Nat. 35.20 (of painting) “afterwards it was not a project for respectable hands” (postea non est spectata honestis manibus). Pliny’s catalogue is an essential text for understanding how the Romans of the second half of the first century CE sought to fit their knowledge of Greek art, especially those works taken to Rome in the course of previous centuries of conquest, into the imperial project of domination. 22 Plin. Nat. 35.19. For further sources on Pictor, see Gruen 1992: 92n36. 20

Introduc ti on

9

from a lost Greek work, confirms further that, suspicions notwithstanding, pleasure, delight, and wonder over Greek art had by no means died out among the Roman elite almost four centuries after Pictor.23 The early stages of Roman philhellenism, in all its ambivalence and self-contradictory complexity, contribute to the irony and play we will remark in some of the first surviving Roman ecphrases (Chapter 1). As the encounter with Greek culture became less a matter of novelty and more a received part of Roman identity, successors to such views, which had grown even more complex, informed the ecphrastic literature of the first centuries bce and ce, the principal focus of this study. Actual Greek art and/or artistic traditions, together with Roman adaptations of them, will in fact figure in every period under discussion in spite of their considerable differences: the iconography of Second-Style wall painting in connection with Catullus 64 (Chapter 2), drinkware and statuary reflecting Hellenistic types in relation to small-scale ecphrasis in “triumviral” literature (Chapter 3), Augustus’s display of Greek statues and its effects upon a wide range of contemporary poetry (Chapters 4 and 5), Greek panel paintings in Petronius’s gallery scene (Chapter 6), and a series of sculptures by the great Greek masters as they appear in the ecphrases of Flavian occasional verse (Chapter 7). Above all, our topic demands sensitivity to the changing meaning of Roman “philhellenism” at these different times and necessitates avoiding reductive formulations such as “the Roman viewer” or “the Roman reader.” The circumstances are simply too intricate for recourse to such generalizations, however convenient they may be. Ecphrasis has in recent decades enjoyed a great deal of attention as a topic of critical theory and of scholarship on both Greco-Roman antiquity and other eras.24 But rather than offering a potentially tedious review of theory (or scholarship) here at the outset, I have chosen to draw upon it where relevant in the chapters themselves, where it will become clear, I hope, how my close readings challenge specific aspects of the “standard” view outlined above. When read in their social, cultural, and political contexts, Roman ecphrases show that notions of a text’s attempts at power and control are, on their own, too limited. Audiences for ancient ecphrasis are assumed both to perceive the image’s visual appearance and experience its nonvisual effects, such as the ideas, values, emotions, and beliefs that it activates.25 Roman ecphrasis often seeks to persuade the listener to be receptive to Greek culture, as imparted through these appearances 23

See esp. Carey 2003: 8 on the likelihood that Pliny’s catalogue derives from lost Greek works filtered through later Greek and Roman authors (cf. Gruen 1992: 139); 13–14 on Pliny’s “overwhelming focus on Greek art” and its role in the development of the now established model of art history; 75–101 on collecting Greek art and its relation to conquest, luxury, and wonder. 24 See above, n4. 25 In ancient terms, ecphrasis (whether of art or other objects) aims, through enargeia, or “vividness,” at phantasia, “the faculty of imagination,” “the re-presentation of appearances or images,”

10

introduction

and effects, and makes such receptivity itself a matter for reflection. Thus where modern theory may emphasize, for example, the text’s tension-ridden encounter with its visual “other,”26 I also find pleasure and sameness being expressed, as in the case of Vergil’s Aeneas transfixed by the Daedalus story. This is an important difference insofar as it helps broaden my readings beyond questions simply of dominance and submission. Individual chapters of this book explore how the simple assumption of a self-asserting ecphrastic text is complicated by comic performance, self-consciously inconsistent narrative, the thematization of civil discord, the numinous powers of Greek religious statuary, the contradictory associations of epic imagery, satiric poetry and the satiric novel, and the author’s relations with a patron. The early examples of Roman ecphrasis treated in Chapter 1, because two of them come from plays, allow me to illustrate what is at stake in my use of the metaphor of “staging” and to distinguish ecphrasis onstage from similar techniques in nondramatic literature, from which most of my examples come. If Roman ecphrasis stages receptivity to Greek culture, the comic texts of Plautus and Terence reveal its function literally as staged behavior within the Roman society of their day. The self-consciously hybrid Greek-and-Roman milieu of Latin literature’s supposed “invention” in the third to second centuries bce makes these early texts a helpful proving ground for my ideas about identity. And the risible, self-undermining ecphrasists of the comic stage complicate the ecphrastic text’s supposed quest for dominance over its image in an obvious fashion. While they literalize my thesis about staging, however, these opening instances also highlight the importance of generic and historical specificity within my discussion. Chapter 2 treats Catullus’s poem 64, his “miniature epic” on the marriage of Peleus and Thetis, in which an extended ecphrasis describes a tapestry depicting the myth of Ariadne. Now the self-subverting ecphrasist is no longer a character in a play, but a self-dramatizing, inconsistent narrator who laments moral decline from archaic Greece to contemporary Rome while expressing an idiosyncratic receptivity to the heroic past. Assimilated to the lover’s persona familiar from Catullus’s or “creative imagination.” For these definitions of phantasia, see LSJ 2a, b, c. On the development of the term in classical thought beginning with Plato, see Watson 1988. The Roman-era Greek rhetorical texts in which the term ἔκφρασις is first defined clarify this connection, and it is appropriate, too, within the strongly rhetorical and performance-based framework of ancient literature. On the functions of ecphrasis (understood in a broad sense) according to the Progymnasmata and rhetorical handbooks and its connection to phantasia, see above all Webb 2009, esp. 51–55, 107– 30. In ancient thought, language itself may depend upon the formation of visual images, as in the Stoic understanding of phantasiai as the very basis of language because of the way they give rise to speech, which then serves as the medium through which the phantasiai are communicated to the mind of a listener (Webb 2009: 114; on Stoic phantasia see further Watson 1988: 44–58). 26 Mitchell 1994: 151–81.

Introduc ti on

11

lyric and elegiac poetry, this narrator treats the Greek past itself as an eroticized object both eminently desirable and ultimately disappointing. With this problematic invitation beyond the here and now, furthermore, Catullus 64 offers both ironic and sympathetic commentary upon a culture of viewing like that implied by the illusionistic tableaux of contemporary Second-Style wall painting. Through an immensely allusive, Alexandrian technique apparently typical of his poetic circle, Catullus encourages his audience to enter as Romans into an archaic Greek world seen in part through Alexandrian eyes. He thus presses his audience to reflect on the broader phenomenon of elite Hellenophilia as it existed in the late Republic. Chapter 3 discusses a series of more diminutive examples of ecphrasis in two texts of the tumultuous “triumviral” period, Vergil’s Eclogues 3 and Horace’s Satires 1.8. While on the surface very different, both of these poems figure the search for elements of Greek culture that might ameliorate Rome’s tendency toward civil violence rather than, as they do for the Catullan narrator, provide a fantastical escape from it. Here the Greekness of the ecphrastic image, rather than being subject to the dominating text, becomes something for both ecphrasist and audience to be transformed by—at least potentially. Like Catullus 64, each poem also opposes itself specifically to the epic tradition, vitiated by its implicit association with civil war, and envisions objects with suggestive parallels in actual art. The set of beechwood cups wagered by a pair of quarrelling herdsmen in Eclogues 3 illustrates Greek cultural figures whose ordering of the cosmos and the natural world signifies a hoped-for return to social order in the face of acknowledged threats. The figwood Priapus of Horace’s Satires 1.8 activates a mix of Greek influences—including Aristophanic social critique, Socratic irony, Callimachean literary refinement, and Epicurean ideals—to reject civil war and those who would precipitate its resurgence. Social regeneration was closely tied, at Rome, to religious renewal. The fourth chapter argues that in the ecphrastic temples of Vergil’s Georgics 3.13–36 and Propertius’s Elegies 2.31, religious statuary comes to the fore as a type of image that it is impossible entirely to dominate by textual means. Octavian’s Palatine temple of Apollo, to which both poems allude, adapted architectural features like those exploited by the Hellenistic dynasts at Pergamum and Alexandria, while his victories over Antony and Cleopatra, celebrated in the temple’s imagery, led to the conquest of Alexandria itself. In asserting renewed control over the Greek legacy, particularly the poetry of Callimachus, Vergil’s and Propertius’s ecphrases become in this way analogs to Octavian’s cultural activity. Writing in about 29 bce, Vergil describes a metaphorical temple that he will “found” for Octavian (most likely a symbol of the forthcoming Aeneid) and where, as athletic victor over Greece, he will drive a hundred four-horse chariots and put on theatrical performances. Octavian’s numinous power, embodied in a cult image of him at the temple’s center, inspires the vision and, Vergil implies, the very text of Georgics 3 in which it appears. A threat to the new order remains, however, in the image of Invidia

12

introduction

(Envy) with which the ecphrasis concludes. Propertius, in an outright description of the Palatine temple a few years later, distances his amatory, elegiac pose from the ambitions of public life and yet also enfolds his Alexandrianism within Octavian’s. Propertius seems to imagine himself receiving inspiration from the temple’s Greek cult statue of Apollo as the true Greek god of song, an epiphany that caps the whole poem’s staging of an attitude toward Greece that is both in harmony and in tension with Octavian’s own. Thus Propertius’s poem especially offers an example of how ecphrasis may not only serve empire but also, in the same gesture, direct scrutiny toward its means and methods. This theme becomes still more prominent in Chapter 5. While the first four chapters treat a variety of non- or antiepic genres, this one turns to the full-length Augustan epics, Vergil’s Aeneid and Ovid’s Metamorphoses. The ecphrastic objects of these poems, I maintain, do not simply belong to heroes and heroines but are themselves “heroic”: they are resistant to the text’s dominance and uncannily self-asserting, even over the texts in which they appear. Comparing the two poets’ much-studied imagery, I show how both complicate the assimilation of Rome as non-Greek (i.e., Trojan) with Rome as supremely Greek, a synthesis evident in a monument such as the Forum of Augustus. Vergil and Ovid thus disclose Augustan literature’s pervasive role in constructing an ambivalent viewer, one who might see both kinds of interaction with Greece as fundamental to Roman culture, and so both appreciate and resist control of Hellenic visual imagery as a part of Roman imperial self-definition. Here epic ecphrasis becomes a showcase for the network of contradictory associations that mythological imagery may create in its viewers, even in spite of the efforts of an Augustus to assert his own dominance over this process. Such irony and ambivalence lend themselves naturally to satire. Chapter 6 discusses Petronius’s gallery scene (Satyricon 83–90.1) as a satire of the effort to produce a version of Greek literary ecphrasis in Latin—and of the very idea that Greek images might require such ecphrasis at all. Playing with the venerable tradition surveyed in previous chapters, particularly Vergil’s Aeneid, Petronius pokes fun at his characters Encolpius and Eumolpus by assimilating their ecphrases of paintings to failed pederastic conquest over Hellenic youths, and thus emphasizes the ambiguous relation of the Greek/Roman dichotomy to that of object/viewer. Although the Satyricon’s date remains uncertain, the gallery scene may have appealed to a renewed Roman interest in the proper role of Greek cultivation, since the tyrannical, extravagant Nero was a flamboyant Hellenophile. Elsewhere in the Satyricon, Petronius skewers Greek affectations in the nouveau-riche freedman Trimalchio, possibly a figure for Nero himself. My reading of the gallery scene thus ties it to the descriptions of preposterous food artifacts produced by Trimalchio’s aptly named cook, Daedalus, in the Cena Trimalchionis episode. Chapter 7 explores the patron’s mediating role in ecphrasis through the occasional poetry of Statius and Martial written under the Flavian emperor

Introduc ti on

13

Domitian (ruled 81–96 ce). In these texts, extant Latin poets for the first time use patrons’ actual art objects as the main focal point for elaborate praise calling attention in particular to the patron’s mastery over Greek culture. The presence of such patron figures problematizes still further the text’s efforts at control or dominance, insofar as the poet’s subordination to his patron informs his ecphrasis to a considerable extent. The first part of the chapter moves from Statius’s Silvae 1.1, on a massive equestrian statue of Domitian placed in the center of the Forum Romanum, to Martial’s epigrams on a statue of Hercules with Domitian’s features set up as a cult image in a new temple to Hercules on the Appian Way. Part II considers similar techniques in a private context through both authors’ poems on a purportedly Lysippan statuette of Hercules owned by Novius Vindex. Offering an especially good illustration of their “occasional” verse’s larger social, cultural, and political import, Statius’s and Martial’s ecphrases reveal the way in which the interrelation of patron and poet may determine even the subtlest details of an ecphrastic text and undermine modern notions of how speaker, object, and audience interact through the trope. True to the book’s emphasis on cultural hybridity, the epilogue proposes two alternative end points for my discussion: Apuleius’s Metamorphoses, one of the last surviving examples of pagan Latin literature, and an actual Greek text, the Eikones (Paintings) ascribed to the Elder Philostratus, a characteristic work of the Second Sophistic. Making an especially appropriate pair with Apuleius, a “Roman sophist” adapting the ecphrastic techniques of his literary forbearers, the Eikones can stand paradoxically as the culmination of the Latin works studied in this book by embodying an extreme, totalizing version of the receptivity to Greek culture that informs them. Philostratus’s text sets in high relief what the Roman examples are not: explicitly didactic appropriations of Greek culture for predominantly Hellenist purposes. Yet Philostratus’s emphasis on the work of becoming receptive to excellence in art calls attention to Roman ecphrasis as implicit acculturation into an ambiguously Greco-Roman identity. Finally, a word on inclusiveness. As should now be clear, this book proceeds through illustrative examples rather than attempting an exhaustive survey, which means that some readers might find I have left out a favorite passage in the name of a more balanced presentation. I have excluded, for instance, Flavian epic, not because I think it would not reward study from my perspective, but because it would make the book unwieldy and risk giving the impression that ecphrasis is simply an epic trope sometimes appearing in other guises, an impression that I wish to avoid. I have also passed over Lucretius, in spite of his poetry’s highly visual tableaux likely informed by contemporary iconography, since explicit descriptions of art objects in his text are brief and rare. If it turns out, however, that my views can apply as well to still other cases both within and beyond the Roman period, I will be only too happy.

1

Staging Ecphrasis in Early Latin Literature From Naevius to Plautus and Terence

Latin literature as we know it begins with adaptation from Greek literature, in a welter of Roman enthusiasm over fashioning a new cultural identity within the broader Mediterranean world.1 The extant Roman ecphrases from this early period reflect such large-scale cultural developments: a rapidly changing sense of Romanness influenced, deeply and pervasively, by its perceived competitor, Greek culture. But already here ecphrasis is so far from being a simple assertion of dominance—of Rome over Greece, of text over image—that one could hardly ask for a better starting point for my argument, all the more so because these texts coincide so closely with an assertion of gross material domination, the era of Roman artistic plunder outlined in the introduction. In Naevius’s fragmentary epic, the Bellum Punicum (The Punic War), an artistic depiction of a primordial Greek myth underscores Rome’s sense of itself as a rising military power. A lead character in Plautus’s comedy, the Menaechmi, appears onstage wearing feminine dress, then proceeds to compare himself jubilantly with paintings of Greek mythological subjects. More darkly, a comic protagonist in Terence’s Eunuchus justifies his rape of a girl via another such painting. For the narrating ecphrasists of these Latin works, the description of art becomes an act of (re)identification with Hellas, a moment of ceding one’s self to something both different from and uncannily the same as oneself, a pleasurable mimesis—however unsettling to us—and a display of such pleasure as much as of tension and conflict. Already here, I claim, Roman ecphrasis stages a complex, ambivalent receptivity to Greek culture. But this chapter gains further in argumentative

1

This is not to avoid the vexed question of whether or not greater knowledge of the literary, subliterary, and performance cultures of archaic Italy would lead us to revise the received notion of Latin literature’s sudden “invention” with Livius Andronicus’s first play (dated to 240 bce). For the ancient sources and an introduction to the debates, see Habinek 1998: 34–68; Goldberg 2005: 1–51.

14

Stag ing Ecphrasis in Early Latin Literature

15

import by allowing me to explore, at the outset, one of my central metaphors: staging. Although most of our extant Latin literature was originally performed in some way, to speak of staging ecphrasis across different genres risks eliding real differences between drama and epic, as well as between these and elegy, satire, or epigram, all of which will feature in the chapters to come.2 We will remark in particular how ecphrasis’s stilling function—the way it casts Hellenic artworks in the role of passive, silent objects while nevertheless intimating their power over Roman subjects—serves distinctly different purposes for playwrights and nondramatic authors. The break in action characteristic of epic ecphrasis, for example, contrasts with the fusion of image, action, and spectatorship in plays, and drama, of course, often directs us toward the metatheatrical aspects of imitation itself. Yet the term staging suggests nevertheless the crucial idea that the ecphrasist, whether author, narrator, or dramatic character, is understood as an actor under observation by audiences internal and external to the text. While real contexts for viewing art change with the viewer, both dramatic plot and nondramatic narrative supply delimited contexts for the viewers and images that they present. Ecphrasis in whatever genre invites certain audiences to attend closely to this performance, while excluding others from it implicitly or explicitly. We will see Roman authors, time and again, drawing upon such techniques to place ambiguous relations with Greek culture center stage. These early texts thus present both an opportunity and a caveat. Naevius’s ecphrasis, perhaps the first description of any art object in Latin literature, not only offers suggestive contrasts and parallels with the dramatists’ methods but also reveals that in early Roman ecphrasis we are dealing with a practice of wider extent than drama alone. In the case of the comedies, we have a literalization of my thesis, yet a generically and historically specific instance of it that must be dealt with as such. On any reading, the subtle cultural dynamics of these supposedly “archaic” Roman ecphrases belie entirely the notion that circumstances of pillage and plunder govern, by themselves, relations with Greek culture as they are here staged. These early authors use the trope to appeal to their audiences through a display of receptivity that is simultaneously a vehicle for Greek culture’s own growing influence over the Roman sense of self. Rome the captor has begun to see itself captivated by Greek art.

2 All genres of ancient literature, both Greek and Roman, were regularly voiced, even when read to oneself, and often performed by a trained reader for an audience. For bibliography on literature and performance at Rome, see Dufallo 2007: 129n1; Lowrie 2009: 13–18. Rich discussions of ecphrasis in ancient dramatic settings can be found in, e.g., Zeitlin 1994 and O’Sullivan 2000 on Greek tragedy and satyr drama.

t h e c a p t o r’s i m a g e

16

I. The View to Olympus: The Gigantomachy Ecphrasis of Naevius’s Bellum Punicum Naevius’s Bellum Punicum, the first Roman historical epic and only the second Latin epic of which we know, told the tale of Rome’s victorious clash with Carthage in the First Punic War (264–241 bce).3 At some point in the first book it portrayed an object—exactly what we cannot tell—depicting a Greek myth of cosmic struggle and victory. This was the Gigantomachy, the Olympian gods’ defeat of the Giants who had sought to overturn their rule:4 There were figures portrayed on it: how the Titans, double-bodied Giants and great Atlases, Runcus and Purpureus, sons of Earth . . . inerant signa expressa, quo modo Titani, bicorpores Gigantes magnique Atlantes Runcus ac Purpureus, filii Terras (Naevius, Bellum Punicum fr. 4 Strzelecki) The fragment, alas, breaks off before its visual narrative has really begun, but we have enough of it to say that the trope of ecphrasis, like the myth it conveys, here signals Naevius’s location of himself and his subject matter in a tradition extending back to the earliest Greek poetry. Hesiod’s Theogony reported the same mythical battle, while Homer’s Shield of Achilles had enshrined ecphrasis, particularly of objects portraying cosmic phenomena, as an epic technique.5 The fusion of Greek and Roman in the Bellum Punicum takes a multitude of forms; more, in fact, than would be worthwhile to address here. Yet a pair of well-known instances will bring our passage into better focus. Naevius’s immediate predecessor in Roman epic, Livius Andronicus, had chosen what is probably a native Italic verse, the Saturnian, to render the dactylic hexameters of Homer’s Odyssey into Latin.6 In the Bellum Punicum Naevius follows suit, as though aware of how the meter signals a nascent tradition through which Rome might gain a footing in a cultural arena long dominated by Hellas. Livius, moreover, and apparently also Naevius, opened their epics with invocations of the Camenae, 3

On Livius Andronicus’s earlier Odussia, see below. On the conflation of Titans and Giants (regular in such contexts), see Fraenkel 1954: 15. 5 For Homer’s Shield of Achilles, see Iliad 18.478–608, and for the possibility that Naevius is in fact describing a shield here, see Fraenkel 1954; for Apollonius Rhodius’s later description of the cloak of Jason at Argonautica 1.721–67 as Naevius’s possible model, see Wimmel 1970: 97–98. The earliest account of the Gigantomachy appears in Hesiod, Theogony 820–85. 6 For the long-running but still unsettled question of the origins of the Saturnian verse, see Kruschwitz 2002. 4

Stag ing Ecphrasis in Early Latin Literature

17

Italian spring goddesses substituted for the Greek Muses.7 This is akin to the way that in the Gigantomachy Naevius chooses the familiar Latin noun Terra (“Earth”) to stand in for the Greek name of the earth-goddess Γαῖα within a context that nevertheless remains marked as Greek through the close imitation of other Greek forms (Γίγαντες, Ἄτλαντες, etc.).8 Ecphrasis, in other words, begins at Rome as strongly Greek and yet distinctively Roman, an assertion of the Roman self over the Hellenic art object, but at the same time an admission of Roman dependence on Greece for the very sense of self that makes ecphrasis relevant within Roman culture in the first place. Within the emergent tradition of a new epic poetry composed in Latin, displaying detailed knowledge of a Greek mythological image while recounting Roman history proves a valued technique to this Roman author. This approach finds further parallels in the Bellum Punicum’s larger narrative, where it in fact forces us to consider an intriguing possibility about the actual Gigantomachy image Naevius may have had in mind. Although the poem’s fragmentary state makes certainty about its story line impossible, ancient testimony to the contents of its various books shows that the Gigantomachy ecphrasis, which appeared in Book 1, must have somehow been connected with the story of Aeneas’s departure from Troy after the Greek sack.9 This was told in the same book as part of a larger Aeneas-narrative embedded in the first three of the poem’s seven books. Book 1 also recounted the Sicilian expedition led by the Roman consul Manius Valerius Messalla (cf. fr. 3 Strzelecki) at the outbreak of the First Punic War, and so is very likely to have described the siege of Agrigentum, an important Roman victory. In fact, the Greek temple of Zeus at Agrigentum depicted, on its pediments, both the Gigantomachy and the Sack of Troy (D.S. 13.82.4), so that a scene in which it was viewed by a Roman character, perhaps Valerius Messalla himself, could have provided the link between the war narrative and the myth: a description of the temple’s mythical sculptures, that is, could have led conveniently into the Aeneas story.10 (Valerius Messalla was later famous for having a painting 7

On the Camenae in Livius and Naevius and as objects of cult, see Goldberg 1995: 64, 82, 89, 117–18, 131. 8 Priscian (Keil 2.198) is incorrect to identify as a Graecism the genitive in –as (Terras), an early Latin form that here may have been felt, rather, as an archaism. 9 Ancient testimonia are supplied in the editions of Stzrelecki (1964) and the more recent edition of Blänsdorf (2011), the former a superior reconstruction of the poem in general (hence cited here), especially in terms of how its mythic portion related to the war chronicle; see esp. Fränkel 1935: 59–61; Strzelecki 1964: xxii-xxiii; Goldberg 1995: 51–52 with bibliography. 10 As Fränkel notes (1935: 59–61), this would help explain the curious plural Atlantes, since the temple had massive supporting figures termed Ἄτλαντες in Greek (telamones in Latin: see Vitr. 6.7.6). A compelling precedent for a temple ecphrasis of a Gigantomachy (and one the playwright Naevius was very likely to have known) comes from Euripides’ Ion (206–19), where the Chorus describes a Gigantomachy on the temple of Apollo at Delphi. Apart from the word Gigantes in Naevius’s text (cf. E. Ion 207: Γιγάντων), the phrase filii terras recalls Euripides’ Γᾶς τέκνων (219).

t h e c a p t o r’s i m a g e

18

of his own military successes displayed on the senate house [Plin. Nat. 35.22], a circumstance Naevius may possibly have prefigured through portraying him as a viewer of these other artworks.) So a real Greek image may have supplied a poetic transition to Rome’s own foundation story. If we accept this very plausible reconstruction, additional implications emerge that further support my argument. The Agrigentum Gigantomachy originally commemorated a Greek victory over Carthage (the Battle of Himera in 480 bce) with the Sack of Troy doubling the symbolism by recalling Greek defeat of the Trojans, an eastern people like Carthage’s own founders, the Phoenicians. And in Greek culture more generally the Gigantomachy and similar myths with which it was often conflated, as in Naevius, trope the repulsion of barbarian invaders: in Pindar’s Pythian 1, Hiero I of Syracuse’s defeat of Carthaginians at Himera, together with his victory over Etruscans at Cumae, is again the comparandum, while the Great Altar at Pergamum alludes through the Gigantomachy to the Attalid dynasty’s resistance to the Gauls, another of Rome’s fiercest early foes.11 It is possible, then, that the Roman commander Valerius Messalla, viewing this temple in Bellum Punicum, Book 1, could have been imagined to see its sculptural program as asserting Greek superiority over Carthaginians and Trojans, and at the same time to see himself, a descendant of the Trojans, as rescuing the western Greeks from Carthaginian encroachment. In this case, Greek/Roman relations in the poem become still less easy to characterize: Rome’s beneficence toward its ancient enemy, Greece, leads to Greek obligation to Rome, but exposes still more strongly the value Rome places on Greece, a people worthy to be rescued in this way. Rome’s close involvement with its “others,” both Greece and Carthage, would come through all the more if such an episode in fact introduced an ancient, mythical aetion of the present war, the story of Aeneas’s erotic involvement with the Carthaginian queen Dido (see below). Rome’s politico-military emergence onto the Mediterranean stage through the Punic Wars, its intrusion into a preexisting history of relations between competing peoples and cultures, would make the act of observing Greek sculpture in such circumstances anything but a straightforward performance. But whether we accept it or not, this reconstruction points up what must have been part of the ecphrasis’s role within the poem’s larger expression of Roman identity. On whatever object it appeared (Aeneas’s ship, his shield, or an embroidered robe have also been suggested by scholars12) the Gigantomachy scene portrayed Olympian victory over defiant creatures doomed to defeat. It thus surely foreshadows Rome’s coming victory over barbarian Carthage.13 11

For these and other examples see O’Hara 1994: 219–20. See Goldberg 1995: 52n47 13 Cf. Wimmel 1970. The name Purpureus perhaps even suggests the Carthaginians through their association with Tyrian purple dye. 12

Stag ing Ecphrasis in Early Latin Literature

19

A Greek image of divine victory infuses with pregnant suspense the entire poem’s narrative of Roman victory. Rome’s very pride in conquest depends on figuring its military success through Greek mythical symbolism, so that to speak of Naevius’s text simply dominating its image, or Roman culture dominating Greek, at once appears too simplistic, an absurd reduction of a far more complex situation. Still other pressing interpretive questions follow from this view. The Gigantomachy, in fact, stands in some tension with both the Aeneas legend and the Punic War narrative. On the one hand, the Carthaginians are comparable with the Giants, but on the other hand, when seen from the perspective of Aeneas’s ultimate battle to subdue the native peoples of Italy, the cosmic myth takes on an incongruous cast, since Aeneas, rather than the Italians, is more readily associated with the Giants as invader. Vergil will exploit this very inconsistency in comparing both Aeneas’s opponents and Aeneas himself to the Giants, a deceptive pattern that pulls the reader toward differing conclusions.14 Naevius, moreover, again like his successor Vergil, calls attention to Aeneas’s close involvement with the Carthaginians, while nevertheless highlighting the same elements that, in the Aeneid, make them most suspect as descendents of the effeminate east (Carthage being a colony of Phoenicia, and effeminacy a trait it shares, in Roman eyes, with the Greeks themselves). We know from Servius Auctus that the Bellum Punicum included the Dido story.15 Fragments assigned to Book 2 show someone, possibly Dido, asking “with charm and cleverness” (fr. 23 Strzelecki, blande et docte) how Aeneas left Troy and gifts perhaps being exchanged (fr. 22 Strzelecki).16 The Gigantomachy, moreover, ends in catastrophic defeat like that to which Aeneas has just fallen victim at Troy—again, the Sack of Troy and the Gigantomachy appeared together on the temple at Agrigentum, whether or not it is Naevius’s direct referent. Moving out to the poem’s main narrative, we recall that the First Punic War, which failed to settle hostilities between Rome and Carthage, was in fact nowhere near as decisive as the Gigantomachy. If, as is often supposed, Naevius was writing late in life, during the Second Punic War, the image’s valence in relation to the Punic War story might have seemed more hopeful than triumphalist, perhaps even cautionary. Self-consciously a version of an earlier Greek persona gazing at a Gigantomachy, the Roman who “views” Naevius’s ecphrasis would then be aware of belatedness as much as promise. Let us resist the temptation to speculate for a moment and return to Naevius’s text. Whatever its exact context and whoever its internal viewer, the Gigantomachy’s imagined effect must have been, in part, wonder. The fantastic 14

O’Hara 1994. He builds upon the earlier treatment in Hardie 1986: 85–156. Serv. Dan. A. 4.9 (1.462 Thilo and Hagen). 16 But see Goldberg 1995: 54–55 on the insecurity of such notions. 15

20

t h e c a p t o r’s i m a g e

epithet bicorpores (double-bodied) and the piling up of the generic titles Titani, Gigantes, and Atlantes subordinate to the main subjects Runcus and Purpureus impart terrifying grandeur, while the lines’ overall crescendo creates a sense of writhing movement stilled in ecphrastic description.17 The words quo modo (in what way) are striking: Naevius sees the image as a story capable of holding an audience’s attention as such, but the act of looking at an artwork necessarily breaks in upon and stills the poem’s main narrative of heroic deeds while generating its visual counterpart. Indeed, through ecphrasis’s formal juxtaposition of internal with main narrative, Naevius concentrates attention on his own poem as a textual artifact both same and other in relation to the visual image it strives to contain. By highlighting, furthermore, how the image is both similar to and different from the text, Naevius encourages his audience to reflect all the more upon its own receptivity, as Romans, to a visual depiction of the Greek mythological past. The invitation is particularly forceful if the Roman consul Valerius Messalla is looking at an artwork on the Greek temple of Zeus, but it is present, too, even if the viewer is a character from the Trojan saga, since at this point in Rome’s self-mythologizing the Trojan Aeneas, as founder figure, has begun to become a proto-Roman. In spite of all its uncertainties, the Bellum Punicum shines a spotlight on Roman self-consciousness about looking at Greek imagery and the cultural implications of so doing. It stages these phenomena in a figurative if not a literal sense. Although we cannot know whether a character viewed the image in the poem, if one did, that personage became metaphorical for the personae of author, narrator, and reader. Like all of them, the viewer would have been understood to interpret the visual image, whether explicitly or implicitly, and relate its signs selectively to a body of received cultural knowledge, so as to reconstruct its narrative in the act of viewing. And since ancient reading was regularly voiced, any internal viewer also figured whoever listened to this performance (even if only the solitary reader). Both conduit and foil for the visualizing Roman audience, an internal viewer, again like his or her metaphorical counterparts, would have guided the audience’s interpretation in this subtle fashion.18 With this suggestive view of Roman ecphrasis’s cultural role before us, let us now consider the heightened “staginess” that results when the trope becomes a comedic technique, and its immediate practitioners the onstage characters of drama.

17 For “stilled” movement as a feature of ecphrasis, see esp. Krieger 1992; cf. Goldberg 1995: 74–75. 18 For a classic treatment of focalization through the spectator figure in Roman ecphrasis (the example is Vergil’s Aeneid), see Fowler 1991. My understanding of the metaphorical effects of the internal viewer figure and their relation to the literary construction of identity owes much to Reed 2007, also on the Aeneid.

Stag ing Ecphrasis in Early Latin Literature

21

II. Trying on Plautus’s “Greek” Culture: Crossdressing, Ecphrasis, and Performance in the Menaechmi Two of the next significant ecphrases from early Latin literature, passages from fully extant comedies of Plautus and Terence, bear out still more strongly my emphasis on cultural competition as opposed to the “standard” view of ecphrasis outlined in the introduction.19 Not only does the notion of staging become still more self-conscious with ecphrases in plays, but pleasure and irony over Hellenic art also emerge more forcefully as issues for a Roman audience to consider. They therefore disrupt still more clearly the simple assumption of a Roman text bent on self-assertion over its image. Like Naevius’s epic, these comedies receive Greek culture by their very nature, since their genre itself, the fabula palliata (a “play wearing a pallium,” a Greek garment), derives from Greek New Comedy of the Hellenistic period. Greek culture thus competes here with Roman in an impersonal sense, but the plays themselves are also competitive displays of Hellenophilia, insofar as the Roman comic playwright sought elite sponsorship in order to have his dramas put on during ludi scaenici (theatrical games) associated with major religious festivals and other public events.20 And characters within the plays enact, in Latin and on the Roman stage, receptiveness to imagined Greek artworks, though in no simple fashion. Plautus’s Menaechmi presents delight over aspects of Greek culture concentrated in ecphrasis as a component of farcical overindulgence, while nevertheless recuperating this indulgence itself as sympathetic rebellion. Soon after the prologue, Menaechmus I (a twin whose brother bears the same name) emerges from his house wearing, under his male cloak (pallium), a woman’s mantle (palla) that he has stolen from his wife. He proceeds to ask his parasite Peniculus if he has ever seen a painting in which an eagle snatches away Ganymede or one in which Venus carries away Adonis (143–44: dic mi, enumquam tu vidisti tabulam pictam in pariete/ubi aquila Catameitum raperet aut ubi Venus Adoneum?). “Often,” Peniculus cheekily replies, “But what do those pictures have to do with me?” Revealing the palla, Menaechmus then inquires whether he bears any resemblance to the paintings that both are now in the process of imagining: “Don’t I utterly remind you of something?” (146, ecquid adsimulo similiter?). 19

In addition to the two passages I discuss here, a brief ecphrasis occurs at Pl. Mos. 832–34, where Tranio asks Theopropides whether he sees a painting in which a crow tricks two vultures, an image he intends to signify his own trickery of Theopropides and Simo (835–40). For further references to painting in Roman comedy see Bettini 1999: 169–75; Germany 2008: 107–14. 20 Funerals, triumphs, and dedications might also involve dramatic performances. On ludi scaenici and the festival context, see Barsby 1999: 6; Leigh 2004: 2–3. The exact date of the Menaechmi is unknown (see below), while the Eunuchus was first performed at the Ludi Megalenses of 161 BCE.

t h e c a p t o r’s i m a g e

22

“What’s that fancy getup you’ve got on?” Peniculus queries. “Tell me I’m the most charming of men,” demands Menaechmus. “Where are we going to eat?” retorts Peniculus, changing the subject, but Peniculus eventually complies with his master’s wishes and pays him the compliments he requests (145–49). Menaechmus, after a row with his overbearing wife, has already announced his plan to use the mantle to help secure the services of a prostitute, Erotium, with whom he intends to spend a day feasting and otherwise enjoying himself (130; cf. 173–75). Eventually snared by the grasping Erotium, he must be delivered by his long-lost brother, Menaechmus II, who has journeyed to Epidamnus from Sicily with his loyal slave, Messenio. Through his brother, however, Menaechmus I is ultimately able to leave his wife as well as to shake off his ties to Erotium and return to his family in Syracuse. The viewer ensnared by Greek art may, it turns out, discover a triumphant deliverance after all. But what sort of viewer is this? Within the thematics of the play, Menaechmus’s transvestitism helps characterize him as an unsuccessful master whom his subordinates, particularly his wife, “unman” and who must therefore reassert himself over powers resistant to his dominance.21 Menaechmus’s reference to divine thefts of boys can of course be taken as an oddly overdetermined reference to his own theft of the palla.22 But the humor of his teasing language (“Don’t I utterly remind you of something?”) depends on at least a simultaneous self-comparison with the boys themselves.23 Neither Ganymede nor Adonis, furthermore, seems at first a ready model for the traditional Roman manhood to which Menaechmus’s ambitions as a dominant, heterosexual lover seem to appeal. Ganymede was the young male love object (eromenos) of a male lover (erastes) while Adonis was a young man beloved by a more mature goddess.24 Later (513), wearing feminine garb is expressly identified as something a cinaedus (the passive partner in male same-sex intercourse) might do. And yet as objects of divine favor—indeed, of cult—the pair of mythical boys nevertheless foreshadows Menaechmus’s comic heroism.25 His identification with them plays on the gender-related ambiguity of the young male love-object already in Greek culture: as eromenos, he is “other” to the older male or female lover, but as an object of same-sex love, his otherness is partly elided. The joint 21

McCarthy 2000: 35–76 building esp. on Leach 1969 and Segal 1987. So Knapp 1917: 152. 23 Gratwick 1993: 151–52 interprets Menaechmus as “facetiously comparing himself with Ganymede and Adonis as types of champion ‘pretty boy’ whom Zeus and Venus respectively as types of the superhuman found irresistibly attractive.” For the ambiguity of the comparison, see Germany 2008: 112–13. Moseley and Hammond 1953: 62 suggest that Menaechmus appears as both captor and captive at once, with his outspread cloak reminiscent of the eagle’s wings and himself in feminine garb the youthful prey. 24 McCarthy 2000: 45–46. 25 McCarthy 2000: 45–46. 22

Stag ing Ecphrasis in Early Latin Literature

23

appearance of Ganymede and Adonis in Plautus thus makes Menaechmus’s gender deviance all the more suggestive from the audience’s perspective, because it is not based on one simple model of otherness, but delineates circumstances in which gender may become more or less a marker of the other, depending upon who desires whom. Plautus invites his audience to see Menaechmus as a hero (his name in Greek, “Breakspear,” already has a facetious heroism about it, while his theft of the palla is comically mock heroic26). With this gesture, he also invites the audience to see the mythical young men as less other or, conversely, to embrace alterity in the way it views their images. While the play “follows the logic of the slaveowner,” it realizes a fantasy intrinsic to Plautine drama, which grants the dominant man “power to be a sympathetic rebel without laying down his real authority.”27 A comical Greek, identifying unmanfully with artistic images in a “translated” drama, nevertheless offers something for the Roman male spectator to identify with in turn. The specific cases of Ganymede and Adonis, moreover, suggest ethnicity itself as something that may become more or less a category of otherness depending on circumstances. Although both their stories had been fully accepted into the body of Greek myth by Plautus’s day, neither youth is in fact of Greek origin. Adonis, a seventh-century bce Greek import from a Mesopotamian pantheon, remains consistently associated with the Orient in our sources. Pseudo-Apollodorus, reporting the words of the fifth-century bce epic poet Panyassis of Halicarnassus, relates what became the accepted version of his birth: his father was Theias, king of Assyria, and his mother was Theias’s daughter, Smyrna. The couple’s incest led to Smyrna’s transformation into a myrrh tree, from which Adonis emerged (Ovid relates this as the story of Cinyras and Myrrha).28 While Adonis’s origins make him still more foreign (i.e., Eastern) than the Greeks, the origins of the Trojan prince Ganymede push him in the opposite direction, toward proto-Romanness. In chapter 6, we will see Petronius, generations later, playing hilariously upon the Roman associations of Ganymede’s Trojan birth to locate his protagonist, Encolpius, ambiguously between Greek and Roman. And given Naevius’s sophisticated use of the Troy story in the roughly contemporary Bellum Punicum, we can only assume that Ganymede’s presence in the Menaechmi is likewise a nod to Troy’s Roman connections. The images Menaechmus has in mind again put both difference and sameness center stage. As both reception of and staged receptivity to Greek culture, the passage is still more complex and multivalent. The pronunciation of Ganymede as Catameitus is a function of his myth’s having passed into Latin through the 26

Gratwick 1993: 138; Germany 2008: 113. McCarthy 2000: 37, 39 (citing Segal 1987: 43). 28 [Apollod.] 2.85–87; cf. Ov. Met. 10.298–514. For further sources and discussion, see Reed 1995, esp. 328–30; cf. Reed 2000. 27

t h e c a p t o r’s i m a g e

24

strongly Hellenophile culture of the Etruscans, whom the Romans had conquered in the fourth century bce.29 Although it is hard to know whether Plautus or his audience was aware of this, the detail highlights for us the fact that we are dealing with an already received Greek culture, shaped by the very historical forces that had long since brought it to the Italian mainland. Far more selfevident is the way Plautus gleefully refers, earlier in the play, to the fact that he is adapting a Greek comedy into Latin. At the play’s outset, he has the Prologus announce, no doubt with a gesture to the actual city of Rome, This city is Epidamnus as long as this play is being staged. When another play is staged, it will become another town.30 haec urbs Epidamnus est, dum haec agitur fabula: quando alia agetur aliud fiet oppidum. (Plautus, Menaechmi 72–73) The lines have the typically Plautine effect of rendering the play’s setting Greek in name only, and as per his usual practice, Plautus fills it with many Roman details, producing a Greek/Roman city of the imagination, “Plautopolis.”31 The Prologus also points out that, while others set their plays in Athens to make them “seem more Greek” (7–9) the Menaechmi has a Sicilian, not an Attic plot (11–12). This emphasis helps enfold the drama and its Greek elements further within a perspective that is already Roman. Sicily had become Rome’s first province after the conclusion of the First Punic War in 241 bce, when Plautus was a teenager. The city of Syracuse, first in the received story of large-scale Roman artistic pillaging, fell to Marcellus in 211 bce after a failed rebellion. Although Erotium says, in a bogus account of Syracuse’s history (409–12), that it is now being governed by Hiero (Hiero II ruled 269–215 bce), the ironic “that’s not incorrect” (412, haud falsa) in Menaechmus II’s response to her and the very fact of such interest strongly suggest that the play dates from after Rome’s takeover in 211, with the events set “within the living memory of most spectators.”32 An especially detailed knowledge of the cultural history of Syracuse would inform Plautus’s choice of a name for his twin protagonists if they recall the Syracusan 29

Gratwick 1993: 152. The translation is Anderson’s (1993: 138). These lines are transposed by Gratwick 1993 to come after 10. 31 Cf. the marvelous wit of ego nusquam dicam nisi ubi factum dicitur (10): either “I shall say what-happened, nowhere except where it is said to have happened” or “I shall say what-happened-nowhere, except where it is said to have happened,” as Gratwick 1993: 134 suggests. 32 Gratwick 1993: 180. Plautus seems to play with this idea, too, in having the Prologus report that he has been told of the close resemblance of the Menaechmi twins to each other by someone who saw them as boys (17–23). 30

Stag ing Ecphrasis in Early Latin Literature

25

mathematician who solved the problem of the duplication of the cube.33 In any case, the Menaechmus brothers would have been equatable, in the audience’s mind, with Greek-speaking Roman provincials inhabiting an island central to Roman military control in the Mediterranean. In this way, too, they are not only other to the Roman audience but also same. Further, as characters “less Greek” than those from plays set in Athens by Plautus’s unnamed poetae, they slide toward (but do not fully become) the non-Greek barbarian, a category in which Plautus famously inserts the Romans themselves. At Menaechmi 236 Magna Graecia is designated with the term “outer Greece” (Graecia exotica), words placing Sicily on the border between Greece and the non-Greek, barbarian world.34 The very fact of being from Sicily may qualify the Menaechmus brothers, from a Plautine perspective, for “barbarian” status in spite of their Greek names. At Rudens 583 Sceparnio insultingly calls the Sicilian Charmides, another Greek-named character, a “barbarian guest” (barbarem hospitem).35 Referring to his own Roman identity, Plautus elsewhere has prologue figures describe his adaptation of earlier Greek plays with the words “Plautus translated it into barbarian [i.e., Latin]” (Plautus [or Maccus] vortit barbare).36 This ironic embrace of non-Greek, outsider status finds a number of parallels in different Plautine versions of the same basic joke about how Greeks see Romans. Miles Gloriosus 211–12 seems to appeal to audience knowledge of a “barbarian poet” (perhaps Naevius) punished for his transgressions. At Bacchides 120–24, Pistoclerus calls Lydus a barbarian for not knowing of gods with comical Latin names such as Suavisaviatio (“Kissykissy”), while at Stichus 193–95, Gelasimus says he will have to learn “barbarian” (i.e., Roman) customs and act as his own auctioneer.37 Together, these passages treat with multivalent humor the same attitude about which the Elder Cato, warning of Greek doctors’ oath to poison barbarian patients, is said to have remarked indignantly, “They call us barbarians, too” (nos quoque dictitant barbaros).38 In Plautus, characters nominally Greek perform Roman self-consciousness about the simultaneous non-Greekness and adaptation of Greekness concentrated in the Roman use of 33

Gratwick 1993: 138. But cf. Maurice, who sees the term rather as “reminding the spectators that while the play is set in the Greek world, Greek is not Roman, and is indeed foreign and exotic, so that the term distances the audience from the play” (2005: 49). Distancing and assimilating interpretations could, of course, have co-existed in the collective audience’s mind. 35 See Leigh 2004: 91n127 for discussion. Evidence for non-Greek conquered peoples on Sicily is cited by Marx ad loc. 36 As. 11; Trin. 19. 37 See further Pl. Cur. 150, Mos. 828, Cas. 748, Capt. 492. For discussion of several of these passages, see Owens 2000: 397; Browning 2002: 262; Leigh 2004: 5; Reed 2007: 104. 38 Quoted at Plin. Nat. 29.14. For more on Catonian irony on this subject, see Reed 2007: 126–27. 34

t h e c a p t o r’s i m a g e

26

the word barbarus. Sicily activates both sets of associations within the Plautine worldview. While Menaechmus’s description of the mythological images plays upon the Roman construction of Greek effeminacy, it simultaneously engages Rome’s growing fascination with all things Greek and participates in “the hate-love, inferiority-superiority ambivalence that characterizes this long and complicated relation.”39 Ecphrasis here stages, in the fullest sense, receptivity to Greek culture, and uses this staging to explore a ramifying series of ambiguities. Plautus has Menaechmus emphasize certain things about the pictures he describes: in his reference to paintings “where the eagle snatches away Ganymede or where Venus snatches away Adonis” (144) the centrally placed verb raperet ( “snatches away”), which governs both subjects, aquila (“eagle”) and Venus, expresses simultaneously Menaechmus’s belief in his own attractiveness, his hopes for liberation, and his vulnerability to self-interested manipulation. Peniculus sees nothing of immediate relevance in the images, and views his master’s strange identification with myth as only an impediment to a good meal. Sympathetic and ironic interpretations compete, as Plautus uses a notionally Greek image to raise the question of how an (already Romanized) Greek identity might be related to its (already Hellenized) Roman counterpart. Menaechmus’s response to the imagined paintings in turn models the audience’s ideal response to the drama they are witnessing, as we discern from a closer examination of how his crossdressing functions within the plot as a whole. When we first see Menaechmus, he has come out of his house into a street of Epidamnus, the Illyrian Greek city where the play is set. At this point he is wearing the palla under the pallium. We know this from the fact that at line 196 he strips off his pallium and hands it to his parasite Peniculus before stripping off the palla in turn and handing it to Erotium (202).40 Menaechmus must reveal the palla at line 145 where he says “come, look at me” (age, me aspice) following his reference to the paintings. This scene alone would be enough to pique our interest in the palla and its symbolism, but this prop will have a still more elaborate function in the sequel. After delivering the palla to Erotium, Menaechmus heads to the agora for some unspecified business while Erotium prepares their meal. Not long after this, however, Erotium, all unawares, encounters Menaechmus’s twin, who has arrived in Epidamnus to look for his brother, and asks him to take the palla to the embroiderer to have it spruced up (425–27). Puzzled but eager to enjoy her himself, Menaechmus II agrees. Two scenes later (3.2), we see him emerging from Erotium’s house with the palla, 39

Anderson 1993: 139 with n9 on Plautus’s varying exploitation of contemporary views of the Greeks; cf. Anderson’s emphasis on features of Plautine comedy as “representative of Greek weaknesses Romans wanted to believe in” (139). 40 Cf. Ketterer 1986: 52.

Stag ing Ecphrasis in Early Latin Literature

27

which he now considers an unexpected windfall. Later (4.2) Menaechmus I is confronted by his wife for having stolen the palla, then denied reentry into his house until he gives it back (662). He, of course, thinks Erotium has it, and goes to get it from her, but when she hears him saying he doesn’t have it she thinks she has been cheated and denies him entry to her house, too (692). All is eventually resolved at the play’s end, when the twins recognize each other and figure out the confusion. The first twin decides to auction off all his belongings, including his wife and, presumably, the palla, and agrees to return with his brother to their familial home. While the palla functions as “a symbol of the varying relationships and changes of character within the play,”41 Menaechmus’s wearing of it stands in part for the audience’s pleasurable but risky “trying on” of the ambiguously Greek-and-Roman culture embodied by the drama itself. Menaechmus performs onstage, in other words, the kind of transgressive identification with an ambiguously Hellenic “other” that the sympathetic audience performs internally, an unseen, unspoken act of boundary-crossing enhancing its pleasure in the proceedings. Menaechmus thus transfers at least some of the anxiety inherent in such crosscultural identification to himself, so as to allow the audience to enjoy the play all the more. Although he thus becomes the focal point for the negative connotations of such a response, he nevertheless escapes this stigma in part as the play progresses. And given the palla’s function in producing such pleasure, it is particularly suggestive that at the drama’s end it appears destined to go the way of all Menaechmus’s auctioned belongings and be turned, quite literally, to profit.42 The theme of (Roman) material gain informs the Menaechmi both locally and in general. As Menaechmus’s performance most likely recalls the actual plundering of art from Syracuse, he himself, a Syracusan, both steals a palla from his wife and constitutes a kind of plunder, having been stolen in his youth (31–33), thus becoming an additional symbol of the Roman empire’s acquisitive reach. He, too, after all, has been delivered to Rome for the delight of Plautus’s audience, like the whole city of Epidamnus where he now resides. His ability to entertain by making artworks come alive announces Rome’s capacity to derive pleasure and enjoyment from the fruits of its expansion, whether paintings, plays, slaves, or anything else the Greek world might have to offer. Along with Greek art comes an imaginary Graeculus to act as a guarantor of both its identity-forming value and its comforting triviality in comparison with the fact of Roman power. On the stage, of course, the very notion of imitating someone else takes on self-referential significance, made all the greater here by the likelihood that the actors themselves might be slaves, possibly Greek ones.43 Thus when the character 41

Leach 1969: 42. Leach 1969: 43. 43 On the servile status of most Roman actors in Plautus’s time, see Christenson 141. 42

28

t h e c a p t o r’s i m a g e

Menaechmus, already imitating a woman by wearing the mantle (as well as imitating, in a different sense, the Epidamnian he is not), gleefully plays with the notion of copying painted figures from myth, the audience can also hear the actor’s voice teasingly playing upon the idea of his own dramatic mimesis. If the actor is a slave or a freedman, Menaechmus’s domestic servitude and deliverance becomes all the more resonant with the play’s Roman context, while his flirtation with Greek myth becomes all the easier for many audience members to laugh at, since his Greek otherness is doubled by the actor’s otherness as a figure of lower social status and perhaps as a foreigner.44 Scene-painting, originally a Greek technique, might have made his remarks on painting all the more amusingly self-referential onstage.45 But a still broader anticipation of the play’s theatricality emerges in the fact that parody of Greek tragedy, whose subjects came directly from myth, was a staple of Plautine drama. In this play it appears especially during Menaechmus II’s mad scene near the conclusion, when he imitates a raving initiate (or victim) of Bacchus (835–71). The episode was sure to recall such “classic” treatments of madness in tragedy as that of Hercules; further, the Bacchic cult, another import from Greek lands, had already begun to claim a growing number of devotees in Plautus’s day, and eventually attracted enough suspicion to be the target of legislation in 186 bce.46 So stage convention again merges with Roman reality. From Menaechmus I’s crossdressing to his brother’s frenzy, Plautus pokes fun broadly at Roman Hellenomania—and this within a wildly popular genre itself derived from Greek tradition! Far more than with a nondramatic work, the actual, staged performance of Plautus’s Menaechmi bestows a visceral immediacy upon the idea of viewing an artwork in this receptive fashion, as it makes ecphrasis itself concretely into a staging. Of course Plautus’s imagined paintings can stand for his play itself, an ambiguous product of artistic craft subject to the engaged spectatorship of a likewise ambiguously construed audience. But the fact of comic performance undoes any simple notions of the text’s self-assertion over the image: Plautus strives to appeal to an audience that can, like Menaechmus, be seen as “crazy” for Hellenic art—subject to its wondrous spell—at least to the sympathetic audience member. Greek culture is given space to compete with Roman, and just as the playwright allows his hero to dress like a painted Ganymede or Adonis, so too the audience caught up in the humor others itself in accordance with the cues such paintings provide. With its multilayered onstage imitations, the play thus embraces a technique of cultural mimesis that Plautus’s successor, Terence, would take up and adapt for his own ends. 44 Freedmen could, of course, be among the audience, and even slaves were apparently allowed to attend Roman comic performances (cf. Pl. Poen. 24–28). 45 On the possibility of scene-painting (skenographia) on the Roman comic stage, see Marshall 2006: 49–50. 46 The famous senatusconsultum de Bacchanalibus. On this and the Greek tragic reference here, see Gratwick 1993: 215.

Stag ing Ecphrasis in Early Latin Literature

29

III. In the Image of Jupiter: Ecphrasis, Exempla, and Rape in Terence’s Eunuchus Written perhaps a generation after the Menaechmi, Terence’s Eunuchus stages a darker receptivity to images of Greek myth. The difference in tone cannot be ascribed simply to societal changes, yet the Rome that had since 200 bce seen the dissolution of the Bacchanalian cult, the censorship of Cato the Elder, and two wars with Macedonia, and that in the very year of the Eunuchus’s first production (161 bce) had witnessed an expulsion of Greek philosophers and rhetoricians, was arguably less carefree in its attitudes toward Hellas than the Rome in which Plautus began his career.47 Terence’s association, nevertheless, with members of an increasingly discerning Hellenophile elite can be deduced from his prologues, the didascalic notices to his plays, and Suetonius’s Vita, while the commercial success of the plays themselves indicates their popular appeal.48 Like Livius, Terence apparently came to Rome as a slave, in this case from Carthage, and was bought, educated, and manumitted by a senator, Terentius Lucanus.49 Although the notion of his participation in a so-called Scipionic circle of cultured philhellenes has long been dismissed, the sophistication of critics who could, for example, take a playwright to task for engaging in the wrong kind of “translation” from the Greek is self-evident.50 If Terence’s social milieu assured his keen, competitive interest in Greek culture, the decades leading up to 161 also saw more Roman victories that placed the spectacle of captured Greek artworks before the eyes of a marveling public.51

47

Cf. Dessen 1995: 126, who also notes the passage of sumptuary laws in 182 and 161 BCE as a sign of unease with the increasing extravagance that resulted from the massive increase in wealth in this period. 48 See Goldberg 1986: 13–15 for discussion of Terence’s elite ties and the popularity of his plays. On the “purer” Greek milieu of Terentian drama, which includes fewer of the explicitly Roman references found in Plautus, see Goldberg 1986: 10–13; conversely, on the ancient notion of Terentian Latin’s linguistic “purity,” see Müller. 49 This is the story transmitted by Suetonius (Vit. Ter. 1.1); but for skepticism over the details of Suetonius’s account see, e.g., Barsby 1999: 1–2. In particular, the presumption of a Carthaginian origin may be a false deduction from Terence’s cognomen, Afer (“the African”), which is shared by an established Roman gens, the Domitii (Barsby 1999: 1n3). 50 See Goldberg 1986: 13–15 on the now universal dismissal of a supposed “Scipionic circle” of Hellenophiles opposed to the anti-Hellenic tendencies of the Elder Cato (the latter’s interests in Greek literature are now also well established in spite of his public stance against contemporary Greece and Greeks). On the sophistication implied by Terence’s references to so-called contaminatio, cf. Martin 1976: 7 and below, pp. 33–34. 51 See the introduction, p. 30 and on this context in relation to the Eunuchus, cf. esp. Germany 2008: 101–07.

30

t h e c a p t o r’s i m a g e

Here art, however, instigates sexual violence. At about the Eunuchus’s midpoint, the young man Chaerea relates his doings after disguising himself as a eunuch and sneaking into the house of the prostitute Thais so as to be close to a girl, Pamphila, with whom he is infatuated. Once inside, he says, he noticed the girl observing a painting of Danae and Jupiter (the god is called this rather than “Zeus”): While preparations [for her bath] are being made, the girl sits in a chamber looking at a certain painting. There upon it was this picture: how they say Jupiter once sent a golden rain into the lap of Danae. dum adparatur, virgo in conclavi sedet suspectans tabulam quandam pictam: ibi inerat pictura haec, Iovem quo pacto Danaae misisse aiunt quondam in gremium imbrem aureum. (Terence, Eunuchus 583–85) Later, finding himself alone with the girl, he rapes her in what he describes as an imitation of Jupiter’s sexual conquest. Speaking of his thoughts while gazing at the image—in a much fuller account than Menaechmus’s response to the Ganymede and Adonis paintings—Chaerea declares, I also began to look at it [the painting], and because he [ Jupiter] had already once played a similar trick, I delighted very much more that a god had turned himself into a man and come secretly onto the roof of another in order to play a trick on a woman through the impluvium. But what a god! “He who shakes the highest regions of heaven with his thundering.” I, a mere mortal, was I not to do it? Indeed that’s what I did—and with pleasure. egomet quoque id spectare coepi, et quia consimilem luserat iam olim ille ludum, inpendio magis animu’ gaudebat mihi, deum sese in hominem convortisse atque in alienas tegulas venisse clanculum per inpluvium fucum factum mulieri. at quem deum! “qui templa caeli summa sonitu concutit.” ego homuncio hoc non facerem? ego illud vero ita feci—ac lubens. (Terence, Eunuchus 586–91) Modern critics have in general condemned Chaerea’s action and often supposed a similarly negative assessment by Terence’s original audience, although some have questioned the extent to which that audience, unlike a modern one,

Stag ing Ecphrasis in Early Latin Literature

31

would have viewed the rape negatively.52 More than one response is, of course, possible: Chaerea’s bragging is simultaneously “the voice of the aggressive male” inviting others to become complicit in his feelings of power over women and a performance that “a more reflective portion of the audience could view . . . in a negative light.”53 In fact, Terence encourages reflection over whether the painting could justify the deed. The particular way, moreover, that Terence uses the idea of amatory paintings inspiring imitation foregrounds Rome’s adaptive emulation of Greece.54 Based on two plays of Menander, the Eunuchus, like the earlier work of Plautus, receives Greek culture in a highly adaptive fashion, and the dialogue between Chaerea and Antipho, we learn from Donatus, results from the transformation of what had been a monologue.55 The Eunuchus is set in Athens, and Chaerea is described as an ephebe absent from his post in Piraeus (290), yet his words in the passage pointedly assimilate his perspective to that of a Roman viewer. He recounts a feeling of enormous pleasure that so great a god as Jupiter had “come secretly onto the roof of another in order to play a trick on a woman through the impluvium.” The impluvium is a Roman detail: Greek houses did not possess this feature. Donatus informs us, moreover, that line 590 contains a parody of Ennius, apparently one of his tragedies, an allusion certainly not 52 Especially because of the scene’s echoes of Roman marriage ritual and the eventual betrothal of Chaerea to Pamphila. See Philippides 1995 with bibliography. The generic conventions of Greco-Roman comedy, she argues, would have left the audience expecting a marriage, and the negative impression is mitigated in other ways as well. For Konstan, “An untroubled empathy with the youth is licensed by the holiday mood of comedy, as well as by the custom of the genre”; his later subjection to “a sense of decorum and shame” allows us to “bracket his outrageous interlude as a transient impulse or aberration” (1986: 387). Contrast Smith 1994, who calls attention to the fact that the freeborn Chaerea defends himself insincerely by saying he thought the girl was a “fellow slave” (858, conservam) and, before entering the house, announces his plan to infiltrate it as a revenge for prostitutes’ exploitation of young men (382–87): his rape differs from those committed by characters in Terence’s other plays, the Hecyra and the Adelphoe, in that “it is broad daylight, not night, Chaerea is completely sober, not drunk, and there is no festival atmosphere in which normal rules of conduct might be or might seem to be suspended” (Smith 1994: 23). James views the rape plots in Eunuchus and Hecyra as standing apart from other examples in Roman comedy in the way that they “deliver a powerful critique of the coercive, self-centered masculine sexuality that characterizes Roman marriage” ( James 1998: 46). On the question of whether Chaerea intended the rape before entering Thais’s house, see Rosivach 1998: 46–50, who maintains that he does, and Germany 2008: 30–65, who elaborates upon an observation of Donatus to reject this assumption (while providing an extensive history of the scene’s reception). On the prevalence of rape as a theme in New Comedy, see Rosivach 1998: 13–50. 53 Smith 1994: 30, 38n51 (citing Richlin 1983: 58–59). 54 Even if, as Barsby 1999: 198 suggests, the idea itself is a literary commonplace. See further Germany 2008: 67–96. 55 Don. Ter. Eu. 539. Extensive discussion of Terence’s Menandrian sources can be found in Lefèvre 2003.

32

t h e c a p t o r’s i m a g e

lost on an audience of avid Roman theater goers.56 And a detail of this kind reminds one of Ennius’s own retrojection of Jupiter’s Roman title maximus (Annales 444 Skutsch) upon Homer’s Zeus himself, the Greek god thus becoming Jupiter Optimus Maximus and the world he rules over a Roman one.57 It is possible that Menander included tragic parody at a corresponding moment in his play, but the Ennian passage was presumably itself based upon a Greek model, and allusion to Ennius allows a slippage of Chaerea’s Greek persona to suggest a Roman youth with theatrical interests. The Hellenophile Terence does not simply characterize Chaerea’s response as an example of Greek weakness, to be looked upon with haughty superiority by the play’s audience.58 Rather, Chaerea, an embodiment of human failings and an (erotically) engaged spectator, is a figure for the audience itself. Yet Terence both encourages and complicates identification with Chaerea as a receptive viewer of art. His comically literal-minded assumption that Jupiter’s disguising himself as a shower of rain meant sneaking into a girl’s house through the impluvium of course lacks the sophistication that some members of Terence’s audience might bring to the interpretation of a mythical artwork, while for others it could have introduced just the right note of irreverence to bring Greek art down to a more approachable level. His seemingly confused description of an image in which Jupiter appears both as a shower of golden rain (Iovem/quo pacto Danaae misisse aiunt quondam in gremium imbrem aureum) and as a man (deum sese in hominem convortisse) could simply suggest Chaerea’s shallow opportunism, or it could reflect detailed knowledge of an iconographic type attested later in Pompeii.59 As in Plautus, the presence of skenographia would bring the art of painting itself into visible proximity, though the audience is free to imagine the painting in the onstage house as a more (or less) technically accomplished example.60 Chaerea, furthermore, encounters the painting in what is essentially a brothel, the house of a self-employed prostitute (382: domum meretriciam). This is not the sort of place with which respectable Romans might readily admit to familiarity, although it is appropriate within the festival atmosphere of Roman comedy.61 The setting colors perceptions of the erotic subject matter and lends the motif of the shower of gold a mercenary connotation it might not 56 See Barsby 1999: 198; Karakasis 2005: 96. For more on the tragic dimensions of this scene, including the possibility of allusion to Euripides’ Bacchae, see Sharrock 2009: 219–26. 57 Feeney 1991: 128. Cf. the expansion of this point at Reed 2007: 109–10. 58 Whether it amounted, for them, to “trite irreverence” or a “repulsive” act on Chaerea’s part: so Konstan 1986: 386; Smith 1994: 27. 59 See Barsby 1999: 197; Germany 2008: 182–88. 60 Cf. Germany 2008: 11, 116. 61 On the ancient practice of matching art to its context, especially in evidence in the use of erotic images in brothels, see Germany 2008: 73 and 67–96 on the theme of imitating erotic pictures in later Roman poetry and painting.

Stag ing Ecphrasis in Early Latin Literature

33

otherwise possess. Indeed, Chaerea seems to miss altogether the point of the painting’s location: even Jupiter, the painting reminds potential clients, had to pay for the favors of Danae, by transforming himself into gold.62 The painting’s theme accords well with Terence’s overall emphasis in this play on the conflict between love and material constraint. In particular, it points up the contradiction in Thais’s character between the true love she expresses for Phaedria and her careful scheming to insure her own financial security.63 And yet Chaerea’s disguise as a eunuch helps motivate the play’s title; he is promised Pamphila in marriage and at the drama’s finale exerts some control over the action by urging Phaedria to share Thais with the soldier Thraso, Phaedria’s erstwhile competitor for her affections.64 Again like Menaechmus, Chaerea is a comic hero, rape notwithstanding. As in the Menaechmi, ambiguous imitation is key to the Eunuchus’s plot and refers in part to the play’s Hellenic theatricality. Chaerea play-acts the part of an enslaved eunuch given as a gift (he is still dressed as such when he performs the ecphrasis), and the mimesis of this uncanny figure doubles the real actor’s depiction of a Greek thus “enslaved” by love. This is all the more true if the actor himself is a (Greek) slave.65 Terence’s own experience of slavery and manumission could of course have drawn him to such topics or influenced perceptions of his work, but we lack evidence for either point. In any case, part of the humor of watching (the actor playing) Chaerea then profess to have copied an image of Jupiter lies in the socially degrading nature of these previous onstage imitations. It is especially easy, in other words, to laugh at such a figure’s misplaced self-aggrandizement given the polar opposition between the first figure he imitates, a castrated male, and the second, the hypermasculine, philandering king of the gods. Drama itself, Terence suggests, is funny because of how ludicrous its imitations are in general, when actors in Rome play Greeks imitating eunuchs and gods, or sometimes, as especially in tragedy, the gods themselves. Indeed, paratragedy may specifically be at work in the Eunuchus ecphrasis, as noted above. Competitive emulation of Greek models is deeply significant for the Eunuchus as a whole (as for Terentian drama in general), since its prologue treats at length the question of whether Terence is to be blamed for having combined Menander’s Εὐνοῦχος with elements of his Κόλαξ (The Flatterer), when previous versions of the latter play, by Naevius and Plautus, were already extant (19–26, 62

Cf. Smith 1994: 27. Konstan 1986. 64 Cf. Smith 1994: 29, although she emphasizes this as an instance of Chaerea’s “selfish, self-centered, and mercenary qualities” (30). 65 On the eunuch scheme as an instance of metatheater, cf. Dessen 1995, Knorr 2007: 173, Germany 2008: 116–88, and Sharrock 2009: 219–26. For more on the thematic significance of Chaerea’s appearance as a slave within this play and Roman comedy broadly, see McCarthy 2004: 112–14. 63

34

t h e c a p t o r’s i m a g e

so-called contaminatio).66 The Prologus declares Terence’s ignorance of these earlier renditions, and denies the charge he claims to have been launched at the playwright by his detractor Luscius Lanuvinus during a preliminary performance before an aedile: that “a thief, not a poet, presented the play” (23–24, furem, non poetam fabulam/dedisse).67 The Prologus retorts further that since all the characters of comedy (the running slave, the parasite, etc.) are familiar from their stock usage in earlier plays, Terence’s use of similar figures cannot be the cause for blame, when “nothing in fact is said that has not been said before” (41, nullumst iam dictum quod non dictum sit prius). Chaerea’s comparison of his own actions with the behavior of Jupiter takes this adage a step further in suggesting that the stage world brings to life the story lines of myth, which was in fact true in the case of certain comedies, such as Plautus’s Amphitruo, while being the very basis of tragedy. Intermingled with this question is that of whether and how life itself imitates art: Terence, like Plautus before him, reveals acute self-consciousness about drama’s claim to be a mimesis of reality, a claim held up for inspection in miniature through Chaerea’s response to the painting. Like the narrator of the Gigantomachy ecphrasis in the Bellum Punicum, Chaerea sees the painting as a story (note quo pacto . . . aiunt [585, “how they say”]), a detail blurring the line between image and narrative in a manner especially pertinent to drama, since plays are in fact stories animated and given visual form onstage. Against Roman imitation of Greek art and drama stands still another kind of mimesis, Romans’ imitation of their own historical or familial exempla. These, too, were enshrined in myth and legend and given visual form in the display of wax ancestral masks (imagines), whose powerful effect on Roman spectators was remarked by Terence’s contemporary, Polybius.68 Indeed, imitation of the ancestors (imitatio maiorum) was a Roman cultural mainstay in which Terence seems to have taken a special interest. No play of his illustrates this better than the Adelphoe (The Brothers), with its plot centering on the question of whether or not the educational example set by a father for his son will influence the latter for better or worse, issues perhaps imbued with Aristotelian connotations of the Golden Mean in Terence’s Menandrian source.69 The Adelphoe invites its spectators to reflect on Demea’s strict, rural upbringing of Ctesipho as a failed model, 66 On the much-studied issue of contaminatio and its status as a characteristically Terentian concern, Goldberg 1986: 91–122 has become a standard discussion; but against his view of the phenomenon see Germany 2008: 189–204. 67 On Luscius, his quarrel with Terence, and the preliminary performance, see Barsby 1999: 15–17, 85–86. 68 Plb. 6.53–54. For discussion of this passage see Flower 1996: 91–127 and for its relevance to the Eunuchus, cf. Germany 2008: 96–101. 69 For helpful discussion of this essential aspect of the Adelphoe and similar features of the Heautontimorumenos (The Self-Tormentor), see Martin 1976: 16–29; Braun 2000. For the possibility that Terence conceived of his plays as having an educational purpose, see Fantham 2004.

Stag ing Ecphrasis in Early Latin Literature

35

an impetus toward the youth’s wild behavior in the city, and also on the effects of the lenient urbanite Micio’s attitude on Aeschinus, who has raped and impregnated a girl while drunk but nevertheless promised to marry her.70 The youths’ behavior, together with Demea’s conversion to leniency at the play’s conclusion and the bachelor Micio’s decision to marry the mother of Aeschinus’s new wife, confutes easy distinctions between good and bad, influential and noninfluential exempla. These plot points call into question the model of old-time severity on which Roman moral discourse often turned, but nevertheless project the whole issue of how imitation makes Romanness onto a milieu that is easy to write off as Greek. So, too, the audience of the Eunuchus is invited to see its own predilection for imitating illustrious models in Chaerea’s citation of Jupiter’s example, but it can still, if it prefers, dismiss such behavior as Greek error or excess. Chaerea’s ecphrasis situates itself in an ambiguous relation with the Roman process of learning from exempla. His comic heroism both sustains and falls satisfyingly short of the Roman ideal. The Romans’ own use of paintings as illustrations of exemplary military conduct reinforces this aspect of Chaerea’s behavior.71 As early as 272 bce the temple of Consus in Rome bore paintings commemorating L. Papirius Cursor’s victories over the Samnites and Tarentines. M. Fulvius Flaccus enjoyed a similar honor in 264 bce, when paintings of his combats at Volsinii graced the temple of Vortumnus. As already noted, M’. Valerius Messalla’s success against Carthaginians and Syracusans in the following year was represented on a wooden panel on the Curia Hostilia. Among such objects, those placed in temples were meant especially to indicate the workings of divine favor behind both the victories in question and the larger phenomenon of Roman expansion. Paintings were also displayed in the processions of victorious generals, as in 211 bce following the defeat of Syracuse or in 188 after that of Antiochus III of Syria, and these may have been based on Hellenistic battle paintings known first from Roman incursions into Magna Graecia.72 Thus from a Roman perspective Chaerea’s imitation of a painting of Jupiter’s conquest could be imagined not only as a comical Greek behavior but also as one suggestive of Roman practice. Paintings of victorious conquest, some of them no doubt created by Greek artists, had long since been a way to express the very essence of Romanness. Turning its central ecphrasis into an allegory of its cultural borrowings, the Eunuchus sets out to show that what might seem like transgression or theft may nevertheless redound for the good, depending on one’s perspective. The behavior of Jupiter that Chaerea claims to have copied is decidedly thief-like—to have 70

See esp. Rosivach 1998: 16–20 on the rape in the Adelphoe. For the following, I depend especially on Gruen 1992: 90. 72 Holliday 1997: 130, 134–37. They are also likely to have been executed by Greek artists (Holliday 1997: 143). 71

t h e c a p t o r’s i m a g e

36

“come secretly onto the roof of another in order to play a trick on a woman through the impluvium”—and this makes Chaerea’s own act akin to theft, since much of the play centers around the question of who possesses Pamphila. But because Chaerea’s rape of Pamphila precipitates his betrothal to her, he can be said not only to “steal” her for himself but also to have this thievery legitimated as marriage. Compare Terence’s own appropriation of Greek material (two Greek plays married into a Latin one, not simply “stolen”). Even if an audience member chooses to find fault with Chaerea, the latter’s own model, Jupiter (not identical with the Greek Zeus), offers a more certain authority. At the heart of the Eunuchus, a Romanized Greek character describes the Roman high god’s conquest of a Greek heroine. Jupiter’s precedent allows the audience of the Eunuchus to fantasize about violence and yet admit its own captivation by Greek culture’s alluring beauty. As in the Menaechmi, so here: the text that achieves a kind of aesthetic control over an imagined artwork points to the image’s overwhelming power over the comical viewer who describes it. This power is linked to Jupiter, whose might of course surpasses the abilities of any onstage character, playwright, or dramatic text. Terence himself shares with the audience his status as the Roman Jupiter’s subject, and insofar as Chaerea adapts Greek material (myth) for his own purposes while falling under its spell, he becomes a comical figure for the Hellenophile playwright as well. Chaerea’s painting even “functions as an epiphany” of the god if we choose to see it as confirmation of his role as tutelary deity, guiding the plot to its resolution as he also oversaw Rome’s continued rise.73 For all that it may suggest the powers of the dramatic text, the ecphrastic painting of the Eunuchus indexes still more forcefully Rome’s broader investment in the identity-forming potential of Greek art and artistry. Terence’s very genre bears witness to an intense Hellenophilia that puts simple notions of dominance and submission profoundly into question, whether they apply to the dramatist, his text, or his subject matter.

Conclusion: Ecphrasis Onstage and Off Already at this point in our study, a comparison of the Menaechmi and Eunuchus with Naevius’s Bellum Punicum yields much to reflect upon. In spite of what we might expect, given the “standard” view of ecphrasis, the cultural dynamics of the trope here far outweigh in importance the programmatic role that would make it simply a symbol or stand-in for the written text. Plautus and Terence share an interest in ecphrasis as a way to concentrate attention upon imitation

73

For Jupiter’s “epiphany” and his possible role as the play’s tutelary deity, see Germany 2008: 61.

Stag ing Ecphrasis in Early Latin Literature

37

and spectatorship in the context of stage drama. Both authors, like Naevius, thereby push the audience to reconsider its relation to Greek culture, especially images of Greek myth, as received in Rome. Indeed, the comic stage, with its capacity to make stories from the past come alive, emerges as an especially good place to think through the content of Greek mythological imagery and the way it exerts influence upon the Roman self, in spite of the fact that later sources deemphasize the content of the paintings, sculptures, and other artworks that the Roman war machine was busy furnishing for its citizens at this time. Plautus’s Menaechmus appears to be nothing but a ridiculous Greek, prancing about in the feminine garb that assimilates him, he believes, to a painted Ganymede or Adonis. His sympathetic rebellion, however, against the feminine and servile forces oppressing him make his identification with the two mythic figures an aspect of his comic heroism. His perspective is both contrasted and aligned with the Roman audience’s through his potential status as a provincial. Terence’s Chaerea, although his rape of Pamphila demeans him with its self-indulgent violence, also succeeds as a comic hero. His imitation of what he considers Jupiter’s exemplary behavior creates a similarly indeterminate relation between himself and the spectators. The ecphrasists of Roman comedy are richly imagined characters within a hybrid, Greek-and-Roman stage world, and their reactions to artworks turn the question of receptivity back forcefully upon their audiences. Such ecphrases, literally stagings, will be essential to recall in the chapters that follow. A similarly receptive audience is assumed by Naevius’s epic, with its image of Greek myth underscoring the importance of Roman victory. But our comedic examples take on a significance both greater and more limited when compared with the ecphrasis in Naevius’s nondramatic poem. Even if we had more of the Bellum Punicum, as an epic it would hardly provide us with the same sense of self-consciously dramatized looking that pervades the Plautus and Terence passages, and this issue, too, should be kept in mind as our study progresses. Naevius’s image may have had an internal viewer, but this character’s scrutiny of it would have appeared, first and foremost, as a break in the epic narrative. In this moment of contemplation, a static artwork momentarily takes over the narrative function, but by its contrast with the main narrative reveals a salient difference between doing and seeing, heroic deeds and inward thoughts. In a play, action and spectatorship are more closely fused: looking, thinking, and describing are doing, both for the characters onstage and for the audience, who is there precisely to perform the function of spectator, to be an observant audience, sometimes in very active manner. The unreflective Menaechmus and Chaerea, in their impulsive amatory outbursts, hardly signal the transition from narration to description the way an epic narrator might. Menaechmus’s question to Peniculus about paintings—“Tell me, have you ever seen . . .?” (143)—creates the most casual of links, while Chaerea is mainly interested in telling how the

38

t h e c a p t o r’s i m a g e

painting spurred him on to disreputable action. The very nature of the mimesis here, together with the brevity of these passages, may signal a bathetic contrast with the epic trope. Again, Naevius’s ecphrasis, if more of it survived, might help to support this view. With Plautus and Terence we leave actual plays behind and turn to other forms. The world of drama, however, will appear again, resituated for us on an (anti-)epic stage, as we consider next the self-conscious inconsistencies informing ecphrasis in Catullus’s “epyllion.”

2

Becoming Ariadne Marveling at Peleus’s Coverlet with the Inconsistent Narrator of Catullus 64

What if a Roman man’s description of art exposed the flaws in his knowledge of Greek mythology? And made him talk like a Greek woman to boot? The first major ecphrasis in extant Latin literature plays upon exactly these notions.1 The passage, from Catullus’s “miniature epic” (“epyllion”), poem 64, depicts an imaginary coverlet illustrating the story of the Greek heroine Ariadne, an object displayed in the house of the Argonaut Peleus on the day of his wedding to the goddess Thetis.2 Unlike any other surviving ancient ecphrasis, this one accompanies elaborate visual description with extended mimicry of direct speech by figures within the image.3 The technique suggests intense identification on the part of the poem’s narrator with a piece of visual art, and such is the effect, too, of its arresting narrative, direct address of its characters, claims to hear the past’s distant music, and praise for the heroic age as a time when justice reigned and gods mingled with men. Yet a considerable gap exists between the narrator who is made receptive to the Greek past in this way and the controlling poet, Catullus, who undercuts his narrator by lacing his story with unavoidable inconsistencies. So Catullus’s dazzlingly brilliant ecphrasis bears out my main contentions in a somewhat different fashion from the earlier Roman examples treated in Chapter 1. Through his self-dramatizing narrator, Catullus self-consciously

1 For earlier, shorter ecphrases in Latin literature, see Chapter 1. Lucretius’s strongly visual Epicurean poem De rerum natura seems to recall contemporary artistic iconography in its descriptive sections and briefly describes golden, torch-bearing statues of youths in a passage (2.24–26) reminiscent of Homer’s ecphrasis of the Phaeacian palace in the Odyssey (7.100–02), but contains no extended, explicit ecphrases of art. Cf., e.g., Gale 1994: 80–84; Fowler 2002: 92–101; Edmunds 2002. 2 The term “epyllion” is employed by convenience to refer to a short hexameter poem in Hellenistic style, although there is no ancient authority for its use in this sense. For discussion and bibliography see Fantuzzi and Hunter 2004: 191–245. 3 Laird 1993: 20.

39

40

t h e c a p t o r’s i m a g e

demonstrates—and undermines—idiosyncratic Roman ways of identifying with Greek mythological imagery. At issue everywhere in Catullus 64, but especially in its outsized ecphrasis, is the cultural competition at work through and within Roman receptivity to the Greek past. If the ecphrasis can stand for the poem, far more important is the way Catullus’s elusive narrator appeals to his audience by inviting it to enter, as contemporary Romans, into an archaic Greek world seen in part through Alexandrian Greek eyes. While Catullus, that is, appeals to the Roman philhellenism of his day, Greek and Roman cultures compete, often in the subtlest terms, for influence over the narrator’s projected sense of self, which remains fluid, ambiguous—and ultimately inconsistent—to the poem’s end. Catullus opposes Greek and Roman types of seeing, cultures of viewing, and holds up for inspection their simultaneous claims upon Roman identity. Thus as in Chapter 1, simple notions of the Roman text’s attempts at dominance over its Greek image are clearly inadequate in this case. Both sympathetic and ironic at once, poem 64 reflects a new cultural moment, the wave of highly sophisticated Hellenism that, following repeated military conquests of Greek lands in the third and second centuries bce, washed over Rome in the following century. We are now a long way beyond any simple tension between Greek and Roman culture, as one might posit for an earlier time, and far into a period of the most nuanced adaptation. Just how nuanced has been shown by recent discussion of the poem’s narratorial persona.4 Poem 64 begins with a vision of the Argo’s first voyage, during which Peleus and Thetis fall in love, then recounts the preparations for the wedding in Peleus’s opulent palace, including the display of the marriage-couch on which the coverlet appears. The great central ecphrasis sets Ariadne’s abandonment by Theseus beside her delivery by Bacchus, with “flashback” interludes recounting Theseus’s arrival in Crete, slaying of the Minotaur, and, in a further embedded flashback, his departure from Aegeus in Athens as well as Aegeus’s eventual suicide upon viewing the dark sails of his son’s returning ship. The main narrative concludes with an epithalamium performed by the three Fates 4 My understanding of this aspect of the poem owes much to O’Hara 2007: 33–54. While extending his readings of inconsistency, I also situate the phenomenon itself in a somewhat different context of Roman identification and appropriation. For the latter, I draw especially, although again with some modification, on Fitzgerald 1995: 140–68 (cf. below), notwithstanding his general skepticism toward “Alexandrianism” as a helpful critical category here and apparent conflation of the personae of poet and narrator, as at 141: “Closing the poem, Catullus regrets that the depravity of the modern age has caused the gods to withdraw themselves from the light of day.” As I do, Fitzgerald compares the poem with Campanian wall painting, but of a later (most likely Augustan) period, while I seek connections between Catullus’s technique and its more immediate Republican visual context. For an introduction to the massive bibliography on the poem, see Schmale 2004: 294–98.

Becoming Ariadne

41

at the marriage ceremony, a song predicting the birth and military prowess of Achilles, before the narrator offers a commentary on decline. But all is not as it seems. The Argo, for example, though introduced as the world’s first ship, turns out somehow to have been preceded by the ships in Theseus’s fleet. The farmland of Thessaly improbably goes to ruin while its rustic custodians attend the wedding. The narrator’s performance is thematically inconsistent as well: he praises the virtutes of the heroes and yet focuses on Theseus’s abandonment of Ariadne as well as the darker side of Achilles’ martial valor, while, in the ecphrasis, seeming to show a pronounced sympathy with the claims of eros and familial bonds. His treatment of Ariadne, moreover, is not only sympathetic but also voyeuristic, and in her flaring passion she comes to suggest the murderous figure of Medea as received through Euripides, Apollonius Rhodius, and Callimachus. If the temptation exists to view the narrator simply as ironic and even to posit identity between him and the poet (especially when the love-struck “Catullus” of the poet’s amatory verse begins to reemerge in a quasi-epic form), this assumption, too, grows problematic as the possibility of real “mistakes” on his part becomes impossible to rule out. These are not just dry Alexandrian puzzles posed for the learned reader, but carry the utmost significance for the poem’s larger cultural attitudes, as Catullus wittily coaxes his audience to ask itself whether, for all its learning, it sees the Greek past in a similarly inconsistent way. To the extent that it identifies with Catullus, the “hidden author” or controlling poet behind the scenes, the audience of poem 64 is offered some sense of mastery over both literary allusion and visual illusion.5 But Catullus’s narrator raises questions about himself that might also apply to the audience (and to Catullus), and so prompts learned reflection on Roman Hellenophilia itself. Even as he undercuts the narrator, however, Catullus also makes him appealing to a Roman audience, not least in his expressed admiration for Greek aesthetic wonders and interest in the mythic tradition. In this regard, the illusionistic panels of contemporary Second-Style wall painting, with their vivid depictions of a resplendent world strongly influenced by Greek traditions, present an invaluable comparandum. Indeed, as we shall see, poem 64 offers a sly commentary on a broader culture of viewing like that implied by these paintings, another highly self-conscious medium for philhellenic self-styling in Rome at this time. In particular the awareness of social limitation built into the grandiose iconography of

5 Catullus, that is, appeals to the audience as “confident consumer and appropriator of Greek culture” (Fitzgerald 1995: 167) as he distances the narrator’s perspective from the audience’s. I am not as ready as Fitzgerald, however, to ascribe the epithet “confident” to this Roman consumer, as will become clear in what follows. For the idea of a “hidden author” in poem 64, see O’Hara 2007: 43; the term was coined by Conte (1996) to describe the elusive authorial presence within Petronius’s Satyricon, on which see below, pp. 177–206.

t h e c a p t o r’s i m a g e

42

the Second Style suggests why Catullus’s audience might be drawn to the narrator’s inadequacies in the first place. A representational tour de force, poem 64 stages Roman verbal art imitating Roman life imitating Hellenic visual art. Its “labyrinthine” temporality, often remarked by scholars, is causally linked to the deep and conflicted identification of the narrator with the Greek past, while its problematic moral “indictment” of the Roman present emerges as only the final scene in this performance of inconsistent receptivity.6 Such is the wealth of cultural invention that Catullus brings to ecphrasis in a poem forever “teasing with paradox.”7

Wondering at the Argo A proper appreciation of the ecphrasis in poem 64 requires us to start at the text’s opening, since the narratorial postures enacted there become the basis for those of the ecphrasis. At the poem’s very outset, Catullus mobilizes a dense, Alexandrian web of allusions to both Greek and Roman authors around his narrator’s account of a wondrous, erotically charged spectacle from archaic myth: the Argo, the first ship (or so it would seem) on its initial voyage (1–21). This, the poem’s brief first movement, recalls and revises no less than Euripides’ Medea, Apollonius’s Argonautica, Callimachus’s Aetia, Ennius’s Medea exsul and other verse by Ennius and Accius.8 The opening detail “are said” (2, dicuntur, in reference to the trees that “swim” to the Phasis river in the form of the Argo) signals the presence of such allusion in an “Alexandrian footnote.”9 The allusions themselves announce the poem as an Alexandrian work “polemically” correcting older Greek and Latin versions of the myths surrounding the Argo.10 The poet’s skillful handling of allusion, however, will be part of what allows him to undercut his visualizing narrator, so that the question of the poem’s Alexandrianism is already becoming more complex than this. Visual lushness abounds: the ship’s oars striking the blue water and churning it to a white foam as its prow cleaves the surface of the sea personified as the goddess Amphitrite (7, 11–13) infuse the scene with an erotic energy fitting in the prelude to a marriage.11 Catullus concentrates attention on the act

6

On the poem’s temporality, see Weber 1983; Janan 1994: 107–12; Gaisser 1995; Schmale 2004; O’Hara 2007: 33–54. For its “indictment” of the Roman present, see Konstan 1977; cf. Curran 1969; O’Connell 1977: 754–55; Gardner 2007: 162–63. 7 Martindale 2004: 98. 8 Thomas 1982. 9 Ross 1975: 78. 10 Thomas 1982. 11 Konstan 1977: 15–18; cf. Fitzgerald 1995: 140–68.

Becoming Ariadne

43

of viewing the Argo by including the “wondering Nereids” (15) as spectators in the text.12 The Nereids in turn, depicted semi-nude (16–18), their wonder having led them to lift themselves “as far as the breasts”13 from beneath the water, become objects of an erotic gaze: that of the narrator but also of the Argonauts and potentially of the visualizing audience. And Peleus and Thetis now each become the love object of the other, although with considerable emphasis on Peleus’s passion and only litotic suggestion of Thetis’s assent: Then Peleus is said to have burned with love for Thetis, then Thetis did not spurn marriage with a mortal. tum Thetidis Peleus incensus fertur amore, tum Thetis humanos non despexit hymenaeos. (Catullus 64.19–20) Here, as throughout poem 64, Catullus’s narrator foregrounds an eroticized receptivity to arresting visual images from the mythical past while the poet appeals to an audience familiar with earlier literature (note the additional “footnote,” fertur, in this last quotation). Delving into some additional details of this opening passage will confirm further my point about its cultural dynamics. Catullus in fact makes the narrator impart his initial wonder through a skillful reworking of the Hellenistic epic poet Apollonius Rhodius, so as to produce a partial syncretism of different viewers and focalizers. Line 9, “[the goddess] herself made the chariot, flying in a light breeze” (ipsa levi fecit volitantem flamine currum), which identifies the Argo as entirely a divine production and so even more deserving of wonder, alludes to Argonautica 1.111–12 on the collaboration of the goddess Athena and the mortal Argus. Catullus’s suppression of Athena’s co-worker serves the purpose of a counter-etymology linking the name “Argo” to the Greek adjective ἀργός (“swift”),14 while calling the ship a “chariot,” a detail not in Apollonius, suggests the inability to assimilate this novel invention on the part of those encountering it for the first time. In a similar vein, Catullus alters Apollonius again to bring wondering deities closer to the Argo itself. In depicting the Nereids around the ship (14–15), he conflates Argonautica 4.930–63, where Thetis and the Nereids swim playfully around the Argo as they guide it through the Wandering Rocks, with 1.549–51, where the Nymphs stand on Mt. Pelion and marvel at the first 12 On Catullus’s thematization of the gaze in poem 64 and its equivalents in later Roman wall paintings of Ariadne, see Elsner 2007: 67–109. 13 But on the difficulty of this expression, see Trappes-Lomax (2007: 173), who regards line 18 as interpolated. 14 Thomas 1982: 149–50.

t h e c a p t o r’s i m a g e

44

ship.15 The transposition of Nymphs on Pelion to Nereids in the water again increases the sense of wonder at the divine. Indeed, the admiration experienced even by the Nereids for the handiwork of another divinity leads to their self-exposure (never repeated: 16–18), part of the erotic buildup to the wedding. The initial events of poem 64 are thus made to depend on the receptivity of already received mythical characters to the first spectacle its narrator, similarly receptive and received at once, offers. At this point, the audience may not have enough information confidently to dissociate narrator from poet as the controlling agent behind the poem’s allusive play, but as the expectations triggered by successive allusions are defeated, suspicions begin to arise about the narrator’s reliability. Indeed, this process has already begun: the poem’s first lines recall the dramatic Medea (both its Euripidean and Ennian versions) and the Argonautica in what then swiftly becomes a marriage poem (20: “then Thetis did not spurn marriage with a mortal” [tum Thetis humanos non despexit hymenaeos]).16 The traditional sequel, however, to the Argonautic journey was in fact the tragedy of Medea, while Peleus and Thetis’s marriage usually came before the Argo’s maiden voyage rather than after it.17 We appear to be at the mercy of a Roman Alexandrian (or an Alexandrian Roman) sowing doubt about where his polyvalent narrative is heading. But before anyone can gain a stable footing, he halts altogether and makes his Greek characters themselves into an audience, as though to fold the whole question of audience response back onto his mythic subject matter itself.

Communing with Heroes In the apostrophe of the Heroes beginning on verse 22, the narrator’s receptive stance takes the form of desire for direct access to the Greek past, while his persona receives more definition as an idiosyncratic mirror in which the Hellenophile Roman audience might find itself reflected. The narrator now steps forward in a first-person address to the Heroes in the second person: O Heroes, born in an age very much wished-for, hail, race of the gods! O good offspring of mothers, hail again . . . You, you I will address often with my song. o nimis optato saeclorum tempore nati heroes, salvete, deum genus! o bona matrum 15

Thomas 1982: 158. Cf. Clare 1996: 62–65. Clare 1996: 64–65. 17 Zetzel 1983: 261; Clare 1996: 64. 16

Becoming Ariadne

45

progenies, salvete iter . . . vos ego saepe, meo vos carmine compellabo. (Catullus 64.22–24) Reminiscences of hymnic convention here enhance the effect of “inner sympathy”18 with the heroes, while a sense of personal emotion comes through similarly in the adjective optato (22). In the same gesture, Catullus assimilates the narrator to himself as poet: we are dealing expressly with a producer of “song” (24). This song only just begun, however, its story is now suspended in a moment of tension focusing the audience specifically on the relationship between the present and the past. The narrator’s desire seems to make the past present—he speaks to the heroes themselves—while his wistfulness (perhaps already ironized)19 toward what he perceives as a better time prepares the audience for the fully articulated comparison at poem 64’s conclusion.20 The lines that follow express receptive admiration for Peleus’s good fortune in the erotic domain. Likewise addressed in the second person, Peleus is “so especially blest by a fortunate marriage . . . keystone of Thessaly, to whom Jupiter himself, the father of the gods himself, conceded his love” (25–27). “Did Thetis most beautiful,” the narrator continues, “daughter of Nereus, embrace you?” (28). Here the suggestion is that Jupiter meant to honor Peleus by conceding Thetis (27, concessit), his own love-object (suos . . . amores).21 Thetis’s “embrace” of Peleus, conveyed by the verb tenuit in the next line, places similar emphasis on Peleus’s erotic success, and this is further echoed in Oceanus’s “embrace” of the earth in the final two verses of the apostrophe, where he and Tethys also “concede” (29, concessit again) to take Peleus as their son-in-law. In the past gods mingled with men, the narrator reminds us. His fantasy of communion with the heroes stands in, through contrast, for the mortal connections with the divine whose dissolution he will later explicitly lament. The apostrophe of the heroes in poem 64 trades on the paradoxes of apostrophe as a poetic trope seeming to focus attention on the addressee, but in fact shining a spotlight on the speaking persona. Apostrophe works here, in Jonathan Culler’s words, as “a device which the poetic voice uses to establish with an object a relationship which helps to constitute him. The object is treated as a subject, an I which implies a certain type of you in its turn.”22 So, as he 18

Kroll 1959: 147; on the hymnic resonance see Schmale 2004: 73. Pointing out the ambiguity of nimis optato, which could mean “very much wished-for” or “too much, excessively wished-for,” Janan remarks, “The decision demands a choice between an age of heroes that either was truly heroic or was overrated” (1994: 108). 20 Cf. Schmale 2004: 72–76. 21 Cf. Schmale 2004: 74. In traditional versions of the myth Zeus was compelled by a prophecy to relinquish Thetis to a mortal so as to avoid producing an offspring who would overthrow him. 22 Culler 1977: 63. 19

t h e c a p t o r’s i m a g e

46

appears to seek out a still closer bond with the heroes, the narrator takes on more definition from the perspective of Catullus’s audience. And if apostrophe helps define him, his self-definition turns the work of self-assessment back on the audience. Indeed, the narrator’s desire for the past is effectively triangulated through the audience, since in addition to standing for the poet he is also an audience-figure, and the assumption of a pervasive Roman Hellenophilia informs Catullus’s construction of him. In this connection, one aspect of the narrator’s language again targets the most sophisticated readers of “neoteric” verse: placed near the beginning of poem 64, the apostrophe of the Heroes alludes to the end of the Argonautica (4.1773–5), where Apollonius’s narrator bids the Argonauts farewell. A typically Alexandrian reversal of poetic beginnings and endings, the allusion sets up a system of implicit comparisons between the Medea story and the love affairs of Thetis and Ariadne, while also suggesting the ensuing narrative as the “sequel” to a Greek poem that its narrator has set out to “correct” in Latin.23 Increasingly idiosyncratic, poem 64 is both revision and extension of stories about a past to which its desiring narrator also claims direct access. This complex dynamic of distance and identification will be central to the coming ecphrasis.

A Lover’s Ambivalence toward the Heroic Past With the arrival of the Thessalian wedding-guests in lines 32–33, Catullus introduces the internal spectators who will actually view the Ariadne coverlet. And in their eventual reactions (267–68), Catullus both stages and thematizes receptivity, just as he has done earlier with the narrator’s vision of the Argo, the wondering Nereids, and the apostrophe of the heroes. Through the way he introduces the Thessalians, however, this condition now appears still more problematic than before. When he first mentions them, the narrator does not in fact emphasize the Thessalians’ interest in any of the marvels of Peleus’s court, but, strangely, the degeneration of the farmland they leave behind as they attend the wedding (38–42). On the appointed day, he says (now with the immediacy of the present tense) all of Thessaly throngs to the palace. The halls are filled with a happy crowd bringing gifts and visibly expressing their joy. The Thessalian towns, by contrast, lie abandoned. Cieros, Tempe, Crannon, Larissa—all are emptied as their people converge on Pharsalus. The picture grows still darker. No one, we are told, tends the fields, the necks of the bullocks grow soft, viticulture, cultivation, and the pruning of trees are forgotten, as rust infects the unused ploughs (31– 42). With its ironic echoes of “Golden Age” rhetoric amid an account of decline, 23

See Clare 1996; Zetzel 1983: 261.

Becoming Ariadne

47

the passage has been read with poem 64’s concluding lament over present-day corruption as insistence that the age of heroes is already corrupt: drawn to the luxuries of material splendor, the Thessalian farmers dispense with their traditional values.24 The account of decline itself, however, seems suspicious. The narrator clearly exaggerates the extent of the interruption. This is especially so in light of the fact that the Thessalians are later described, through a prominent simile comparing them to waves in the dawn, as returning home (277, ad se quisque . . . discedebant) immediately after viewing the tapestry (267–77). That they do so “with wandering foot” (277, vago . . . pede) protracts their homeward journey somewhat, but we do not learn by how much. Nor is there indication of this earlier, in the account of agricultural decline itself. The passage is no simple condemnation of the heroic age on moral grounds, but is a further example of the narrator’s inconsistency. Catullus’s original audience might hardly have been surprised by ambivalence over the mythical spectacle to which his narrator feels so powerfully drawn. The negative tone of the passage on agricultural decline (e.g., deseritur . . . linquunt . . . nemo . . . non . . . non . . . non) is jarring after the predominantly wondering tone of the poem’s opening, yet the emotional instability it suggests is a familiar feature of the Catullan lover.25 In poem 51, for example, the lover expresses ambivalent feelings roused by seeing Lesbia in proximity to another man, in an adaptation of Sappho’s similar lines on a desirable girl (Sapph. fr. 31 Lobel and Page). Lesbia, whose very name here makes her stand for Sappho’s Greece, becomes the focus of intense emotional turmoil. The lover’s mute tongue, burning limbs, ringing ears, and darkened sight lead him to the observation that “leisure” (otium) is ruining him, just as it has previously ruined kings and fortunate cities (13–16). In poem 8 (“Wretched Catullus, stop being foolish”) the lover’s vivid memories of happiness with Lesbia compete with his resolution to forget her and threaten to undo his composure, while in poem 11, the insulting image of Lesbia copulating with three hundred men (17–20) cedes suddenly to that of a flower grazed by the plow, a symbol of Catullus’s unrequited love (21–24). So, too, in poem 64 the narratorial pose becomes unstable just as powerfully attractive Greek objects of desire (symbolic, as in 51, of regal otium)26 begin to come into view. The exaggeration of the passage on agricultural decline undercuts the narrator’s very attraction to his principal objects. By now, moreover, it has come to seem characteristic of poem 64 to drop the hint that such notions may be baseless (i.e., the wedding guests may in fact return to farming): this 24

Konstan 1977: 31–36; for the Golden Age echoes, cf. Reckford 1958: 81. The specific knowledge of Catullus’s other poetry possessed by the audience of poem 64 must remain speculative, since exact publication dates are unknown. For a recent discussion of book-divisions and circulation of the poems, see Hutchinson 2003. 26 Munich 2003: 55n20. 25

t h e c a p t o r’s i m a g e

48

only undercuts the narrator’s position further. In a similar way, the poet of the Lesbia poems makes the Catullan lover a figure of pronounced inconsistency while nevertheless elevating this lover’s experience. This is nowhere more true than in the famous paradox “I hate and I love” (85.1, odi et amo). Indeed, at this point in poem 64, odi et amo is beginning to seem like an apt characterization of the narrator’s attitude toward the heroic Greek past itself.

A Roman Villa in Archaic Greece As the poem turns, next, to an account of Peleus’s palace and its decorative objects—including the Ariadne coverlet—the narrator’s posture produces a new kind of temporal disjunction. Roman details now emerge in the midst of archaic Greek opulence. The palace is splendid, shining, wherever it “extends backward” (43, recessit), with gold and silver (44). As we shall see in a moment, this detail reproduces through verbal means the effect of trompe-l’oeil colonnades in Second-Style painting, which lead the viewer into a painted world of ambiguously Greek and Roman design. Likewise, the visual opulence of Peleus’s palace recalls in general that of the Second Style: ivory glistens on chairs, cups gleam on the table, and the whole house “delights” in its regal treasure. The couple’s marriage bed, also worked in ivory, stands in full view, and the coverlet, soon to be described at length, rests upon it (45–49). As often noted, the bed’s placement “in the middle of the palace” (48, sedibus in mediis) corresponds entirely to Roman ritual and is unprecedented in any Greek context. Although it is called a pulvinar (47), a word belonging “to the high epic vocabulary” and denoting the ritual couch of a divinity (appropriate in the case of the goddess Thetis), the adjective genialis in the same line would have reminded Roman audiences of the lectus genialis placed in the atrium of a Roman house at a wedding.27 In fact, the whole description of Peleus’s palace is likely to have recalled the Roman luxury villa with its splendid decorations and furnishings, so as to foreground the narrator’s difficulty in contrasting a negatively weighted Roman present with an idealized heroic past.28 The narrator glorifies, however problematically, the Greek past at the same time as he invites an aesthetic response conditioned by the cultural circumstances known to his Roman audience. And because Roman Hellenophilia also expressed itself through domestic luxury, with Greek fashions and techniques transformed in the process of adaptation to Roman space, the connection is apt.29 A richly allusive and self-conscious visual culture, capable even of irony over its own predilections, is one reason Catullus’s 27

Thomson 1997: 400. Schmale 2004: 89. 29 Cf. Schmale 2004: 89 and the discussion of Second-Style wall painting below. 28

Becoming Ariadne

49

audience could easily identify with the narrator’s wonder over the aesthetic glories of Greece in spite of doubts that the poet has begun to raise about both the Greek past and him.

Late-Republican Aspirations: Beyond the Here and Now in Second-Style Painting and Poem 64 Before turning, therefore, to the art object at the heart of poem 64, it will be useful to consider the art Catullus’s audience actually saw within the opulent domestic interiors of the Roman elite, a place where viewership was explicit. In its ambivalent gestures toward the broader philhellenism of its audience, Catullus’s use of ecphrasis to manipulate his viewing narrator is a way of commenting on a culture of viewing like that suggested by the Second Style. Appreciating this will help both explain Catullus’s poetic choices and locate the poem within late-Republican Hellenism as opposed to the Augustan, Neronian, and Flavian contexts discussed in subsequent chapters. Just as Catullus has his narrator, a product of his poetic art, issue a qualified invitation to enter into an archaic Greek world received through earlier literary treatments, so the anonymous painters of the Second or “Architectural” Style produced artworks that invite the viewer beyond the immediate space of viewing into a world of the imagination, sometimes a world centered in Greek myth and literature. Like Catullus’s poem, the trompe-l’oeil iconography of the Second Style appeals not only by creating an illusion in which the viewer can become immersed but also by eliciting discernment, by raising questions about what is being represented and how it relates to the whole social context of viewing.30 Staging receptivity is here, too, of central importance: the painter’s technique, rather than the reading or performance of an ecphrastic poem, displays receptivity to earlier traditions, while becoming the basis of the viewer’s own receptive response. At some point following the massive influx of plundered Greek art to Rome in the second century bce, and the transportation, along with it, of Greek artists familiar with trompe-l’oeil illusionism already developed in Greece (see the introduction31), well-off Romans adopted a style of domestic wall painting in which a row of supporting columns seems to stand out from the wall surface, which consequently appears located in the background behind it. The earliest Italian example of this technique, characteristic of the so-called Second Pompeian Style, actually comes from Rome: the House of the Griffins 30 31

Cf. Stewart 2008: 49. Pp. 6–7 above. Cf. Clarke 1991: 45 on Greek trompe-l’oeil painting.

t h e c a p t o r’s i m a g e

50

Figure 2.1 Early Second-Style wall painting from the House of the Griffins. Rome, Palatine Hill, 100–80 bce. Koppermann, Neg. D-DAI-Rom 1966.0008.

(100–80 bce) on the Palatine Hill (figure 2.1). This style’s exact relation, at Rome, to the technique known as the First Pompeian Style, in which stucco was molded to resemble blocks of ashlar masonry and painted in bright colors as if it were valuable marble, is still a matter of debate.32 The paintings in the House of the Griffins use trompe-l’oeil instead of molded stucco to suggest masonry, valuable stone, and other architectural features. They indicate new possibilities for connecting the actual domestic space with imaginary realms capable of enhancing the status claims of the house owner. Along with other aspects of domestic decoration, they express the status competition typical of the late-Republican elites. The mature Second Style exploits the space behind the wall in ways immediately suggestive of Catullus’s poetic technique. The objects and architectural forms it depicts convey luxury, opulence, and sophistication, even if their exact cultural reference points often remain a source of scholarly uncertainty. In panels like those in Room 8 of the Villa of the Mysteries, Room 15 of the Villa Oplontis (figure 2.2), or Cubiculum M from the Villa of P. Fannius Synistor at Boscoreale, a colonnade (porticus) recedes into the distance behind other forms

32

See Leach 2004: 63–64.

Becoming Ariadne

51

Figure 2.2 Illusionistic colonnade from Second-Style wall painting in Room 15 of the Villa Oplontis. Ca. 40s bce. Sichtermann, Neg. D-DAI-Rom 1974.2691.

of illusionistic architecture, itself located behind the characteristic trompe-l’oeil column screen. Within these impressive tableaux, objects with diverse associations, including dramatic masks, peacocks, tripods, and exotic stone, can be linked to the traditions of Hellenistic stage painting, the splendor of Hellenistic palaces and public religious sanctuaries, and the luxury of the grandest late-Republican houses in contemporary Italy.33 Theatrical associations are underscored by the presence, sometimes in the same house, of other panels unmistakably reproducing the scaenae frons structure that formed the background in ancient theaters (both Greek and Roman), as in Room 23 of the Villa Oplontis (figure 2.3), the Room of the Masks in the House of Augustus at Rome, or the theatrical

33

For the debate (still unresolved) over the images’ associations, see Leach 2004: 298n140; cf. O’Sullivan 2007: 507–08.

t h e c a p t o r’s i m a g e

52

Figure 2.3 Stage background (scaenae frons) motif from Second-Style wall painting in Room 23 of the Villa Oplontis. Ca. 40s bce. Sichtermann, Neg. D-DAI-Rom 1974.2680.

room of the House of the Cryptoportico at Pompeii.34 And here, too, Catullus’s general aesthetic in poem 64, which joins visual opulence to expressly theatrical traditions, comes to mind.35 Whatever the exact meaning of Second-Style theatrical references, they are of a piece with other signs of visual artifice, simultaneously drawing the viewer in and “boasting of their own unreality,”36 while interacting with other elements to focus attention on the act of viewing itself. As we shall see, poem 64 compares a woven figure of Ariadne, who gives speeches like the tragic Medea, to

34

For comprehensive discussion of the motif in its social context, see Leach 2004: 93–122. So Wiseman 1985: 129. 36 Stewart 2008: 49. 35

Becoming Ariadne

53

a statue of a Bacchante. Similar juxtapositions structure Second-Style art.37 The north wall of Room H in Fannius Synistor’s villa at Boscoreale depicts statues in the distance behind the central figure of Venus, with shuttered pinakes—painted paintings—located on the illusionistic entablature. Pinakes accompany dramatic masks in Room 15 of the Villa Oplontis, and appear similarly above statuesque figures with Dionysian themes in the alcoves of Room 4 in the Villa of the Mysteries.38 A dense concentration of such representational self-reference occurs in a Second-Style fragment from Herculaneum, where statues stand within a painted painting placed next to a mask on an illusionistic ledge.39 More famously, a theatrical mask is displayed prominently in the megalographic frieze of the Room of the Mysteries from the Villa of the Mysteries, one of the most discussed—and yet still most controversial—of all Pompeian artworks (group D: figure 2.4).40 Here the column screen is dispensed with, and the illusionistic space filled with near life-size images of what are apparently celebrants in a Dionysiac religious rite associated somehow with marriage, while in their midst appears Bacchus himself reclining in the lap of a female figure.41 This could in fact be Ariadne, in which case we have a direct correspondence with Catullus’s poem. But if the great uncertainty about the painting’s contents prevents such close thematic comparison between the Mysteries frieze and poem 64, the painting’s inviting visual rhetoric nevertheless suggests connections. The divine and human figures interacting in a space beyond that of the room draw the viewer into their midst, just as the Catullan narrator invites his audience into a time and place where gods and mortals interact.42 And like the viewer, the painted figures are spectators: seeming to look in our direction and at each other, they guide our perceptions of the scene and make us aware of how we look at them.43 The Catullan narrator, and with him Nereids and wedding guests, focuses his audience’s attention both on particular objects and on their own critical gaze.

37

With what follows cf. Vitr. 7.5.2, which mentions stage settings together with paintings of statues as features of art in this period. 38 See Longfellow 2000: 119–123. 39 National Archeological Museum of Naples, no. 9244, reproduced, with commentary and bibliography, in Nava, Paris, and Friggeri 2007: 84. 40 Recent discussion and bibliography can be found in Stewart 2004: 78–79. See further Gazda 2000. 41 Ariadne, Venus, and Semele have all been suggested. See Longfellow 2000. 42 Martin 1992: 155 expresses the idea of invitation somewhat differently: “Having abolished the wall, Roman artists invited gods, goddesses, and heroes into the homes of their patrons, mingling mortals and immortals in the same way Catullus does in poem 64.” In spite of differences, I am indebted to his valuable remarks on poem 64 and the Second Style (154–56). 43 Clarke 1991: 100 fig. 33 diagrams the interplay of gazes between the viewer and the figures and among the figures themselves.

54

t h e c a p t o r’s i m a g e

Figure 2.4 Megalographic Second-Style frieze (group D) from the Room of the Mysteries, Villa of the Mysteries, Pompeii. 50s bce. Photo Credit: Scala/Art Resource, NY.

An imaginary world with origins in Greek myth—and here, expressly, literature—also appears in what may be the sole surviving example of a subgenre within the Second Style: the Odyssey Landscapes, a series of panels from the Esquiline Hill at Rome.44 The allusive manipulation of epic narrative forms a compelling link between these paintings and Catullus’s “epyllion.”45 Behind an illusionistic portico of square pillars appear scenes derived from Homer’s Odyssey, some of the very scenes, it so happens, presented by an internal narrator, Odysseus, within Homer’s poem: the meeting and violent confrontation with the Laestrygonians, the visit to Circe’s palace, the summoning of the dead from the underworld (nekyia), and the encounter with the Sirens. In spite of their Homeric reference point, the paintings’ incorporation of landscape imposes its own constraints on the narrative, leaving the viewer often in need of interpretive 44 Vitr. 7.5.2 speaks of paintings that depicted “battles of Troy or the wanderings of Odysseus through the landscape”: see O’Sullivan 2007 for recent discussion. 45 My initial thoughts about the epic connection between poem 64 and the Odyssey Landscapes were sparked by a paper delivered by Agnès Rouveret at a seminar on Hellenistic aesthetics held at the University of Cincinnati on September 29, 2007.

Becoming Ariadne

55

work to discern its correspondence with Homer’s text.46 In certain cases the paintings even include details not in Homer: Tityus, Sisyphus, and the Danaids, for example, appear in the painted underworld but not in the Odyssey.47 Spectator figures again guide and make self-conscious the viewer’s perceptions. In panel 2, which depicts the arrival of Odysseus’s men at the Laestrygonian palace, a seated figure, usually identified as the personification of the local spring, watches peacefully, imparting a deceptive mood of tranquility that enhances the jarring effect of the Laestrygonians’ subsequent attack (panel 3).48 In the bottom right-hand corner of panel 5, three personifications of the shoreline calmly watch Odysseus’s ship escape from the Laestrygonians (figure 2.5). Their attitude suggests momentary deliverance, but the figure on the right points in the direction of Circe’s palace in panel 6 (figure 2.6), and so calls attention to the progress of the narrative toward further danger.49 Increasing the pull into the mythical world are the rich effects of spatial depth, most evident in the mysteriously receding architecture of Circe’s palace, a replica, in fact, of Second-Style porticus panels.50 Here multiple planes of architecture and landscape, from the initial column screen before a round aedicula and colonnaded hemicycle back to the mountains emerging over the top of the palace, draw the viewer’s eye into the depths of the mythological scene. At the same time, the trompe-l’oeil pillars that divide the visual narrative foreground painterly artifice and act “as a mediating device, an interpretive lens,” eliciting the full critical powers of the Roman viewer.51 Verbal representation has its part to play here as well. The Greek titles inscribed in small white letters within the paintings are icons of the images’ dual effect, rewarding close scrutiny with the pleasure of knowing who acts within this hauntingly beautiful world and yet activating the scene as reception of Homer’s text. Second-Style art operates through a complex system of visual allusions requiring interpretation at the same time that it aims at illusion, a technique analogous to Catullus’s overall approach in poem 64.52 If Catullus’s learned audience took sophisticated pleasure in perceiving the allusive artifice surrounding his use of ecphrasis, the audience of the Second Style could appreciate the house owner’s “power to command visual transformation”53 as an aspect of his social status. 46

See Leach 1988: 42–56, esp. 43. Cf. O’Sullivan 2007: 521–22. 48 O’Sullivan 2007: 511. On the larger phenomenon of “supernumeraries” who observe the main action in Roman wall painting, see Elsner 2007: 175 with bibliography. 49 Cf. O’Sullivan 2007: 513. 50 O’Sullivan 2007: 514. 51 O’Sullivan 2007: 525–26. 52 Leach 2004: 88–89, esp. 88: “the fictive prospects of these paintings are far too tolerant of inconsistencies and incoherences to produce genuinely convincing illusion.” Cf. Elsner 1995: 74–85; Wallace-Hadrill 1988: 69–77. 53 Leach 2004: 89. 47

Figure 2.5 Panels 4 and 5 of the Odyssey Landscapes. Vatican Museums, 50–40 bce. © Vatican Museums.

Becoming Ariadne

57

Figure 2.6 Panel 6 of the Odyssey Landscapes. Vatican Museums, 50–40 bce. © Vatican Museums.

But we may in addition read the paintings for clues as to why Catullus’s audience might be interested in a self-subverting narrator, one whose intense identification with Greek myth and opulence produces irresolvable contradictions. The Second Style’s “tendency to fantisise” expresses the aspirations of a late-Republican elite whose luxuries were in fact limited, not necessarily the ambitions of the wealthiest classes, who could afford to produce the visual grandeur the paintings only imitate.54 In transforming the confined spaces of dining rooms and bedrooms, the Second Style necessarily called attention to the constraints of the dwellings it adorned, even if those limitations could sometimes be recuperated as active choices—conspicuous modesty—on an owner’s part.55 It is no surprise that Catullus hints at the inadequacies of his fictional narrator when we assume that an implicit awareness of representational inadequacy in fictive decorative contexts was an everyday feature of his audience’s social experience. Catullus’s poetry in general shows acute consciousness of social jockeying and ambitious self-representation among both men and women of various standings 54 55

Stewart 2008: 50. Cf. Leach 2004: 89. Stewart 2008: 50; Leach 2004: 89.

t h e c a p t o r’s i m a g e

58

within the Roman elite. His acquaintances included members of prominent families and grasping arrivistes, politicians and orators as well as poets, and any of these might have exploited Hellenophile display as a status marker, a means of social self-advancement. Ambivalently drawn to the Greek mythical past as though to a love-object, poem 64’s narrator exposes the imperfections in his own cultivated fantasy. Whatever any individual Roman felt for the fantastic worlds of the Second Style, the effect of its representational means was the same. Romans of the late Republic found their identity, a sense of self as it extended to the social, cultural, and political realms, both affirmed and subtly undermined by the receptivity to Greek culture they embraced in ecphrastic literature and visual art. The chinks in their Hellenophile self-presentation, the contradictions in their self-asserting Romanness, manifested themselves in their visual culture, and it is just this vulnerability that Catullus dramatizes, if sympathetically, in the Ariadne ecphrasis, to which we will now turn.

Becoming Ariadne: The Ecphrasis Though the culmination of the narrator’s performance as receptive viewer, the Ariadne ecphrasis underscores still further his ambiguous placement—and that of the audience seeing through him—as both “insider” and “outsider” to the mythical past. The emotionality of the narrator’s rhetoric continues to suggest an intense bond, just as the narrative as a whole foregrounds the Greek past as a series of interrelated “frames” through which to see the Roman present. And yet the narrator digresses on background events not likely to appear in the imagery itself, as he blurs the boundaries between a convincing description, the whole context of stories surrounding Ariadne, and his own reflections.56 It is often unclear, moreover, whose perspective counts most: that of the wedding guests, the external audience, or the narrator. Such effects begin at once. The coverlet is, perhaps, “worked variously with figures of ancient men,” but through a transferred epithet it is literally “worked variously with ancient figures of men” (50, priscis hominum variata figuris),57 while the first image described includes Theseus’s departure “with his swift fleet” (53, celeri cum classe). A glaring anachronism emerges: if the Argo is really the first ship, there is no way that Theseus, even assuming him, improbably, to have built and launched his fleet in the interval between the Argo’s maiden voyage and the 56

For scholarly attempts to visualize the coverlet, see, e.g., Wiseman 1985: 179; Martin 1992: 155–56; cf. Gardner 2007: 168. On the confusion of Ariadne’s voice and the narrator’s, see esp. DeBrohun 1999. 57 On the Apollonian allusion and the effect of the hypallage, see O’Hara 2007: 36n4.

Becoming Ariadne

59

wedding, can be “ancient” from the perspective of the guests, whose viewpoint nevertheless seems evoked by the present-tense verb at line 51, indicat (“[the coverlet] depicts . . .”).58 Making things still more difficult is the sophisticated audience-member’s cultural knowledge: by Catullus’s day mythographers had in general accepted the priority of the Argonautic voyage to Theseus’s Cretan journey, and yet Apollonius reversed this chronology in the Argonautica (4.430–34), as did Callimachus in the Hecale, another widely popular Alexandrian poem to which the ecphrasis itself may allude.59 As he begins the description in lines 50–51, then, the narrator is made to stumble over a particular point of Alexandrian contrarianism in the explication of a notionally Greek art object for a contemporary Roman audience, but, simultaneously, to perform a fantasy of what it was like to view the object in its distant mythological setting. Recall the analogous effects of the Second Style: Catullus affords his audience the pleasure of being both Thessalian wedding-guest and learned Roman connoisseur at once. Poem 64 brilliantly develops this twofold perspective as the narrator starts to bring Ariadne’s image to life. A description of her external appearance accompanies that of her internal condition so as to suggest the visual and emotional impact on a wedding-guest viewing the coverlet directly. Indeed, as a highly engaged viewer herself, Ariadne both tropes and resembles the wedding-guests. Depicted “with wondrous art” (51, mira . . . arte) the woven Ariadne seems to perceive her own situation: “gazing out” (52, prospectans) from the shore of Naxos, she “looks at” (53, tuetur) Theseus’s departing fleet, just as the female figures on the shoreline of the Odyssey Landscapes guide the viewer’s attention toward the escaping ships of Odysseus at the same time as they point—quite literally— elsewhere within the visual narrative. (Later Pompeian paintings of Ariadne herself in fact emphasize her role as spectator.60) Strong emotions come to the fore: bearing in her heart “wild passions” (54, indomitos . . . furores) Ariadne does not yet believe that she sees what she sees (55). Only at this moment awakened from a “deceptive” slumber, “she perceives herself, miserable, on the deserted sand” (57, desertam in sola miseram se cernat harena). For his part, Theseus, “forgetful” (58, immemor), sails off, “leaving his unfulfilled promises to the windy gale” (59, irrita ventosae linquens promissa procellae), these last lines suggesting 58

On this much-studied problem, see recently O’Hara 2007: 34–41, with a helpful summary of earlier approaches at 37–38. Cf., e.g., Weber 1983; Gaisser 1995: 592; Schmale 2004: 149–52; Gardner 2007: 170–71. 59 Catul. 64.217, reddite . . . nuper mihi (“recently returned to me”) perhaps recalls Call. fr. 234 Pfeiffer “you have come back unexpectedly” (παρὲκ νόον εἰλήλουθας). In each case, Aegeus describes Theseus’s return to Athens. See Weber 1983: 265; Gaisser 1995: 604–5; Schmale 2004: 150–51; O’Hara 2007: 39. For another possible allusion to the Hecale in Catul. 64, see Hunter 2006: 98–100. 60 Elsner 2007: 91–109.

t h e c a p t o r’s i m a g e

60

Ariadne’s feelings of abandonment through recollection of Theseus’s unfulfilled vows, while providing a glimpse of his mental state as well. The same details, however, simultaneously encourage a distanced, critical regard, engaged further by what follows, in the audience of this particular poem. Ariadne, upon waking, sees Theseus speeding away on his ship; so, earlier, Thetis had first seen Peleus on board the speeding Argo. As a spectator figure, Ariadne helps focus attention on the imagined coverlet; but the Nereids had earlier done the same for the poem’s opening spectacle. The wedding guests, one might suppose, come to the festivities sensitized to the notion of erotic passion and immediately feel it emanating from the image of Ariadne. The external audience, however, has heard directly of the passion kindled in Peleus for Thetis in verse 19, where Peleus “is said to have been inflamed with love” (incensus fertur amore). Here, as we have already remarked, the “Alexandrian footnote” “is said” strikes a distancing note. Moreover, virtually as soon as the narrator has invited his audience to see Ariadne as a living, impassioned woman, he reminds them that she is an artwork—if only to intimate, through raising the question of artistic representation once again, the ways in which this woven Ariadne will destroy all expectations for a silent, static object. In verse 61, Ariadne is compared with a statue of a Bacchante: standing fixed in one place, she looks out, ah, looks out like the stone image of a Bacchante, and seethes with great waves of distress. saxea ut effigies bacchantis, prospicit, eheu, prospicit et magnis curarum fluctuat undis. (Catullus 64.61–62) Such statues, represented for us by the famous Vatican Ariadne (a Roman adaptation of a Hellenistic original), would have been familiar to Catullus’s audience.61 And as we have seen, Second-Style art relished the insertion of painted statues within paintings: a Bacchante in an alcove of Room 4 in the Villa of the Mysteries may be a “statue” of this kind.62 Images of Bacchantes might in addition have been something of a Hellenistic literary touchstone for contrasting the fixity of the artistic image with the mobility of the emotions it depicted, as in an epigram by Antipater of Sidon (AP 9.603).63 But none of the rustic farmers of 61

See esp. Klingner 1956: 31–43; cf. Schmale 2004: 152. Longfellow 2000: 121–22. 63 Cf. Syndikus 1990: 141–42n168; Laird 1993: 20–21; Gardner 2007: 164. The learned audience may also hear an allusion to Euripides’ Hecuba (557–61); see Gardner 2007: 166–67. But cf. Gardner 2007: 174 on the significant contrast between the static image of the statue and the moving crowd of Bacchantes who accompany Bacchus. 62

Becoming Ariadne

61

archaic Thessaly could be thought to make such comparisons. Of course, Ariadne is doubly like a statue in standing still and actually being an artistic image (triply if we think of her also as a product of poetic art).64 In the Bacchic reference, furthermore, the audience hears a witty allusion to Ariadne’s subsequent union with Bacchus, included, in fact, at the end of the ecphrasis. The simile, like the lines that precede it, creates a tenuous sense of artistic unity. This double effect continues.65 In the vivid lines 63–67, Ariadne appears denuded, her garments having fallen off her and lying in the water around her feet. At the mention of her “milky breasts” (65), the audience recalls the bare-breasted Nereids, a sight unavailable to the internal audience of wedding guests. All this is a result, again, of Ariadne’s inner turmoil: heedless of herself, she cares only for Theseus “in her whole soul, her whole mind” (70, toto animo, tota . . . mente). The narrator’s address of Theseus at lines 69–70 recalls his earlier apostrophe to the heroes. Similarly, the background story of Theseus’s journey to Crete contains details linking it to the main narrative of the ecphrasis, and these suggest both stories as projections of the narrator’s idiosyncratic imagination.66 Theseus’s self-sacrificing act of bravery in going to Crete “for his dear Athens” (81, pro caris . . . Athenis) recalls his implicit designation as a “hero” in line 51, a notable inclusion, since in the image itself he seems to appear only as a deserter. Emphasis on the “desirous gaze” (86, cupido . . . lumine) and “burning glance” (91–92, flagrantia . . . lumina) Ariadne directs toward Theseus when she first sees him is reminiscent of the likewise impassioned gaze she directs at his departing ship when the audience first sees her at the opening of the ecphrasis. The same trope focuses attention on Theseus as both hero and deserter. The protective “soft embrace” (89) of Ariadne’s mother in Crete contrasts with Ariadne’s vulnerability and exposure on the open shore of Naxos, while the apostrophe to Cupid and Venus in verses 94–102 returns the audience to Venus Erucina in line 72. As a further apostrophe recalling the address of the heroes, the Cupid and Venus passage deserves special comment. The narrator’s rhetoric as he describes Ariadne’s billowing emotions suggests especially deep identification with the heroine herself. The art object overwhelms its “viewer” all but completely with a sentiment that is at basis religious, however colored by literary conven-

64

Cf. Schmale 2004: 154. Much has been said about the correspondences between the internal narrative of poem 64’s central ecphrasis and its external frames, both the descriptive passage that immediately precedes it (52–75) and the Peleus and Thetis story. My purpose in recalling these connections is, again, to emphasize how the poet both foregrounds and calls into question a certain kind of receptive response. See esp. Putnam 1961; Gardner 2007: 165–66. For temporal “circularity” as a feature within Ariadne’s lament, see Gardner 2007: 170–75. 66 Cf. Schmale 2004: 188 on Ariadne’s speech as a “Projektion des Erzählers.” 65

62

t h e c a p t o r’s i m a g e

tion. Indeed, cult titles suggestive of prayer (96) punctuate these verses, while anaphora, assonance, and alliteration build to a crescendo: O holy boy distressingly stirring up furious longing with merciless heart, you who mix delights with men’s cares, and you, lady, who rule Golgi and leafy Idalium, in what billows you tossed the girl, her mind inflamed, sighing often for the golden-haired guest! What great fears she bore in her weakening heart! By how much more than with the brightness of gold she often paled, when Theseus, desiring to contend against the fierce monster, sought either death or the rewards of glorious action! heu misere exagitans immiti corde furores sancte puer, curis hominum qui gaudia misces, quaeque regis Golgos quaeque Idalium frondosum, qualibus incensam iactastis mente puellam fluctibus, in flavo saepe hospite suspirantem! quantos illa tulit languenti corde timores! quanto saepe magis fulgore expalluit auri, cum saevum cupiens contra contendere monstrum aut mortem appeteret Theseus aut praemia laudis! (Catullus 64.94–102) This close focus, however, on Ariadne’s amorous furores actually looks back (which is to say, forward) to the indomitos . . . furores (54) she feels when abandoned. Similarly, the epic heroism of Theseus in killing the Minotaur (complete with a Homeric-style simile at 105–11 comparing the vanquished monster to a felled tree) is reminiscent of his earlier designation as a hero, while his departure from the labyrinth (112–15) corresponds to his unheroic departure from Naxos at the very opening of the ecphrasis (53). Just as in Second-Style painting, any sense of “reality” within the representation winds up calling attention to the fact of representation. The more we are encouraged to identify with an image, the more its status as image comes into view. Commencing the last part of this flashback, Catullus wittily has the narrator draw attention to its digressive nature, and simultaneously to the question of his own identity once more: sed quid ego a primo digressus carmine plura/ commemorem . . .? (116–17, “But why should I digress from the first part of my song and relate more . . .?”). The remark, a typically Alexandrian self-interruption, challenges the audience to ask itself about the relevance of what it is hearing to the image it has been encouraged to see. The inclusion, moreover, of the first-person pronoun ego seems an almost impudent reminder of how much the narrator’s persona has been placed in doubt thus far. “Just who is this ‘I’?” we continue to wonder, but to little avail. The passage then concludes with details

Becoming Ariadne

63

of the background myth leading up to the very moment depicted in the image. We are now taken from the moment of Ariadne’s departure from her family to her arrival on Naxos and the onset of the sleep from which she wakes to witness Theseus’s ships receding into the distance.

“Catullus” or Not? “Just who is this ‘I’”? Nothing in poem 64 answers this question for us definitively, so that the sense of self it creates for its narrator is left highly indeterminate: seemingly Roman but heavily influenced by Greece, and in doubtful control of ecphrasis itself. But one way that Catullus, the controlling poet, teases his audience on this score is to draw subtle parallels with “Catullus,” the lover’s persona of the Lesbia poems (also famously thrown into question in poem 16, with its obscene insistence that a poet should never be identified with the lascivious voice speaking in the poetry). With Ariadne’s lament (132–201), “Catullus” the lover becomes still more suggestive than before as a reference point for the narrator’s ambivalent intensity toward his objects of desire.67 The violent swing of Ariadne’s feelings for Theseus—first love, then hatred—point to the lover’s emotional instability, and so create a bridge between him and the narrator for an audience familiar with both. Ariadne, moreover, even as she emerges as a Bacchante avant la lettre, already suggests Catullus as lover, her surging passions like those described in poem 50, where “Catullus” is “wild with passion” (11, indomitus furore; cf. 64.54, indomitos . . . furores) over his friend and fellow poet, Licinius, or 38, where he expresses his intense anger over Cornificius’s failure to reciprocate his amores. From the moment the narrator begins to speak as Ariadne, s/he uses the language of broken fides central to “Catullus’s” agony: “Thus, false one, on the desolate shore have you abandoned me, carried off, false Theseus, from ancestral altars? Departing thus, the gods’ will slighted, all forgetful, ah, do you bear home your cursed perjuries? Could nothing sway the intention of your cruel mind? Was no mercy yours, that your pitiless heart might wish to pity me? But not these promises did you make me at one time with sweet voice, not these did you bid me, wretched, hope for, but happy marriage, a desired wedding, all of which the airy winds tear to pieces unfulfilled.” “sicine me patriis avectam, perfide, ab aris, perfide, deserto liquisti in litore, Theseu? 67

In what follows I have relied especially on Putnam 1961, although my emphasis falls on narratorial personae rather than the poet’s biography and underlying emotions. Cf. Gardner 2007.

t h e c a p t o r’s i m a g e

64

sicine discedens neglecto numine divum, immemor a! devota domum periuria portas? nullane res potuit crudelis flectere mentis consilium? tibi nulla fuit clementia praesto, immite ut nostri vellet miserescere pectus? at non haec quondam blanda promissa dedisti voce mihi, non haec miserae sperare iubebas, sed conubia laeta, sed optatos hymenaeos, quae cuncta aerii discerpunt irrita venti.” (Catullus 64.132–42) Here, Theseus is perfidus (twice) and immemor. Ariadne bewails her abandonment on a deserted shore and Theseus’s broken promises, which, she says, have been carried off by the wind.68 All these sentiments also appear in poem 30, in which “Catullus” complains of abandonment by Alfenus, and similar language returns in poem 76, in which he laments the fides that Lesbia has broken between herself and him. The emphasis on Theseus’s promissa echoes the lover’s reminder to Lesbia in poem 72 (1–2) as well as poem 70’s assertion that a woman’s words to her lover ought to be written “in wind and swift water” (4). Just as Ariadne bewails her hoped-for marriage with Theseus, the lover complains of a promised marriage (70.1–2). Further on, Ariadne asks whether Theseus was born from a lioness, the sea, Syrtis, Scylla, or Charybdis, that he has made such poor returns for her devotion (154–57). The same topos governs poem 60, which likewise posits birth from a lioness or Scylla as the cause for the addressee’s neglect of the speaker’s desperate pleas. In Ariadne’s concluding curse on the sea-faring Theseus (192–201), we can recognize, mutatis mutandis, the persona of poem 11, who, imagining a long journey, curses Lesbia and laments the destruction of his amor. Within the Catullan universe, fides is associated with presence, and this makes Ariadne’s lament an allegory of the narrator’s own travails as he tries to identify with an absent, distant past of his own devising. Poem 64 creates a world where “faithfulness,” whether in love or representation, is nowhere to be found. Further heightening the sense of these particular idiosyncrasies governing the narrator’s view are the correspondences that link Ariadne to Aegeus as victims of a loved one’s forgetfulness. Failure of memory here both motivates broken faith and, like it, allegorizes the loss of stable reference points in Catullus’s poetic labyrinth. Like Ariadne when we first see her, Aegeus gazes anxiously at the sea as he waits for Theseus’s return (241–42).69 A series of words used to describe Ariadne’s situation, including the verb conspexit, the adjectives 68

For more on the “language of abandonment” in poem 64, e.g., the forms of relinquo at 117, 123, 200, 213, see Gardner 2007: 169. 69 Putnam 1961: 185.

Becoming Ariadne

65

anxius and assiduus, and the noun luctus recur in the account of his suffering. Forgetfulness, denoted by the adjective immemor, characterizes Theseus’s attitude toward both Aegeus and Ariadne, and this emphasis represents a significant departure from received versions of his myth, since Ariadne’s curse is made the cause of Aegeus’s suicide.70 Ariadne herself makes the connection explicit in her closing request to the Eumenides: but through the same [forgetful] mind in which he left me alone let him pollute with death both himself and his family. sed quali solam Theseus me mente reliquit, tali mente, deae, funestet seque suosque. (Catullus 64.200–201) Aegeus, in spite of the careful instructions we hear him administer (225–37) will commit suicide when he views the dark sails that Theseus forgets to change to white (241–45). The narrator’s equation of Theseus’s broken familial bond to Aegeus with his broken erotic bond to Ariadne may seem surprising, but in poem 72 the lover makes this same connection in asserting he loved Lesbia “as a father loves his sons and sons-in-law” (4, pater ut gnatos diligit et generos). In poem 68 the Catullan lover connects eros and family ties through a simile comparing Laodamia’s love for Protesilaus to the love felt by an aged grandparent for a late-born grandson who can be his direct heir (119–24). Laodamia, like Ariadne, is robbed of marriage through her beloved’s departure overseas, and this loss further corresponds, within the poem, to Lesbia’s unfaithfulness. Aspects of Aegeus’s situation also suggest “Catullus’s” travails over the loss of his brother, which is itself given an erotic undercurrent by being linked, in poem 68, to Laodamia’s loss of Protesilaus and, through her, to the lover’s agonizing love for Lesbia. The departure of “Catullus’s” brother for “Troy” (perhaps Roman Asia Minor, where “Catullus” tells us he himself served in the cohort of a praetor [10.9–13]) echoes that of Protesilaus, the first Greek warrior to die in the Trojan War. The general emotional cast of Aegeus’s farewell to his son, described as “more delightful than a long life” (64.215, longa iucundior . . . vita) parallels the lament at 65.10–11, where the brother is “more beloved than life” (10, vita . . . amabilior). Aegeus’s pained words to the departing Theseus at 64.218–19, “since my fortune and your burning valor are tearing you away from me, unwilling” (quandoquidem fortuna mea ac tua fervida virtus/eripit invito mihi te) suggest still more closely “Catullus’s” expression of dismay at his brother’s death in poem 101, 70

Schmale 2004: 196.

t h e c a p t o r’s i m a g e

66

since fortune carried off your own self from me, O wretched brother undeservedly taken away from me . . . quandoquidem fortuna mihi tete abstulit ipsum, heu miser indigne frater adempte mihi. . . (Catullus 101.5–6) The phrase quandoquidem fortuna even appears in the same metrical position in each passage. The heroic past disappoints the narrator in an emotionally complex way analogous to the way Lesbia disappoints the lover: as elusive objects, both activate the pain inherent in the larger realm of emotional attachment signified by the family. But again, it is never clear that the narrator of poem 64 is “Catullus” (let alone Catullus). Indeed, refractions of the lover in still other Catullan personae come into play through the figures of Ariadne and Aegeus, as though to highlight the impossibility of making such an equation. Consider the echoes here of Callimachus’s Lock of Berenice as translated by Catullus in poem 66. Catullus’s decision to translate this poem, which describes the anxiety of Berenice II over the departure of her husband and half-cousin Ptolemy III on a military expedition to Syria, has often been seen to reflect his sorrows over both Lesbia and his brother.71 Because Ptolemy’s father, Ptolemy II Philadelphus, married his own sister Arsinoe II, his son, too, came to be styled as his wife’s “brother,” and so eros and the bonds of family, commingled again, find simultaneous expression in Catullan verse: And did you [Berenice], deserted, grieve not your destitute bed, but the tearful separation from your dear brother? et tu non orbum luxti deserta cubile, sed fratris cari flebile discidium? (Catullus 66.21–22) The overall similarities to Ariadne’s situation are self-evident, but this is an actual translation, where the borrowed voice is manifestly not that of “Catullus” or Catullus “himself.” And given the original Lock’s own dazzling series of displacements and collocations, whereby the Greek features of the composition are infused with aspects of classical Egyptian myth and cult,72 a mise-en-abîme 71

Thomson 1997: 448. Selden 1998: 326–54. For more on the place of Egypt in Alexandrian poetics (in this case in a possible model for Catullus 64), see below on Theocritus’s Idylls 15. 72

Becoming Ariadne

67

opens up, linking the Hellenistic queen with the Roman brother and lover, as well as with the various personae of poem 64, and making it hard to distinguish between “originals” and “copies” within this network. Was Catullus simply drawn to Berenice’s story because it reminded him of his own, or did Catullus style “Catullus” in part on Berenice? Both, of course, could be true at once, but this opens still further the possibility that the narrator of poem 64 is, in part, another such imitation. In a different fashion, Attis, the self-castrated priest of Cybele whose lament occupies much of poem 63, bears similarities to Ariadne in outward circumstances and inner turmoil, and so therefore to all of the other personae just mentioned. Like Ariadne, Attis awakens on a beach to feelings of sorrow, and both gaze out at the sea as full cognizance of their circumstances hits home. Both bewail separation from homeland and family, Ariadne having left hers to go with Theseus, Attis having traveled from Greece to Phrygia in service to Cybele. In each case, “the excessive furor of devotion”73 motivates a journey, while a pose of abject servitude becomes a cause for regret: Ariadne says she would have gone with Theseus as his servant (64.160–64), while Attis serves Cybele, who is his “mistress” (63.13, dominae).74 Attis’s self-inflicted violence, often read as a symbol of the lover’s being “unmanned” by his subservience to Lesbia, alerts us to the darker side of Catullan erotic passion as portrayed in poem 64 and elsewhere. Yet we cannot be certain that the narrators of poems 63 and 64 are identical, any more than that they are the same as “Catullus” or Catullus. Even within poem 64 itself, the narrator’s identification with his alter ego, Ariadne, is far from straightforward. Allusions to Medea contribute to the darker shading of Ariadne’s character and render the narrator’s intense identification with her still more deeply problematic, even self-subverting. A series of correspondences, for example, link the opening of Ariadne’s lament to Medea’s great speech to Jason at Euripides’ Medea 465–519. Ariadne’s insistence on broken faith, unfulfilled promises, and dashed hopes for a happy marriage recall similar emphases in Medea’s rhetoric. Ariadne specifically reminds Theseus that she saved his life, just as Medea reminds Jason at Med. 476. Ariadne wonders aloud where she will turn; Medea poses the same question to Jason (502), with each, similarly, recalling her departure from home and country. Further, Ariadne’s rhetorical question about Theseus’s birth, which, as we have seen, employs a trope found in Catullus’s poem 60, also echoes Medea’s angry words later in Euripides’ play (Med. 1342–43). Once again Second-Style painting comes to mind for the way it resituates the lived experience of its viewers within the illusionism of stage drama, while “boasting” of its own unreality.

73 74

Putnam 1961: 169. Putnam 1961: 177–78.

t h e c a p t o r’s i m a g e

68

Like her Euripidean predecessor, furthermore, Apollonius’s Medea wonders angrily where she is to go when it appears that the Argonauts are abandoning her and recalls her loneliness and departure from home (Arg. 4.355–78). Her words here, spoken under threat of abandonment, make an even closer parallel with Catullus 64.177–81.75 The Apollonian Medea invokes the Erinyes against Jason at the end of her speech (Arg. 385–86), while Catullus’s Ariadne, at the end of hers (Catul. 64. 192–97), invokes the Eumenides against Theseus. And Apollonius’s Medea, again anticipating Catullus’s Ariadne, hurls charges of forgetfulness (Arg. 4.356), broken oaths, and unfulfilled promises (4.356–59; cf. Catul. 64.135, 139). “One gradually perceives that [Ariadne’s] mounting fury is a form of madness,”76 nowhere more directly than when she describes herself as “helpless, burning, blind with maddened fury” (Catul. 64.197, inops, ardens, amenti caeca furore). Like Medea, she hovers between being a “woman to be pitied” and “a woman out of control, a woman to be feared.”77 But crystalizing her threat is the fact that Callimachus’s Hecale, a poem whose possible influence on poem 64 we have already remarked, included a version of the myth in which Medea tried to poison Theseus after their arrival in Athens. The assimilation of Ariadne’s anger to Medea’s murderous impulses complicate in the extreme the narrator’s seeming attachment to the erotic domain over the heroic ethos.78 Here again, the narrator falls into inconsistency, just as the expression of his bond with the past seems to reach its most intense pitch. The final scene on the Ariadne tapestry, which depicts Bacchus’s arrival with his retinue of Bacchantes, enhances the portrait of a maddened Ariadne even as it seems to provide resolution to her story. Forming a ring-composition with her earlier depiction as a sculptured Bacchante, the passage ought to provide a natural transition back to the main narrative, since joyous marriage with Bacchus was the normal conclusion of her myth.79 But in fact the narrator does not mention the marriage, and his account focuses instead on the sights and sounds of the Bacchantes’ frenzied worship. The Bacchantes “were raving” (254, furebant), we are told, “with frenzied mind” (lymphata mente). Throwing back their heads, they cry “euhoe” and shake their thyrsoi. Some of them hurl the dismembered limbs of a sacrificial victim in the ritual of sparagmos, while others wrap themselves in serpents, and still others perform “mysteries.” The ecphrasis concludes with their cacophonous music and an especially unsettling reference to the “raucous booms” and “horrible song” emerging from their horns and “barbarous” pipes. 75

Clare 1996: 75. Pavlock 1990: 122. 77 Clare 1996: 76. 78 Thus a reading of the poem “whereby Ariadne is granted moral superiority over Theseus through allusion to a woman who . . . is a would-be murderess of Theseus, clearly subverts itself ” (Clare 1996: 76). 79 See Syndikus 1990: 165–66 on Nonn. D. 47.419–71. 76

Becoming Ariadne

69

The incongruity of this with, for example, a passage such as Nonnus 47.419–71, which emphasizes Ariadne’s joy at her divine marriage, prompts Schmale to suggest that the narrator actually reads “against the grain” of what the audience is to imagine was actually depicted on the tapestry; that is, a “marriage” appropriate to the setting.80 Be this so or not, the scene surely does represent a culminating moment of idiosyncrasy, its extended description of music, complete with onomatopoeia, seeming to refute the insistence of Argonautica 1.763–77 on the silence of the images on Jason’s cloak.81

Seeing Like Alexandrians Catullus’s narrator knows things that a Hellenophile Roman would know and makes “mistakes” that such a figure might also make, and so prompts both a sympathetic and an ironic response from other Romans. But even here Catullus adapts the Alexandrians’ own syncretistic techniques, adding further layers of complexity to the narrator’s cultural affiliations. Again, poem 64’s Alexandrianism comes through not simply in the imitation of themes and formal patterns, but in the staging of receptivity, the interaction between viewer and object concentrated by ecphrasis. Structurally and thematically, poem 64’s ecphrasis has often been compared with that of Moschus’s Europa (37–62) as well as the Cloak of Jason from Apollonius’s Argonautica (1.721–67), the latter poem, as we have seen, a primary model for Catullus’s as a whole.82 Moschus’s “epyllion” frames an image of Io within the story of Europa and connects the two myths especially through erotic motifs: Zeus’s pursuit of Io and her journey, as a heifer, through the sea anticipate his abduction, now in bovine form himself, of Europa and their sea voyage. Jason’s cloak, with its combination of epic and erotic themes in a series of mythical images, helps Apollonius call attention to the innovative portrayal of his hero in similar terms. Both images are presented as wondrous, yet neither ecphrasis thematizes the expression of receptive wonder or explores its implications for the characterization of the viewer in the same way as the Catullan narrator. For this, a closer parallel exists in Theocritus’s Idylls 15, which recounts the visit of two Syracusan women to the palace of Ptolemy Philadelphus during the festival of Adonis and their expressions of amazement over images they find there. A number of scholars have proposed Idylls 15 as among Catullus’s direct models, but even if there is no conscious intertext here (the grounds for positing one 80

Schmale 2004: 205–07. Laird 1993: 24. 82 For discussion see Schmale 2004: 117–20, 124–26. Thomas 1983b: 112–13 suggests the influence of Callimachus’s lost Victoria Berenices. 81

t h e c a p t o r’s i m a g e

70

are somewhat uncertain83), Theocritus’s poem reveals Hellenistic preoccupations anticipating Catullus’s technique. The concluding passages of Idylls 15 provide the closest existing precedent for wonder-filled ecphrasis of a tapestry located within a royal palace during the celebration of the union between a mortal and a divinity.84 The Syracusan housewives, Gorgo and Praxinoa, their words filled with sensuous appreciation, describe woven images including one of the young Adonis lying on a silver couch (80–86). Using conventional aesthetic terms and some that seem distinctly Callimachean,85 they marvel at the embroiderers’ skill and the images’ verisimilitude. Praxinoa’s language in particular is decidedly eroticized, lingering on the early growth of beard on Adonis’s cheeks (85) and culminating, with swelling emotion, in the epithet “thrice-beloved” and the clause “who is beloved even in Acheron” (86). Idylls 15 ends with a further ecphrasis in the form of a hymn to Adonis sung by an “Argive woman’s daughter” and describing an image of his “marriage” to Aphrodite.86 In the midst of her description, however, Praxinoa is interrupted by a stranger who chides the two Syracusans for their talkativeness and mocks their Doric accents (87–88). She in turn defends Gorgo and herself as descendents of Corinthian Greeks (Syracuse being a colony of Corinth), with whom they share their dialect (89–95), before Gorgo directs her attention to the singer (96–99). The meaning of the Theocritean encounter has been much debated. Clear enough, however, is its intense concern with measuring both difference and sameness within the Greek-speaking population of Alexandria. Gorgo and Praxinoa situate this group in relation to its Greek-speaking ruler, Ptolemy, as well as in opposition to the local Egyptians, who appear in Idylls 15 as nothing

83

The apparent echo of Idylls 15.100, the first line of the Theocritean ecphrastic song, “you who love Golgi and Idalium” (ἅ Γολγώς τε καὶ Ἰδάλιον ἐφίλησας) in Catul. 64.96, “and you who rule Golgi and leafy Idalium” (quaeque regis Golgos quaeque Idalium frondosum) very well may, as Kroll (1959: 157) and Clausen (1982: 201) posit, indicate that Catullus had Theocritus’s verse in mind, but since Catullus’s text is both emended and suggests a formulaic cult title due caution is warranted. I thank Benjamin Acosta-Hughes for pointing this out to me. On the emendation, see Thomson (1997: 406), who also accepts a Theocritean allusion. Sebesta (1994: 35) argues that Idylls 15 is Catullus’s principal model for the setting of the Ariadne ecphrasis. Cf. Wheeler 1964: 126. 84 The tradition of mime suggests further, if less direct, Hellenistic parallels. From a scholiast on Idylls 15 (15.arg.7–8 Wendel) we learn that Theocritus drew on a mime of Sophron in which women may have observed the Isthmian festival (see Skinner 2001: 204 with n10). Herodas’s Mimiamb 4 portrays two women marveling over artworks in a sanctuary of Asclepius. See Skinner 2001: 216–21 for an illuminating discussion of this work’s relation both to Idylls 15 and to the ecphrastic epigrams of Nossis. 85 Goldhill 1994: 218. 86 It is worth noting that the passage corresponds in placement to the incipient marriage between Ariadne and Bacchus depicted at the end of the Catullan ecphrasis.

Becoming Ariadne

71

more than petty thieves (47–50).87 Gender accompanies language as an ironic aspect of the women’s response, since their enthusiastic reactions to the royal celebration of Adonis’s divine “marriage” stand out against their earlier complaints about unhappy marriages to inadequate husbands.88 But their remarks are genuinely appreciative, pointing to their sympathies with the proceedings, whether or not we ascribe sophistication to their assessment or the subsequent ecphrastic song. Whatever the idyll’s exact significance for its original audience, its identification of the Doric dialect and Corinthian Greeks as both same and other to the regal court of Ptolemy anticipates Catullus’s ambiguous situation of his Latin-speaking narrator vis-à-vis the Greek world of Peleus’s wedding. In both works, cultural identity ironically signaled by language, as well as the influence of gender and eros on perceptions of a given political formation (cf. poem 64’s “political” conclusion treating civil war, etc.) are made to reflect on viewers’ responses to visual art. In both, viewers’ wondering receptivity is both affirmed and undercut by a learned poet. The viewing female characters of Idylls 15, like the narrator of Catullus 64, stand in for—and can be read as manipulated versions of—both the poet and his trained audience.89 They, like their Catullan successor, press an audience to rethink not only its own critical judgments but also its sense of self.90 Simon Goldhill helpfully remarks the “ironic twist to the language of poetics” here, “to have two housewives adopt ‘Callimachean’ terminology.” “As the women evaluate the depiction of Adonis,” Goldhill points out, “the work of evaluation—of the women’s response—is turned back on the reader (as the reader formulates a response to the poetic work of art).”91 The same can be said of the way poem 64, a text also inflected with Alexandrian aesthetics, turns the work of self-evaluation back on Catullus’s audience. A specific allusion to Idylls 15, but in its absence a more general reminiscence of Hellenistic techniques that this poem embodies, would help Catullus underscore the ambiguities of his narrator’s position. Theocritus’s Syracusan women occupy a similarly ambiguous place with respect to Ptolemy’s court as the Catullan narrator does in relation to the imaginary court of Peleus.

Conclusion: An Ambiguous Critique of the Roman Present The remainder of poem 64, which reverts to the frame narrative of the wedding of Peleus and Thetis, enhances the ambiguities of the narrator’s position primarily 87

Stephens 2003: 242–43. For more on Theocritus’s play with Greek language and identity in Idylls 15, see Griffiths 1979: 82–86; Hunter 1996: 116–23; Reed 2000; Fantuzzi and Hunter 2004: 371–75. 88 Goldhill 1994: 218. 89 See Griffiths 1979: 84; Skinner 2001: 211–16. 90 Goldhill 1994: 216–23. For more on this important theme, see Hunter 1996: 117. 91 Goldhill 1994: 218.

t h e c a p t o r’s i m a g e

72

through the Song of the Fates, whose prophecy for the couple’s future takes the place of a traditional epithalamium.92 In the interests of space, however, let us jump to the poem’s conclusion. A perennial problem for critics, the “political” ending was once seen as an unnecessary addendum,93 but from our perspective the logic of its inclusion emerges in a new light. Sensitized, especially through its ecphrasis, to poem 64 as a self-conscious staging of inconsistent receptivity, we can only read its conclusion as the final scene in this production, rather than mere parody or a simple statement of a moral position, even an ambivalent one. The discrepancies, that is, concentrated in the poem’s ecphrasis lurk here as well: the narrator elevates the heroic age over the present as a time when pietas (386) governed human action and gods mingled with men (384–96), in spite of the fact that he seems to have rejected aspects of the heroic ethos, particularly the way it conflicts with erotic fidelity, and has even gone so far as to suggest that the heroic age is corrupt in its love of material splendor leading to agricultural ruin. Catullus, again, stages a performance in which such contradictions emerge. Once the earth, he says, was “stained with unspeakable crime,” justice was forgotten: brother shed brother’s blood, sons no longer grieved for dead parents, fathers wished for their sons’ deaths so as to be able to enjoy their young brides, and mothers slept with sons (397–404). “All things speakable and unspeakable commingled in wicked fury” (405, omnia fanda nefanda malo permixta furore) banished divine justice, and the gods no longer deigned to visit the earth (406– 408). Contemporary tropes for civil war and social degeneration, particularly the collapse of religion, locate the narrator in Rome of the mid-first century bce.94 But even here Catullus declines to pin things down in an unambiguous fashion. If an audience recollecting the violence of the Sullan period, the Social War, and the suppression of the Catilinarian conspiracy would be likely to hear its own predicament described, Catullus still does not say “Roman.” This, too, in another passage whose most famous antecedent comes from a Greek poem: the account of the Ages of Man in Hesiod’s Works and Days 109–201.95 We struggle to position the story securely within a network of ethnic affiliations, both Roman and Greek. Such irony must have been all the more intense for Catullus’s original audience, to whom the Greekness within Romanness was a lived phenomenon in addition to a literary one. Thus, as poem 64 ends the narrator retreats from his unsettled fantasy of a bond with an objectified Greek past yet settles on no coherent substitute. Yet on many levels we might expect such a conclusion. “Apostrophic poems,” remarks 92

Cf. O’Hara 2007: 47–54. For a helpful overview of scholarship, see Schmale 2004: 275–81. 94 See Schmale 2004: 280 on, e.g., Cic. N.D. 2.8; Sal. Cat. 12.3–4. For the trope of civil war as fratricide, see esp. Lucr. 3.72. 95 See Schmale 2004: 262–64. 93

Becoming Ariadne

73

Culler, “display in various ways awareness of the difficulties of what they purport to seek,” often concluding “in withdrawals and questions.”96 In an analogous fashion, poem 64 as ecphrasis must ultimately confront the impossibility of embodying a material image in its own verbal form, a task that Murray Krieger associates with nostalgic yearning for static presence thwarted by the contingency of a text’s existence in linear time.97 The dissolution of stable identity between hero, community, and the contents of ecphrasis that for Page duBois characterizes the trope in its Roman instantiations is unquestionably operative—indeed, thematized—here.98 From the point of view of the Roman present, Catullus implies, the world of archaic myth must appear to be a better time. Any attempt, however, to assert unqualified control over images of this world through ecphrasis—to dominate them by means of the descriptive text—begins on the same uncertain footing that it ends. By contrast, the powerful, lingering influence of Greek culture on a Roman sense of self often takes center stage. Poem 64, in short, does not emerge from or produce a stable impression of image or text, past or present, Greekness or Romanness, but offers its audience an ironic and yet sympathetic perspective on a kind of cultivated play blurring these categories as they intersect with others: objecthood and subjecthood, love and heroism, the personal and the political. The very attitude that produces the Catullan narrator’s strong, idiosyncratic identification with the Greek past reveals inconsistencies similar to those of “Catullus,” the embattled lover and pushes the audience to reflect on its own attitudes. Although he pronounces on decline, the narrator’s position must remain morally and politically ambiguous, since it is not clear whether, in deploring aspects of Roman society, he has perceived clearly the relation between past and present, Greek and Roman. Finding expression for the most intense aspects of a lover’s travails in his pose as Ariadne, he cannot help bringing to bear a contemporary Roman perspective upon his representation of her and other mythical characters. The pain—and the pleasure—he seems to feel in telling their stories stems from a divided loyalty, an interplay of competing cultures, that the poem leaves unresolved. Poem 64 thus undercuts the “standard” modern view of ecphrasis through a massive display of the ancient trope’s cultural implications. We will now see how ecphrastic poems from the generation following Catullus, though composed on a far smaller scale, achieve similar effects by yet another means: the theme of civil discord itself.

96

Culler 1977: 64. Krieger 1992. 98 Cf. duBois 1982. 97

3

The Challenge of Rustic Art Ideals of Order in Vergil, Eclogues 3 and Horace, Satires 1.8

Ecphrasis, W. J. T. Mitchell writes, is stationed between the object described and a listening subject. Its implicit “social structure” is thus triangular rather than binary, and its textual image is often offered as a gift to the listener or reader. “Relations of self and other, text and image,” are “triply inscribed” in a scene that receives its “fullest poetic representation,” Mitchell hypothesizes, in Theocritean pastoral with its rustic singing competitions.1 The ecphrastic poetry treated in this chapter, whether referring directly or indirectly to the pastoral tradition, bears out Mitchell’s emphasis on the receptive audience for ecphrasis, as we shall see. But Vergil’s Eclogues 3 and Horace’s Satires 1.8, texts written during Rome’s tumultuous “triumviral” period of civil war and upheaval (43– 28 bce), reveal, too, the limitations of such a perspective, insofar as they both assume and construct an audience that is far from unitary. “Triply inscribed” must here become “inscribed manifold times,” since the poems insist on the changing, uncertain nature of their audiences’ composition, knowledge, and political allegiances.2 And just as a unified audience for Vergil’s and Horace’s ecphrases is unthinkable, so, too, is a unified attitude of the texts themselves toward the images they create: a posture that could be reduced simply to the self-asserting text’s attempts at control or domination, as Mitchell and others would have it. Latin literature, when read in context, continues to challenge such influential assumptions.

1

Mitchell 1994: 164. For his view of pastoral, Mitchell draws in part here on unpublished remarks of Joshua Scodel. 2 Mitchell is not alone, of course, in his desire to posit a unified audience for the literary artifact; one thinks, for example, of Wolfgang Iser’s “implied reader,” a unified figure who comes to seem, on closer inspection, merely a feature of the text. For an introduction to Iser’s thought within the context of reader-response criticism, see Tompkins 1980; for Iser’s implied reader as “imminent in the form of the text itself,” see Morris 2003: 122.

74

The Chal l e ng e o f R usti c A r t

75

The beechwood cups of Vergil’s poem and Horace’s crude Priapus statue use ecphrasis to scrutinize, assess, and play with the idea of Hellenic cultivation as a potential remedy for Roman social discord. Vergil’s cups, wagered by a pair of quarrelling herdsmen, illustrate Greek cultural figures whose ordering of the cosmos and the natural world signifies a hoped-for return to social order in the face of acknowledged threats. Horace’s Priapus activates a mix of Greek influences, including Aristophanic social critique, Socratic irony, Callimachean literary refinement, and Epicurean ideals, to reject civil war and those who would precipitate its resurgence. Both poems, furthermore, pit their small-scale Hellenistic aesthetics against the conventions of Greco-Roman epic, a genre vitiated, in these circumstances, by its associations with martial bloodshed. Vergil and Horace thus express a newly unsettled sense of Roman identity, and an awareness of suffering beyond the confines of elite privilege, by recalling the damage wrought by fresh civil wars from the 40s to the early 30s bce. Their rustic art bears the potential to make the Roman world a more unified place, to reconfigure Roman identity for the better—if only an ideally receptive audience can be found. Thus while this poetry holds out hope of social order through the Hellenic cultural references that pervade it, the poetry frames aesthetic control itself as doubly problematic. On the one hand, the ecphrastic text here confronts the possibility that the image’s Hellenic traits might have a transformative effect on the Roman subject, seen as belonging to a historically specific group that includes both author and audience. On the other hand, through their overt self-implication in the rough-and-tumble of literary patronage, both poems evoke the social divisiveness that they simultaneously decry and, nevertheless, serve. Such paradoxes lie at the heart of the poems and help explain their ambivalent engagement with Roman philhellenism. Cultural ambiguities similar to those that earlier chapters have remarked in Naevius, Plautus, Terence, and Catullus surface again here, but in different social and political circumstances, more urgent configurations of power, pleasure, and authority centered on Greek culture and its place in Roman life. As before, art history will help ground these features of the poetry within the actual viewing experience of contemporary audiences: the circumstances of real visual display. And because each poem envisages a specific patron figure as among its ultimate addressees—in Vergil’s case, C. Asinius Pollio, and in Horace’s, C. Maecenas—it will be helpful to focus on these two men as a way of understanding both the political history and the wider Roman philhellenism of the triumviral period, the rich but unsettled cultural atmosphere to which Vergil and Horace respond.3 Pollio and Maecenas brought the worlds of 3

For the term triumviral period used to describe the years 43–28 bce (with reference especially to literary production), see Osgood 2006: 4–5.

t h e c a p t o r’s i m a g e

76

Roman art and literature together in the cultural activity they fostered, and it is no coincidence that they also wind up among the imagined audiences for poetic ecphrasis at this time. But again, neither of them is the whole audience for the poem in question. A full account of ecphrasis’s “social structure” in this case demands that we consider other audiences: some central, some peripheral, but all potentially interested in being “in the know” about the Greek culture that Vergil and Horace handle so skillfully. This kind of cultural competition for the benefit of individual patrons now emerges clearly beside the more impersonal way that Greek and Roman culture compete through the staging of ecphrasis, and so signals a further development of my larger argument. Vergil and Horace, that is, make us still more keenly aware than before of how ecphrasis implies an author’s standing within the Roman social hierarchy.

I. Hellenic Marvels in a Landscape: Vergil’s Beechwood Cups C. Asinius Pollio and Vergil’s Eclogues At Eclogues 3.84–89, Vergil has his herdsmen refer to his own contemporary, C. Asinius Pollio, in a clear intrusion of Roman reality upon the imagined pastoral world of the Eclogues. The cups that these herdsmen wager, the function of these objects in Vergil’s poem, and Pollio’s own role as internal viewer and listener will preoccupy us in what follows. But as a way of setting the stage for Vergil’s ecphrasis—and appreciating it as a staging, complete with “actors” and audience—let us focus first on Pollio himself and the Roman society of his day, an essential background for understanding the Vergilian passage in cultural terms. Pollio’s brilliant career placed him at the very center of Roman politics, culture, and, more darkly, civil war.4 After becoming an associate of Julius Caesar, Pollio crossed the Rubicon with him in 49 bce and served as one of his commanders at the Battle of Pharsalus in 48 bce, Caesar’s watershed defeat of his great rival, Cn. Pompeius Magnus. After serving with Caesar again in Africa (46 bce), Pollio commanded troops against Pompey’s renegade son, Sextus, in Spain, and was thus occupied during the year of Caesar’s assassination (44 bce). Pollio’s decision to become Mark Antony’s associate might have spelled his eventual demise on the political scene, but a shrewd posture of alternating allegiance and independence kept his fortunes alive. His nonintervention at Perusia, where Lucius Antonius, Antony’s brother, lay besieged by Octavian’s 4

On Pollio, see André 1949; Bosworth 1972; Nisbet and Hubbard 1978 ad Hor. Carm. 2.1; Osgood 2006 passim.

The Chal l e ng e o f R usti c A r t

77

troops, is a key example of this approach. Indeed, in spite of losing his father-in-law to the bloody proscriptions enacted by the Second Triumvirate of Octavian, Antony, and M. Aemilius Lepidus, Pollio worked as Antony’s superintendent in Gallia Transpadana where, readers of the Eclogues have long speculated, he may have restored the farm that Vergil supposedly lost—and so have provided a real-life experience informing that of the fictional herdsman Tityrus in Eclogues 1.5 In response to hostilities between Octavian and Antony, Pollio brokered, in 40 bce, the peace of Brundisium between them, the short-lived agreement resulting in Antony’s marriage to Octavian’s sister, Octavia. It is apparently this union, and the hope of its progeny, to which Vergil alludes in Eclogues 4, a poem lauding Pollio’s consulship of the same year as a new Golden Age. Pollio celebrated a triumph in October of 39 bce for his victories in Illyria (perhaps commemorated at Eclogues 8.7), before retiring from politics in favor of cultural pursuits. His earlier bond with Antony prompted him to stay neutral again during the Battle of Actium in 31 bce, Octavian’s decisive victory: Pollio is said to have declared that he would stand apart from the quarrel and be the “prize” awarded to the victor.6 Pollio’s wit while still only a “boy” had been praised by Catullus (12.8–9). His sorrow over the bloodshed of the civil wars, his strong desire for peace and liberty, and his eagerness to return to cultural pleasures shared among friends emerges clearly from his surviving correspondence with Cicero.7 In a letter written from his Spanish campaign of 43 bce he insists, “My own nature and pursuits . . . pull me toward a desire for peace and freedom” (Cic. Fam. 10.31.2, natura . . . mea et studia trahunt me ad pacis et libertatis cupiditatem). Pollio goes on to describe himself as “most desirous [cupidissimus] for peace” (10.31.5) and to express envy over Cicero’s discussions with a cultivated friend (10.31.6), probably the poet C. Cornelius Gallus (prominent in Eclogues 6 and 10), to whom another of Pollio’s letters refers by name (Cic. Fam. 10.32.5). Still another letter in June of 43 bce expresses the deepest regret over the bloodshed at the Battle of Mutina and particular distress over the general “desolation of Italy” (10.33.1, vastitatem Italiae) evident not only at Mutina but also in the slaughter at Forum Gallorum and the sack of Parma (10.33.4). Like so many other Romans of his time, Pollio was profoundly affected by civil war’s atrocities, a legacy of violence that leaves traces, too, in every major Latin author of the period. And like a number of other cultivated peers, Pollio sought solace in the redemptive pleasures of literature and art. 5 See Osgood 2006: 112–13, 125–26. The identification of Vergil and Tityrus goes back at least to Servius (ad Buc. 1.1). 6 Vell. 2.86.3 7 In what follows, I have relied especially on Osgood 2006: 55, 194, 253–55, 296. For skepticism over Pollio’s reputation as a republican, see Bosworth 1972. He would come to be seen as one of Cicero’s detractors (see André 1949: 93–98).

t h e c a p t o r’s i m a g e

78

Pollio’s direct experience of civil war, proscription, and the land redistribution that accompanied them would have made him acutely sensitive to Vergil’s disguised recollections of such events in Eclogues 1 and 9.8 But Eclogues 3 assumes his participation in a different milieu, that of the cultural patronage to which he would later devote himself more fully. Himself an orator and author of (now lost) poetry, drama, criticism, and an influential history of the recent civil wars, as well as the first Roman to institutionalize poetic recitation for assembled audiences, Pollio used the wealth gained in Illyria to restore the Hall of Liberty near the Forum, and in it established Rome’s first public library.9 Julius Caesar, before his assassination, had planned such a public benefaction after witnessing the royal library at Alexandria.10 But Pollio’s project, not subject to the suspicions of regalism that doomed Caesar, could go further in democratizing the library as an institution, and is thus an important precedent for Octavian/ Augustus’s use of this and similar elements in his cultural program.11 Pollio’s library put his Roman philhellenism on the most prominent display, for it was really two: one containing Latin works, the other Greek.12 Pollio showed the same Hellenophile predilections in the impressive art gallery he also housed in the Hall of Liberty. The idea for a public art gallery had likewise been anticipated— Pompey had established a celebrated one in the portico adjoining his theater—but Pollio’s was grander still, and his collection of Greek art treasures probably more dazzling than any other Roman’s of the first century bce.13 Pliny the Elder lists its most important pieces: sculptures by Scopas, Praxiteles, and other Hellenistic masters, works highlighting erotic and Dionysian themes.14 Two artists represented in the collection, Arcesilaus and Stephanus, were Pollio’s contemporaries, and the latter, a resident of Rome, may have worked by his commission.15 Some have identified the “Farnese Bull,” a massive sculptural group depicting the punishment of Dirce and attributed to the Rhodian artists Apollonius and Tauriscus, as the only surviving masterpiece from Pollio’s gallery.16 If so, it is a rare example of a preserved Hellenistic art object Vergil himself is sure to have known. 8

For recent discussion, see Osgood 2006: 108–34. For Pollio’s poetic fragments, see Courtney 1993: 254–56; Hollis 2007: 215–18. 10 Osgood 2006: 252. 11 I thank David Potter for pointing this out to me. On Octavian’s library in the temple of Palatine Apollo, see below, pp. 109–110. 12 Cf. Zanker 1988: 70 for Pollio’s library as expressing “a radically changed public attitude toward Greek culture in Rome.” 13 Cf. Pollitt 1986: 163; Osgood 2006: 252–53. On Pompeius’s gallery, see recently Evans 2009 with bibliography. 14 Plin. Nat. 36. 23–25, 33–34; on the collection cf. Pollitt 1978: 170; Stewart 1990: 308; Isager 1991: 163–67; Osgood 2006, 253. 15 See Osgood 2006: 253 citing Plin. Nat. 36.33: Arcesilaus also created the cult statue in the Temple of Venus Genetrix in the Forum of Caesar (Plin. Nat. 35.156). 16 Osgood 2006: 253; for discussion see Ridgway 2000: 273–77. 9

The Chal l e ng e o f R usti c A r t

79

Vergil’s adaptation of Theocritus’s Idylls in particular, with their emphasis on the enjoyment of song in a tranquil, rustic setting, corresponds at once to Pollio’s love of culture, pacifism, and concern for the well being of rural Italy evident in his correspondence with Cicero. Pollio’s disappointment over the wars’ interruption of his own cultural pursuits lends special poignancy to the threatened cessation of pastoral song that haunts the Eclogues both here and elsewhere. The inclusion of art objects, moreover, in Eclogues 3—Vergil’s only ecphrastic eclogue in our sense of the term—is especially appropriate in light of the art-loving Pollio’s presence as addressee, though his gallery itself had not yet been built. Most would date the Eclogues’ publication to 39 bce, the year after Pollio’s consulship.17 Whether or not he should be regarded as its sole patron (the first poem, as is well known, praises a young man [42] generally assumed to be Octavian, who thus becomes the ostensible dedicatee),18 Vergil’s serious play in Eclogues 3 with the notion of Hellenophile cultivation, and the subtle, elusive way he links it both to darker themes of conflict and to the contemporary Roman context, take on sharper focus as a production meant in part for Pollio’s eyes. The questions Vergil raises about Greek culture’s value in such circumstances are ones we can imagine Pollio and others who shared his tastes and experience asking themselves as well.

Signs of Cultivation: Vergil’s Learned Rustics If Pollio illustrates elite Roman Hellenophilia in the period, the Eclogues themselves embody a notoriously complex reception of Greek literature and culture. They also, as in Eclogues 3, thematize and display receptivity to Greek culture, a condition both claimed by Vergil and ascribed to the fictional and nonfictional characters within the poetry. Vergil links this entire dynamic to the question of identity in the personae of his herdsmen: far more explicitly than in the case of poem 64’s narrator, their identity, and that of their world slide between Greek and Roman, to an extent rivaling even that of Plautus’s hybrid dramatic realm nearly two centuries before.19 Vergil’s rustics are Greek-named, Latin-speaking figures inhabiting a Theocritean landscape, though one identified with the non-Theocritean Arcadia in Eclogues 10 and never outright with Sicily, the setting of Idylls 1.20 In Eclogues 1 and 9 they resemble the inhabitants of rural Italy, while at other times, as in Eclogues 3, they address themselves to Vergil’s urbane Roman contemporaries, seem to possess a Greco-Roman cultural sophistication 17 See Osgood 2006: 109n8 for a convincing dismissal of 35 bce, favored by Bowersock 1971 and Clausen 1994: xxii. 18 But cf. Osgood 2006: 109n8. 19 Plautine comedy is in fact an influence on the collection: see Currie 1976; Karakasis 2011: 87–124. 20 See Clausen 1994: 289–90.

80

t h e c a p t o r’s i m a g e

far beyond their lowly status, and even uncannily recall (Roman) Alexandrian poets in their deployment of arcane literary knowledge in song. Thus their role in Eclogues 3 as internal “viewers” of the beechwood cups that they wager (36–47) already frames these objects in ambiguous terms where Roman identity is concerned. To what extent, Vergil’s audience is encouraged to wonder, do these objects, like the herdsmen who describe them, straddle the line between a Greek fantasy world and the Eclogues’ Roman shadow world? In fact the herdsmen’s ecphrases focus attention directly on the idea of Roman receptivity to Alexandrian Greek culture, perhaps in a more concentrated fashion than any other passage in the Eclogues. The cups described by Menalcas depict the only identifiable Alexandrian cultural figure, the astronomer Conon (40), named anywhere in the collection.21 Conon, moreover, a Samian who worked at the court of Ptolemy III Euergetes, had the special distinction of having described the Lock of Berenice, the subject of a portion of Callimachus’s Aetia translated by Catullus, and one line of Menalcas’s ecphrasis, “he who marked out with his pointer the whole vault of heaven for the races of men” (41, descripsit radio totum qui gentibus orbem) recalls Callimachus’s description of Conon as “he who sees the whole of heaven in his charts and how the stars move” (fr. 110.1 Pfeiffer, πάντα τὸν ἐν γραμμαῖσιν ἰδὼν ὅρον ᾗ τε φέρονται; cf. Cat. 66.1, “he who discerned all the stars of the great universe” [omnia qui magni dispexit lumina mundi]).22 Conon’s very presence as an image on Menalcas’s cup thus signals and embodies both Catullus’s adaptation of Callimachus and Vergil’s “neoteric” leanings in his own adaptation of Theocritus. But such intense concentration on receptivity in the ecphrasis (more elaborate even than this, as we shall see) focuses a concern strikingly prevalent in the rest of the poem.23 Menalcas’s doubt over the name of the other astronomer on his cups (40–42), while having elicited much scholarly debate over this figure’s true identity, has also sparked recognition that Alexandrian learning is itself at issue here: Menalcas’s hesitation over the “unseen” verbal image prompts a learned audience to think about the completeness of its own knowledge (a tip-off seems to lie in the fact that the unnamed astronomer is the one actually indicated 21 His is only one of two nonmythological Greek names in the Eclogues, the other being the unidentified Biainor (see Lipka 2001: 173–75). 22 Catullus’s translation of the Coma has, by this point in Eclogues 3, already been brought to mind by the allusion to Cat. 66.47, “What will locks do, when such things yield to iron?” (quid facient crines, cum ferro talia cedant?), at Eclogues 3.16, “What will masters do, when thieves dare such things?” (quid domini faciant, audent cum talia fures?). See Cassio 1973; Clausen 1994: 96, 102; Hubbard 1998: 72; Lipka 2001: 82, 93; Karakasis 2011: 90–91. Note Lipka 2001: 136 on the Roman colloquial color of the last-quoted lines in Catullus and Vergil. 23 Cf. esp. Barchiesi 1997: 275: “The interpretation of the ornament of the cups in Eclogue 3 is the subject of a question in the text, and, like the riddles at the end of the poem, the correct reading of the images presupposes learning and interpretation, setting a challenge to the reader.”

The Chal l e ng e o f R usti c A r t

81

by the Callimachean line describing Conon).24 What figure, Eudoxus, Aratus, Archimedes, or some other, makes an appropriate pair with Conon, and why? Although both herdsmen stake cups by “the divine Alcimedon,” each seems more interested in a heifer as a prize; and their humorous lack of appreciation becomes a further benchmark against which the audience is invited to measure its own position. The pair of riddles, moreover, with which the poem ends (104–07), a feature for which there is no exact parallel in Theocritus, likewise puts front and center the question of the learned audience’s knowledge, for Alexandrian sophistication in particular may be what is needed to supply their answers.25 That Vergil is said to have described the first riddle as a “trap” (crux) for his interpreters suggests his own self-consciousness about the scrutiny of knowledge, rather than simply the provision of answers, as part of the point here, too.26 The enigmatic third eclogue, then, is itself a riddle about the appreciation of Theocritean pastoral beauty and order in the disrupted world of triumviral Rome. As he has done already in Eclogues 1, Vergil presses his audience to ask whether such a vision is appropriate to the times and subtly suggests how it must to be altered to suit present circumstances. But with these questions he poses others still more pointed: who among his audience (Pollio, for instance) will correctly perceive the Eclogues’ neo-Theocritean poetic vision? And might Vergil’s ambitious re-ordering of Theocritus, so perceived, lead to greater order in the same way that the cultural figures on the cups, astronomers describing celestial motion to the benefit of reaper and ploughman alike (40–42), and Orpheus leading the woods with song (46), produce order on earth and in the cosmos?27 Can philhellenic poetry help bring peace? Eclogues 3 itself sows doubt. The cups, described by the herdsmen and then seemingly lost sight of both by them and by their chosen arbiter, Palaemon, who nevertheless appears to appreciate pastoral beauty far more keenly than they do, raise a hope of order that receives no firm confirmation.28 While the poem alludes to no specific political 24

See Clausen 1994: 102 for bibliography on the unnamed astronomer. For the focus on audience reaction cf. Henderson 1998: 221: “If your Menalcas really forgets the name, he comes a cropper, tripping over his own vanity and in his fall suturing and re-establishing the uncrossable divide between herdsmen and urban(e) culture. Perhaps Virgil, and Virgil’s educated readers, will find it no trouble to supply the name that Menalcas does not recall; if not, this could be because his clues are inconclusive, as gauche as he is, too clumsy by half. Or that could be a problem for those among us who do not already know who it is that belongs in a pair with Conon . . .” etc. 25 On the riddles see Segal 1967: 301–02; Coleman 1977: 125–27; Dix 1995; Lipka 2001: 93–94. On their link to the ecphrases see Barchiesi 1997: 275 and Karanika 2006: 112, and to the beginning of the poem Clausen 1994: 116. 26 Servius ad loc. reports that Vergil made this remark to the grammarian Asconius Pedianus. 27 With this interpretation of the cups’ iconography cf. Segal 1967: 290. 28 Boyle 1975: 194: “The works of art [on the cups] . . . fail to affect the discord manifest in the herdsmen’s world. From the artist’s values and perceptions, embodied in his created work, the herdsmen learn nothing.” Cf. Henderson 1998: 221–22; Schultz 2003.

82

t h e c a p t o r’s i m a g e

event, its internal address to Pollio locates Vergil himself in a context of rivalry over poetic patronage, and its imagery links this to threats to the pastoral landscape like those accorded a political dimension in Eclogues 1 and 9. Menalcas’s invective against the rival poets Bavius and Maevius (90–91), while tempered by ascription to a fictional character, nevertheless recalls a cultural practice with strong ties to social discord, making Vergil himself at least an indirect participant in a milieu corrosive of the very order toward which his cups gesture. One can hope for poetry’s beneficial influence, but such an outcome is, alas, uncertain.

The Cups of Alcimedon: Seeing and Not Seeing Vergilian Pastoral Like the insults they trade, the prizes they stake in the song contest ambiguously define Menalcas and Damoetas—together with Vergil’s own poetry—both with and against Theocritean models, while calling attention to problems in the herdsmen’s relation to the Greek legacy itself. The cups can perhaps stand for the poem (and to this extent what I have called the “standard” view of ecphrasis would be affirmed), but the ambiguous cultural import of this correspondence far exceeds in importance that of any simple generic hierarchy it might imply. Damoetas first wagers a heifer that can be milked twice a day. With allusion to Idylls 8.15–16 (where a Menalcas also speaks), Menalcas says he would not dare stake an animal, since his father and “unjust” stepmother keep careful count of the goats and kids (32–34). Instead he offers “what you yourself will admit is much greater” (35, id quod multo tute ipse fatebere maius), a pair of beechwood cups made by “the divine Alcimedon” (37), objects he does not in fact carry with him but keeps “stored up” elsewhere (43). Menalcas describes his cups before Damoetas counters with his own pair of (absent) cups, which he claims to be by the same artist: M. But—what you yourself will admit to be much greater (since you’re inclined to such madness)—I will stake beechwood cups, the engraved work of the divine Alcimedon, on which the soft vine added by the easy lathe clothes the berry-clusters strewn on pale ivy. In the middle are two images: Conon and—who was the other, he who marked out with his pointer the whole vault of heaven for the races of men, the times which the reaper, which the bent ploughman should observe? Nor yet have I moved them to my lips, but keep them stored up. D. For us, too, the same Alcimedon made two cups, and encircled their handles with soft acanthus, and in the middle placed Orpheus and the woods following him. Nor yet have I moved them to my lips, but keep them stored up. If you look at the heifer, though, you’ll have no reason to praise the cups.

The Chal l e ng e o f R usti c A r t

83

M. verum, id quod multo tute ipse fatebere maius (insanire libet quoniam tibi), pocula ponam fagina, caelatum divini opus Alcimedontis, lenta quibus torno facili superaddita vitis diffusos hedera vestit pallente corymbos. in medio duo signa, Conon et—quis fuit alter, descripsit radio totum qui gentibus orbem tempora quae messor, quae curvus arator haberet? necdum illis labra admovi, sed condita servo. D. Et nobis idem Alcimedon duo pocula fecit et molli circum est ansas amplexus acantho, Orpheaque in medio posuit silvasque sequentis; necdum illis labra admovi, sed condita servo. si ad vitulam spectas, nihil est quod pocula laudes. (Vergil, Eclogues 3.35–48) An audience steeped in Theocritus would have instantly recognized Vergil’s transformative engagement with this source.29 In Idylls 1, a goatherd offers Thyrsis a cup decorated, like Menalcas’s (Ecl. 3.39), with a pattern of entwined ivy at its lip (Theoc. 1.29–31) and, like Damoetas’s (Ecl. 3.45), with acanthus over the whole (Theoc. 1.55–56). The goatherd’s cup accompanies an offer to milk (although not possess) a she-goat who, like the heifer initially offered by Damoetas, gives milk twice a day (Theoc. 1.25–26). But even these details in Eclogues 3 signal a departure from Theocritus, since Vergil pairs his ivy with a grape vine (Ecl. 3.38–39), absent from the cup in Idylls 1, and his cups are an alternative to receiving the heifer as a prize, not its accompaniment. The cup of Idylls 1 is itself richly allusive: a response to both Homer’s Shield of Achilles and the pseudo-Hesiodic Shield of Heracles, it embodies epic themes of conflict, rivalry, and destruction as well as a Theocritean poetics of small size and delicacy.30 Each of Vergil’s herdsmen lays claim to certain aspects of this Theocritean object apportioned between their cups. To my knowledge, however, no earlier critic of Eclogue 3 has noted the remarkable play with temporality, reminiscent of Catullus 64’s “labyrinthine” effects, that appears when we consider exactly how Vergil’s herdsmen stand in for Theocritus’s. If the Daphnis mentioned by Damoetas at Eclogues 3.12 is the same Daphnis who has died and is mourned by Thyrsis at Idylls 1.64–142 (a connection confirmed, it would seem, by the singing contest between a Daphnis and a Damoetas in Idylls 6), Vergil’s cups temporally precede Theocritus’s, since 29 30

Cf. esp. Segal 1967: 287–92. See Hubbard 1998: 21–22 for an introduction and bibliography.

t h e c a p t o r’s i m a g e

84

Daphnis has evidently not died yet in Eclogues 3. Confirming this in part is the fact that Vergil’s lament for Daphnis comes in Eclogues 5, where Menalcas is treated as older than Mopsus (4, tu maior); that is, presumably older than he was in Eclogues 3. So the Greek “model” and its Roman “copy” trade places. By locating himself in Theocritus’s own literary past, Vergil constructs himself as more Greek than the Greek poet, and this upsets in the extreme any facile notion that his Roman text is “dominant” over its Hellenic images. Indeed, Vergil’s ingenuity in avoiding such an effect is simply dazzling. Vergil departs from Theocritus, moreover, in the way he situates his ecphrases in relation to epic, a difference with social, cultural, and political overtones as well as simply literary. The cup in Idylls 1 aggrandizes Theocritean poetry by implicitly elevating its contents to the level of epic, even as it dissociates pastoral from epic itself. Vergil’s cups, by contrast, contain no outward reminiscence of epic themes, which are nevertheless present in the surrounding account of the herdsmen’s confrontation (for example, in Menalcas’s epic-inflected boast to Palaemon about the song-contest’s importance, “the matter is not a small one” [54, res non est parva]; cf. 35, maius). In this way the cups stand apart from their frame. Now rivalry in the form of a quarrel and a song contest (the latter, with its amorous content, recalling specifically the contest on Theocritus’s cup) has been transferred from its earlier site within bucolic ecphrasis to the Vergilian herdsmen’s verbal exchanges. The disjuncture of frame and ecphrases emerges in other ways as well, notably by the absence of the actual cups.31 In occupying, moreover, fewer lines together than the single Theocritean description, the two small-scale ecphrases suggest a rejection of large-scale “epic” contests, a choice that resonates poignantly with the bitter reminiscences of civil war elsewhere in the Eclogues. Even if the ecphrases are short because the herdsmen’s true interests lie elsewhere, Vergil’s own formal decision, as controlling poet, can still be read on its own terms for such connotations. From this perspective, it comes as no surprise that Menalcas and Damoetas, their attention focused on quarrel and contest, fail to appreciate such non-epic cups as are supposedly in their possession. Vergil further prepares his audience for this outcome through an additional Theocritean echo in Menalcas’s ecphrasis. Unlike the cup in Idylls 1, Vergil’s are wagered in a song contest, not given as a gift in return for song, and their maker, Alcimedon, although otherwise

31

Note also the future tense of Menalcas’s pledge “I shall stake” (36, ponam) and the insistence by both shepherds that they have never even used the cups for drinking, a Theocritean detail (cf. Theoc. 1.59–60) that perhaps takes on a different connotation in light of the herdsmen’s relative lack of interest in the cups (Schultz 2003: 214). On all these features of the poem, cf. Saunders 2008: 11. Henderson 1998: 218–19 interprets the cups’ absence (and so their possible non-existence) as indicative of the poem’s overall stress on the herdsmen’s tricky rhetorical maneuvering; cf. Powell 1976: 115.

The Chal l e ng e o f R usti c A r t

85

unknown, is named.32 The model here is Idylls 5.104–05, in which the goatherd Comatas, engaged in a contest with Lacon, says he has a wine bowl fashioned by Praxiteles, the renowned Hellenistic sculptor. Comatas’s boast has been taken as a sign of his rusticity: in ignorance of Praxiteles’ actual work, he “assigns it to the most eminent artist he has ever heard of.”33 Theocritus plays to the sophistication of his Hellenophile audience, who recognize Praxiteles as a master sculptor and find it humorously absurd that a rustic goatherd should claim to possess a cup made by him. Something similar is true of Menalcas’s reference to Alcimedon, as hinted at by the deficiencies in his actual description of the cups. Although his cups are made of beechwood, Menalcas calls them a caelatum opus (37), words usually describing “chased” metalwork rather than wood, the incongruity of the adjoining words “beech” (fagina) and caelatum further enhanced by the connotative links between the word caelatum and caelum (“heaven”) underscored by Varro’s putative etymology.34 This may be a learned allusion to the epic ecphrastic tradition, in which the objects are in fact metal, but the misapplication of the phrase becomes humorous in Menalcas’s mouth, especially in light of Comatas’s grandiose reference to Praxiteles in Idylls 5.35 Similarly, there is incongruence in the word superaddita (38, “added”) used to describe an image actually carved out from a wooden surface.36 Both details, further, anticipate Menalcas’s forgetting the name of the astronomer who appears with Conon. And in this light, the very name “Alcimedon” (Gk. “bold in art”) begins to appear “suitably pretentious.”37 The cups seem simply to exceed Menalcas’s capacities, and their description is to this extent governed more by the herdsmen’s overall tactics of rhetorical feint-and-jab than by the recollection of some “real” objects.38 The herdsmen use ecphrasis to show off and best each other. Vergil’s audience could have seen this as flattery of their own capacity for “purer” aesthetic appreciation or as a subtle reminder of the social competitiveness they, too, expressed through cultural pursuits. Praxiteles’ work, it is worth recalling, formed an important part of Pollio’s art gallery in the Hall of Liberty. 32 But for the possible identification of this Alcimedon with a hero who lived in a cave near a plain bearing his name in Arcadia, and whose daughter bore a child to Heracles, see Lipka 2001: 184. 33 Gow 1952: 2.110. 34 Saunders 2008: 10, 13, 21 citing Var. L. 5.18. On Vergil’s possible wordplay with the Latin fagina “beech” and the Greek φηγός “oak,” see Lipka 2002. 35 Faber 1995: 411n3; Schultz 2003: 213. 36 Schultz 2003: 214. As she notes, Henderson 1998: 220 sees this as Menalcas’s insistence on Alcimedon’s “magical creativity,” a claim that Damoetas then undermines by producing his own cups by the same artist! 37 Clausen 1994: 100. 38 Henderson 1998: 218–23; cf. Powell 1976.

t h e c a p t o r’s i m a g e

86

Damoetas, for his part, is already suspect in claiming that he also has a cup by Alcimedon, since simply one-upping Menalcas, has, by this time, emerged as his basic approach to the confrontation. His description of Orpheus likewise raises doubts, since he does not make it clear whether this figure appears “in the middle” of one or both of his cups (44, duo pocula; 46, in medio). But still better evidence for his failure to appreciate such objects is the fact that a heifer, rather than cups, is the first thing he tries to make Menalcas stake (29–31).39 Menalcas, too, seems to value a herd animal, which he will not risk losing, more than cups, which he will. And Damoetas does not actually accept his offer and say, “I’ll stake my cups,” but simply points out that he also has a pair of cups by Alcimedon, then, after describing the image of Orpheus, tries to direct the wager back toward the heifer (48).40 Through the ambiguous reactions of both his herdsmen, Vergil transforms the Theocritean Comatas’s rustic boast into a more complex assessment of what it might mean to be truly receptive to the images on the cups and what they stand for. Insofar as they suggest Vergil’s adaptation of Theocritus, the herdsmen’s behavior draws pointed attention to Roman audiences’ potential response to the Eclogues, with reference especially to their Greek background. As one scholar puts it, “[t]he personalities of both herdsmen are a debasement of the symbolic value of the cups, whose ideal implications are left to the reader.”41 At least to judge by the reactions of modern readers, the discovery of such “ideal implications” involves finding an impetus toward greater order. Vergil directs his ideal audience, that is, toward a vision of order encompassing the natural world and the cosmos, and by extension the Roman social and political realms as well, even if his exact symbolism remains characteristically elusive. So Orpheus can be seen as the original symbol of pastoral poetry, with the image of the “woods following” him (46) a figure for the divinely inspired ordering of the natural world. The astronomers correspondingly embody the ordering of the heavens— both pointing toward the establishment of a harmonious relation to the cosmos.42 And Eclogues 4, tellingly, inverts the sinister bucolic imagery of Eclogues 39

Cf. Schultz 2003: 212–13. Schultz 2003: 214–15. She also helpfully points out the change here from the basic pattern of Idylls 8. 41 Leach 1974b: 175; in Henderson’s terms, Eclogues 3 offers an inconclusive “dramatization within the text of the fundamental question of what is at stake” in Vergil’s representation of a Theocritean pastoral world (Henderson 1998: 214). I share his emphasis on how the poem turns questions of assessment back on the audience, although I see these less as questions about literary representation for its own sake (see esp. Henderson 1998: 228) than about the particular relation of Vergil’s poetics to their social, cultural, and political context. 42 Segal 1967: 290; Orpheus’s ability to make even the trees “follow” him, furthermore, alludes to the larger power with which he is explicitly credited elsewhere in the Eclogues, especially 6.27–30. 40

The Chal l e ng e o f R usti c A r t

87

3 in a vision of supreme order in heaven, nature, and at Rome.43 It is as though Vergil himself models here the discovery of order that the herdsmen of Eclogues 3 cannot see in their environment but at which the cups hint.44 If Orpheus organizes nature and the astronomers organize the heavens, Vergil does both and more by leading his audience from the “divine” cups to the new Golden Age. Such poetry might help restore peace by revealing the sources of disorder that cloud Theocritean pastoral beauty and pointing the way to an expansion of the Theocritean vision to universal scope.

Pollio as Ideal Audience But will this ideally receptive audience for the cups and for Vergil’s poetry, an audience capable of not only seeing but also cherishing and preserving order, emerge within the course of Eclogues 3? Or ever? Vergil prompts this question by introducing a judge, Palaemon, to oversee the herdsmen’s contest, but raises it still more pointedly through Pollio’s implied view of artistic ideals like those symbolized on the cups. As Damoetas asserts, Pollio loves our Muse, although it is rustic. Pierian ones, fatten a heifer [to sacrifice] for your reader. Pollio amat nostram, quamvis est rustica, Musam: Pierides, vitulam lectori pascite vestro. (Vergil, Eclogues 3.84–85) Damoetas’s belittling of his own efforts reads as flattery, especially in light of the fact that, as we then learn from Menalcas, Pollio also writes “new poems” (86, nova carmina).45 “Merely to imply that the shepherds’ muse is rustica in Pollio’s eyes is to place him in an urbane, sophisticated world,” a point emphasized by the fact that he reads what is, for them, an oral art form.46 Pollio thus stands at a distance from the threatening pastoral world in which the herdsmen struggle, his “love” for their poetic art closer to the attitude of the Muses and Apollo earlier in the poem than to that of the herdsmen themselves, one of whom has broken the tools of Daphnis while the other remains resentful over having failed to win Damon’s goat through song. Pollio’s may be the ideal perspective the 43

Cf. Segal 1967: 304. Cf. Schultz 2003: 210: “both bucolic and golden worlds embody a certain serenity and harmony among the elements they comprise. The images of Eclogues 4 reinforce the idea that in Eclogue 3 Damoetas and Menalcas do not perceive the harmony in the environment around them.” 45 Perhaps his tragedies. Cf. Coleman 1977: 122; Courtney 1993: 255. 46 Putnam 1970: 129. 44

t h e c a p t o r’s i m a g e

88

poem envisages. The cups, which he “witnesses” as an external audience for the poem, gesture toward appreciation for a renewed, Vergilian pastoral order like that with which he is here credited: Damoetas says he loves “our” Muse; that is, Vergil’s Muse as the inspiration for these particular herdsmen as opposed to Theocritus’s. Even before this, however, Vergil’s cups have brought Pollio’s urbane, Hellenophile world to the fore. Although they might seem to be purely literary constructions, they are not unlike objects that a Roman aristocrat such as Pollio might have actually known or possessed, and so enhance the uncanny effect of mirroring and differentiation that now links his circle in particular to the herdsmen. In their overall pattern, Vergil’s poculae recall a well-attested type of GrecoRoman silver cup (often appearing in pairs) on which a figured medallion appears in the center of the bowl.47 Figure 3.1 shows a famous example of this type, the so-called Africa dish, a silver vessel perhaps dating to the first century bce and one of the most impressive objects from a rich horde found at Boscoreale in 1895. Its central emblema depicts a female figure wearing an elephant scalp helmet and accompanied by other symbols associated with Alexandria or its rulers so as to suggest an allegorical representation of the city.48 If this is correct, the object creates an intriguing visual parallel to Vergil’s poem, since it concretizes the literary theme of Roman receptivity to Alexandrian Greek culture. No metal cup depicting Hellenistic astronomers has survived (and one would not expect a wooden cup to have done so), but Menalcas’s cups have a further, roughly contemporary parallel in an ecphrastic epigram by Antipater of Thessalonica (AP 9.541), which is voiced by a pair of “cunningly wrought” (1, τεχνήεντα) cups depicting the constellations as described by Aratus in the Phaenomena, apparently real objects given as a gift to the distinguished aristocrat L. Calpurnius Piso Frugi (cos. 15 bce) by a certain Theogenes.49 The cups, which describe themselves as two halves of a sphere (two concave bowls that, if joined, would form a globe), do not say that they portray Aratus himself, but declare that there is no longer any need to consult his poetry (5, ἀλλὰ σὺ μηκέτ’ Ἄρητον ἐπίβλεπε): one of them depicts the southern constellations, the other the northern (3–4), so that when they are drained one perceives “all 47

Clausen 1994: 100. The elephant scalp refers, at origin, to the campaigns of Alexander against India and figured in the iconography of the Ptolemaic dynasty; the cobra and the Horn of Plenty also have associations with Alexandria. Other proposals for the figure’s identity include Africa itself, Cleopatra VII (possibly figured as Alexandria), and Cleopatra’s daughter by Mark Antony, Cleopatra Selene. For discussion, see Strong 1966: 151; Linfert 1984; Baratte 1986: 77–81, 90; Walker and Higgs 2001: 312. For thematic resemblances between the cup of Theoc. 1 and surviving Hellenistic cups, see Zanker 2004: 14 with n25. 49 For commentary, see Gow and Page 1968: 2.54–56. I follow their text at Gow and Page 1968: 1.38. 48

The Chal l e ng e o f R usti c A r t

89

Figure 3.1 Silver dish from Boscoreale. First c. bce–first c. ce. Photo Credit: Réunion des Musées Nationaux/Art Resource, NY.

the Phaenomena (6, πάντα τὰ Φαινόμενα). Aratus’s own presence in Eclogues 3 has been noted above. A more integral connection, however, to Vergil’s poem lies in the cups’ visual rhetoric according to Antipater. Just as Menalcas’s cups test the learned audience through the omission of an astronomer’s name (possibly “Aratus”) from Vergil’s text, Piso’s cups take the place of an absent text of Aratus one might consult. They, too, encourage the testing of a Roman viewer’s knowledge against information gained from the Alexandrian astronomical poem. Pollio, like Piso, might have delighted in the opportunity to display his cultivation through similar objects. Antipater’s cups describe themselves as two halves of a sphere. Vergil’s interest in the cosmos as a sphere or globe emerges plainly in Menalcas’s description of “he who marked out with his pointer the whole vault of heaven (orbem) for the races of men” (41).50 It is possible, therefore, that we are meant to understand Menalcas’s cups as concave half-spheres evocative of the heavens as well as depicting the astronomers themselves. And artistically crafted spheres with figural images of the constellations were another feature of Hellenistic culture 50

See Segal 1967: 305 for the focus on Hellenistic learning and astronomy in Ecl. 3.

90

t h e c a p t o r’s i m a g e

taken over by the Romans, with a famous example by Archimedes (an associate of Conon, and so one of the names proposed to supplement Menalcas’s failure at 40–42) on public display in the Temple of Virtue, thus surely known to Pollio. After the sack of Syracuse, Cicero tells us, Marcellus brought this and another sphere by Archimedes, who died in the siege, to Rome (Rep. 1.14.21– 22). The latter sphere, the only Syracusan treasure Marcellus kept for himself, was a solid object (also duly famous, Cicero reports) and the former a mechanical planetarium whose motions accurately portrayed those of the celestial bodies.51 Elsewhere Cicero describes still another revolving planetarium created by his friend, the Greek philosopher Posidonius (N.D. 2.88), and it is telling that both planetaria have been invoked to answer the riddle Damoetas poses at Eclogues 3.104–05: Dic quibus in terris . . ./tris pateat caeli spatium non amplius ulnas. Because these objects reside at Rome, the “land” in which “the space of the sky extends not more than three ulnae” is, perhaps, “Rome.”52 If Vergil did intend his audience to think of Hellenistic celestial spheres at the poem’s end, it would not be out of place for Meanalcas’s cups to evoke them, too. Like the suggestion of real cups, this would have made the herdsmen’s objects seem to belong all the more to Pollio’s cultivated world.

Shouldering Out Others Pollio loves pastoral from a lofty distance, but even he is not entirely immune to concerns afflicting the herdsmen. His perception risks being clouded by others vying for his attention, an additional menace to pastoral order. In striking anticipation of Eclogues 4, Damoetas intimates that a new Golden Age, here signified by miraculous natural abundance, has dawned for Pollio and his associates: Let him who loves you, Pollio, come where he rejoices that you, too, have come; let honey flow for that man, and let the rough bramble bear spice. 51 Cicero associates the solid sphere with what was thought to be the first such object, the sphere of Thales of Miletus marked with the constellations by Eudoxus of Cnidos (ca. 360 bce). On these and other passages attesting to Archimedes’ importance in the Roman imagination, see Jaeger 2008. On the so-called Antikythera mechanism, discovered in an ancient shipwreck in 1900 and identified as part of a larger, highly complex geared device displaying the celestial bodies and their motions, see Jaeger 2008: 177n2. 52 See Wormell 1960; Segal 1967: 298; Clausen 1994: 116–17. Serv. ad Buc. 3.105 defines an ulna as a measurement of approximately six feet. Several examples of solid celestial globes have survived from antiquity. One, the Kugel globe, has been claimed as possibly a Hellenistic object (see Cuvigny 2002), while two of the best known are from the Roman imperial period: the Mainz Globe and the globe carried by the Farnese Atlas (on both see Künzl, Fecht, and Greiff 2000). See further Jaeger 2008: 177n2.

The Chal l e ng e o f R usti c A r t

91

Qui te, Pollio, amat, veniat quo te quoque gaudet; mella fluant illi, ferat et rubus asper amomum. (Vergil, Eclogues 3.88–89) The image is a hopeful one, but in light of the doubt cast earlier about Damoetas’s singing abilities, these words, with their thinly veiled request for an audience, suggest pretentious striving after social and literary favor.53 Damoetas seeks to gain entry from outside the charmed circle (just as Horace’s famous “Bore” hopes for an introduction to Maecenas in Satires 1.9). This is certainly how Menalcas hears his rival’s words. He caps Damoetas’s plea with a withering comparison of him to a certain Bavius and Maevius: He who does not hate Bavius, let him love your songs, Maevius, and let the same man yoke foxes and milk he-goats. Qui Bavium non odit, amet tua carmina, Maevi, atque idem iungat vulpes et mulgeat hircos. (Vergil, Eclogues 3.90–91) The pair were contemporary stock-types of the poetaster (or perhaps even real individuals), one of whom, Maevius, is likewise pilloried by Horace in Epode 10, published about a decade later.54 Someone, Menalcas implies, who favors Damoetas would go in for any kind of versified trash, and misunderstands pastoral just as “an awkward city slicker in the country”55 might make hilarious gaffes as a farmer. Menalcas, although his own abilities have, too, been called into doubt, hopes Pollio will not misdirect attention toward an ambitious rival. This, he implies, would leave the pastoral world in still more disarray. All may seem comical here, but the stakes of Vergil’s own interest in Pollio’s patronage were real, even if we have no certain record of the material benefits—for instance, the supposedly lost Mantuan farm—Pollio may have conferred.56 Menalcas’s remark is, moreover, directly akin to the literary invective in which Horace trades in the Epodes and Satires (as evinced by the recurrence of Maevius’s name), where its association with the political divisiveness of the late Republic and triumviral period is, as we shall soon see, easier to recognize. Literary and political wrangling had gone hand in hand at Rome at least since 53

Leach 1974b: 178. Bavius’s death in Cappadocia in 35 bce is reported by Jerome in the Eusebian Chronicle entry for that year; see Clausen 1994: 112. The publication of Horace’s Epodes and Satires, Book 2, is regularly dated to 30 bce. 55 Leach 1974b: 179. 56 Cf. Farrell 1992. 54

t h e c a p t o r’s i m a g e

92

the bilious attacks on prominent contemporaries by the founding Roman satirist, Gaius Lucilius, in the second century bce. And Vergil shows up the necessarily competitive atmosphere of Pollio’s recitation hall for what it is: a venue for poets perhaps “every bit as cantankerous, prickly, and bitchy as the herdsmen.”57 But Vergil cannot maintain a completely distanced perspective on Pollio’s world as he clears the field, so to speak, of those who fail to appreciate his ecphrastic cups. Ecphrasis implies not one but many viewers, some of whom Eclogues 3 constructs as inadequate to the task, their limitations confining them to a divisive social context Vergil himself does not entirely escape. Within this poem, ecphrasis, too, is a crux to catch the unwitting. In the cups of Eclogues 3, images of inspired Greeks ordering both the cosmos and earthly nature through science and song go unappreciated—or not fully so— within a troubled, contentious Greek-and-Roman landscape. There is hope of a better reception and better times to come; Eclogues 4 will sound a note of still greater optimism for Pollio’s Golden Age. Here, however, Vergil leaves it to the audience to respond with the appropriate sensitivity to the new ideals of pastoral order glimpsed on the cups, objects quite literally at stake in an unsettled world.

II. (Not) Measuring Up: Horace’s Figwood Priapus C. Maecenas and Horace’s Satires, Book 1 In 35 bce, four years after the likely publication date of the Eclogues, the Roman world had by no means settled. But in Rome itself power now centered on the circle of Octavian, particularly on his long-time associate Gaius Maecenas, who effectively controlled the capital while his absent leader prosecuted wars with Sextus Pompeius (36 bce) and the Illyrian tribes (35–33 bce).58 Antony, his relations with Octavian all but severed, nourished his bond with Cleopatra in the East, while the third triumvir, Lepidus, lived away from Rome in forced retirement. Recognizing the propaganda value of cultivating authors who might sing Octavian’s praises, Maecenas gathered around himself a group of poets including Vergil and Horace. The winter of 36/35 very probably saw the publication of Horace’s first book of sermones, satiric hexameter poems whose outward posture of nonambition and gentle ridicule belies their implication in the Octavianic circle’s highly politicized self-definition as urbane sophisticates in the Hellenistic manner. Friendship, fairness, and peace—a rejection of the ambitious power politics that had produced so much suffering—were among the group’s professed values. And yet these were the years leading up to Actium, when the 57 58

Henderson 1998: 226. On Maecenas, see recently Osgood 2006, passim; biblography collected in Evenepoel 1990.

The Chal l e ng e o f R usti c A r t

93

balance of Roman power would shift for good. If Pollio’s Hellenophilia accompanied a tendency toward political independence, Maecenas’s own posture of political disengagement and luxurious Hellenism, censured as decadence and effeminacy by his detractors, was much involved with Octavian’s power play. Horace’s Satires, Book 1 names Maecenas as its dedicatee (1.1), and many of its poems touch in some way upon aspects of his dual career as Octavian’s political emissary and literary agent. Satires 1.7 looks back to the eve of Philippi, where Horace fought on the losing side, only to be pardoned in the aftermath, while Maecenas, fighting with the victors, evinced early support for Octavian’s cause. 1.5 describes a journey to Brundisium made by Maecenas and his poets, with certain details recalling, through suggestive ambiguity, not only the treaty he brokered here with Antony’s representative, Pollio, in 40 bce but also the later pact forged between the triumvirs at Tarentum in 37. 1.6 recounts Horace’s acceptance into Maecenas’s circle, for which it expresses the deepest gratitude, while 1.9 relates Horace’s encounter with a “Bore” who seeks entry into this company. 1.10 concludes with the names of many whom Horace regards as his docti amici (cf. 87): Maecenas as well as Pollio, Vergil, Varius, and other writers. Here as in the collection as a whole, the group emerges as “sophisticated, cultured and intelligent men who are humane in their attitudes to others,” “the ideal Roman citizens,” their literary tastes tending, like those of the “neoterics” before them, to the Alexandrian standards of Callimachus.59 Satires 1.8 commemorates, in Priapus’s voice, one of Maecenas’s signal cultural undertakings: the creation of his Esquiline gardens, the horti Maecenatis, over the site of a paupers’ graveyard.60 A byword for their luxury, they were visible to the public from the ancient rampart that Maecenas converted into a promenade (Hor. S. 1.8.14–16) and after his death became a favored imperial property. Written sources tell us little about their actual contents: Dio Cassius (55.7.6) mentions a heated swimming pool that they likely enclosed, while their lofty tower became a landmark, from which Nero supposedly watched the great fire of 64 ce (Suet. Nero 38.2). Centuries of excavation in this vicinity have produced much artwork of Hellenic inspiration.61 The so-called Auditorium of Maecenas, a sunken, delicately painted Nymphaeum that almost certainly lay within the gardens’ bounds, connects the area to Alexandrian literary activity through the verses of Callimachus (AP 12.118) found inscribed on one of its walls, and the fragments of Maecenas’s own poetry recall the Alexandrian proclivities of Catullus.62 Maecenas’s decision to surround his house with a hortus may in some way have been connected with his Epicurean interests, since 59

DuQuesnay 1984: 19, 28, 57. On which see Richardson 1992: 200–01. 61 For catalogues of sculptural finds, see Stuart Jones 1926: 155–70, pl. 55–59; Häuber 1991. 62 On the “Auditorium of Maecenas” see Richardson 1992: 44–45. For the fragments of Maecenas’s poetry, see Courtney 1993: 276–81; Hollis 2007: 314–25. 60

t h e c a p t o r’s i m a g e

94

the garden as a locus for intellectual and cultural pursuits was suggestive of Epicurus’s original meeting place for his followers.63 A likely setting for Horatian poetic performance as well as, in Satires 1.8, its subject matter, Maecenas’s gardens were a place of culture and ostensible removal from public life, but one where poetry and politics mingled nevertheless.

Reinterpreting Signs of Violence: A Priapus of a Different Sort Satires 1.8 helps present Horatian satire, with its deployment of Callimachean brevity and polish, Aristophanic social critique, Socratean wisdom, and Epicurean ethics, as a genre both deeply imbued with Greek culture and aimed at the amelioration of Roman social ills. Thus, again, while Priapus’s statue may stand for Satires 1.8 itself or for Horatian satire in general (a tamed, less aggressive version of satire as opposed to the more virulent satire of Horace’s great predecessor, Lucilius), the poem’s details push us toward a broader, cultural interpretation at odds with any simple notion of a self-asserting text. The statue’s very location, a graveyard transformed into a garden, tropes social renewal in the wake of the recent civil wars. If the witches, with their necromantic magic, threaten a return to the dissention of the past, Priapus’s triumph over them posits a solution in Horace’s self-deprecating outlook. This Priapus refrains from outright violence, yet his indecorous fart nevertheless signals the aggressive defense of Maecenas’s circle against opponents, an acknowledgment, however disguised, of the factionalism still dividing Roman society. Horace himself seems uncomfortable with this arrangement, and even hints, through the statue’s self-disfigurement, at his own vulnerability to the very techniques he employs. Recalling the programmatic tag of Satires 1.1.24, “one telling truth while laughing” (ridentem dicere verum), Priapus makes one wonder whether he can, for all his absurdity, nevertheless articulate and symbolize lasting social change. As in Eclogues 3, Greek culture ambiguously received through ecphrasis accompanies both hope for social order and doubt over the effectuation of implicit ideals. Like Vergil’s herdsmen, Horace’s Priapus relates the cohesiveness of Rome itself—a future untroubled at last by civil discord—to the audience’s perception of a humble artwork, in this case himself. Since the beginning and end of Satires 1.8 are the sections most genuinely descriptive of the statue’s physical appearance, it is here that I will concentrate my own reading. The poem’s opening leads the audience to expect a reference to sexual violence through its reminiscences of poetic antecedents in Greek epigram: Once I was a figwood trunk, useless wood, when a craftsman, uncertain whether to make a stool or a Priapus, preferred me to be a god. Thereafter I was a god, the greatest terror of thieves and birds; for my 63

See Wallace-Hadrill 1998: 4–5.

The Chal l e ng e o f R usti c A r t

95

right hand and the ruddy stake protruding from my indecent groin restrain thieves; but the reed fixed in on my head terrifies the troublesome birds and prevents them from settling in the new gardens. Olim truncus eram ficulnus, inutile lignum, cum faber, incertus scamnum faceretne Priapum, maluit esse deum. deus inde ego, furum aviumque maxima formido; nam fures dextra coercet obscenoque ruber porrectus ab inguine palus; ast importunas volucres in vertice harundo terret fixa vetatque novis considere in hortis. (Horace, Satires 1.8.1–7) By the poem’s end, however, a new perspective on the traditional epigrammatic icon has emerged, a satiric counter-example to received notions. “Why should I recount details?” (40), Priapus asks, For with a sound as great as a burst bladder I farted, fig wood with split buttock; but they ran into the city. With great laughter and jest you would have seen fall Canidia’s false teeth and the tall wig of Sagana, and the herbs and enchanted knots from their arms. nam displosa sonat quantum vesica pepedi diffissa nate ficus: at illae currere in urbem. Canidiae dentis, altum Saganae caliendrum excidere atque herbas atque incantata lacertis vincula cum magno risuque iocoque videres. (Horace, Satires 1.8.46–50) The statue says he wards off thieves with his “right hand” (4; this would traditionally hold a sickle), and his erect member (5), while with the “reed” (harundo) affixed to his head he terrifies the “troublesome birds” (6). The expectation of violence here is underscored by precursors such as APl. 261, by Leonidas of Tarentum (third century bce), in which a sculpted Priapus at a crossroads threatens to make a thief “weep” after “receiving” his penis, to which he refers threateningly as his “club” (2, ῥόπαλον).64 The Latin Priapea 64

A number of other Priapic epigrams from the Greek Anthology, Hellenistic and later, include similar threats. Cf. APl. 236–237, 240–41, 243, and 260, likewise spoken by statues of Priapus watching over gardens or farms. The fact that Horace’s Priapus is made of figwood produces its own expectation of coarseness and violence, given the anal associations of figs in such works. See Hallett 1981.

t h e c a p t o r’s i m a g e

96

echo such violence and take it to new extremes. Five poems (6, 10, 25, 56, 63) recall Horace’s emphasis on Priapus as a humble wooden statue, whose maker is described simply as “rustic hands without art” (manus sine arte rusticae) in 63.10; in 25, Priapus declares he will penetrate the “entrails” of a thief “as far as my own pubic hair and balls” (6–7).65 The highly aggressive basis of Greco-Roman male sexuality finds expression here, in marked contrast with Horace’s “impotent” Priapus, whose fear of the witches renders his traditional “weapon” useless: his terrified fart (cf. 45, “I shuddered,” [horruerim]) rather than his erection frightens his adversaries away.66 So ficulnus turns out to have a comically different connotation—that is, it refers to a different use of the anus—from what it might elsewhere.67 By thus eschewing typical Priapic violence, Horace offers a less aggression-centered way to envision the god’s social function. The gesture is no simple prescription for greater harmony (let alone of tolerance for undesirables such as witches, thieves, etc.). But in its departure from precedent, Satires 1.8 at least reframes such violence as a matter for fresh consideration, rather than a paradigm adopted unthinkingly from sources such as epigram.

A Callimachean Icon Leonidas hints at his play with Hellenistic convention in the name of the person who “set up” his Priapus statue—Theocritus (APl. 261.3)—and his concern with the low and everyday (as in the statue’s much-frequented location) conforms to broader Hellenistic interests.68 Horace situates his Priapus in this tradition by having him say he has been fashioned from “useless wood” by a craftsman who nearly decided to make a stool, the epitome of an everyday object, instead. But if Satires 1.8’s opening recalls in a general way a familiar type of Hellenistic epigram, the poem as a whole embodies Horace’s adaptation of a specifically Callimachean form: polished, anti-epic, aetiological poetry describing humble circumstances. Callimachus in fact wrote at least one poem featuring a small, ithyphallic statue: the highly fragmentary Iamb 9 (fr. 199 Pfeiffer), on a Herm. Here an interlocutor, identified by the Diegesis as the lover of a handsome youth named Philetades, questions the statue about its 65

In 6 Priapus threatens to penetrate a female addressee “to the seventh rib” (6); in 56 he threatens to call a master who will punish the thief with oral rape. Cf. Columella 10.31–34, and for other parallels, see Parker 1988: 10–31; Schlegel 2005: 91–93. 66 On the particular importance of the “Priapic paradigm” in Roman culture and society, see Vout 2007: 17–20. 67 Cf. Hallett 1981; Schlegel 2005: 93. 68 See Zanker 1988: 162–63 on Leonidas’s interest in the low and everyday as shared Alexandrian concerns. Theocritus includes a Priapus statue at Idylls 1.21, to which Vergil possibly alludes at Ecl. 7.33–36 cited above.

The Chal l e ng e o f R usti c A r t

97

state of sexual arousal.69 Iamb 7 (fr. 197 Pfeiffer) is in the voice of another Herm, the cult statue of Hermes Perpheraios in the Thracian city of Aenus.70 Both of these Callimachean poems provide local aetiologies, just as Satires 1.8 explains the origins of Maecenas’s garden and Priapus’s presence in it. This has led at least one scholar to call Iambs 7 and 9 “perhaps the closest Greek parallels” to our poem.71 Further, the reed affixed to Priapus’s head tropes Callimachean literary delicacy. The idea of the reed as a delicate instrument worthy of a Callimachean poet had been enshrined only a few years before this in the “slender reed” (tenui . . . harundine) of Eclogues 6.8, the musical instrument on which Vergil says he will “meditate the rustic Muse” in a reminiscence of Callimachus’s most famous anti-epic statement, the Aetia prologue.72 Harundo can also mean “reed pen,” and by the time he mentions the reed, Horace’s Priapus has already intimated his literary affiliations through the substitution of dextra, “right hand”— that is, writing hand—for falx, “sickle” in verse 4, so as to direct the audience specifically toward the written word as analogue for his activities. The harundo used to ward off “troublesome” foes, the birds, and prevent them “from settling in [Maecenas’s] new gardens” (6–7) thus already suggests Horace the Callimachean satirist defending his patron’s circle. In Priapus’s opening assimilation of his phallus and his reed, moreover, we find confirmation of Sharland’s amusing interpretation of Priapus’s “Callimachean fart.” Priapus’s use of his anus, Sharland notes, rather than his penis to frighten the witches is a satiric anticipation of a later Callimachean statement in Satires 1.10, where Horace orders aspiring writers to “turn the stylus” so as to use its blunt end to erase what they have just written (1.10.72–74). Employing, as it were, his own blunt end, the antipode of both his reed-adorned head and his phallus, Priapus brings his satire, the shortest poem in the sermones, to its nonviolent conclusion (cf. 1.10.9, “there is need of brevity” [est brevitate opus]).73 Ethics and poetics merge in the distinctive application of Callimachus that Horace invites his audience to consider in social terms. So Priapus’s opening selfdescription becomes the key to a Callimachean undercurrent running through the whole poem’s social and political allegory, and linking it to Horace’s wider satiric practice.74

69

For discussion, see Acosta-Hughes 2002: 300–03. See Acosta-Hughes 2002: 294–300. 71 Anderson 1982: 76n9. 72 Cf. below, p. 119; for the Eclogues’ dominant role in creating the terminology of Latin bucolic musical instruments, see Lipka 2001: 154–57. 73 Sharland 2003: 106–07. 74 Satires 1 as a whole endows its Callimachean subtext with ethical content. See esp. Freudenburg 1993 passim; Zetzel 2002: 40–45. 70

t h e c a p t o r’s i m a g e

98

Recognizing Old Comedy This Priapus is a Hellenistic god in Roman guise. But Horace’s implicit social critique and literary self-positioning gain considerably in force from an earlier Greek literary tradition, the invective techniques of Old Comedy, that Horace has by this point singled out as the very basis of Lucilius and claimed to have modified in his own poetry. At the beginning of Satires 1.4, Horace describes Eupolis, Cratinus, Aristophanes, and the other Old Comic poets as authors who, if any man deserved to be called a thief, adulterer, or murderer, “marked him with great license” (5, multa cum libertate notabant). “All of Lucilius, who followed these authors, derives from here,” Horace continues, “with only the meter and rhythm altered” (6–7, hinc omnis pendet Lucilius, hosce secutus/mutatis tantum pedibus numerisque). Lucilius follows Old Comedy, that is, in his “free” attacks on contemporaries. Later in the same poem, Horace specifies how his own satiric methods differ from this. Returning to the topic of thieves, Horace differentiates himself from a certain Sulcius and Caprius—informers or perhaps other satirists—who are called “a great source of fear for thieves” (67, magnus timor . . . latronibus) for their own uncontrolled attacks:75 Be you like the thieves Caelius and Birrius, I need not be like Caprius or Sulcius: why should you fear me? ut sis tu similis Caeli Birrique latronum, non ego sim Capri neque Sulci: cur metuas me? (Horace, Satires 1.4.69–70) Horace thus implies that he will not engage in the vicious behavior he goes on to describe (78–103), which includes slandering absent friends: poisonous methods also implicitly ascribed to Lucilius.76 And in general Horace’s targets, stock types, persons masked behind pseudonyms, or names taken from Lucilius himself, have proven hard to identify, or if they are known, seem to belong to factions hostile to Octavian, so that to this extent his self-definition rings true.77 At the opening of Satires 1.10, Horace recapitulates his admiration for the Old Comic poets, this time for their use of humor to address serious issues: “in this they achieved success,” he affirms, “through this they are to be imitated” (17, hoc stabant, hoc sunt imitandi). In a genre thus characterized as descended from Old Comedy, Priapus’s very presence underscores this lineage, since the erect penis figured prominently in 75

Cf. Anderson 1982: 81 with n17. Anderson 1982: 81n16. The imagery here also recalls iambic invective. See Keane 2006: 79. 77 See DuQuesnay 1984: 53–56. 76

The Chal l e ng e o f R usti c A r t

99

Old Comic humor, its actors wore leather phalluses, and its origins were associated by Aristotle with phallic processions (P. 1449a11–13). Priapus’s scatological preoccupations, too, both with his own fart and the excremental punishments he calls down on himself (Hor. S. 1.8.37–38), repeat an Aristophanic staple. Similarly, given Satires 1.4, Priapus’s concern here with thieves recalls Old Comic themes, and in what follows Priapus will also stigmatize the witches for both murderous and erotic associations of their magic: they trouble dead souls, employ a wax image that stands as though “about to perish” (33, peritura) and wear “knots” (50, vincula) suggesting love-charms.78 But as prescribed by his earlier self-differentiation from Sulcius and Caprius, Horace here employs stock types or pseudonyms.79 Canidia, who recurs in Epodes 3.8, 5, and 17 as well as Satires 2.1.48 and 8.95, is identified by the scholiasts, rather implausibly, with the metrically equivalent Gratidia, a perfume-dealer from Naples who was Horace’s old flame. Sagana, who appears with Canidia at Epodes 5.25, is almost certainly a type (cf. Lat. saga, “sorceress”). Elsewhere in the poem Priapus ridicules the paupers Pantolabus (Gk. “Grab-all”) and Nomentanus (11), a stock wastrel in Book 1, and, should he lie, hopes to be pissed and shat upon by the otherwise unknown Julius (presumably a freedman of the gens Iulia), the “feeble” Pediata (his name suggestive of the pathic role in same-sex intercourse), and the thief Voranus (cf. Lat. vorare, “to consume, devour”) (37–39). Priapus’s monologue is Greek Old Comic social critique repackaged, and challenges its audience to perceive this. Just so, the statue itself represents in a new form the Old Comic god Πρίαπος. The proud, comical repetition of the word deus in line 3 is an implicit reminder that we are dealing with an entity whose existence spans Greek and Roman culture.

(Un)covering Socrates More subtly, the opening of Satires 1.8 evokes still another source of Greek irony and wisdom, Socrates, but recommends a new understanding of his philosophic model in terms of the way Horatian verse satire collapses poetic “surface” and “depth” so as to work upon its audience. Horace’s Satires generally recall the persona of Socrates as ironist and moral teacher.80 Much later in his career, in the Ars Poetica, Horace himself would praise the “Socratic pages” (i.e., the writings of Plato, Xenophon, and other fourth-century bce philosophers) as the fountainhead of “understanding” (309, sapere) that makes for good writing.81 But already in Satires 1, sapere and sapientia are fundamental to Horace’s self-differentiation from the unrestrained exuberance of Lucilius, 78

See Oliensis 1998: 70 on the erotic connotations of Canidia’s “puppets.” With what follows, cf. Brown 1993: 171, 172, 174. 80 Anderson 1982: 13–49. 81 Anderson 1982: 14 on Hor. Ars 309–11. 79

t h e c a p t o r’s i m a g e

100

particularly in the creation of a “thoughtful conversational style” opposed to Lucilian garrulity.82 The Socratic techniques of irony and ordinary conversation on vital ethical topics, as well as Socrates’ practice of surrounding himself with talented friends, all find analogues in Horace, even if the moralizing of Horatian personae becomes at times long-winded and absurd, the butt of its own satire.83 The opening of Satires 1.8 furthers the Socratic effect by recalling Alcibiades’ famous dual comparison of Socrates to, on the one hand, artisan’s statues of Silenuses that reveal images of gods when opened and, on the other, the satyr Marsyas (Pl. Smp. 215a-b).84 The Platonic opposition between a humble, lusty, excessively corporeal exterior and a divine interior is transmuted into an object that is all surface with no interior (or one containing only the worthless gas expelled at the poem’s conclusion) but nevertheless divine. Silenus and the satyrs are minor divinities regularly portrayed as wine-loving, ithyphallic, and ugly; Socrates shares in their desires and physical traits while possessing a divine soul. Like both, Horace’s Priapus is lusty and physically unprepossessing but also a “god.” Rather than being a symbol of the philosopher’s wisdom, he embodies the satirist’s poetry, which “employs the outward shape of Socratic irony as a device to win his readers’ sympathy, and indeed to make them expect . . . deep wisdom in his satire, without necessarily expressing that wisdom.”85 In the place of pure wisdom buried beneath life’s complex, unpromising surface, Horace makes surface complexity both his subject and the object of his poetic mimesis. A Greekand-Roman object with no “inside,” Priapus may nevertheless do important work within his Roman audience.

An Epicurean Garden Undone The philosophical associations of Socrates accompany hints of Epicurean ideals in the way the disrupted garden setting of Satires 1.8 suggests, again through satiric inversion, the Epicurean garden as a site of ataraxia or “freedom from care.” If, as is often supposed, Maecenas himself was a convinced Epicurean, the connection is especially apt. But Horace, whose Epicurean sentiments have frequently been remarked, has earlier gestured toward Epicurean notions, as in the first three satires of Book 1, with their allusions to Lucretius amid a burlesque of Cynic diatribe, or in a passage such as the end of Satires 1.5, where Horace, again reproducing Lucretian phraseology, asserts that “the gods live a 82

Anderson 1982: 27. Anderson 1982: 37–41; Freudenburg 1993: 3–51; cf. Dufallo 2000. 84 See Plaza 2006: 208–11. 85 Plaza 2006: 208. 83

The Chal l e ng e o f R usti c A r t

101

life free from care” (101, deos . . . securum agere aevum).86 Satires 1.8 also recalls Lucretius and Epicureanism, but in such a witty, self-conscious way that any implicit Epicurean “message” becomes another object of ambivalent scrutiny rather than a simple prescription for social betterment. The Epicureans themselves in fact recognized the persuasive power of Epicurus’s portrait sculptures: the “sculpted word” whose visual rhetoric, including facial expression, posture, and gesture, aimed at creating new converts and encouraged disciples to ponder the master’s teachings.87 Horace’s satiric ecphrasis, accompanied by an account of anxiety, fear, physical suffering, and their alleviation (all experienced by a god uncontrollably governed by sexual arousal) is here likewise an impetus toward reflection of a quasi-philosophical nature. Priapus calls attention to his existence in a state that is the opposite of ataraxia when he says, while for me the thieves and wild animals accustomed to trouble this place are not as great a source of care and distress as the women who conjure human souls with their songs and poisons: these I can in no way destroy or prohibit. . . . cum mihi non tantum furesque feraeque suetae hunc vexare locum curae sunt atque labori, quantum carminibus quae versant atque venenis humanos animos: has nullo perdere possum nec prohibere modo. . . . (Horace, Satires 1.8.17–21) The Lucretian vocabulary and themes of these verses cast Priapus’s predicament in a decidedly Epicurean light. He claims that witches who call up the spirits of the dead are, for him, “a source of care . . . and distress” (18, curae . . . atque labori) greater than the thieves and wild animals accustomed to “trouble this place” (hunc vexare locum). The noun cura is Lucretius’s ubiquitous means of indicating the state Epicureans hope to avoid (cf. above on securum), while the verb vexo and the noun labor occur together famously at the opening of De rerum natura, Book 2, where an observer enjoys the storm-tossed sailor’s labor, not because there is pleasure in seeing another person “being troubled” (vexari) but because it is sweet to remark the evils from which one is free (1–4). Only a few 86 On Satires 1.1–3, see Freudenburg 1993: 19, who notes, however, the “common pool” of themes found in the Cynic writer Bion of Borysthenes and the differing hexameter adaptations of his diatribes by Lucilius, Lucretius, and Horace. Cf. Freudenburg 2001: 15–44. On Satires 1.5.101–03 as reproducing an Epicurean dictum also expressed at Lucr. 5.82, see Clay 1998: 29. 87 See Frischer 1982.

102

t h e c a p t o r’s i m a g e

lines later, Lucretius asserts that Nature cries aloud for the alleviation of bodily pain so as to enjoy pleasure in the mind “with care and fear removed” (19, cura semota metuque). Priapus hopes for a similarly tranquil existence.88 Lucretian as well is our poem’s concern with traditional notions of dead souls, which, if they motivate the witches and cause fear in the credulous Priapus, are nevertheless, as in De rerum natura, held up to ridicule. And Horace’s Epicurean emphases here become a marked feature of his implicit politics, if it is correct to link the denigration of necromantic magic to anti-Pythagorean rhetoric advocated by the circle of Octavian.89 Lucretius targets traditional beliefs in life after death and rejects in particular views popular among Roman Pythagoreans of his day.90 Nigidius Figulus, the best known of this group, was associated with the Pompeian cause, and the Elder Pliny reports a story, perhaps part of Sextus Pompeius’s own propaganda, that the reanimated corpse of Octavian’s lieutenant Gabienus predicted victory for the Pompeian forces in the Sicilian conflict.91 Under Maecenas’s watch, Agrippa would soon (33 bce) reactivate long-standing Roman suspicion of astrologers and magicians by expelling them from Rome.92 Satires 1.8 reflects similar attitudes. Priapus’s flatulence itself discredits Pythagorean beliefs, since avoidance of flatulence, together with a belief in the soul’s ability to pass from one body to another after death (metempsychosis), were hallmark Pythagorean tenets. From this perspective, Priapus’s fart satirically becomes not only a cause but also a component of the expulsion of “cares” attendant on the witches’ departure. The playfulness of Horace’s treatment stands out, but its potential import is that Epicurean rather than Pythagorean ideas are a salubrious influence on Rome: Maecenas’s garden rather than a graveyard.

Priapus as “Anti-Art” Does Horace’s poem refer to a real statue in the horti Maecenatis?93 If so, his studied rusticity would emerge as a feature of Maecenas’s own cultivated self-presentation; if not, this is a feature that Horace adds for effect. We will probably never be able to answer the question definitively, since, as with any “real world” counterparts for Vergil’s cups, no comparable wooden Priapus 88 Priapus’s Lucretian pedigree is further underscored, through contrast, by the lovely visual evocation of the Epicurean garden at De rerum natura 2.29–33 (before which golden statues are in fact singled out as a luxury inappropriate to the Epicurean [24–25]). 89 See DuQuesnay 1984: 38–39. 90 See Gale 1994: 93. 91 DuQuesnay 1984: 38–39. 92 Agrippa’s action finds major precedent one hundred years earlier in the expulsion of the Chaldeans (139 bce) but clearly represents part of a long-term trend of fascination and revulsion before and after Horace’s day. See Dickie 2001: 142–61. 93 For a recent argument in favor of this view, see Edmunds 2009.

The Chal l e ng e o f R usti c A r t

103

statues have survived from the ancient world. In any case, Latin poetry’s recurring concern with such objects has led many to assume that rustic herms of Priapus littered the Roman countryside, a throwback to an early moment in the development of Roman art. Peter Stewart, however, has cast appropriate doubt on this view by examining the god’s wider depiction in both art and literature of the late Republic and early Principate, a period when Priapus enjoys a “vogue” at Rome.94 Whether or not we posit a real statue behind Horace’s poem, Stewart’s discussion of Priapus as “anti-art”—a figure whose image is regularly seen to contrast with more sophisticated kinds of representation—helps reconstruct the audience expectations to which Horace plays in using him as an ambiguous vehicle for Greek cultivation. The conceit in itself becomes a familiar one in the early Principate, where it reproduces more general attitudes toward “fine” and “coarse” art discernible in Cicero and Varro as well as aspects of Hellenistic epigram. Comparing the Priapea with passages of Columella and Martial, Stewart notes the deliberate opposition of Priapus herms to works by the canonical Greek sculptors.95 In Priapea 10, for example, the rustic statue informs a girl laughing at his humble wooden form that he was not made by Praxiteles or Scopas or “polished by the hand of Phidias” (2–3). Similarly, Columella, De re rustica 10.29–32 recommends the trunk of an old tree hewn into the form of Priapus over the products of “Polyclitus’s or Phradmon’s or Ageladas’s art,” while a Priapus in Martial, Epigrams 6.73 ironically (and with more obscene humor) asserts that his sculpted erection is “worthy of the hand of Phidias” (8). Although Horace’s Priapus does not mention Greek sculptors, sophistication in Greek verbal if not visual artistry is a comparandum kept at every moment before his audience’s eyes. Real visual parallels compiled by Stewart, however, fill out this context.96 An opus sectile panel from the House of the Colored Capitals in Pompeii (figure 3.2) shows a Priapus herm standing on a simple pile of rocks placed directly alongside a more sophisticated statue with a proper base. The image’s self-consciously rusticating construction of Priapus recurs in a number of other urbane artworks from the first centuries bce and ce, including wall paintings from the Pompeian House of Julius Polybius and the Villa of the Mysteries. Still more emphatically, a set of early imperial reliefs from what may have been an altar base contrast a Priapus statue standing on a rock with statues of Hercules and Spes on cylindrical bases, and set off this difference through the otherwise marked parallelism between the Dionysiac and sacro-idyllic scenes in which the 94 Stewart 2003: 72–77 with the more detailed discussion in Stewart 1997; cf. O’Connor 1989: 24 on Priapus as a recent import; Richlin 1992: 143 on the vogue for Priapus in late Republic and early Principate. 95 Stewart 1997: 575–77; translations are his. 96 See Stewart 1997; 2003: 76.

t h e c a p t o r’s i m a g e

104

Figure 3.2 Opus sectile panel from the House of the Colored Capitals. Pompeii. Koppermann, Neg. D-DAI-Rom 1964.1828.

statues appear. “No figure in Roman art or literature,” Stewart concludes, “is so expressly defined as an outcast from the norms of cult, art, and culture. This Priapus is anti-art, and as such, he serves to elevate and underpin proper cult, proper statues, proper art.”97 So in Satires 1.8, Horace has zeroed in on the one figure from the Greco-Roman pantheon most likely to set his cultivated satiric “message” in high relief. The audience listening to Horace’s poem with real art in mind would possess a crucial interpretive clue to its cultural significance.

Trouble in Paradise Connecting Priapus to Horace evokes the pair’s analogous service to Maecenas.98 If Priapus and his surroundings have been rehabilitated through Maecenas’s munificence, so, too, has Horace, who, after fighting on the losing side at Philippi was nevertheless accepted into Maecenas’s circle. This is the domain Horace, in Satires 1, takes it upon himself to defend, just as Priapus protects Maecenas’s actual gardens. In Satires 1.5, for example, Horace maintains a tactful silence about the political wrangling between Octavian and Antony that forms the backdrop for the journey to Brundisium.99 In Satires 1.9, Horace repels the 97

Stewart 2003: 77. Gowers 2003: 83; Sharland 2003: 104. 99 See Gowers 1993; Sharland 2003: 104. 98

The Chal l e ng e o f R usti c A r t

105

selfish efforts of the “Bore” to insinuate himself, and characterizes Maecenas’s company as a bastion of modest, self-effacing regard for what is “pure” (49–52). There, Horace insists, “each has his own place” (51–52, est locus uni/cuique suus). Priapus, too, has found his own place in the new gardens, and his role in defending them—by humorous rather than violent means. But threats remain. If Satires 1.8 “drives off the bugbear of civil war” in the violence it rejects, it simultaneously expresses hesitation about the stability of this arrangement: “Just as the garden still bears the bones of its former existence [16, ossibus; 22, ossa], so one need not scratch too deep beneath the surface of the new order to find the skeletons of civil war.”100 The witches embody the threat of the past both in their magic and in its generic associations with the Lucilian tradition of bitter invective.101 They dig for bones in the earth (where the dead lie buried), hide charms there, conjure up ghosts, and call upon the cthonic powers Hecate and Tisiphone (17–45). Although these two witches flee “into the city” (47, in urbem) at the end of the poem, Priapus has earlier expressed his inability to “destroy” them or “prohibit” this sort of person from gathering bones and harmful herbs in the garden after nightfall (21–22). Their rites are violent: they tear apart a lamb with their teeth so as to pour out its blood for the shades (27–28), and stage a woolen effigy’s “punishment” of a smaller wax one, which is eventually burned (30–33, 43–44). Hellenistic poetry, which displays a similarly marked fascination with arcane magic, is an influence here, too, but far more significant is the fact that the witches’ poisons (19, venenis) recall the venomous words associated with Lucilian invective in Satires 1.4.100–01.102 This is language designed to harm other, friendly parties and so is easily associated with civil discord, especially that of the Republican past. The witches are all the more like invective satirists in using carmina, a word denoting not only spells but also (satiric) poems, lampoons, and other types of utterance. And the suggestively ambiguous nature of Canidia’s language is magnified especially by Horace’s subsequent reference to her in Satires 2.1, where he contrasts, only to equate in part, his own “good” carmina with “evil” ones (48, 82–84).103 This last example especially shows how the self-definition that distinguishes Horatian satire from its antitypes, including the witches’ magic, simultaneously assimilates it to that which it purports not to be. And such assimilation undermines in part the implicit claims of Satires 1.8 to mobilize Greek culture in the cause of social change. Even more than the Eclogues, Horace’s Satires participate 100

Pagán 2006: 56. Cf. Anderson 1982: 81; Schlegel 2005: 100–02. 102 Anderson 1982: 81. 103 On the differentiation/assimilation of satire and magic in Horace, see esp. Oliensis 1998: 64–101; cf. Anderson 1982: 81; Schlegel 2005: 98–107. 101

106

t h e c a p t o r’s i m a g e

in the factional politics that they apparently dismiss. For all its insistence on the cultivated, harmonious relations among Octavian’s followers, Satires 1 raises the threat of a “regulatory violence” of language, delineating exclusive boundaries and harming enemies.104 So Priapus, as we have seen, colors his language with reminiscences of Callimachean iamb and Aristophanic comedy, genres including aggressive and polemical elements.105 Similarly, the departure from decorum in Satires 1.8 “not only compensates for, but ultimately inverts” the apparent impotence of Horace’s Priapus in comparison with his epigrammatic counterparts. If Horace makes Priapus impotent, he himself has written a poem that “will always, every time, put the reader in an awkward position.”106 Such awkwardness applies even to Maecenas himself; if the tactic elevates Maecenas’s tolerant circle over competitors, it nevertheless reveals Horace’s uncertain position as someone who relies on the violent methods that he ascribes to others, and that may “backfire” on their users.107 The harmful effects of satire even leave their mark on the statue’s own posterior, which splits in fear when he farts (46–47).108 Like the witches, who are shown up as frauds at the poem’s end, their false teeth and wigs dropping from them in their flight (48–49), both Horace and his opponents must fear the deflating effects of the satirist’s pen. The closed, philhellene world around Maecenas cannot remain “pure.” Its very symbols, artistic and otherwise, are part of its political vulnerability.

Conclusion: Social Competition and Cultural Competition in Triumviral Ecphrasis In conclusion let us sum up this chapter’s argument about the competitive display of Greek cultivation as a feature of ecphrasis for patron figures: namely, that we now see such display operating clearly beside the more impersonal way that Greek and Roman culture can be said to compete through the trope. Even a partial reconstruction of the poems’ likely effects on their actual audiences makes the case in point. Imagine for a moment how divergent an impact these 104 Keane 2006: 55–56; cf. 46–47 on Lucilius as both “valuable precedent” and “liability” for Horace and the other later satirists, who “indicate that they are restraining their own attacks in the face of the more limiting political conditions of the early Principate.” 105 See Rosen 2007: 7–8 (on Horace and Old Comedy) and 172–206 (on Callimachus). 106 Pagán 2006: 63. 107 Welch 2001; Keane 2006: 55–56. Welch argues that Priapus’s impotence reflects the tensions between Horace’s position of subservience to Maecenas and the idea of invective poetry. The normally aggressive Priapus finds himself “unmanned” in Maecenas’s gardens, just as Horace, Welch proposes, is unable to target Maecenas, his patron. And yet “by dramatizing the pressure Maecenas’s friendship puts on his poetry, Horace manages to poke fun at him as well” (189). 108 Keane 2006: 56.

The Chal l e ng e o f R usti c A r t

107

ecphrases might have had on those within and outside the poets’ chosen circles of amici. For Pollio and his friends around 40 bce, Vergil’s beechwood cups are a dazzlingly witty puzzle, confirming Pollio’s cultural prowess via both the brilliance of the poet whose attentions he has attracted and the expectation that, if no one else, Pollio himself can master their complexity, answer their riddles, help calm their factious world. Thus Pollio sheds upon his associates the glow of a supreme Hellenophilia, the impetus behind his library and art gallery, and the basis for the new pastoral order at which the cups hint. But what of Bavius and Maevius, or the real-world versions thereof? To them Vergil’s sophistication is an enviable passkey. They are potential participants whom ecphrasis excludes: their vision is flawed, they cannot see what “we” do, and so on. They are pitiable, perhaps, like the herdsmen locked in quarrel, but their emblem is more the “chill snake” than the “work of divine Alcimedon,” and if real Greek culture produces harmony in others, it leaves them cold. A Hellenophile Golden Age may be at hand, but if not, they are responsible in part for its failure to materialize. Rome as whole, Vergil suggests, still cannot grasp the symbols of truly civilized, philhellene behavior. Some five years later Maecenas, ensconced in his gardens, pulls the strings of power in the capital. He can laugh with his Hellenophile circle at Horace’s Priapus, both antitype and satiric icon of the group’s professed urbanity. Even Horace, however, knows that self-assertion can bring revenge. Priapus still fears the witches. And for those outside the gates a warning: even our trifles are effective; don’t make us use our weapons. We’re laughing at you, and you don’t really get the joke, so keep out. Behind a ridiculous and vulnerable statue embodying the humility of the poet figure before Maecenas, Octavian’s political might seems all but assured. The humility of this art object allows Horace, like Vergil, to call attention to the social implications of culture, to real social hierarchy, a technique far less pronounced in the familiar Hellenistic trope on which both poets draw. As with the Eclogues, the dissolution of Roman civic cohesiveness is both the poetry’s target and, ironically, its potential effect; Greek culture is a possible remedy that may also exacerbate the symptom. Ecphrases of rustic yet Hellenic art here gesture toward a “social structure” at once more and less than the real thing. In these two small-scale ecphrases we have come some distance from Catullus 64’s elaborate visual fantasy. In the following chapter we turn to grander objects still more reflective of Octavian’s—and Augustus’s—growing dominance, the ecphrastic temples of Vergil’s Georgics 3 and Propertius’s Elegies 2.31. But now another source of Greek influence, the numinous power of religious iconography, comes to the fore.

4

Describing the Divine The Ecphrastic Temples of Vergil, Georgics 3.13–36 and Propertius, Elegies 2.31

Ecphrasis of religious images has been a problem for ecphrastic theory at least since Lessing’s famous denunciation of descriptive poetry as virtually synonymous with the worship of idols, and a technique unworthy of true poetic genius. “Religion,” for Lessing, “exercised . . . constraint upon the old artists. . . . Superstition loaded the [devotional images of] gods with symbols” such that all works “that show an evident religious tendency, are unworthy to be called works of art.” And poets who emulate artists’ techniques, as through the description of such attributes, do so at their peril.1 In the most-cited ecphrastic theory to follow Lessing, religiosity per se is in fact seldom encountered, while classical examples seem drawn from an antiquity strangely emptied of its religion, as though it conformed unproblematically to the secularizing terms retrojected onto it. The antireligious tendencies of theory, however, must be countered head on in my argument, since the ambitions of Roman ecphrasis clearly extend to divine iconography, a domain where the text’s presumed attempts at domination of the image are highly doubtful, especially when the god in question inspires the text itself. Here, too, furthermore, the competition and interplay between Greek and Roman culture is of paramount importance, since the Greek gods—as Greek gods—often inspire Roman cultural endeavor more broadly, and inform the Roman sense of self. 1 Lessing 2005: 62–63, 67–70; cf. Mitchell 1994: 155. And yet in criticism of more recognizably “religious” eras one finds ecphrasis and religion easily inhabiting the same intellectual space, as in Kustas’s description of Byzantine ecphrasis as expressing “the inner experience of the beholder before the sacred objects of his religion”—a kind of “revelation” (Kustas 1973: 58)— or in Wandhoff ’s account of religious iconography in medieval ecphrasis as connecting the reader simultaneously to political and spiritual realms (Wandhoff 2003: 134–40). On religion and (Greek) ecphrasis, however, in the classical period, see esp. Platt 2011: 1–27 and 170– 211 and on the secularizing tendencies of modern text/image theory, see above all Squire 2009: 15–193.

108

Describing the Divine

109

This chapter will consider two ecphrases, the opening of Vergil’s Georgics, Book 3 and Propertius’s Elegies 2.31, where statues of divinities, described within larger accounts of the temples over which they preside, are in some way responsible for the poet’s philhellene literary efforts (the ancients often making no firm distinction between gods and the statues representing them2). Vergil and Propertius in this way create poetic analogues—though by no means simple or straightforward ones—to the cultural activity of their de facto ruler and patron, Octavian. For each ecphrasis refers, directly or indirectly, to one of Octavian’s central cultural undertakings, the Palatine temple of Apollo with its adjoining precinct and library, a sacred complex built to honor Octavian’s own patron deity. Vergil’s poem describes an imaginary edifice evoking the Palatine temple (then being built) only by suggestion, while Propertius’s is a direct account of the temple complex soon after its official dedication.3 Octavian’s project in fact represented a victory over the Hellenistic Greek world that was nonetheless a display of deep dependency and Hellenophilia, since while the new temple commemorated, through its suggestive imagery, his defeat of Antony and Cleopatra and acquisition of Alexandria itself, it was nevertheless modeled in part on the great sanctuaries of the Hellenistic world including Pergamum and Alexandria. Octavian even appropriated actual Greek statues to adorn his new temple; one of these, the cult statue of Apollo, will play an especially important role in Propertius’s poem, as we shall see. But key to understanding both Vergil’s and Propertius’s ecphrasis is perceiving how each poet mimics and responds to the cultural dynamics of Octavian’s temple, which resituated Hellenic art within a Roman context to offer religious validation for imperial conquest and Octavian’s victory in a devastating civil war. Although the two ecphrases are on the surface very different, closer inspection reveals their similarities when read from this perspective. Vergil’s temple is made all but explicitly to stand for a poem (most likely the unwritten Aeneid), but focuses even greater attention on the act of cultural appropriation. Vergil imagines a triumph over Greece inspired by the divine Octavian, who is the temple’s resident deity. Yet the whole passage is clearly modeled on Greek, as well as Roman, poetic antecedents. Propertius, in a poem alluding similarly to Greek models, concludes with an image of Apollo’s Greek cult statue singing. This seems to be the epiphany behind his poem, and by suggesting this possibility, Propertius plays with a typically Hellenistic synthesis 2

See, e.g., Elsner 2007: 11. We do not know exactly when construction of the Palatine complex began (dates ranging from 36 to 29 bce have been proposed), but it was dedicated in October, 28 bce. See Miller 2009: 185 with bibliography. The Georgics were completed in 29 (see Thomas 1988: 1.1). Propertius 2.31 explicitly situates its narrative directly after the temple’s dedication (1–2). 3

t h e c a p t o r’s i m a g e

110

of religious transcendence and artistic connoisseurship.4 Elegies 2.31 appears to depart from Georgics 3 in creating tension between its genre—elegy—and the imperial message that the Palatine monument embodies, since Propertius’s is a different sort of receptivity to Greece, one foregrounding eros and sympathy for war’s victims. But even here no clear-cut dichotomy exists: between Vergilian praise, that is, and Propertian blame (precursors to reductive “pro-” and “anti-Augustan” attitudes long dismissed by Latin literary scholars5). Each poem mixes elements of panegyric with other features less easily categorized as such. In fact, by focusing on the poems’ ecphrases of religious objects, we see how both poets engage the ambiguities surrounding Octavian’s own use of religion at this time. Vergil’s “divine” Octavian must be represented by a fictive statue in an explicitly “notional” ecphrasis referring to the future, since Octavian, though aligning himself closely with Apollo as patron deity, had been careful to avoid overt claims to divinity at Rome in this period.6 The reference to Invidia (Envy) with which Vergil’s ecphrasis concludes reactivates anxiety over civil war like that which we have remarked in earlier Roman poetry, and focuses this concern specifically on the act of temple building, long associated at Rome with elite self-promotion and display. Propertius’s ecphrasis, the structure of which suggests the progress of a celebrant, conforms to the ostensible religiosity of Octavian’s building of the temple, and yet calls attention to Octavian’s identification with Apollo as both martial and poetic god, while Propertius’s emphasis falls squarely on the latter. But the idea of drawing poetic inspiration from (Octavian’s) religious art conjoins our two texts both thematically and in terms of technique. As before in this study, the cultural valence of ecphrasis will be kept foremost in what follows: it is more important, again, than the literary programmatics that would subordinate visual art merely to the status of a symbol or stand-in for verbal art.

The Palatine Temple of Apollo: Ecphrastic Object and Cultural Analogue Since the Palatine temple of Apollo figures so prominently in my argument, an overview of its architecture, imagery, and social context is in order at the outset. The key point to grasp when considering the links between Octavian’s temple and the ecphrases is the utility of Greek iconography and Hellenistic architectural forms as media in which Roman ideas about civil conflict and foreign 4

See esp. Platt 2011: 170–211 on this combination as a technique characteristic of Hellenistic ecphrastic epigram. 5 A key text is Kennedy 1992. 6 Cf. the prospective tone with which Vergil invokes Octavian’s divinity at G. 1.24–42.

Describing the Divine

111

conquest can be held in suspension. That is, while the temple’s imagery and form subtly acknowledge Octavian’s conflict with Antony, they encourage an onlooker to interpret Octavian’s victories as a pious defeat of a barbarous foe that is at the same time an expansion of Rome’s claim over highly cultivated areas of the Hellenic world—all without ever spelling this out in explicit terms. Octavian had vowed the temple before his defeat of Sextus Pompeius at Naulochus in 36 bce and claimed Apollo’s special favor both there and at the more important victory, over Antony, at Actium in 31. The new sanctuary, which outclassed in sheer size and expense all similar, previous efforts by ambitious Roman elites, pushed Rome ever further toward the status of a major Hellenistic capital in its own right, while foregrounding both Octavian’s connection to Apollo and his political dominance. Vergil and Propertius, by using this temple to celebrate their poetic appropriations of the Greek and Hellenistic cultural legacy, similarly suspend ideas about civil war and empire in a strongly visual, religious medium: temple-ecphrasis itself. Both the monument and the poetry display a new receptivity to Greek culture connected in this complex fashion with recent events. Apollo’s Palatine temple, now largely vanished, must have offered the grandest of venues for the celebration of the god and his mortal protégé, Octavian.7 Like the multipurpose sanctuaries of the Hellenistic dynasts, the Palatine complex included a combination of impressive elements. Surrounding the temple itself, which overlooked the Circus Maximus, was a portico housing sculptures of the mythical Danaids (figure 4.1).8 A sculptural group near the central altar depicted Apollo and four cattle by the great Greek artist Myron. A library with both Greek and Roman authors, like that built by Asinius Pollio some years before, stood nearby, and contained clipeate portraits of the great writers as well as their works. The temple’s doors depicted Apollo’s victories over the Gauls and the children of the impious Niobe. On its roof were sculptures by the sixth-century bce Greek artists Bupalus and Athenis (Plin. Nat. 36.13), while inside stood its three cult statues of Apollo, Diana, and Latona, by the fourth-century Greek sculptors Scopas, Timotheus, and Cephisodotus, respectively (Plin. Nat. 36.24–26, 32; see figure 4.2).9 Like so much Greek art in Rome, these works are a testament both to a Roman sense of dominance over the Greek world—the sculptures have been transplanted from their original locales—and Rome’s continued emulation of this world in the cultural domain. Notions of artistic and religious renewal are implicit in still other aspects of the sanctuary’s sculptural program: two statues 7

See esp. Miller 2009: 185–96 with bibliography. Our principal ancient sources for this monument are Prop. 2.31 and the much shorter descriptions at Ov. Am. 2.2.4, Ars 1.73–74, and Tr. 3.1.60–64. For more on the Danaid sculptures see below, p. 115. The sons of Aegyptus are mentioned by the scholiast to Pers. 2.55–56 as accompanying the Danaids, but the authority of this source has been doubted. For a reconstruction of the archaeological remains, see Quenemoen 2006. 9 For discussion see Miller 2009: 187–89. 8

t h e c a p t o r’s i m a g e

112

Figure 4.1 Female herms in Nero Antico, possibly three of the Danaids from the portico of the Temple of Apollo. Rome, Museo Palatino, first century bce. Su concessione del Ministero per i Beni e le Attività Culturali—Soprintendenza Speciale per i Beni Archeologici di Roma.

of Apollo, one with a lyre in the courtyard and the cult statue of Apollo citharoedus, depicted artistic creation, so as to suggest that Octavian “rehabilitates and re-enables the Greek arts.” Similarly, the reuse of Greek statues as cult images, attested here first in the Roman record, accords well with Octavian’s overall emphasis on the “rehabilitation of a failing piety.”10 But while the Palatine sanctuary evoked those of Hellenistic rulers, its architecture also reveals Octavian elaborating traditions already in use by the Roman elite,

10

Welch 2005: 85.

Describing the Divine

113

Figure 4.2 Altar base. Sorrento, early Imperial period. Depicted are the three cult statues in the temple of Palatine Apollo: Artemis/Diana by Timotheus, Apollo by Scopas, and Leto/Latona by Cephisodotus. The Sibyl sits at Latona’s feet. Neg. D-DAI-Rom 04500.

and so emphasizing his elevated status as a member of this group.11 Due to the precedents set by structures such as Sulla’s sanctuary of Jupiter Anxur at Terracina as well as the temple complexes of Pompey and Caesar, Octavian could expect that his project would suggest “his conformity with Roman tradition.”12 As we have seen, conjoined temples, porticoes, and libraries were features of earlier Roman building projects, in which they, too, recalled architecture from the Hellenistic world. Octavian’s cultural emulation of Alexandria was at the same time political self-assertion over his defeated rival, Antony, who had proclaimed himself a ruler at Alexandria, and this fact is also reflected elsewhere in Octavian’s art and monuments. A cameo gem, for instance, probably dating from just after the actual seizure of Alexandria in 30 bce, links Octavian to the iconography of Ptolemaic god-kings earlier exploited by Antony. In time, an Egyptian obelisk, its inscription announcing it as a monument to the conquest of Egypt, would stand as the gnomon of the vast 11 12

See esp. Quenemoen 2000. Welch 2005: 83.

t h e c a p t o r’s i m a g e

114

sundial Augustus built in the Campus Martius.13 On the Palatine itself, Octavian located his house directly adjacent to the temple and had the two structures connected in some way: a clear act of self-association with Apollo. Thus Octavian’s cultural exploits came to signal “the novelty of his political situation.”14 If Octavian employed techniques long in use, it is also true that a simple association with Hellenistic royalty would have been too risky for him, not only because of Antony’s behavior but also in light of the unbridled monarchic ambitions of which Julius Caesar had been suspected. Octavian’s identity was not to be that of a Hellenistic king, but of a powerful Roman leader capable of restoring social and political order. Thus the Palatine complex contained traditional Roman architectural features together with its many Greek elements, as in the temple’s Roman frontal design and high podium. The emphasis on public space even within Octavian’s adjoining house conforms to Vitruvius’s recommendations for the proper elite Roman house in the period.15 It is no accident, furthermore, that the casa Romuli, the supposed hut of Rome’s founder, stood nearby: Octavian presented himself, if not as a king tout court, then nevertheless as a second Romulus, and the rustic humility ascribed to Rome’s first king acquired a renewed cultural value in the Augustan Age (Octavian’s house was even comparatively modest by contemporary elite standards). A careful choice of Greek and Roman allusions underlies Octavian’s sanctuary, as we shall see it does both Vergil’s and Propertius’s poems. Although there is no scholarly consensus as to the exact political message of the sanctuary’s mythological imagery (and no definitive ancient statement on this topic), the unavoidable conclusion is that its Greek myths associated with Africa and Egypt, as well as others exploited previously by Antony, would have suggested both Octavian’s recent victory at Actium and his annexation of Egypt as a new Roman province, two of the three conquests celebrated in his triple triumph of August, 29 bce.16 As one scholar observes, the iconography of the Palatine complex is polysemous and resists a single, monolithic interpretation, but rather “brilliantly combines several threads, among which is the victory at Actium.”17 The diversity of modern interpretations does not contradict, but 13

On the cameo and the obelisk, see Galinsky 1996: 114–15, 146. Welch 2005: 83. 15 Quenemoen 2000: 274–75 on Vitr. 6.5.2; cf. Welch 2005: 83. 16 Cf. Welch 2005: 81–82 following Zanker 1983. The other conquest celebrated in the triple triumph was that over various tribes of Illyria, Germany, and Gaul. This came first in the three-day celebration, with Actium celebrated on the second day and the capture of Egypt on the third. Noting Octavian’s well-attested desire to portray Actium as a victory over a foreign queen rather than a civil conflict with Antony, Gurval (1995: 87–136) has denied any Actian resonance in the Palatine complex, much of which, because it was vowed before Naulochus in 36 bce, had in fact been completed previous to 31 bce. His views, however, have not found general acceptance; see Miller 2009: 191–93. 17 Welch 2005: 82. 14

Describing the Divine

115

rather enhances this point. Between the columns of the portico stood sculptures of the fifty Danaids, their father, Danaus, standing nearby with his sword raised.18 Three black stone female herms now in the Museo Palatino (figure 4.1) may be surviving elements of this imposing visual sequence, their mixture of Archaic and Early Classical Greek styles suggesting the Neo-Attic sculpture characteristic of the triumviral period.19 Danaus and Aegyptus were Greek brothers who lived in Egypt (their father was Belus, Egypt’s king); when Aegyptus insisted upon a fifty-fold marriage to his sons, Danaus fled with his daughters to Argos, where his brother and nephews pursued them. There Danaus consented under duress to the wedding but, still resistant, instructed each daughter to kill her husband on the wedding-night. The Danaids, then, are available as allusions to another royal, Greek-speaking resident of Egypt, Cleopatra, who likewise spelled disaster for her husband (Cleopatra is in fact the name of one of the Danaids in some ancient sources), and their punishment symbolizes her defeat at Octavian’s hands.20 At the same time, the punishment of Aegyptus’s offspring can suggest Rome’s punishment of Antony and Cleopatra both.21 The whole myth, moreover, points ruefully to the killing of kin, since the fraternal quarrel between Danaus and Aegyptus is the impetus for the murders.22 Each of these explanations has found proponents among modern scholars, and together they well illustrate the monument’s thematic complexity.23 Among surviving fragments of the temple’s decoration, the so-called Campana reliefs are a further instance of Greek myth used with Egyptian and Actian resonance. At least two motifs from these terracotta panels suggest Octavian’s conflict with Antony and Cleopatra: Perseus and Athena with the Gorgon’s head and Hercules and Apollo fighting over the Delphic tripod.24 As a monstrous but defeated female adversary, Medusa stands easily for Cleopatra. Because Antony, moreover, had identified himself with Hercules in his own propaganda, Apollo’s struggle with the demigod tropes Octavian’s with his rival, and the presence of both mythical figures signals Octavian’s seizure of an adversary’s symbols. Egyptianizing elements, moreover, frame the reliefs: lotus flowers, elephants, and Isis with sphinxes. But while immediately topical, such 18

For ancient sources, see above, n8. Quenemoen 2006: 229; cf. Zanker 1983: 27; 1988: 239–63; Candilio 1989: 85–90; Tomei 1990. For further bibliography on these statues see Strocka 2008: 49n34. 20 See Kellum 1993: 80–81. 21 Lefèvre 1989: 14; Klodt 1998: 4; Spence 1991: 15; but cf., contra, Fowler 1991: 30, who notes the consistent association of the Danaids with their guilt in our sources. 22 Zanker 1983: 30; 1988: 86; Kellum 1993: 81. For further views, see the helpful summaries at Putnam 1998: 199; and Strocka 2008: 48–49. 23 For “thematic complexity” as characteristic of the Palatine sanctuary, see Welch 2005: 89, who rightly echoes Fowler’s insistence (1991: 30–31) that such an approach is as appropriate to this monument as to any work of literature. 24 Welch 2005: 87–88; Miller 2009: 192–93. 19

t h e c a p t o r’s i m a g e

116

imagery also possesses a broader Hellenistic significance: the Gorgon’s head in particular was a symbol used by Hellenistic monarchs, and so, like other features of the monument, situates Octavian both within and in opposition to this whole venerable tradition.25 Thus the Palatine complex celebrates Apollo and Octavian as victors over barbarous opponents (symbolized by the Gauls, the Danaids, Medusa, etc.), while nevertheless representing a display of piety that was at once a political and cultural claim over the Hellenic world. So, too, the ecphrasis of religious imagery became an occasion for such claims in Vergil’s and Propertius’s poetry.

I. A Triumph over Greece: The Temple of Georgics 3 The Cultural Frame: Leading Home the Muses, Founding a Temple, Celebrating Games The ecphrastic temple of Georgics 3.13–36 is undeniably a symbol for a poem, Vergil’s “most extensive, and most complex, programmatic statement.”26 But as such it is an especially valuable test case for my thesis, since its cultural framing is so self-evident, and, as I argue, more important to understanding its function in the poem in which it appears. For this ecphrasis simply to dominate its image and assert itself as text over that image is indeed an impossible proposition. Vergil’s temple may be a (future) text, but it is, more significantly, a statement of the numinous power with which Octavian, through his link to Apollo, fills both Rome and Vergil’s own poetry. The Georgics as a whole elevate farming as a symbol of mankind’s ordered relation to the universe, circumstances to which it is hoped that Rome will now attain with Octavian at its head. Within the world of the ecphrasis, the divine Octavian inspires a pious cultural victory over Greece that is simultaneously a reabsorbing of Greek culture into Romanness, just as Octavian’s own exploits brought the Hellenistic cultural capital Alexandria itself into the Roman empire and celebrated this fact through the Palatine temple. Thus at least here, Octavian’s whole philhellene self-styling as a Roman ruler becomes a basis for Vergil’s as

25

Strazzulla 1990: 41–43; Welch 2005: 88. Thomas 1988: 2.36. For my understanding of the passage, I have relied especially on Wilkinson 1969: 165–72; Buchheit 1972: 92–159; Putnam 1969: 165–74; Gale 2000: 11–15; Nappa 2005: 116–24; Meban 2008; Lowrie 2009: 150–57; Miller 2009: 2–6. I have found Miller’s discussion particularly illuminating for its analysis of the poetic, religious, and political themes concentrated in Vergil’s allusion here to the Palatine temple of Apollo. 26

Describing the Divine

117

a Roman poet. And insofar as Vergil’s temple recalls Octavian’s, Vergil draws upon and intensifies the fusion of Octavian with his patron deity that was already implicit in the temple complex itself: I will sing of you, too, great Pales, and you, famed shepherd from Amphrysus, and you, woods and rivers of Lycaeus. All the other themes, which might have charmed idle minds with song, are now commonplace: who does not know either harsh Eurystheus or the altars of accursed Busiris? By whom has the boy Hylas not been sung, and Latonian Delos and Hippodamia and Pelops, famous for his ivory shoulder and keen with horses? A path must be tried by which I, too, might be able to lift myself from the ground and fly, as a victor, through the mouths of men. I first, returning to my country, will lead down the Muses with me from the Aonian peak, provided life shall last; I first will carry back Idumaean palms to you, Mantua, and I will found a temple of marble in a green field, near the water, where the river Mincius wanders, vast with slow bends, and edges its banks with the tender reed. In the middle I will have Caesar and he will possess the temple: for him I as a victor and distinguished in Tyrian purple will drive a hundred four-horse chariots beside the river. For me, all Greece, leaving the Alpheus and the groves of Molorchus, will compete in running and with the rawhide glove. I myself, my head adorned with leaves of the trimmed olive, will bear gifts. Now, even now, it delights to lead the solemn processions to the temple and to see the slain bullocks, or how the scene withdraws with the sets’ rotation, and how the woven Britons lift the purple curtain. Te quoque, magna Pales, et te memorande canemus pastor ab Amphryso, vos, silvae amnesque Lycaei. cetera, quae vacuas tenuissent carmine mentes, omnia iam vulgata: quis aut Eurysthea durum aut inlaudati nescit Busiridis aras? cui non dictus Hylas puer et Latonia Delos Hippodameque umeroque Pelops insignis eburno, acer equis? temptanda via est, qua me quoque possim tollere humo victorque virum volitare per ora. primus ego in patriam mecum, modo vita supersit, Aonio rediens deducam vertice Musas; primus Idumaeas referam tibi, Mantua, palmas, et viridi in campo templum de marmore ponam propter aquam, tardis ingens ubi flexibus errat Mincius et tenera praetexit harundine ripas.

t h e c a p t o r’s i m a g e

118

in medio mihi Caesar erit templumque tenebit: illi victor ego et Tyrio conspectus in ostro centum quadriiugos agitabo ad flumina currus. cuncta mihi Alpheum linquens lucosque Morlorchi cursibus et crudo decernet Graecia caestu. ipse caput tonsae foliis ornatus olivae dona feram. iam nunc sollemnis ducere pompas ad delubra iuvat caesosque videre iuvencos, vel scaena ut versis discedat frontibus utque purpurea intexti tollant aulaea Britanni. (Vergil, Georgics 3.1–25) From the beginning, Vergil presents his temple as an imitation of Octavian’s in at least three interrelated ways: first through association with Apollo, second as an appropriation of Alexandrian Greek culture analogous to Octavian’s conquest of Alexandria, and third as verbal mimicry of the Palatine temple’s Greek and Alexandrian visual elements. From this perspective, the first thing to remark about the passage is its invocation of both Pales and Apollo. Pales, the Italian deity whose sphere of influence, flocks and herds, prefigures the major topics of Georgics 3, receives pride of place in a gesture appropriate to Vergil’s Roman position and interests. But she is accompanied by the “shepherd from the Amphrysus” (2), the Greek Apollo in his role as Νόμιος (“Shepherd”), a title recalling the time when he tended the mares of Admetus on the banks of the river Amphrysus in Thessaly. The invocation sets the coming ecphrasis squarely in an Apolline context, and contains the first of many allusions in this book (as throughout the Georgics) not simply to Greek parallels and precedents, but specifically to the Alexandrian master poet Callimachus (cf. Ap. 47–49).27 There is no need to belabor Vergil’s well-known Callimachean borrowings in the lines that follow. In the temple ecphrasis as a whole “reference to Callimachus . . . constitutes precisely a rejection of the letter of Callimacheanism.” Vergil, that is, announces a “departure from the chief authors and influences that have inspired his career to date” on grounds that “are partly Callimachean.”28 Lines 27

Thomas 1988: 2.37. Thomas 1988: 2.36–37. I make no attempt to engage here the question of whether or not the Georgics as a whole is a predominantly Callimachean poem (a view of which Thomas is the chief proponent). Very different emphases have characterized work since Thomas’s first exposition of this perspective. Farrell, for example, finds in the poem an “allusive structure . . . arranged in such a way as to build” to a “great Homeric conclusion” (1991: 256). Similarly, Morgan 1999 emphasizes Homeric allusion in the Proteus episode in Book 4 as the vehicle through which Vergil achieves the poetic triumph forecast in the proem to Book 3. Among studies affirming the importance of Hesiod and Lucretius, see esp. Farrell 1991: 131–206; Schäfer 1996; Nelson 1998; Gale 2000. On the whole question of intertextuality in the Georgics, see further Nappa 2005: 13–14. 28

Describing the Divine

119

4–5 are a case in point: the opening sentiment here, omnia iam vulgata, while recalling in general the Aetia preface, is virtually identical to that of Callimachus, Epigrams 28.4, “I hate all common things” (σικχαίνω πάντα τὰ δημόσια).29 Vergil thus lays claim to the Callimachean preference for the unfamiliar, only to reject, in the very next line, a myth recounted at the end of Aetia 2, that of the Egyptian king Busiris, killed by Heracles for his unacceptable practice of sacrificing strangers.30 Similarly, the rejected myth of Hylas (6), the handsome youth loved by Heracles but abducted by nymphs in Mysia, was a theme treated by Theocritus and the Roman “neoterics,” poets whose influence on the Eclogues we have remarked earlier, as well as by Vergil himself in Eclogues 6.43–44.31 Vergil next predicts his ascendancy as a Roman poet able, like Octavian, both to appropriate and to surpass the Greek tradition. This idea receives the greatest emphasis in the Ennian/Hesiodic claim that the way to “lift myself from the ground and fly, as a victor, through the mouths of men” is to “lead down the Muses with me from the Aonian peak” (Helicon, a mountain in Boeotia thought to be a home of the Muses). In the context of victory, the verb deduco possesses the fully Roman, military connotation of “to lead back in triumph”: the Greek Muses will become Vergil’s captives.32 In the sense of “to compose in an attenuated style,” the verb was also a touchstone of Roman poets adapting Callimachus’s “delicate Muse” (Aet. 1, fr. 1.24 Pfeiffer, Μοῦσαν . . . λεπταλέην) and so develops further the Alexandrianism implicit in the book’s first lines. The whole theme of victory combined with the temple metaphor resituates recognizably Pindaric motifs in the Roman context, while apparently also pointing to the third book of the Aetia.33 But such allusions to Greek praise poetry written for ruler-patrons inevitably shine a spotlight on Octavian’s political superiority in relation to both Pindar’s fifth-century patrons and the Alexandrian rulers whose entire realm Octavian has annexed to himself.34 Indeed, a cunning topical reference may reside in the fact that Idumaea was an eastern region sought after by Cleopatra but denied to her by Antony: by bringing “Idumaean palms” home to his native Italian city of Mantua, Vergil implicitly mimics Octavian’s influence over this eastern land.35 29

Thomas 1988: 2.38. Thomas 1988: 2.38. 31 Thomas 1983b: 94; 1988: 2.38; Ross 1975: 80–81 for the possible Gallan precedent. For further Callimachean references here, see Thomas 1988: 2.38–39. 32 On the triumphal language of this passage, see Buchheit 1972: 100–03; Boyle 1986: 46; Nappa 2005: 117. 33 Thomas 1988: 2.40–41. 34 Cf. Barchiesi 2005: 300. 35 I thank David Potter for pointing this out to me. On Idumaea, see Roller 2010: 121–22. The region was controlled, through a governor, by the client-king Herod the Great under both Antony and Octavian. Thomas 1988: 2.41 suspects a further Callimachean allusion in the word Idumaeas. On the significance of the Italian setting, see further Kraggerud 1998: 9; Barchiesi 2005: 299–300; Nappa 2005: 121. 30

120

t h e c a p t o r’s i m a g e

Vergil now begins to describe the temple that he says he will build on the banks of the river Mincius in Mantua. At its center will be a statue of Octavian, designated simply as Caesar (with no explicit mention of the statue as a work of art), and Caesar “will possess the temple” as its god. If the temple is a poem, this poem is in some way already Octavian’s: the point here is obviously not to insist on every aspect of this textual temple as a visual image subordinated to the controlling text of the Georgics, but to see the coming poem it represents as made possible by and belonging to the divine Octavian, just as Apollo made possible the Palatine temple and now possesses it. But in this way Octavian’s divine inspiration is already being felt behind the text of Georgics 3, especially since he has now in some sense adopted the role of Apollo Musagetes (“Leader of the Muses”) in presiding over the imaginary temple where Vergil will deliver the captive Muses. This is underscored by a reminiscence of Ennius’s Annales, whose concluding account of M. Fulvius Nobilior’s exploits seems to have ended with the latter’s foundation of the temple of Hercules Musarum in the Campus Martius.36 Vergil, in other words, picks up where Ennius left off, and bestows upon Octavian an Apolline function similar to the role played by Hercules (Antony’s patron) in the Annales.37 The ecphrasis of the Georgics, with its Greek and Hellenistic allusions, is presented as evidence of the Roman cultural achievement for which the Hellenophile Octavian provides the necessary circumstances and impetus. The idea of Hellenophilia is appropriately magnified and mythologized into a Roman god’s assumption of Greek poetic culture itself through the activities of his poetic “victor” and captor of the Muses, Vergil. In what follows, Vergil’s victory takes on an additional, athletic dimension appropriate to one who has staked a claim over the very people who founded the great Panhellenic Games. But, again, this is an imitation of Octavian: as a component of his planned victory celebration Octavian established Actian games meant to rival the Greek games in grandeur.38 Vergil says first that as “victor” he will drive (i.e., cause to be driven) a hundred four-horse chariots beside the river (17–18). The audience has been prepared for the conceit with the verb deduco earlier: Vergil’s chariot-driving, too, represents a kind of triumphal procession, but it is at the same time a chariot race through which Vergil claims the role of president of (and, proleptically, victor in) ceremonial games. Further, his Italian games will surpass the great Greek games just as Vergil himself outdoes the Greek poets. And Vergil will wear the quintessential Greek 36 Mynors 1990: 181; Meban 2008: 150–51; for this reconstruction of the end of the Annales, see Skutsch 1968: 18–20. 37 Cf. Meban 2008: 151 citing Skutsch 1968: 20. 38 Miles 1980: 173. For the general correspondence between the games described in Georgics 3 and those that regularly accompanied the dedication of a Roman temple, see Meban 2008: 153.

Describing the Divine

121

symbol of victory as he makes offerings: the crown of olive leaves awarded to winners at Olympia. Vergil’s imagery continues to suggest Octavian’s actual exploits through the introduction, now, of both stage drama and subject barbarian peoples. Imagining himself already present at the proceedings, Vergil expresses his pleasure in leading “solemn processions” and observing sacrifices at his temple (22–23) as well dramatic performances. Ludi scaenici or theatrical games often accompanied the dedication of a new temple at Rome, and this was almost certainly the case at the Palatine temple’s opening as well.39 Here the “woven Britons” appear to “lift” the rising curtain that marks the play’s end, the Roman custom. The Britons, moreover, symbolize the furthest extent of the empire and an area of hoped-for expansion following Julius Caesar’s invasions in 55 and 54 bce: Octavian planned a British campaign in 34, then again in 27 and 26.40 Vergil’s fictive Britons do Rome’s bidding in the task they seem to carry out. Their appearance on the curtains, moreover, is fitting in light of the way Vergil has sought to leave a Roman imprimatur on Greek practices and customs. Just as Roman references recontextualize Greek allusions at the outset of Georgics 3 so as to signal Vergil’s “victory” over Greece, the sign of a hoped-for Roman victory marks the conclusion of a stage drama, a genre in which Roman playwrights had long emulated Greek predecessors.

Doors and Statues: Shared Means As already mentioned, the Palatine temple had doors with elaborate relief-work images depicting the god’s victories, and the sanctuary housed a large number of impressive statues. So, too, Vergil’s temple has highly ornamented doors and lifelike, “breathing statues” (34) although these portray Octavian’s victories and Trojan lineage, subjects more suggestive of the Aeneid than of the Palatine temple itself. The painted Britons have prepared Vergil’s audience for the military exploits he now says he will depict: On the doors I will fashion from gold and solid ivory the battle of the eastern peoples and the arms of victorious Quirinus, and here the Nile billowing with warfare and flowing with a great stream and columns standing high with naval bronze. I will add the conquered cities of Asia and Niphates repulsed and the Parthian trusting in flight and arrows

39

Meban 2008: 153. Whether or not Vergil had specific performances in mind, Octavian is known to have enjoyed stage drama and to have cultivated it during his rule (Suet. Aug. 45). 40 See Mynors 1990: 183; Thomas 1988: 2.44. For Augustus’s planned expeditions (all three called off ), see Dio 49.38.2; 53.22.5, 25.2.

t h e c a p t o r’s i m a g e

122

fired backward, and trophies taken by twice-victorious hand from far-flung enemies and peoples from either shore triumphed over twice. in foribus pugnam ex auro solidoque elephanto Gangaridum faciam victorisque arma Quirini, atque hic undantem bello magnumque fluentem Nilum ac navali surgentis aere columnas. addam urbes Asiae domitas pulsumque Niphaten fidentemque fuga Parthum versisque sagittis; et duo rapta manu diverso ex hoste tropaea bisque triumphatas utroque ab litore gentis. (Vergil, Georgics 3.26–33) A combination of recent and hoped-for victories informs these lines. While the adjective Gangaridum (27, lit. “of the people of the Ganges”) refers in general to eastern peoples (perhaps those who fought with Antony at Actium), the Nile and “columns standing high with naval bronze,” that is, ships’ beaks, suggest more specifically the victory at Actium and the rostra set up before the temple of Divus Iulius.41 Vergil links Octavian to his adoptive father’s divinity, while stressing Octavian’s own victories rather than dwelling on the figure of Julius Caesar himself. The reference to the “cities of Asia” and the Niphates, a mountain range in Armenia, recall Octavian’s consolidation of eastern areas after Actium (or perhaps the latter is symbolic of a future eastern conquest), while victory over Parthia, not achieved until the return of Crassus’s standards in 20 bce, was both much desired and forecast repeatedly by Augustan authors.42 Lines 32–33 then revert back to generality: the “peoples from either shore” recall tribes mentioned previously. But the phrase also conjoins, through inversion, Vergil’s victory over Greece, a people near the center of the known world, with Octavian’s victories over those at its perimeters. Egypt, the first specific locale described on the temple doors, provides the transition, not surprisingly, since with the seizure of Alexandria Octavian could claim one of the greatest of the Greek-speaking capitals as his own. Alexandrian poetic allusion underscores this effect. Mention of the Nile “flowing with a great stream” (28) looks to Callimachus’s comparison of epic with the Euphrates (Ap. 108), a programmatic intertext perhaps already activated by Vergil’s Mincius “flowing vast with slow bends” (14). But here Vergil’s insistence on appropriating the epic poetics that Callimachus rejects is still more obvious: the full-flowing Nile will appear on the temple that represents Vergil’s 41 42

Thomas 1988: 2.44–45; Mynors 1990: 184. Thomas 1988: 2.45; Mynors 1990: 184–85.

Describing the Divine

123

forthcoming epic; rather than being a detriment, it will there help assure that Vergil’s poetry lives on in “the mouths of men.” Vergil now turns to his temple’s statuary: images of Rome’s Trojan ancestors and of Cynthian Apollo (34–36): And there will stand Parian marbles, breathing statues: the offspring of Assaracus and the illustrious members of the race sent down from Jove, and father Tros and the Cynthian founder of Troy. stabunt et Parii lapides, spirantia signa, Assaraci proles demissaeque ab Iove gentis nomina, Trosque parens et Troiae Cynthius auctor. (Vergil, Georgics, 3.34–36) In several ways these lines are the culmination of the whole ecphrasis’s engagement with Hellas. First, the statues will be made of marble from the island of Paros, a stone prized by the great Greek masters themselves. The detail constitutes as clear a symbol as one could wish of Vergil’s inclination toward Greek “material” as the basis of his Roman poetic endeavor. Second, as we shall be reminded yet again in Chapters 5 and 6, the Troy story is the principal mythic narrative linking Rome to the Greek world while at the same time differentiating the two.43 The Aeneid is of course the supreme cultural document articulating this richly ambivalent relationship, already operative in some of the earliest Latin literature. Here, only Assaracus, grandfather of Anchises, and his father Tros, a great-grandson of Zeus (cf. 35, ab Iove), are named, but this is enough to recall Ennius’s adaptation (Ann. 28–29 Skutsch) of the Trojan genealogy related by Homer’s Aeneas (Il. 20.215–40).44 The “race sent down from Jove” is the Julii, Octavian’s adoptive line. Vergil’s genealogy itself, then, tropes the literary filiation linking him to the greatest Roman and Greek epic poets. Further, the adjective Cynthius, which appears before this only in Callimachus and the Eclogues (and here in the reworking of the Aetia preface at Ecl. 6.3), is used to describe a statue of Apollo in an unassigned fragment of the Aetia (inc. lib., fr. 114.8 Pfeiffer).45 Thus the Apollo statue in Vergil’s poem-as-temple, the last image in his ecphrasis proper, appears to come directly from Callimachus while at the same time refocusing the audience’s attention on Octavian’s link to the god—this, too, in a temple over which Octavian presides as a deity himself and which, as poem, will assure Vergil’s own literary “immortality.” Vergil here concentrates the whole field of associations connecting his neo-Alexandrian 43

On the Troy story, see further pp. 17–20, 23 above. Cf. Thomas 1988: 2.46. 45 Thomas 1988: 2.46; Miller 2009: 2–3. 44

t h e c a p t o r’s i m a g e

124

poetics with the divine Octavian’s Rome-as-Hellenistic-capital.46 As the ecphrasis concludes, the Hellenistic aspirations embodied by real and fictive temples coincide, a gesture we will see the Aeneid itself repeating in another temple-ecphrasis: that of Apollo’s shrine at Cumae, placed, like the Georgics passage, at the virtual center of its poem.

Darker Notes: A Sight to Banish Envy So far in this chapter I have emphasized cultural synthesis—between Greek and Roman, religious image and text—as an unproblematic phenomenon producing coherent results. But previous chapters should alert us to the potential difficulties of this view. A reading of the opening of Georgics 3 as unalloyed panegyric falters especially on the fact that the three lines following Vergil’s ecphrasis proper contain a barely disguised reference to the circumstances of political competition about which much has been said in the last chapter (and whose shadow is readily discernible in Georgics as a whole: the key passage is 1.466–514 on the assassination of Julius Caesar and civil war). In fact temple building itself was one way the Roman elite had competed among itself, long and intensely, for popular favor. At Actium and after, Octavian had eliminated his main competitors, but his Palatine temple could nevertheless have been viewed as the most extravagant production yet in this ongoing struggle. Vergil himself implies as much: Dire Envy will fear the Furies and the grave stream of Cocytus and the twisted serpents of Ixion and the huge wheel and the rock that cannot be overcome. Invidia infelix Furias amnemque severum Cocyti metuet tortosque Ixionis anguis immanemque rotam et non exsuperabile saxum. (Vergil, Georgics 3.37–39) The splendor of Vergil’s temple will be such that envy—he hopes—can touch neither poet nor ruler.47 Yet new temples had certainly been the cause of envy 46

Cf. Miller 2009: 5: “Virgil makes politics and poetics converge perfectly in the rich phrase Troiae Cynthius auctor.” 47 For further discussion of the invidia theme here, see Meban 2008: 154, 156 with bibliography. The prominence of envy as a theme in Pindar’s epinicians suggests Pindaric allusion here again (Balot 1998: 86–87). Callimachus’s Epigrams 21.4 declares that in his poetry Callimachus “surpassed envy”; the conclusion of his Hymn to Apollo depicts Apollo conquering Envy, who must subsequently dwell “below,” just as Vergil’s Envy cringes in Hades (Thomas 1988: 2.46–47; cf. Thomas 1983b: 99–100).

Describing the Divine

125

among Rome’s political elite; their splendor had paradoxically come to stand in part for the forces destroying Roman social cohesion, in spite of the fact that the piety they expressed was based on hopes for social order and concord. And for all Vergil’s talk of triumph, warding off envy was actually ritualized in the triumphal procession though the ribald jokes shouted by the returning soldiers and the triumphator’s possession of various ornaments including a phallus.48 The unchecked military ascendancy of individual generals had been the prelude to civil war since the Sullan era. With Octavian at its center, the temple Vergil envisions may somehow overcome the disaster attendant on excessive invidia and usher in a new era of Roman peace and primacy in both the political and the cultural arenas. In 29 bce, however, no one could be sure that Octavian’s rule would last. Thus just as in the poetry discussed in Chapter 3, Greek culture, specifically in its religious dimensions, is a potential remedy that may also exacerbate the symptoms of social unrest. Octavian’s rise may mean Rome’s continued rise on the world stage, and a new era of Roman achievement in cultural domains. Or the gods could destroy Roman stability once again; Invidia could win. Octavian’s numinous power, hinted at in Eclogues 1 and invoked more forcefully at the opening of Georgics 1 itself (24–42), is in our passage an Apolline force inspiring the ecphrastic text itself and endowing its philhellenism—literally competitive—with a far broader import, just as the temple of Palatine Apollo would help to make a Hellenistic capital of Rome. In Propertius’s ecphrasis of this temple, to which we now turn, the divine power of Apollo is still more clearly an impetus for the text, even as Octavian’s actual achievements fade and blend into an elegiac background.

II. Propertius’s Temple of Apollo The Elegy of a Celebrant: Wit, Eros, Epiphany Propertius’s Elegies 2.31 is the narrative of a visit to the Palatine complex, with the features of the monument described in the order a religious celebrant would experience them, and culminating with a description of the cult statues in the temple’s inner sanctum.49 All this is wittily cast as an excuse to Propertius’s lover, 48

Meban 2008: 154. On the poem see esp. Boucher 1980: 48–51; Scivoletto 1981: 29–30; Laird 1996: 83–86; Klodt 1998: 3–11; Hubbard 1984; Welch 2005: 89–96; Cairns 2006: 340–41; Keith 2008: 28–29, 148–49; Bowditch 2009; Miller 2009: 196–206. I am especially indebted to Miller’s discussion for its emphasis on the subtle interactions between Propertius’s poem and the Palatine temple’s imagery, as well as on the poem’s concluding epiphany. 49

126

t h e c a p t o r’s i m a g e

Cynthia, for being late for a date: Propertius was so impressed by Octavian’s temple that he just lost track of time—but while he was there he experienced an epiphany. It would be naïve, however, to assume either that Propertius was unaware of the sanctuary’s political message or that the poem’s religiosity is merely a superficial joke. Propertius’s final couplet is of special interest to us, since here for the first time in extant Latin literature we find the notion of divine revelation—a manifestation of the god Apollo “himself ” (15, ipse)— used with reference to viewing a visual artwork.50 Although this idea is fleeting and not unambiguous, the whole poem expresses Propertius’s religious link to Apollo (part of what would develop into his role as priestly vates [“prophet”]) as a sign of his refined philhellenism, a posture both in harmony and in tension with Octavian’s own:51 You ask why I come to you rather late? The golden portico of Phoebus has been opened by great Caesar. So impressive was it to see, arranged with Punic columns between which stood old Danaus’s many daughters. †Here† a marble Apollo that seemed to me †at any rate† fairer †than Phoebus† himself looked as though he was opening his lips in song, his lyre silent. And around the altar stood the herd of Myron, four cattle, signatures of the artist, lifelike statues. Then in the middle the temple rose in brilliant marble, dearer to Phoebus even than his homeland Ortygia. On this, above the pediment, was the chariot of the Sun, and its doors were a masterwork of Libyan ivory: one lamented the Gauls cast down from the peak of Parnassus, the other the deaths of the children of Tantalus’s daughter. Then, between his mother and his sister, the Pythian god himself, dressed in a long robe, makes his songs resound. 50 Bacchus appears in an epiphany to Ariadne in the tapestry of Catullus 64 (cf. chapter 2), but this is the experience of an internal character rather than a viewer. On epiphany in Latin literature, see Feeney 1988: 104–07. On epiphany in Greek literature, including ecphrastic poetry, see above all Platt 2011. 51 My text of Elegies 2.31 is essentially that of Fedeli 2005, though the appearance of Heyworth’s radical edition of Propertius (Heyworth 2007a) has focused renewed attention on a pair of textual debates that in my view remain inconclusive: the unity of this poem with all or part of 2.32 and the possibility of a lacuna before line 5, which as printed above would seem to require, as Heyworth 2007b: 247 suggests, “some introductory explanation.” As he points out, an emended text reading hic equidem Phoebus visus introduces a doubtful homoeoteleuton. See Fedeli 2005: 876–77 for the difficulty of hic Phoebus Phoebo visus, which requires the elimination of the transmitted equidem. I follow Heyworth in reading artificis rather than artifices in line 8 and translating “signatures of the artist” (Heyworth 2007b: 247). For the possibility that the opening of Octavian’s porticus, which Propertius’s poem celebrates (1), occurred some years after the dedication of the temple itself in 28 bce (perhaps in 25 or 24), see Balensiefen 1995: 206–07, but contra, see Fantham 1997: 127 following Dio 53.1.3.

Describing the Divine

Quaeris, cur veniam tibi tardior? aurea Phoebi porticus a magno Caesare aperta fuit. tanta erat in speciem, Poenis digesta columnis, inter quas Danai femina turba senis. †hic equidem Phoebo† visus mihi pulchrior ipso marmoreus tacita carmen hiare lyra; atque aram circum steterant armenta Myronis, quattuor artificis, vivida signa, boves. tum medium claro surgebat marmore templum, et patria Phoebo carius Ortygia, in quo Solis erat supra fastigia currus, et valvae, Libyci nobile dentis opus: altera deiectos Parnasi vertice Gallos, altera maerebat funera Tantalidos. deinde inter matrem deus ipse interque sororem Pythius in longa carmina veste sonat.

127

5

10

15 (Propertius 2.31)

Ample discussion exists of Propertius’s interest here in poetic inspiration and the powers of his own genre: the elegiac text’s self-assertion over visual art.52 But even on first reading, the poem’s philhellenic cast is far more self-evident: the dazzled narrator, certainly to be imagined as the poet/lover Propertius “himself ” (see below), praises the lifelike beauty and power of the site’s artworks. These, he specifies, include sculptures by a Greek master as well as others depicting Greek history and myth, to which he refers in elliptical, Alexandrian fashion, with allusions to Callimachus rewarding the sophisticated audience’s scrutiny. Apollo, he says, will prefer this temple even to his Greek homeland in Delos (Ortygia). Thus from the beginning, Propertius prepares his audience to see the poem’s concluding epiphany as a sign of his link to Greece in relation to Octavian’s: hearing the voice, as it were, of this religious statue in particular, a work of Scopas, caps Propertius’s experience of the precinct’s other Hellenic objects and implicitly constructs Octavian’s philhellenism as both enabling context and foil for Propertius’s own. There is certainly acknowledgment of Octavian’s victories and imperial conquests: “Punic” columns and Libyan ivory recall the domination of Africa, and the poem’s Alexandrianism suggests, as in Georgics 3, the capture of Alexandria itself. But Elegies 2.31 testifies to Propertius’s special, elegiac connection to the Greek god Apollo in the latter’s new, Roman dwelling place. Rather than fervent celebration of Octavian, Propertius offers a glorification of his 52

See above, n49.

t h e c a p t o r’s i m a g e

128

religious sanctuary that nevertheless reads it in the terms already established by the genre of Roman love elegy. What Propertius claims to see is in part a monument to the travails of eros and sympathy with victimhood, as though Octavian had unwittingly glorified elegy in building the sanctuary the way he did! And insofar as Propertius identifies Apollo with his cult statue, he grants the latter a special power over the poem itself, a power incompatible with a simple notion of the ecphrastic text’s aesthetic dominance, since Apollo’s fleeting “appearance” at the poem’s end presumably inspires what Propertius has written. The first half of Elegies 2.31 (lines 1–8) unmistakably identifies Propertius’s experience of the site as that of a love poet—Hellenophile and Callimachean—in Rome. The opening question, “You ask why I come to you rather late?” (1, Quaeris, cur veniam tibi tardior?), recalls a situation familiar, by this point, to the poet’s audience: Propertius has come later than expected to his mistress, Cynthia, who has demanded to know the reason (cf. esp. Elegies 1.3 and 2.29). Further reinforcing this interpretation of the scenario is the fact that the poems immediately before and after (Elegies 2.30 and 2.32) address Cynthia by name, as do many other poems in Book 2. Propertius’s next words quickly reveal the cause of his lateness, as though he feels anxious to preempt the kind of tongue-lashings he receives, for instance, in Elegies 1.3.35–46 and 2.29B.31–38. The poet draws attention immediately to one of the Palatine site’s most splendid features, its “golden” portico (1–2, aurea Phoebi/porticus a magno Caesare aperta fuit). The epithet aurea here recalls elements of the actual structure: either the yellow giallo antico stone of the columns or, more probably, the gilded roof tiles and coffered ceiling.53 But for an audience thinking of the poem itself as an excuse delivered to a cross meretrix (courtesan) by an anxious Propertius, there is already something comical about the “great” Octavian’s massive building project being used for this purpose, its “gold” offered up, as it were, to appease a woman of this kind, and one whose wayward infidelity has been a recurring theme of Propertius’s earlier poetry. Any unseemly mercenary connotations, however, are mitigated here by the presence of Phoebi in line 1 and magno Caesare in line 2: if the notion of lateness signals a familiar erotic trope, Propertius foregrounds the site’s religious significance and acknowledges Octavian’s stature. A similarly playful mix of the erotic and reverential emerges from the very fact that the poem opens with this porticus. While likely to be the first part of the monument encountered by a visitor, and so helping to assimilate the poem’s narrative to the experience of a celebrant, the detail nevertheless subtly contributes to the poem’s overall erotic cast. Not long before this, in Elegies 2.23, Propertius has singled out porticoes as the first place where a nervous lover might bribe someone else’s slave to look for his grasping mistress, a girl eager to receive gifts but sparing with her favors (3–11, esp. 5–6, “what portico covers her now?” 53

Richardson 1977: 302; Fedeli 2005: 875.

Describing the Divine

129

[quaenam nunc porticus illam/integit?], spoken by the lover to the slave). In Elegies 2.32.11–16, verses that might in fact belong together with 2.31 as part of a longer poem mistakenly broken up by Renaissance editors, Pompey’s porticus, a place frequented by prostitutes, is where Propertius wishes the errant Cynthia would walk so he could keep an eye on her.54 The association of Pompey’s porticus with prostitution is famously memorialized in poem 55 of Catullus to whom Elegies 2.32.45 clearly alludes. The narrator of the Catullan poem, on the hunt for his dear friend Camerius, encounters a group of whores (10, pessimae puellae), one of whom exposes her chest and remarks tauntingly “look, he’s hiding here in my rosy breasts” (12). While the first lines of Elegies 2.31 suggest nothing sordid about Octavian’s new porticus, the wider connotations of such locales seem inescapable in a poem already linked to Propertius and Cynthia’s amorous encounters. But a contrast between Cynthia’s usual haunts and Octavian’s new portico can also be heard here, and in this case the expressed reverence of magno Caesare remains intact. As Propertius continues his account, his diction underscores a characteristically elegiac reading of the monument. He chooses to focus next (3–4) on the “Punic” columns (i.e., African giallo antico from Numidia) and the statues of the Danaids between them, images to which he refers, literally, as “the feminine crowd of old Danaus” (4, Danai femina turba senis). It has been assumed that “Punic,” that is, “Carthaginian,” is here simply a synonym for “African,” but this usage is in fact a rare one.55 (Ovid, by comparison, in his brief description of the site at Tristia 3.1.59–62 simply refers to their material as “foreign” [61, peregrinis columnis]). Long before the time of Augustus, of course, the original Carthage had ceased to exist: the city had been razed by Scipio Aemilianus in 146 bce and a new province, called Africa, created from Carthaginian territories. Gaius Gracchus, however, had founded a colony there, and although it was unsuccessful, a further colony planned by Julius Caesar was founded after his assassination and reinforced by Octavian in 29 bce. If Propertius recalls such reconstructive endeavors with the word “Punic,” he also associates Apollo’s porticus with a people famously defeated by the Romans and so anticipates his later emphasis on the sorrows of the defeated Gauls at line 13.56 In this way lines 3–4 contribute to a posture of distance from Octavian’s designs that the subsequent verses will only magnify. 54

On the possible unity of Elegies 2.31 and 2.32 see the recent discussion in Heyworth 2007b: 246. For their thematic connections, see Hubbard 1984; Welch 2005: 89–96; Bowditch 2009. On Pompey’s porticus and its representation in literature, see Kuttner 1999; Evans 2009. 55 Cf. Richardson 1977: 302. 56 On the nuances of Roman attitudes toward Carthage, see recently Gruen 2011: 115–40. It may be going too far to hear the word “Punic” as a veiled reminiscence of Dido’s bitter love story, appropriate in an erotic context, but the fact that the Aeneid was assuredly on Propertius’s mind as he completed book 2 makes such an echo seem at least possible.

t h e c a p t o r’s i m a g e

130

Still more obviously elegiac are the connotations of the phrase femina turba. To begin with, the word turba (“crowd”) has erotic associations at every other place it is used in Elegies, Book 2. A turba of men gathers around the Menandrian prostitute Thais at 2.6.3. In 2.29, a turba (3) of cupids accosts and binds a drunken Propertius as he wanders through Rome at night and compels him to return to Cynthia. In 2.32.8, the turba of Cynthia’s male admirers prevents Propertius from trusting her, and in the same poem (37) a turba of Hamadryads witnesses Venus and Anchises having sex. Given the male and/or divine identity of these elegiac turbae, Propertius’s reference to a femina turba calls particular attention to the incongruity of a group of mortal women on display in a public place. Whether or not this would have been enough to recall the association between porticoes and prostitutes evident in Catullus 55, it certainly aligns Elegies 2.31 with a broader elegiac interest in women who stand outside accepted rules of Roman conduct. The Danaids themselves, furthermore, are certainly at home in the world of erotic misfortune that also characterizes the genre of elegy.57 With the Danaids, Propertius contributes not only to the poem’s elegiac coloring but also to the focus on Greek culture that will characterize its remainder. One might think that referring to the daughters of Danaus was simply descriptive and unmarked in this context, until one realizes that Propertius did not have to refer to them in this way: in both the Ars Amatoria (1.74) and the Tristia (3.1.62), Ovid describes these same sculptures as “female descendants of Belus” (Belides), an epithet recalling a king of Egypt, while Danaus is simply their “cruel father” (Ars 1.74, ferus . . . pater) or “barbaric father” (Tr. 3.1.62, barbarus . . . pater).58 The Tristia’s combined emphasis on Egypt and barbarity is perhaps a recollection of the sculptures’ Actian resonance, appropriate enough in a poem of entreaty written to Augustus from exile.59 Propertius’s mention, however, of Danaus, the eponymous ancestor of the Argive Greeks (Homer uses both “Danaan” and “Argive” to refer to the Greek forces at Troy), foregrounds the legacy of Greece rather than of Egypt. The Greek emphasis becomes a trend in what follows. Although the next verse as transmitted in the manuscripts is notoriously corrupt, it is clear that its couplet contains an allusion to Callimachus’s Hymn to Apollo. The latter poem opens with Callimachus purporting to describe the god’s epiphany at his shrine in Delos. As part of a demand for holy silence during the paean or ritual hymn, the poet enumerates three wondrous examples of entities that become silent during this song (18–24): the sea, Thetis lamenting her dead son Achilles, 57

Cf. esp. Ov. Her. 14. But cf. Ov. Am. 2.2.4, “the battle-line of Danaus” [agmen Danai]). 59 Cf. Barchiesi 2005: 284: “Ovid’s downfall as a love poet is all in the trajectory from Danai agmen to ferus pater to barbarus pater: the monument is still the same, but readers are being taught different approaches.” 58

Describing the Divine

131

and the weeping rock into which Niobe was transformed as she grieved over the deaths of her twelve children, slain by Apollo and Artemis because of their mother’s boasting. Callimachus describes Niobe as “a marble stone like a woman gaping out some woeful sound” (24, μάρμαρον ἀντὶ γυναικὸς ὀïζυρόν τι χανούσης). Propertius adapts this description of Apollo’s victim to Apollo’s own marble statue in saying that it seemed “to open its mouth in song [lit. ‘gape out a song’] with its lyre silent” (6, marmoreus tacita carmen hiare lyra).60 Thus Propertius recalls and renews his claims in earlier poems to be a Roman Callimachean. The transposition of referent from Callimachus’s Niobe to the statue of Apollo in line 6 may seem jarring, yet it accords well with what we have already observed about the poem’s opening. The gesture is both Hellenophile and characteristically elegiac at once, picking up on both the Greek connotations of Danai in line 4 and a series of generic self-references diff used throughout lines 1–4. Propertius’s experience of the sanctuary will be that of a Roman Callimachean (note the stress on Propertius’s particular experience in 5, visus mihi). Further, the moment in Callimachus’s hymn to which he calls the most attention is the description of Apollo’s victim, Niobe. Her image in fact also appeared on the Palatine temple’s doors, as Propertius will be sure to emphasize only a few lines later (14). The subject of victimhood, already broached in the adjective Poenis (3), subtly resurfaces, even as Propertius offers unqualified praise of the site’s artwork. And here Propertius’s knowledge of Callimachus specifically, that is, the Roman poet’s refined Greek culture, provokes the typically elegiac response. Even as we read line 6 for its cultural subtext, we cannot overlook the way its couplet exalts “poetic artistry and its effect on our reception of the things that surround us.”61 Rivalry between poet and visual artist does perhaps run through these verses, via the conventionally ecphrastic emphasis on the wondrous verisimilitude of the art-object in question. Propertius apparently claims that the Apollo statue in the courtyard seems more beautiful than Phoebus himself. Although made of marble (marmoreus) and with its lyre “silent,” it seems to him to “open its mouth in song” (5–6, carmen hiare). Propertius thus calls attention both to his own role in exalting the statue’s beauty and to what he, as poet, is capable of (a real, audible carmen) as opposed to what a visual artist can do (create a silent work in marble). An ecphrastic text, displaying its own ambitions, asserts a form of aesthetic control over the image it describes. 60

Propertius follows Callimachus here both in the epithet marmoreus (cf. μάρμαρον) and in the rare use of hiare with the internal accusative (cf. ὀïζυρόν τι χανούσης). On the allusion see Heyworth 1994: 57–59; Barchiesi 2005: 285; Fedeli 2005: 878; Miller 2009: 202–4. 61 Welch 2005: 94.

132

t h e c a p t o r’s i m a g e

But the way Propertius handles the convention here actually sets up a contrast with the final image of epiphany—significant because while the cult statue at the end of the poem was (or was at least thought to be) a Greek work, the courtyard statue was very likely a Roman one.62 Propertius is careful to specify that the courtyard sculpture remains a “silent” thing of beauty and uses it to slip in a programmatic Callimachean allusion only enforcing the notion of “silence”: the rock that was Niobe, Callimachus says, becomes silent at the god’s approach. Thus, no matter how beautiful and lifelike, this “singing” statue remains strongly bound by the conventional notion of “silent” art enshrined in earlier Greco-Roman ecphrasis (note the direct juxtaposition of tacita and carmen in line 6). So Propertius creates a paradigm that the cult statue, which is Phoebus himself and does sing, will both recall and surpass: the first, non-Greek, image is useful as a vehicle for allusion to a Greek poet and for the exaltation of Propertius’s own poetic artistry, while the second, Greek, image becomes the agent of communion with the divine. The couplet that follows develops still further the poem’s ever more prominent philhellenism—now by deemphasizing the rivalry between the arts and focusing on praise of a Greek artist. Describing the sculpted cattle surrounding the courtyard altar, Propertius actually refers to their creator, Myron, a name he gives special prominence by placing it at the end of the hexameter (7). Propertius says, literally, that the animals “had stood still” (7, steterant), a verb that makes the statues seem as though they are real cattle made motionless by art, and so implicitly constructs them as wondrously realistic images. Propertius goes on both to underscore this effect by calling them “signatures of the artist” (8, artificis) and “lifelike statues” (8, vivida signa), words that call attention simultaneously to their mimesis of reality and the power of the artist who created them, a sculptor who was in fact famous for the verisimilitude of his animal figures.63 Of course the poet, as another artifex, can also be said to endow the sculptures with this (frozen) motion and life, and Myron’s sculptures, unlike Scopas’s at the poem’s end, do not produce the overwhelming effect of epiphany blurring the boundaries between visual and poetic art. Propertius treats Myron, 62 By “Roman” here, I mean a statue of recent construction made specifically for the Palatine complex (whatever the ethnicity of its artist). Pliny the Elder, who catalogues a number of genuine Greek pieces at the site including the cult statue (see above, p. 111), does not mention a statue in the courtyard among them. Heyworth, speculating that a lacuna before this couplet contained lines indicating that this statue was an image of Apollo with Augustus’s features, would increase the extent of Callimachean allusion here to include Callimachus’s panegyric of an unnamed Ptolemy (Heyworth 1994: 57–59; cf. 2007b: 247; Barchiesi 2005: 285). For a rejection of his views see Miller 2009: 204, who points out that the only such statue that we know of at the site was in the library, not in the courtyard. 63 Richardson 1977: 303 citing Plin. Nat. 34.57 and Petr. 88; Welch 2005: 92. I adopt Heyworth’s translation of artificis (2007b: 247).

Describing the Divine

133

however, with appreciative emulation rather than marked rivalry. The images in question are not directly associated with poetry or song, and the sculptor’s own abilities are the focus of praise. Propertius’s words take on a still more Hellenic cast insofar as they echo features of the numerous Greek epigrams on Myron’s most famous cow sculpture, originally placed in the Athenian agora.64 Within the thirty-six such poems preserved in the Greek Anthology (AP 9.713–42, 793–98), a number ascribed to authors either earlier than or roughly contemporaneous with Propertius, the two most common words applied to the image are “living” or “breathing” (ἔμπνους) and “to mold” or “to craft” (πλάττειν), terms easily heard to inform Propertius’s artificis, vivida signa, boves.65 The epigrammatists, further, often describe the lifelike cow as previously in motion and then stilled or fixed to its base by Myron (e.g., 9.719–20, 723, etc.), a conceit foreshadowing the implication of stilled motion in Propertius’s steterant. The tone of mixed reverence and wit that characterizes nearly all the epigrams also anticipates Propertius’s own: the great Greek sculptor has produced a “herd” (7, armenta) that looks as though it might amble off. And while a few of the Greek poets underscore the cow’s silence by saying, for example, that they expect it to low (724.1), others omit the theme of silence or take pains to undercut this effect, for instance, by insisting that the image’s silence is the fault of the bronze rather than Myron (728.2) or that the bronze actually lows (740.3). Not only does Propertius choose to emphasize the sculptures’ Greekness by naming Myron, but the way he praises them, passing over their silence and calling attention to their lifelike appearance, is recognizably Greek as well. Just as Myron’s sculptures have come from Greece to Rome, so Apollo himself has traveled in the couplet that follows (indeed, for Myron’s cow sculptures to come to life would be consistent with an epiphany of the god, since positioned before his temple they presumably represented sacrificial animals). Propertius now describes the Palatine temple itself rising in “bright marble” (9), a shrine “dearer to Phoebus than his homeland Ortygia” (10). Propertius thus emphasizes Apollo as a Greek god in Rome, and here too with a reminiscence of a Greek literary tradition as old as the Homeric Hymn to Apollo, which specifies (143–48) that out of all his temples, the one at Delos is the god’s favorite.66 Propertius’s earlier manipulation of a Callimachean intertext assures our awareness of both this additional motif and the Roman poet’s revision of it: Callimachus’s Hymn to Apollo begins with Apollo arriving at Delos, while Propertius’s poem concludes with Apollo’s epiphany in the Roman temple he 64

It was still in Athens in Cicero’s day (Cic. Verr. 2.4.135) though it was later taken to Rome. For recent discussion, see Squire 2010. 65 Welch 2005: 93. 66 For Fedeli 2005: 880–81 Propertius’s words “correct and update” this tradition.

t h e c a p t o r’s i m a g e

134

now prefers. Building on the notion of Apollo’s journey (and with further premonition of the epiphany) Propertius describes the sculpture on the temple’s roof as “the chariot of the Sun” with no indication of its material or status as an artwork: in quo Solis erat supra fastigia currus (11). The sun god has, as it were, already appeared in propria persona, and Apollo may not be far behind, given the two deities’ close association.67 The next couplet continues to foreground both Greek allusion and the imminent presence of the divine. Propertius first describes the doors as “a masterwork of Libyan ivory” (12, Libyci nobile dentis opus); as supreme art, the provenance of the material perhaps recalling verse 3’s emphasis on Carthage (Libya being virtually synonymous with Carthage as one of its earliest conquests). This emphasis on the defeated continues in what follows. One of the doors, Propertius says, “lamented the Gauls cast down from the peak of Parnassus,” the other “the deaths of the children of Tantalus’s daughter [i.e., Niobe]” (13–14). To have the personified doors lament their subject matter constitutes a further elegiac self-reference.68 At first glance, then, these lines may seem like another assertion of the text’s power to bestow a voice on mute art objects (cf. above on 6). The doors themselves, of course, do not speak, but Propertius’s poetry can make them do so. But note the Callimachean origins of each story: the Gauls’ thwarted attack on Delphi in 279 bce is the subject of a famous passage in the Hymn to Delos (171–85), while the grieving rock that was Niobe falls silent as Apollo approaches in the Hymn to Apollo, as discussed above. That is, in Propertius’s emphatically Greek source for this part of the ecphrasis, Niobe is neither exactly an artwork nor actually silent until Apollo draws near. Her petrifaction is itself a result of her contact with Apollo’s godhead, and to the extent that Propertius, in his Latin poem, renders her once again incapable of speech (the doors themselves taking up the lament ascribed to her in Callimachus), her silencing only points still more strongly to the notion of the Greek Apollo’s approach. With an abrupt switch to the present tense (16, sonat) the last couplet of Elegies 2.31 suggests that both the poet and his poem have passed into a different realm of experience, as well as into a new, transcendent relation to Greece. For now the image that anyone familiar with the temple would recognize as 67

Cf. Bowditch 2009: 411. Carole Newlands points out to me the relative stillness of the epiphany at the end of Propertius’s poem in comparison with the natural disturbances that accompany Apollo’s arrival at the opening of Callimachus’s hymn. The contrast should not surprise us, in my view, since Callimachus’s Apollo is the god to whom he is addressing a hymn, while Propertius’s Apollo is in some way Octavian’s first and Propertius’s second, so that the claim to have had real contact with this god has to be more muted. 68 Even if the verb maereo appears only once before this in Propertius’s work (2.9.11), mourning for the dead is one of his perennial motifs, as especially in the concluding poems of book 1. For more on mourning for the dead in Propertian elegy, see Dufallo 2007: 74–98.

Describing the Divine

135

the cult statue by Scopas—Apollo placed “between his mother . . . and sister” (15, inter matrem . . . interque sororem)—becomes “the god himself” (15, deus ipse). Here is Apollo citharoedus dressed in the traditional “long robe” (16, longa . . . veste) associating him with the world of poets.69 And this god “makes his songs resound” (16, carmina . . . sonat), as though no artistic medium, verbal or visual, separated Propertius from the divine source of his inspiration—the source, Propertius suggests, of the present carmen itself, which concludes with this verb.70 In the poem’s final carmina . . . sonat, the poet’s voice briefly merges with the god’s, the former becoming simultaneously a record and an echo of the latter and a display of its inspirational potency. Sonat is here a self-referential gesture to the sound of this poem as performed or read aloud, all the more so because ancient poetry was regularly voiced, even by a solitary reader.71 And this Apollo is unmistakably Greek, the oracular “Pythian” (16, Pythius), as emphasized by the first word of the poem’s final line. Thus to the extent that Propertius’s voice is Apollo’s, his entire poem hangs suspended between Greek and Roman, just as the Greek Apollo himself takes on a hybrid character in now preferring Rome to his native land. Octavian’s choice of the oracular god (confirmed by the presence of the Sibyl on the Sorrento base; see figure 4.2) was surely intended to signal the inauguration of a new era, a “golden” age of peace following years of bloodshed.72 Propertius, however, makes no mention of this: just as the poet/lover stands both inside and outside the public, political world of Octavian’s monument, so, too, his tie to Greek culture is both in harmony with the site’s purpose and imagery and subtly at odds with them. A statue of Apollo recalls the Callimachean Niobe; reliefs of Apollo’s victories conjure sympathy for the god’s victims. To an audience listening for a dissonant message, the pathos of civil war could be sensed in such details. But Propertius is careful never to say this in anything like explicit terms: rather, through an epiphanic connoisseurship, he responds to Octavian’s piety and political ideology as one who responds to Greece differently. This posture culminates with the idea of an inspirational self-revelation by the oracular Greek god of poetry himself.

Conclusion: Greek Culture, Octavian, and the Religious Image in Ecphrasis Just as the young Octavian, after his victories at Actium and Alexandria, co-opted religious art and architecture to lend credibility to his one-man rule, 69

See Fedeli 2005: 883. Even if Elegies 2.31 should be thought of as a unity with part or all of 2.32, the ecphrastic portion of the poem ends on line 16, a significant division in any case. 71 See above, p. 15. 72 Fedeli 2005: 883–84. 70

136

t h e c a p t o r’s i m a g e

so, too, Vergil and Propertius used ecphrasis of religious objects to help place their poetry on a new footing in Octavian’s Rome. Both the monuments and the poetry look to Greek and Hellenistic culture as a fountainhead of inspiration, and the gods who channel such effects become in turn the objects of ecphrasis as well as a generative source. This means that the Latin text by no means simply asserts itself over the image or seeks to dominate and displace it. Vergil’s imagined temple, even though it stands for a poem, recalls Octavian’s Palatine sanctuary of Apollo closely enough to suggest that the numinous power of Apollo’s protégé, Caesar, infuses the ecphrasis itself. Similarly, Propertius’s Palatine Apollo is, in the same moment, both a statue and the god who inspires an ecphrasis of his new, Roman dwelling-place, a home preferable to his favorite shrine on Delos. Ecphrasis of religious imagery here encounters the brute facts of civil war and imperial conquest that brought new regions of the Hellenistic world under Roman sway, but suspends final conclusions about these brute facts, and makes this the occasion for a new infusion of Greekness into Romanness, a re-articulation of Roman identity in Hellenic terms. If religion, empire, and Greek culture are inextricable in Vergil’s and Propertius’s temple ecphrases, it is because Octavian was rendering them newly so through his emergent cultural efforts. We turn now to Augustan epic’s complex engagement with the emperor’s still more developed cultural program.

5

Heroic Objects Ecphrasis in the Aeneid and Metamorphoses

Anyone reflecting for a moment upon Augustus’s use of Greek culture in his public program must observe a latent paradox. For Augustan Rome was to be, on the one hand, the glorious culmination of a struggle that began with Aeneas’s escape from Troy’s Greek conquerors—the embodiments of alien force and guile—and on the other hand, the Hellenistic capital par excellence, gathering nothing less than the entire gamut of Greek cultural forms into itself and instilling them with new life. A myth of laborious self-distinction from Greece exists in the public realm alongside competitive emulation of Greece, nowhere more plainly than in the visual iconography of Augustus’s rule. Consider, for example, that most characteristic of the emperor’s projects, the Forum Augustum (figures 5.1, 5.2). In its porticoes, very likely inspired by the Aeneid itself, stood sculptures of a Roman “Hall of Fame” including, on the left, Aeneas and members of the Julian gens and, on the right, Romulus. At the same time, the whole monument embodied “an innovative synthesis of Greek and Roman elements,” from its layout as a late Hellenistic temple square, to a pair of large paintings of Alexander by Apelles on public view, right down to column bases modeled on those of the Athenian Propylaea.1 Rome as non-Greek stands beside Rome as supremely Greek. Both ambitions are self-evident and inescapable. In Augustus’s vision, perhaps there was no paradox here.2 But the great Augustan epics, Vergil’s Aeneid and Ovid’s Metamorphoses, use ecphrasis to 1 Galinsky 1996: 199–200. For a detailed introduction to the Forum Augustum’s mixture of Greek and Roman elements, see his discussion at 197–213 (with 206 on the possible influence of Aeneid 6 and 8); cf. Zanker 1988: 210–15 and passim. 2 That is, his citizens could simply benefit from thinking of themselves as endowed simultaneously with Trojan pietas and worldly Hellenism, contradictions between the two notwithstanding. Cf. Gruen 1992: 31: the Troy myth both “enabled Rome to associate itself with the rich and complex fabric of Hellenic tradition” and “announced Rome’s distinctiveness from that world.” Some elite Romans in fact claimed Greek ancestry, while a Greek author living at Rome in the Augustan period, Dionysius of Halicarnassus, sought to prove that Trojans and Greeks were actually

137

Figure 5.1 Forum Augustum. Rome, late first century bce. © Shutterstock, ID: 10158994.

Figure 5.2 Reconstruction model of the Forum Augustum showing porticoes. Photo Credit: Alinari/Art Resource, NY.

Heroic Objects: Ecphrasis in the Aeneid and Metamor phoses

139

construct and appeal to an ambivalent viewer, one who might distinguish these two kinds of interaction with Greece as contradicting each other, and so both appreciate and resist control of Hellenic imagery as a part of Roman imperial self-definition. Indeed, Vergil and Ovid dramatize such ambivalence by having their ecphrastic images themselves resist control by the text; again, the images do not function merely as the text’s symbol or stand-in, but help to stage a complex receptivity of this kind. Like no Roman leader before him, Augustus sought publically a clarified (and correspondingly simplified) version of the relationship that has preoccupied us in this study, that between the Hellenic image and the Roman viewer.3 Vergil and Ovid, however, present a far less straightforward relation between images of the mythic past and the Augustan present. Here Greek culture’s power as a source of inspiration for Augustan aims meets with both confirmation and ongoing interrogation, as though the strictures imposed by official iconography were simply too confining for the Roman epic poet’s particular kind of engagement with Hellas. To capture this state of affairs, I propose that we think of Augustan epic ecphrases as offering “heroic objects.” That is, the armaments, sculptures, and other art objects depicted here do not simply belong to heroes and heroines within the text or represent heroism, but are themselves heroic because of their resistance to control: both by Augustan ideology and by the text itself. Vergil and Ovid, rather than attempting to fix, once and for all, the often contradictory meanings of the textual images they create, produce ecphrases whose very power lies in the breadth of associations they may conjure up in individual audience members, understood as subjects of a regime with more delimiting intentions over the Greek cultural legacy.4 The poets’ unrivaled regard on the epic tradition in fact leads them to perpetuate a notion of visual culture itself as a far less governable, and far more richly associative, realm than Augustus himself tried to render it. Thus if the ecphrastic image’s ramifying associations enhance an audience’s pleasure, such enjoyment derives in part from the way one set of

one people, so that any individual subject of Augustus might have chosen to read his monuments through a strictly genealogical lens (cf. Zanker 1988: 12–14; Gruen 1992: 7–8; Erskine 2001: 21, 24–27 and passim). On notions of Greek participation in Rome’s founding, see Gruen 1992: 6–51. 3 Cf., e.g., Zanker 1988: 112–13 (on Augustan culture in general): “The key messages were quite simple, and they were reiterated on every possible occasion, from festivals of the gods to the theater, in both words and pictures. Even the rich decorative program of the Forum of Augustus was built around a very few images.” 4 Barchiesi 2005: 293: “[S]ome of the intertextual energy of Augustan poetry is precisely about hammering together unbearable contradictions: you cannot really ‘say’ that Augustus is at the same time destroyer and protector, monarch and restorer of the Republic, but you can use traditional models to express this idea through allusion and recycling, and learn how to view the monuments through this aura of ambiguity and polysemy.”

t h e c a p t o r’s i m a g e

140

images, the epic text’s, competes with another, that of public monuments, and even seems to express its own sort of paradoxical agency over author, narrator, and viewing audience.5 By “agency” here I mean the image’s capacity to produce such effects above and beyond what the author or narrator seems at first to be saying, or what the audience seems initially encouraged to perceive. But let me offer some specifics. I will begin with close readings of the Aeneid’s major ecphrases, those of the temple of Juno in Dido’s Carthage, the temple of Apollo in Cumae, and the Shield of Aeneas, with reference as well to Vergil’s shorter description of the artwork on Pallas’s baldric. My main purpose will be to show how Vergil complicates Aeneas’s receptivity to these objects, especially to the aspects of Greekness they embody. We can trace the development of this Vergilian technique from Aeneas’s quasi-Odyssean response to the temple murals of Book 1 and the wrath of Achilles they convey, to the uncannily Aenean story of the Greek wayfarer and primal artist Daedalus portrayed in Book 6, to the Shield’s account of the battle of Actium, a scene recalling the earlier Greek ecphrases of shields belonging to Achilles and Heracles while portraying conquest over an ambiguously Hellenic foe. Response to Greek art remains problematic through the very end of the Aeneid, as a glimpse of Pallas’s baldric incites Aeneas to kill Turnus. Vergil sets in tension Aeneas’s progress away from Greek defeat toward a new Roman future and his transformation into a hero at once different from the Greeks and increasingly like them. Ecphrasis facilitates such derailing of the poem’s main narrative thrust by the way it stills narrative and opens up imaginative space for reflection, in this case on the implications of Aeneas’s resemblance to heroic figures preceding him in literary history. Insofar as Aeneas tropes the Roman viewer, and the images he examines recall Augustan monuments, Vergil thus suggests the ways in which Augustus’s use of archaic myth might undercut the message he himself intended, and the power of the Hellenic image overcome even the overmastering authority of the Princeps to control it. Turning to the Metamorphoses, I will show how its three proper ecphrases, the Palace of the Sun, the tapestries of Minerva and Arachne, and the cup of Anius, delineate an attitude remarkably similar to Vergil’s, in spite of the differences in the two poets’ overall techniques and outlooks. Even when Ovid is not recounting the Aeneas legend, as in the first two examples, he foregrounds issues of direct relevance to the way its narrative of self-distinction functions in a larger mythic context. He also alludes to the Aeneid’s own ecphrases, as well as to at least one of the Augustan monuments they recall, the temple of Palatine 5

On the nonpassivity of art in Augustan epic, cf. esp. Elsner 2007: 113–31, e.g., 115 (on Pygmalion’s statue in Ovid’s Metamorphoses): “The ivory statue (which we may view as a figure for the poem) may have been created by Pygmalion the sculptor, but it generates him as a viewer-lover, just as the Metamorphoses generates us as its readers.”

Heroic Objects: Ecphrasis in the Aeneid and Metamor phoses

141

Apollo. Thus the Palace of the Sun (Sol or Phoebus as closely identified with Apollo in Augustan poetry) presents images of cosmic order that seem to go unheeded by the reckless Phaethon, engrossed in his own self-defining—and self-annihilating—quest, while Arachne’s fateful, likewise self-destructive challenge to Minerva in a tapestry evocative of the Metamorphoses itself puts front and center the issue of art’s independence from “official” myth-making. These two passages prepare Ovid’s audience for the cup of Anius, which sheds doubt upon the seemingly incontrovertible value of Aeneas’s self-sacrifice, by juxtaposing it with another self-sacrificial narrative, that of the Orionids, and beyond this with the pathetic story of Anius’s own daughters. In Ovid’s universe, no mythical image, even one crafted by a deity, can exist “in isolation from a web of stories and connotations”6 to which Ovid provides sometimes dizzying access. Ecphrasis, here fused with the device of metamorphosis though “the shared feature of illusionism”7 offers an ideal vantage point from which to perceive this aspect of Ovid’s poem. Ultimately these images’ startling resistance to the text’s dominance both creates and derives from a distinctive set of paradoxes, none exclusive to epic, but concentrated here to a degree not found elsewhere. Thus they are simultaneously the “same” as the text and its visual “other.” Mere objects, they seem to “live” and “breathe,” possessing a wondrous life-force not simply bestowed by the self-celebrating text but also emanating from that of gods and heroes actually worshipped by Augustan Romans. They are both sacred and profane, even obscene, and semidivine. They create their own trajectories through the epic’s narrative world, and “trespass” within it, both hastening and diverting its overall course.8 As microcosms of the text, they do enhance its order, yet as journeyers, as it were, from earlier texts, especially Greek epic, they disrupt this order. Redemptive and dangerous, they are culture-bearers and culture-destroyers. They serve mighty rulers (including Augustus) and resist their overmastering authority. They are, in short, heroic. This chapter thus develops further our understanding of a theme broached in the previous one: ecphrasis, far from being simply the servant of empire, may through its supposition of the image’s primacy subject empire’s very tools to intensive reassessment.

6

Barchiesi 2005: 293–94, who notes that this is also a feature of Greek art objects in Augustan Rome. 7 Hardie 2002: 176. On the centrality of ecphrasis to the Metamorphoses, cf. Hardie 2002: 177 (treating the parallelisms between Ovid’s cosmogony and Homer’s Shield of Achilles—on which cf. below, pp. 150, 153–54, 162): “If this universe is a work of art, of that magical, Hephaestean and Daedalean, kind endowed with the power of movement, then all particular narratives and descriptions within the universe are examples of ecphrasis.” 8 Cf. Boyd 1995.

142

t h e c a p t o r’s i m a g e

I. The Backward Glance: Ecphrasis in Vergil’s Aeneid Transfixed by the Trojan War: Dido’s Temple to Juno A tension like that which I have described between self-differentiation and rivalrous self-transformation comes momentously to the fore in the Aeneid’s very first ecphrasis, depicting the murals on Dido’s temple to Juno in Carthage (A. 1.453– 93).9 For here the Trojan, proto-Roman Aeneas, shipwrecked in the course of his struggle to reach Italy, observes pictures of the Trojan War, and in his response to them combines aspects of the Homeric Odysseus with a salient foreshadowing of his own Achillean role once landed on Italian soil. The ur-moment, in other words, of Roman conflict with Greek culture is made to coincide both with a considerable setback in the Trojans’ journey and with one of Aeneas’s major steps toward becoming, at least potentially, a new, transformed version of a Greek epic hero. Yet as Aeneas “reads” his own story and even views an image of himself—a detail without parallel in other classical ecphrases—he both feels himself elevated as a non-Greek and becomes more Greek, at least in the eyes of a learned audience keen to compare him with Homeric and other Greek models. So the images’ own ambiguously heroic role takes shape. A staging that highlights the “spellbinding effect of ‘wonder’ on the viewer at both beginning and end”10 (456, miratur; 494, miranda), the episode could have reminded Vergil’s audience of its own experience of public imagery, particularly the Troy story as already incorporated into official propaganda.11 Vergil’s choice of this subject for an ecphrasis perhaps even creates the expectation of a paean to Rome’s combined uniqueness and competitive success in areas formerly dominated by Greek achievement. While Augustus, however, might insist on the smooth complementarity of these two positions, Vergil problematizes their coexistence in the extreme. To grasp this feature of the passage, we do well to consider its place in the overall narrative of Book 1. The Aeneid as a poem of self-distinction from Greece in fact starts off on very uncertain footing indeed. The Trojans’ story begins with their sailing from Sicily to Italy (1.34–35). The crossing should be brief 9 Out of the vast bibliography on the passage I have found the following especially helpful: Johnson 1976: 99–105; Segal 1981; duBois 1982: 32–35; Clay 1988; Leach 1988: 311–19; Williams 1990; Fowler 1991; Lowenstam 1993; Barchiesi 1994; Putnam 1998: 23–54; Beck 2007; Lowrie 2009: 161–62; Dekel 2012: 9, 83–86. It has become conventional to refer to Dido’s images as “pictures” (cf. 1.464, pictura), “paintings,” or “murals,” though in fact Vergil does not specify what medium we are dealing with here. For discussion, see Boyd 1995: 81–83. 10 Segal 1981: 76. 11 Although the Forum Augustum would not be dedicated until 2 bce, the Aeneas myth was already a part of Octavian’s visual propaganda, and before him of Julius Caesar’s. See Zanker 1988: 35–36.

Heroic Objects: Ecphrasis in the Aeneid and Metamor phoses

143

and bring them significantly closer to their goal, but it is interrupted by Juno, who instigates the storm that drives them off-course to Carthage. We learn later that the Trojans have in fact landed on Italy before this, only to be warned off by an omen of four horses grazing in a field (3.537–38). Anchises interprets this sight as predicting war with the local population (3.539–40), who are Greek colonists, perhaps of the “evil” sort feared by the prophet Helenus (3.398): “We leave behind the homes of Greeks and the suspect lands” (3.550, Graiugenumque domos suspectaque linquimus arva), Aeneas reports to Dido in Book 3. Thus the trajectory carrying Aeneas away from the Greek destruction of his city toward the Roman future actually begins at a moment of serious uncertainty: not simply in medias res but stalled, stilled, its momentum halted without any certain notion of how it will resume. And Aeneas, when we first see him, is far along in the process of becoming a Trojan version of the Odyssean wanderer at his most vulnerable, threatened by storm, sea, and strange ports of call (cf. 1.32, “and for many years they wandered” [multosque per annos/errabant]).12 The temple ecphrasis of Aeneid 1, however, helps construct Aeneas’s transformation into a second Odysseus as not only a factor contributing to the Trojans’ delay but also the best hope for its resolution. Carthage will, as every reader knows, be only a way station for the wandering Aeneas, just as his foreign ports are for Odysseus. The invisibility that permits Aeneas to enter Carthage recalls most directly the shroud of mist wrapped around Odysseus by Athena as he approaches the Phaeacians’ city (Hom. Od. 7.14–15), and in his immediate reactions to the images on the temple of Juno, Aeneas becomes especially like Odysseus after the latter, visible once more, has been received graciously at the Phaeacian court.13 Aeneas will remain invisible as he surveys Dido’s images, yet Vergil’s hero unmistakably reenacts the emotional outburst of his Homeric predecessor encountering his own story rendered in song.14 While at the Phaeacians’ banquet, Odysseus hears the bard Demodocus sing three songs on the Trojan War, of which Vergil’s ecphrasis reworks and expands upon the first and third.15 Vergil calls attention to the parallel in especially self-conscious terms, since Aeneas’s perusal of the temple’s Trojan War images tropes his own and the audience’s scrutiny of the relevant Greek texts, not only Demodocus’s songs in the

12

See esp. Clay 1988: 197–98. Cf. Clay 1988: 198; Dekel 2012: 83–86. 14 Many have noted the parallel. See esp. Knauer 1964: 166–67, 376; Putnam 1998: 47–54; Beck 2007. 15 While Knauer 1979: 166–67 and Putnam 1998: 47–54 emphasize only the connections between Vergil’s ecphrasis and the first song of Homer’s Demodocus, Beck 2007, rightly presuming that “[s]everal different strands of Homeric allusions can and often do exist simultaneously in one passage of the Aeneid” (541n31), has recently pointed out its allusions to the third. See Beck 2007: 543–45. 13

t h e c a p t o r’s i m a g e

144

Odyssey and the narrative of the Iliad, but also other moments from elsewhere in the Epic Cycle and in other Greek poetry.16 Demodocus’s first song tells of the quarrel between Odysseus and Achilles that delighted Agamemnon because Apollo had foretold it, presumably as indicative of Greek victory (Od. 8.72–82). The third song tells of the Trojan Horse and the destruction that follows its entry into Troy (499–520).17 After each of these songs, which include mention of his own exploits, Odysseus weeps, and this is Aeneas’s reaction, too (Verg. A. 1.459, lacrimans), as soon as he sees the murals that likewise include a representation of himself (488). Although Odysseus reacts to a song and Aeneas to visual images, for an audience steeped in Homer, Aeneas’s Greekness determines in part how the Troy story affects him. In other ways, however, Aeneas’s reactions diverge from the Homeric model. For while Odysseus experiences overwhelming grief at the memory of war’s suffering, the weeping Aeneas feels both sadness and, for the first time (primum), hope and confidence (450–52), since the fame of Trojan “hardship” (labor) appears to him to have spread to every corner of the world (459–60). Dido’s images, he asserts, are “virtue’s own reward” (461, sua praemia laudi), and the Trojans’ fame, he feels certain, will now bring them safety (463). In this context he utters the nearly untranslatable words sunt lacrimae rerum et mentem mortalia tangunt (462, “tears fall for men’s lot, and mortality touches the heart” [Austin]), and bids his companion Achates “put aside” fear (463). All of this would be unthinkable in a Homeric warrior, to whom the idea of gratification from widespread awareness of one’s failure and suffering would be completely alien. Duly pius, Aeneas thinks first of his compatriots (note esp. 463, tibi), and his words seem to indicate the consciousness of a greater purpose linking him to the Roman future.18 To this extent, the images’ effect on him is heroic in a positive sense, an impetus toward further struggle in which he will ultimately succeed. Yet Aeneas is surely deceived, bested by the evocative power of these objects. As often observed, Vergil’s audience, aware of Juno’s anger against the Trojans from the opening lines of the poem, must find it inherently unlikely that a temple erected to the goddess would celebrate the victims of her wrath, even if Vergil’s extended account of Aeneas’s words and emotions does not so much finalize his “misinterpretation” as distance his viewpoint from the 16

For the non-Homeric events, see Austin 1971 ad loc. Aeneas also recounts the sack of Troy to Dido in Aeneid 2, while Demodocus’s second song, on the love affair between Hephaestus and Aphrodite, is hinted at by the cosmic song of Iopas at Dido’s banquet at Aeneid 1.742–46 (see Putnam 1998: 47–54). 18 “What, precisely, is Aeneas so happy about?” was the question aptly posed by W. R. Johnson in his influential discussion of the passage ( Johnson 1976: 103). With his emphasis on Aeneas’s self-deception, cf. Clay 1988: 197n7 on Aeneas’s having discovered the “traces of humanity” in the far-flung land of Carthage. 17

Heroic Objects: Ecphrasis in the Aeneid and Metamor phoses

145

audience’s own.19 Readers of Chapter 3 may recall Propertius’s arch, subjective reading of victimhood on the temple of Palatine Apollo, and the indeterminate place between Greekness and Romanness it helps the elegiac poet construct for himself. In Aeneid 1, the very lines that perhaps most distinguish the Trojan Aeneas from the Greek Odysseus nevertheless fail to locate him securely in a proto-Roman sense of mission. Aeneas is even too Odyssean (i.e., too Greek) here, if his sympathy with the defeated Troy recalls Odysseus’s grief even after hearing of his own success (Od. 8.519–31), a sentiment suggesting his “pity for all war’s victims” and identification with those experiencing “suffering that he himself has caused.”20 Significantly, Homer compares Odysseus, as he weeps in response to Demodocus’s third song, with a woman weeping over her husband fatally wounded while defending his own city (523–30), just as Aeneas sees himself doing in the images on Dido’s temple. If Aeneas were not misguided, Vergil suggests, by Odyssean sympathy with the victims (including himself), he might see that the images express Juno’s favor for Carthage, a divine connection indicating danger, not safety, for the Trojan refugees.21 Vergil complicates still further any simple identification of Aeneas with Greek, Trojan, or Roman models by the way he forecasts Aeneas’s coming battle in Italy through the details of the ecphrasis itself. Greek savagery and Trojan suffering dominate the pictures, although Vergil gives no sure indication whose perspective, his own or that of Aeneas, governs their nonchronological presentation as described in the text (according to Vergil’s narrator, they appear “in order” [456, ex ordine] on the temple itself).22 Achilles’ bloodthirsty wrath is operative in all but three of the passage’s eight visual episodes, as though to foreground explicitly Vergil’s coming adaptation of the Iliad (the poem of Achilles’ μῆνις) in Aeneid 7–12.23 And through such intimations of the narrative to come, the changes in Aeneas continue to compete with his epic “progress” as the force driving Vergil’s poem. The subtext of the murals is indeed “the metamorphosis of Aeneas” into the Achillean hero of the poem’s second half: his coming “mutation into a destructive Greek.”24 Vergil’s brilliant use of the ecphrastic convention that transforms still images into a motion-filled visual narrative may be an implicit commentary on the power of his textual art, but it more clearly enhances the sense of ambiguous 19

See Fowler 1991: 32–33 responding to the discussions of Stanley 1965, Johnson 1976: 99–114, Leach 1988: 311–19, Clay 1988, Horsfall 1990, and O’Hara 1990a: 35–39. 20 Garvie 1994: 339. 21 In Homer, the woman to whom Odysseus is compared fears for her own safety, specifically the threat of being enslaved by her husband’s killers (Od. 8.527–29). 22 For the implication here of Aeneas’s “participation in the construction of what he is viewing” (Bartsch 1998: 337), see esp. Segal 1981; Leach 1988: 318; Laird 1996: 89; cf. Clay 1988: 202. 23 Noted especially by Stanley 1965; Clay 1988: 200–05; Bartsch 1998: 336–37. 24 Putnam 1998: 43.

t h e c a p t o r’s i m a g e

146

metamorphosis just described. Ecphrasis, by endowing the still images with life, foreshadows the transfixed Aeneas’s own inexorable motion toward his Achillean behavior in the poem’s latter books.25 Achilles first appears at the end of the three-line overview at 456–58, “the Atreidae and Priam and Achilles savage toward both” (458, Atridas Priamumque et saevum ambobus Achillem). Here his wrath, though emphasized, is conveyed in relatively static terms, but with his second appearance at the conclusion of the battle-description ending the first episode, Achilles’ violence has been put in motion: “Crested Achilles assails [the Trojans] in his chariot” (468, instaret curru cristatus Achilles). The words are the last unit of a chiasmus heightening the effect of motion-from-stasis by juxtaposing the fleeing Greeks chased by the young men of Troy with the Trojans fleeing before the most terrifying Greek warrior (467–68). Stillness becomes chaotic activity. The account of Rhesus intervenes, before Achilles appears again as opponent of the mortally wounded Troilus, “unequal to Achilles in combat” (475, impar congressus Achilli). In this, perhaps the most striking image of the passage, Troilus’s listless body is dragged behind his chariot, while “the dust is inscribed by his inverted spear” (478, versa pulvis inscribitur hasta). The detail intimates the role of inscription or writing in creating the very image it describes, but within this image, another kind of writing, pathetic and ineffectual, is central.26 Greek violence, moreover, here creates such useless writing by the hand of Troilus, whose very name was connected in antiquity with “Troy,” specifically its downfall.27 It is as though Vergil fears that the outbreak of Greek-style destructiveness, conveyed by the images, has the power to render his own poem and its Trojan narrative impotent. They risk becoming a mere “empty picture” (464, pictura inani), as he has his narrator refer to Dido’s murals while the groaning Aeneas “feeds his mind” (animum . . . pascit) upon them (Troilus’s chariot is also called inanis at 1.476).28 In this connection it is no accident that Troilus’s death at Achilles’ hand looks forward to Aeneas’s wrathful slaying of the young Lausus in Aeneid 10.815–20.29 The inscribed trail, that is, of Troilus’s demise leads the audience onward in Vergil’s written text toward a moment of violence threatening to invalidate the entire epic’s promise of a brighter future for the Trojans 25

With my discussion, cf. esp. Putnam 1998: 27–36. Cf. esp. Heffernan 1993: 27–28; Putnam 1998: 30–31; Lowrie 2009: 161. 27 For Troilus’s “writing” as “useless,” see Putnam 1998: 31. For the ancient association of “Troy” and “Troilus,” see Boitani 1989: 4–5. 28 The words pictura inani here have drawn much comment. With my reading cf. esp. Beck 2007: 539: “The images are literally without physical substance, but the spectrum of uses for inanis in the Aeneid implies that they are also somehow not living up to the hopes or expectations that Aeneas has expressed about them.” See further Boyle 1972: 143; Arkins 1986: 42; Bartsch 1998: 336; Putnam 1998: 25. 29 Gransden 1984: 103–05. 26

Heroic Objects: Ecphrasis in the Aeneid and Metamor phoses

147

and for Rome. In their threat to the Aeneid itself, the “empty” temple images especially evince the ruinous, culture-destroying side of their heroism. The mobilized wrath of Achilles continues to infuse the images on Dido’s temple—and, through their anticipatory valence, the Aeneid itself—as Vergil brings his first ecphrasis to a close. In its fifth episode, Achilles appears dragging the dead Hector around the walls of Troy and ransoming his body to Priam (483–87). The circular path of Achilles’ chariot tropes the visual scene’s circumscriptional force, especially since Vergil will rework it at the poem’s very conclusion, when Turnus, now in the role of Hector to Aeneas’s Achilles, supplicates him just as Priam does Achilles.30 The seventh episode (489) depicts Memnon, the Ethiopian king killed by Achilles while attempting to aid the Trojans, while the eighth and last (490–93) portrays the Amazonian queen Penthesilea, also slain by Achilles after falling in love with him. Vergil likens her story, which anticipates that of the warrior maiden Camilla in Aeneid 11, to Dido’s by transitioning directly from the ecphrasis to Aeneas’s first, dazzled sight of the Carthaginian queen (494–97), which is also the moment Aeneas becomes visible once more. So Aeneas changes again, this time into the soon-to-be vanquisher of Dido, even as Vergil transfers our own gaze from the image to the poem’s internal “reality.”31 As the Aeneid’s initial ecphrasis concludes, Aeneas’s destructive, Achillean course has already begun. Suspended between Odysseus and Achilles, between Troy, Greece, and Rome, Aeneas in Book 1 already bears the contradictory pressures of his ktistic mission and Vergil’s rivalry with Homer. But unlike Homer’s Odysseus at Scheria, Aeneas at Carthage must contend with the special interpretive problems of visual art, and through ecphrasis, Vergil dramatizes the pressing “need to interpret” that still, silent images impose upon the narrating subject, not least the Augustan Roman faced with mythological iconography.32 Aeneas’s confusion as he gazes on the images in Dido’s temple is enough to call into question the very interpretation of such objects.33 For while Homer depicts Odysseus as neither isolated nor confused in his interpretation of Demodocus’s songs, the invisible Aeneas stands virtually alone (except for the silent Achates) as he misinterprets the representation of Greek victory and Trojan defeat. In contrast with the world of the Odyssey, that of the Aeneid is one in which “understanding art is a problematic endeavor that may in fact be impossible.”34 Already in Book 1, Vergil raises the question of whether or not Augustan Rome might also be such a world. 30

See below, p. 158. Putnam 1998: 43. 32 For the phrase “need to interpret,” I am indebted to Fowler 1991: 28. 33 Beck 2007. 34 Beck 2007: 534. 31

t h e c a p t o r’s i m a g e

148

Contemplating Daedalus With the Aeneid’s second major ecphrasis, Aeneas has begun to fare better as proto-Roman journeyer. He has reached Italy, and Anchises’ ghost has spelled out in the clearest terms yet both his larger labor, war with a “tough people” (5.730, gens dura) in Latium, as well as his more immediate task: to find the Sibyl and proceed with her to Anchises in the underworld, where he will learn of his “whole race and what walls will be given” (5.737). Vergil might have sped his hero directly on to this goal, but Aeneas becomes engrossed by the images on Apollo’s temple at Cumae, sculptures fashioned by the originary Greek artist, Daedalus.35 Though in some sense primal, even Daedalus’s identity cannot be thought of as a universally fixed Greek reference point against which to assess Aeneas’s emergent proto-Roman one, since as a figure from myth Daedalus appears to have Near Eastern roots. Though he is presumably Cretan, moreover, in the Shield of Achilles episode from the Iliad (see below), Homer’s lack of specificity about his origins allowed him to be repatriated to Athens in the fifth century bce.36 For his part, the migrant Easterner Aeneas now “reads” a story not his own, but strangely similar to it: the autobiography of Daedalus himself, a tale of murder, sacrifice, unbridled passion, escape, and wandering. In a text alluding both to the Ariadne ecphrasis of Catullus 64 and Callimachus’s Hymn to Delos, Vergil foregrounds Aeneas’s intense interest in what he sees. And through reference to Icarus’s absent image, Vergil draws still more pointed attention than in the Carthaginian ecphrasis to the perspectives of poet, narrator, and audience. Thus at a moment where Aeneas’s Odyssean role would seem to be aiding his progress toward a Roman destiny—Aeneid 6 as a whole repeats the nekuia of Odyssey 11, in which Odysseus receives essential guidance from the ghost of Tiresias—Aeneas’s receptive interest in the Daedalean images allows a pair of new Greek models to cast a problematic shadow on his whole undertaking: Daedalus, failed artificer and tragic wayfarer, and Theseus, perfidious deserter of a hapless love. The temple ecphrasis of Aeneid 6 unmistakably links Aeneas’s viewing to the experience of an Augustan audience through a double allusion to the Palatine temple of Apollo, which the last chapter has treated in detail. The Sorrento base (figure 4.2) depicting the Palatine cult statues also includes an image of the Sibyl, whose oracular power undergirds the Octavianic/Augustan notion of a new saeclorum ordo, heralded in Eclogue 4 and formalized at Augustus’s celebration of the Secular Games in 17 bce.37 Reinforcing this association is Aeneas’s vow to 35

On the passage, I am indebted especially to Segal 1965: 642–45; Boyle 1972: 116–19; Pöschl 1975; duBois 1982: 35–41; Fitzgerald 1984; Bartsch 1998: 335–36; Putnam 1998: 75–96; Barchiesi 2005: 282; Lowrie 2009: 158–60; Miller 2009: 133–49. 36 On Daedalus’s Near Eastern origins and appropriation by Athens, see Morris 1992. 37 Augustus actually moved the Sibylline books from the temple of Jupiter Optimus Maximus to that of Palatine Apollo. For discussion of this event in connection with Augustan epic, see Feeney 1991: 217.

Heroic Objects: Ecphrasis in the Aeneid and Metamor phoses

149

build a temple to Apollo in Latium at Aeneid 6.69–70 (soon after the ecphrasis), words “associatively aligning” Augustus’s temple and Daedalus’s Cumaean shrine.38 Further, the “Roman tradition of Apolline and Sibylline authority,” appropriated afresh by Octavian in the dedication of this temple, receives pregnant anticipation in Aeneas’s entire Cumaean visit.39 Indeed, an audience of Vergil’s Augustan contemporaries might have had every reason to anticipate the episode itself as a tangible step toward Romanness that is simultaneously an improvement upon Greek precedents, since the Odyssean Aeneas approaches the Cumaean shrine, in reality the site of an archaic Greek site of prophecy, “in the guise of a Greek settler looking for a colonization oracle.”40 And the Sibyl does in fact direct Aeneas to the next stage of his journey (82–97). Such positive associations, however, are swiftly undermined. The very first adjective used to describe Daedalus, “fleeing” (14, fugiens), conjures up immediately the darker aspects of his story, about to come into view on the temple doors. The Greek context of Daedalus’s fugitive journey, beginning in “the kingdom of Minos” (14, Minoia regna) and ending in the “Chalcidian citadel” (17, Chalcidicaque . . . arce, Cumae having been founded by Euboean settlers from Chalcis) casts Greekness itself in a doubtful light. So, too, does the hint of artisanal transgression, soon to be much magnified, in Daedalus as one “having dared to entrust himself to swift wings” (15, praepetibus pennis ausus se credere) and the strange phrase “oarage of wings” (19, remigium alarum) denoting his ingenious conveyance. (Compare the way that Catullus in poem 64 signals the transgressiveness of the first ship, another Greek invention, with the unexpected word currum [9, “chariot”].) The first word indicating an image on the temple doors is even “death” (20, letum), here the fate of Androgeus, Minos’s son killed by Aegeus at Athens, the cause of the ghastly tribute of seven youths and maidens demanded as victims for the Minotaur. Greek art and artistry—ambiguously heroic objects once more—embody a resistant, countervailing force to that impelling Aeneas toward the fulfillment of his goal, even as the artistic narrative in question portrays an earlier Greek’s completed journey to Italy, and is in this sense an impetus toward success. The parallels with Daedalus in particular require further comment. Like Daedalus, Aeneas is an exile who has recently come from circumstances of forbidden love, though while Aeneas has performed the Thesean act of abandonment, Daedalus, we now learn, had been both implicated in deception through his creative powers and moved to pity, traits connecting him far more closely to Vergil and his narrator. The description of the second temple door, 38

Miller 2009: 137. Barchiesi 2005: 282. 40 Barchiesi 2005: 282. On the oracle or μαντεῖον and its association with Greek colonists, see Norden 1957: 117–18. 39

t h e c a p t o r’s i m a g e

150

depicting events at Crete, begins with an image of the “cruel love” (24, crudelis amor) felt by Pasiphae for the bull, the “deception” (furto) of the artificial cow that Daedalus crafted to allow their coupling, and that “hybrid birth and two-formed offspring” (25, mixtumque genus prolesque biformis), the Minotaur. The words mixtum genus here threateningly suggest the perils of miscegenation in an episode where hybridity between Greek and Roman is of the essence. Another, literally forbidden, mingling of races is recalled by the Minotaur’s epithet, “reminder of an abominable love” (26, Veneris monimenta nefandae) which looks back to the love of Dido and Aeneas, prevented by Jupiter and in this sense nefas.41 Certainly Daedalus’s role in abetting the monstrous union and then hiding its murderous progeny in the labyrinth (27) can only reflect negatively on Greek ingenuity, and perhaps all Greek artistry, at this point in the Aeneid. This is all the more true for the contrast that Vergil thereby creates with Daedalus’s role in the preeminent Greek ecphrasis, the Shield of Achilles in Iliad 18, which he will soon imitate directly in the Shield of Aeneas. In Homer’s ecphrasis, Daedalus is recalled for having fashioned a dancing floor for Ariadne (Il. 18.590, χορὸν), to which Homer likens the one Hephaestus crafts on Achilles’ armament. A site of unqualified joy, harmony, and beauty, as finely dressed youths and maidens dance in circles and rows, while a “great crowd” (603) takes pleasure in their performance, the Homeric dancing floor could not be more different from the Daedalian inventions described by Vergil. The Greek master craftsman clearly has two sides to his creative talent in Greek tradition, of which Vergil calls particular attention to the one associated with transgression, guile, and pain. But what this means for Vergil’s own project is never fully spelled out: is the Greek background of the Aeneid, which as a poem necessarily involves its own kind of verbal deceptiveness, thereby indelibly tainted by this, one of the epic’s central images? Does Vergil, as Daedalean artist, thereby consign himself to failure? Or does he eventually surpass Daedalus at the Aeneid’s conclusion, by shining a glaring spotlight on his own hero’s responsibility for Turnus’s death, to which he, as poet, denies the protective concealment of art?42 The Cumae ecphrasis forces such questions upon the audience. Yet Vergil’s narrator seems profoundly, though ambivalently, receptive to one positive aspect of Daedalus’s character: his pity for tragic love, both Ariadne’s for Theseus and perhaps even Pasiphae’s for the bull. Not surprisingly, it is at precisely this moment that his voice comes most to resemble that of Catullus’s narrator in poem 64, whose deep, conflicted identification with Ariadne and other thwarted lovers we have examined in Chapter 1. As is well known, Aeneid 6.27–30 allude clearly to Catullus 64.112–15, set within another ecphrasis 41 42

Cf. Bartsch 1998: 334–35; Putnam 1998: 86. I extend here the argument at Putnam 1998: 95.

Heroic Objects: Ecphrasis in the Aeneid and Metamor phoses

151

recounting Ariadne’s story.43 In particular, Vergil’s odd designation of Ariadne as regina (she was, of course, a princess, not a queen) reinforces the parallel between herself and Dido, while the word’s appearance just after the account of Pasiphae and before any direct mention of Ariadne allows for the suggestion that Pasiphae, the real queen of Crete, is the object of Daedalus’s pity.44 We can agree that if deceit “is the chief impulse behind Daedalus’s initial fabrications, pity rules him in their undoing.”45 So, too, pity for a Greek mythological character overwhelms the Vergilian narrator, stricken, now, by the absence of Icarus’s image. In an another gesture reminiscent of Catullus 64, the narrator resorts to apostrophe of Icarus himself to convey the pathos of his fate: “You also would have a great part in so great a work, did grief, Icarus, allow” (30–31, tu quoque magnam/partem opere in tanto, sineret dolor, Icare, haberes; compare the apostrophe of the Greek heroes at Cat. 64.22–30 and above, pp. 44–46). The pathos mounts still higher as we are asked to imagine Daedalus’s unsuccessful attempts to include his son’s death in the sculptural tableau: “twice he had tried to fashion [Icarus’s] fall in gold; twice the father’s hands fell” (A. 6.32–33, bis conatus erat casus effingere in auro,/bis patriae cecidere manus). Just as Catullus’s deeply emotional ecphrasis included lengthy accounts of what did not, prima facie, seem to have been included in the imagined Greek artwork, so, too, Vergil dramatizes “a model for viewing” in which the narrator “supplies the picture (and us) with that which is not on its surface through a subjective act of interpretation.” The “absent” Icarus, who is not really absent thanks to this intervention, turns out to play a “great part” in the ecphrasis, if not in the sculpture it imagines, through the swelling emotions he creates in the receptive viewer.46 Bartsch, moreover, notes Vergil’s distance here from the “optimistic and perhaps claustrophobic Augustan view of art as a force to contain political unrest.” Another pictura inanis, breaking “the bounds of all ideological binaries,” Daedalus’s sculptures invoke “the impossibility of dictating artistic interpretation even as Augustus begins his turn to an ideological artistic program at Rome.”47 The pathos of a young Greek’s exilic journey cut short casts its shadow on Aeneas’s own, all the more so since Aeneas has recently lost his own father, Anchises, while en route, as well as his beloved Dido. Daedalus’s loss of his son, moreover, anticipates Vergil’s coming emphasis on Augustus’s loss of his own heir, Marcellus, at the conclusion of the underworld “Parade of Heroes” later in book 6 (860–86).

43

Casali 1995: 5–8 with bibliography. Fitzgerald 1984: 56; Putnam 1998: 86–87. 45 Putnam 1998: 88. 46 Bartsch 1998: 336. 47 Bartsch 1998: 336, 339. 44

t h e c a p t o r’s i m a g e

152

Yet even pity for the victims cannot detain Vergil, his narrator, or his hero for long. As though through an awareness of the threatening example posed by Daedalus’s sculptures to Aeneas’s progress, Vergil now has him whisked away by the Sibyl, who appears, accompanied by Achates, before Aeneas and his companions have had a chance to see everything on the temple doors: Yes, even would they have scanned everything sucessively with their eyes, if Achates, sent ahead, had not returned and together with him the priestess of Phoebus and Diana, goddess of the crossroads. quin protinus omnia perlegerent oculis, ni iam praemissus Achates adforet atque una Phoebi Triviaeque sacerdos. (Vergil, Aeneid 6.33–35) The Sibyl seems mainly preoccupied with observing the correct time for sacrifices to Apollo (38–39), yet something in her tone also suggests dismissiveness over Aeneas’s visual fascination: “‘This is not the time for such sights’” (6.37, “non hoc ista sibi tempus spectacula poscit”), she insists upon arrival. Scholars have long speculated about the Sibyl’s motivations, which is to say Vergil’s, in insisting upon a hasty departure. Casali, however, was first to remark that Ariadne’s abandonment, the next stage in her myth as recorded in Catullus 64, could plausibly be imagined as what is kept from Aeneas’s—and the audience’s—sight. But whether or not the missing images should be construed as “what the voice of Virgil cannot say about Aeneas,”48 there is no doubt that Vergil prompts the audience to fill them in, just as the narrator has filled in the story of Icarus a few lines earlier. Vergil encourages his audience to become receptive in this way; but what larger conclusions they are to draw he again passes over in silence. Such is the dual force of Daedalus’s heroic artworks. Their labyrinth contains both Minotaur and guiding artifice compelling Aeneas on to the future in store for him.

Stayed by the Shield The extension of that future into Roman history is the subject of the Aeneid’s longest ecphrasis, the Shield of Aeneas at 8.626–728.49 So much of its imagery,

48

Casali 1995: 7. I have benefitted especially from Becker 1964; Anderson 1969: 72–74; Williams 1981; duBois 1982: 41–48; Hardie 1986: 97–110, 336–76; Farrell 1997: esp. 224–225; Putnam 1998: 119–88; Bartsch 1998: 330–32; Faber 2000; Barchiesi 1997: esp. 274–75 and 2005: esp. 282–83; Reed 2007: 56–57, 123–24; Lowrie 2009: 164–66. 49

Heroic Objects: Ecphrasis in the Aeneid and Metamor phoses

153

in fact, is specifically Roman—from the infanthood of Romulus and Remus to Augustus’s Actian victory and triple triumph—that many of its details lie beyond the immediate scope of my discussion here. Nevertheless, as a potent foreshadowing of Aeneas’s victory over his foes in Latium at the same time as it constitutes a massive symbol of the fact that he is “becoming another Achilles,”50 the passage demands our attention, especially insofar as the Achillean metamorphosis it heralds seems confirmed by the Aeneid’s final—and shortest—ecphrasis, the description of the baldric of Pallas at 10.497–99. As in the Daedalus ecphrasis of Book 6 though now more emphatically, Vergil opens a gap between Aeneas, wondering at the shield in ignorance of what its images mean, and the Augustan audience, who can appreciate both the object’s Greek allusions and its Roman historical narrative. Such appreciation comes, however, even as the audience watches Aeneas slotted into the role of the supreme Greek warrior, an identity that he never fully escapes and that famously returns all but to engulf him in his final act of battle fury, triggered by a glimpse of Pallas’s baldric, at the poem’s finale. The Homeric Shield of Achilles, moreover, shares its place as a prime Greek intertext with the Pseudo-Hesiodic Shield of Heracles, on which the theme of battle’s destructiveness is still more in evidence. Reminiscences of this latter object lend Aeneas still another Greek identity as a second Heracles destined to triumph over impious adversaries, though at devastating cost. Thus a disturbing receptivity to a Greek trait, the blood lust already attributed to Achilles in the Aeneid’s first, Carthaginian, ecphrasis, begins to come full circle in the “untellable fabric” (625, non enarrabile textum) of the Vergilian shield’s circular form. And through this pattern of transfixed viewers and transfixing Hellenic objects, Vergil poses momentously the question of whether Rome, like Aeneas, will repeat its own battle fury so as to precipitate further civil wars, or Augustus’s regime, with all its claims to surpass Greek precedent, will in fact bring peace. The Shield of Aeneas confirms what our earlier readings in this chapter should have led us to suspect: the Aeneid’s ecphrases are not ancillary to, but deeply involved with, the poem’s most pressing themes and problems, not simply as a poem but as an Augustan epic, a response to the emperor’s broadest ambitions for the Roman empire as successor to the Hellenic world. Aeneas’s shield, like Achilles,’ comes to him at a crucial moment as he prepares to return to a battle in which his side will ultimately prevail (though Achilles, of course, dies before this outcome, while Aeneas lives to found a city, Lavinium, named after his Latin wife). While Homer, however, has us watch Hephaestus/Vulcan at work on Achilles’ shield, Vergil brings forth the god’s fresh handiwork fully formed, and so focuses our attention on the “moment of

50

Farrell 1997: 225.

t h e c a p t o r’s i m a g e

154

reception,” the “receiver’s act of contemplation.”51 And Aeneas’s “wonder” (619, 730, miratur) over the depicted events he knows nothing about comes emphatically to trope that of an informed Augustan audience, since the Shield’s narrative culminates in Augustus’s supremacy, even placing him at the object’s very center (675, in medio). Vergil’s audience can identify not only events but also particular localities portrayed, especially the Palatine, such a major focal point, as we have seen, of Octavian/Augustus’s self-promotional building program. Indeed, Aeneas’s shield employs the Palatine as framing device, since the Lupercal, the cave where the she-wolf suckled the twins, was thought to have been located at its base, while the temple of Apollo, actually depicted in the ecphrasis’s final scene, graced its summit.52 As we shall see, one feature of Apollo’s Palatine sanctuary about which we have had much to say in connection with Propertius 2.31, the Danaid portico, will have a special bearing on our discussion. With the heroic Shield of Aeneas’s Palatine associations in mind, let us pass directly to its longer, second movement (which is to say, its central zone), narrating Actium and its aftermath. The first great contrast to draw with the Shield of Achilles concerns the relative placement of analogous features on the two objects, as well as their overall emphases. Homer’s Shield depicts two unnamed cities, one at war, one at peace, while the stream of Ocean forms its encircling motif. An ecphrastic icon of the poem in which it appears, it thus sets the poetic narrative’s dominant focus on a particular war within a larger perspective emphasizing in general terms the activities “of prosperous, settled societies at peace.” Its separate vignettes depict not only battle but also marriage, judicial arbitration, farming, vine-tending, song, and dancing, “the Homeric picture of the good life.”53 All this, moreover, is fitted into a vision of an impersonal cosmic order signified by the image of the cosmos with which the ecphrasis begins (Il. 18.483–89), as well as by Ocean at its conclusion (607–08). Putnam well describes Vergil’s modification of this basic pattern: by placing a single historical event, the Battle of Actium, at his Shield’s center, he intimates that “Actium alone, in its importance for the poet’s new cosmos, deserves the preeminence that Homer accords to all human endeavors.” Underscoring the compliment is Vergil’s transfer of water imagery from the Homeric shield’s rim to his own shield’s center, to become the fluid medium in which the epoch-making conflict takes place. Actium’s “forceful particularity” is here shown to have “preserved mankind for what Homer’s shield illustrates as its ordinary existence.”54

51

Putnam 1998: 168. Putnam 1998: 121. 53 Taplin 1980: 12. 54 Putnam 1998: 137. On the universalizing tendencies of Vergil’s Actium-scene, see further Hardie’s discussion (1986: 97–110, 336–76) of its Gigantomachic imagery and overall “blend of cosmic allegory and political allegory” (342). 52

Heroic Objects: Ecphrasis in the Aeneid and Metamor phoses

155

Vergil’s water imagery, however, is among his more self-evident borrowings from the generally far more gruesome scenes on the Shield of Heracles.55 There, an image of silver dolphins rushing through the water in pursuit of terrified fish (Scut. 207–12) falls between two extended narratives of blood-spattered conflict and carnage. Vergil begins his Actium narrative, marked likewise by pronounced imagery of bloodshed, with a nearly identical image of dolphins “sweeping the sea with their tails and cutting through the seething deep” (A. 8.674).56 The specific conflict, moreover, depicted after this on the Shield of Heracles, Perseus’s battle with the Gorgons (Scut. 216–37), anticipates Vergil’s construction of Actium as a hero’s fight with a monstrous foe. For Antony and Cleopatra’s “barbarous” forces (A. 8.685, barbarica), backed by “monstrous shapes of gods of every kind and barking Anubis” (698, omnigenum deum monstra et latrator Anubis), recall that “terrible monster” (Scut. 223, δεινοῖο πελώρου) the slain Gorgon whose head Perseus carries, and her “terrible and unspeakable” sisters, the other Gorgons, who rush in pursuit.57 In particular the two snakes of death that pursue Cleopatra (A. 8.697) mirror the two serpents that hang down from the Gorgons’ girdles (Scut. 233–34), while the numerous allegorical figures including Fear, Terror, Battle-Din, Murder, Fate, and others, present in both the Perseus episode and elsewhere on the Shield of Heracles anticipate the similar beings Bellona (an archaic war goddess), Mavors (the archaic name for Mars), Discordia, and the Dirae, who populate Vergil’s Battle of Actium.58 Actium proves to be far more than a focal point for Homeric generalities; it is also a sinister reprise of the specific conflicts portrayed by Pseudo-Hesiod. One could not ask for a more trenchant illustration of the ambiguities of Greek culture as received in Augustan Rome. For Cleopatra was, of course, last in the great line of Hellenistic rulers who had presided over Alexandria’s heyday as a Greek cultural and literary capital. Among the Aeneid’s most stunning occlusions is Vergil’s unrelenting characterization of Cleopatra and her forces as merely “barbaric”; this concerning a woman whose intelligence and sophistication, as Plutarch attests, were among her irresistible charms (Plu. Ant. 27). On the opposing side, Vergil groups Greek, Italian, and Roman together almost as one identity, so as to make the Roman virtually an impersonation of the Greek within this part of the Shield’s narrative.59 And recollections of the Gigantomachy underscore the connection between Greek conceptions of cosmic ordering and Augustus’s victory over a barbaric foe, with Augustus playing the role of 55

See Faber 2000. Faber 2000: 54. 57 For the “unspeakable” as another prominent Vergilian borrowing from the Shield of Heracles, see Faber 2000: 49. 58 Only three such figures, Strife, Tumult, and Fate, are on the Shield of Achilles (Il. 18.535); cf. Faber 2000: 54–55. 59 Cf. Reed 2007: 105. 56

t h e c a p t o r’s i m a g e

156

Zeus/Jupiter in defeating the champions of chaos.60 Thus Cleopatra’s husband, Antony, first appears “with barbaric host” (A. 8.685, ope barbarica) as “victor from the peoples of the East” (686, victor ab Aurorae populis) and leads “Egypt and the power of the Orient and furthest Bactra” (687–88, Aegyptum virisque Orientis et ultima . . . /Bactra). Cleopatra is his “Egyptian wife” (688, Aegyptia coniunx), who summons her forces with the Egyptian rattle, the sistrum (696) associated with the worship of Isis, while the episode’s most memorable tableau consists of the Egyptian pantheon arrayed against the Greco-Roman one, monstrous variety epitomized by “barking Anubis” pitted against the ordered forces of Neptune, Venus, and Minerva (698–99). At the sight of Actian Apollo, “every Egyptian and Indian, every Arab, all Sabaeans turned their backs in flight” (705– 06) as the Nile waits to receive its conquered minions within its “secret streams” (713). Nothing Greek here, only the alien East. Nor anything Roman—to this extent Vergil cedes to the official presentation of Actium as a triumph over barbarism rather than—what in fact it was—a decisive moment of civil war. But such clear-cut dichotomies between barbarian and Greek, barbarian and Roman, and East and West, cannot stand, and the receptivity to Greek culture staged around the Shield of Aeneas can only be described as ambivalent. For all that the Shield absorbs the “heroic glory that is associated with Achilles’ weapon,”61 it must convey the Pseudo-Hesiodic implication that war is “monstrous, irrational, an activity proper to beasts in which man also engages.”62 Vergil tempers its triumphalist glance toward the Roman future (brought out especially by its contrast with the brief description of Turnus’s shield [7.789– 92], which depicts only the myth of Io, his distant ancestress) by loading it with contradictory associations from the literary past.63 If Aeneas as a second Heracles has acquired a civilizing, beneficent stature through the story of Cacus earlier in Aeneid 8, Vergil now (if not already in the Cacus myth itself) casts this Greek parallel, too, in a questionable light.64 Indeed, the Greekness of Aeneas’s

60

See above, n54. As Nelis 2001 points out, Gigantomachic imagery is also one vehicle by which Vergil introduces recollections of the Cloak of Jason ecphrasis in Apollonius’s Argonautica, so as to foreground the inseparability of the martial and the erotic, one of Apollonius’s key themes. Allusions to Apollonius further suggest Vergil’s awareness of an interpretive tradition that read the Shield of Achilles as an allegory of Empedoclean Love and Strife, a tradition of which Apollonius also seems to have been aware (Nelis 2001: 345–59). On the Gigantomachy elsewhere in Roman ecphrasis, cf. above, pp. 16–20. 61 Faber 2000: 57. 62 Thalmann 1984: 63. 63 On Vergil’s brief ecphrasis of the Shield of Turnus in the context of the Aeneid’s other ecphrases, see esp. Breen 1986. See further Gale 1997. 64 Putnam 1998: 162 notes the further problems in the verbal parallels that Vergil draws between Hercules and Cacus and the connection he creates between Cacus as Vulcan’s son and Aeneas’s armor as the god’s handiwork.

Heroic Objects: Ecphrasis in the Aeneid and Metamor phoses

157

shield is bivalent in manifold ways: Heraclean and Achillean, constructive and brutal, authorizing and doubt-producing. As an artwork embedded in the Greek epic past, it is rendered still more suspect by having been produced by Vulcan after a night of lovemaking with his seductive but unfaithful wife, Venus.65 Like similar aspects of the poem’s earlier ecphrases, the Shield’s Greekness confronts the Augustan audience with a far wider scope of cultural references than their ruler’s visual program was designed to convey. And Aeneas’s assumption of an Achillean identity, through accepting the shield as his armament, remains problematically linked to his status as a viewer of Hellenic art even as the Aeneid concludes. Upon killing Pallas, Turnus strips from his body “the tremendous weight of his baldric with its imprinted sacrilege” (A. 10.496–97, immania pondera baltei/impressumque nefas).66 Scholarly consensus has identified this image as the Danaids’ murder of their bridegrooms, although in characteristically Alexandrian fashion it goes unnamed as such: on a single marriage-night the band of youths foully slaughtered and the blood-strewn bedrooms, which Clonus Eurytides had chased with much gold. una sub nocte iugali caesa manus iuvenum foede thalamique cruenti, quae Clonus Eurytides multo caelaverat auro. (Vergil, Aeneid 10.497–99) As with Daedalus’s sculptures, we are in the presence of Greek handiwork: the artist’s Greek name, “rout-of-battle, son-of-Eurytus,” signals his origins and perhaps his identity as one of the Greek colonists brought by Evander.67 We do not learn what Turnus thinks of the image—or whether he notices it at all—but the Vergilian narrator is moved to direct comment, again, just as he was with Daedalus’s reliefs. Bewailing the “ignorant mind of men,” knowing nothing of the future and of moderation in success, the narrator predicts that Turnus will wish for “an untouched Pallas bought at great price” and rue “these spoils and this day” (504–05, spolia ista diemque). This emotional outburst prompts the most careful consideration of the baldric’s thematic resonance.

65

Cf. Putnam 1998: 169–80. For discussion of this object see especially duBois 1982: 31–32; Barchiesi 1984: 33–34, 71–72; Conte 1986: 185–95; Breen 1986; Spence 1991; Bartsch 1998: 334–35; Putnam 1998: 189–207; Reed 2007: 54; Lowrie 2009: 162–63. 67 For possible identifications of the Eurytus in question, see Putnam 1998: 241–42n16. 66

t h e c a p t o r’s i m a g e

158

Vergil richly rewards such scrutiny. For in addition to its foreshadowing of Pallas’s untimely death, his baldric becomes the impetus for Turnus’s slaying, the victorious act through which Vergil’s audience nevertheless comes to see Aeneas “reincarnating Achilles killing Hector in the guise of Turnus.” Turnus himself, conversely, here becomes an “image of Priam before Achilles, save that the conquering hero now shows no mercy to his petitioner.”68 Aeneas, is of course, carrying his Achillean armament as he approaches the poem’s final confrontation (cf. “fierce in arms” [A. 12.938, acer in armis]). Certainly its hostile, Greek energies can be felt as he encounters the baldric, another object of a previous ecphrasis. Hesitating as to whether to spare the plaintive Turnus or not, Aeneas becomes “inflamed with fury and terrible in wrath” (946–47, furiis accensus et ira/terribilis) “after he drank in with his eyes the reminder of savage grief, the spoils” (945–46, oculis postquam saevi monimenta doloris/exuviasque hausit); that is, the baldric. The words oculis hausit indicate the most violent response to an artistic image of any treated in this book, the culmination of a series of different reactions to art on Aeneas’s part: first mingled grief and hope (Carthage), then fascination (Cumae), then ignorant wonder (the Shield), and now finally murderousness.69 And it is here that Aeneas fails most unequivocally to live up to the Roman ideal voiced by his father Anchises’ ghost, “to spare the submissive” (6.853, parcere subiectis), words spoken, significantly, within the most categorical distinction between Greek and Roman values in the entire Aeneid, the famous delegation to “others” (i.e., Greeks) of the arts of sculpture, rhetoric, and astronomy (6.847–50). “Anchises is right to attempt such a banishing” insofar as “art does not work, can never work, solely for the purposes of the ideology that produces it, as the very text of the Aeneid makes clear.”70 Aeneas’s reactions at the epic’s conclusion are the case in point. The Aeneid intimates a place for Greece, even for a Greek bloodline, within the mingling of peoples that will assure the Trojans’ future in Italy, since Evander’s subjects “will be absorbed into the new nation that is eventually to become Rome.”71 Too Achillean, however, while at the same time reminiscent of Roman violence in the civil wars, Aeneas’s final response to a Greek artwork sits most uncomfortably beside such objects’ use in the Augustan program, all the more so given the inclusion of the baldric’s central myth within a monument memorializing Actium, to which Vergil has earlier directed Aeneas’s uncomprehending gaze (cf. figure 4.1).72 The Aeneid links the name “Pallas” with Evander’s

68

Putnam 1998: 203. On the significance of this progression cf. Bartsch 1998: 339. 70 Bartsch 1998: 339. 71 Reed 2007: 38. 72 Note incomprehension as a theme linking the Shield of Aeneas with the baldric of Pallas (A. 8.730 rerumque ignarus; 10.501: nescia mens hominum). 69

Heroic Objects: Ecphrasis in the Aeneid and Metamor phoses

159

city Pallanteum and therefore with the Palatine, while Pallas is actually related to Aeneas through Atlas, himself the ancestor of the Italian Dardanus, founder of Troy and Aeneas’s direct forbearer.73 On the other side, Vergil’s Turnus, as a descendant of Inachus through Io, is related to the Danaids as well as to both Dido and Cleopatra, their conjuncture aligning Inachid victims of Rome’s rise.74 Dardanids and Inachids, however, are both sprung from Zeus, and so ultimately relations.75 With the Aeneid’s conclusion, Vergil thus adds, as it were, his own commentary on the Danaid portico to propagandistic interpretations like those offered by modern scholars.76 The Greek maidens whose deed could signify conquest over Egypt or a regretful glance backward at civil war now become the object of a far more disturbing mimesis: Turnus as sacrilegious Danaid (in the killing of Pallas) cedes this role, too, to his Trojan killer.77 The progress of empire, troped by Aeneas’s bloody act, depends upon the reactivation of the Greek Danaids’ murderous example within the Trojan, proto-Roman project, rather than its relegation to the safety of the past. That the mythical daughters act out of filial piety, and Aeneas, too, shows pietas in avenging Pallas, is one of the Aeneid’s quintessential ambiguities, recalling the loyalty that Octavian could claim to have shown in avenging Julius Caesar’s murder. Aeneas’s reaction, nevertheless, is genuinely ambiguous, and the Greekness of wondrous Greek art undeniably dangerous at the end of the Aeneid. Vergil, however, has long since begun to prepare his audience for this outcome, above all through Aeneas’s own account of that most deadly and effective Hellenic artifice, the Trojan Horse.78 The Aeneid, we have seen, complicates Augustus’s use of Greek culture in his public visual program by interspersing the narrative of Aeneas’s ktistic journey with a series of ambiguous ecphrases—descriptions of “heroic objects”—intimating both success and reversion to destructive Greek prototypes. Self-defining struggle is set at odds with outdoing the great Greek precedent, since the hypnotic force drawing Aeneas and/as the Augustan viewer back into the patterns established by Greek mythological imagery dogs the attempt to forge a novel, proto-Roman identity at nearly every turn. One cannot ultimately control, Vergil 73 A. 8.51–54, 134–42; Reed 2007: 199. Serv. A. 8.51 asserts an actual etymological link; cf. Spence 1991: 16. 74 See Reed 2007: 69–70. 75 Cf. Reed 2007: 10. Of course, this makes Augustus ultimately related to his own victim, Cleopatra. 76 For modern interpretations of the Danaid Portico, see above p. 115. For the connection with Pallas’s baldric, see esp. Spence 1991; Putnam 1998: 198–201, 211; Strocka 2008. 77 Putnam 1998: 197–98; for Turnus as “the ephebe cut down on his wedding night,” see Hardie 1993: 33. For a helpful summary of views on the significance of the Danaids vis-à-vis Aeneas, see Strocka 2008: 43–44. 78 On the Trojan Horse as an image of threatening Greek artistry in the Aeneid, see esp. Bartsch 1998: 324. See further Chapter 6 below, on Petronius’s use of the Vergilian paradigm.

t h e c a p t o r’s i m a g e

160

suggests, such iconography’s effects on a viewer, or for that matter an epic poet, and these may lead inexorably into the labyrinth of such ambiguity. Given the Metamorphoses’ title and celebratory culmination in the Rome of Augustus, one might expect Ovid’s ecphrases to tell a different tale focusing on continuity through change, as in many ways they do. And yet the theme of self-differentiation, linking Ovid’s Augustan poem to its Vergilian predecessor, here too becomes a complicating force, as we shall now explore in detail.

II. Looking Askance: Ecphrasis in Ovid’s Metamorphoses The Ineffectual Message on the Palace of the Sun In a poem so clearly alluding to the Aeneid throughout its entire scope, above all in the Trojan story of Metamorphoses 13–14, Vergil’s clear influence on Ovidian ecphrasis may not surprise us. The Aeneid, as is well known, had become the definitive Roman epic already by Ovid’s day, its achievement the avowed benchmark by which subsequent efforts in the genre were to be judged. Indeed, Ovid’s quest for poetic glory regularly takes the form of a dialogue with Vergil, as we shall see especially in the case of the Anius episode in the “little Aeneid” below. Even in light, however, of the deep connections between the two texts, we may still be struck by the way that Vergil’s hesitations over the capacity of art to convey its ostensible message, particularly public artworks or those sponsored in some way by rulers and gods, find a close analogue in the work of his successor. Both poets, that is, insist on epic ecphrasis as a showcase for the dense, indeterminate, sometimes self-contradictory web of associations that mythical imagery may create in its viewers in spite of the best efforts of a figure such as Augustus to bend this process to his political will. Each author presumes an audience able to recognize the two-sided nature of Augustus’s public attitude toward the Greek legacy and assess each side against the other, to hold appreciation and questioning in constant tension. This is true already in the starkest terms of the Metamorphoses’ first ecphrasis, the description of the Palace of the Sun at 2.1–18.79 For here the image of a desired cosmic order embodied in a version of “official” art simply has no discernible effect on the eruption of disorder—Phaethon’s disastrous ride— that immediately follows it. Details of the ecphrasis, moreover, subtly recall the earlier dissolution of order in the worldwide flood and anticipate the coming havoc induced by Phaethon, and in this way highlight the fact that such stories

79

On the passage, see esp. Galinsky 1975: 49–51; Bass 1977; Lausberg 1982: 120–23; Brown 1987; Wheeler 2000: 37–40; Hardie 2002: 177; Graziani 2003; Vial 2010: 424–30 and passim.

Heroic Objects: Ecphrasis in the Aeneid and Metamor phoses

161

are inextricably linked by the mythic tradition. Cosmic order reverts to chaos twice in the Metamorphoses’ early books, and the opening ecphrasis of Book 2 marks the threshold of the second decline, for once Phaethon, having entered his father’s palace, has received Phoebus’s promise to let him drive the solar chariot, catastrophe becomes inescapable. The Palace of the Sun, which also recalls the Palatine temple of Apollo, thus puts front and center the question of public art’s capacity to change the behavior of its viewers for the better, however much it may embody welcome order in the face of chaos. This is an urgent issue in the Aeneid, too, as we have seen. Ovid does not answer this question definitively, since we are not told whether or not the images on Phoebus’s palace are described through Phaethon’s eyes, or indeed whether he is even aware of them.80 The brilliance of Ovid’s ecphrasis and the heroism of these art objects inhere rather in the way they both insist upon and challenge the import of mythological imagery already in the mythological past, and the “need to interpret” that they thereby press upon an Augustan audience. Seen within the entire Phaethon episode, the Palace of the Sun concentrates the audience’s attention on whether art’s imitation of the world corresponds to the essence of that world, and this begins even before the images on its double doors come into view. The first lines of Book 2 present the edifice as “high with lofty columns” (Met. 2.1) and “bright with flashing gold and bronze imitating flames” (2, clara micante auro flammasque imitante pyropo). Imitation immediately takes center-stage: the palace’s external characteristics obviously symbolize the nature of the life-giving sun.81 For a learned audience, however, such doubling must already appear sinister. Phaethon himself will soon seek to imitate his father, Sol, whose borrowed flames will all but destroy the world, an outcome Jupiter himself has in fact anticipated in Book 1 (254–58). Ovid magnifies this threat though his Vergilian subtext. The ecphrastic opening of Metamorphoses 2 is modeled directly on that of Aeneid 6, where Daedalus’s artworks on the doors of another temple of Phoebus symbolically “represent Aeneas’s entrance to the world of Italy and foreshadow his role in the untimely death of young men,” just as Ovid’s palace and its doors “function as a portal to the new world that Phaethon enters,” only to initiate the chain of events leading to his own Icarus-like death.82 Daedalus’s artworks inevitably conjure up the memory of his dead son, Icarus, although the latter’s image does not appear within them. So, too, the flamelike brightness of Sol’s palace, rather than conveying simply the radiance that animates the ordered cosmos we will soon see on the doors, foreshadows the demise of his son without actually depicting it.

80

See Wheeler 2000: 40. Brown 1987: 212–13. 82 Wheeler 2000: 37 with n107. 81

t h e c a p t o r’s i m a g e

162

As Ovid begins to describe the palace’s artworks, one of Vergil’s own major intertexts, the Shield of Achilles, together with more intratextual allusions to other parts of the Metamorphoses, further problematizes the status of Greek precedent.83 Ovid’s survey of the images, crafted, he tells us, by Vulcan, commences with an overview of an ordered, tripartite universe: For Mulciber had depicted there the seas surrounding the lands in their midst and the circle of the earth and heaven which is poised over the world. nam Mulciber illic aequora caelarat medias cingentia terras terrarumque orbem caelumque, quod imminet orbi. (Ovid, Metamorphoses 2.5–7) Homer opens his shield ecphrasis with Hephaestus’s similar handiwork: “And on it he fashioned the earth, on it the heavens, on it the sea” (Il. 18.483, ἐν μὲν γαῖαν ἔτευξ’, ἐν δ’ οὐρανόν, ἐν δὲ θάλασσαν).84 The Homeric allusion, to an object interpreted in antiquity as an icon of creation, underscores the notion of Sol as a deity presiding over an ordered cosmos for which he provides nourishing energy.85 But such order already comes under threat in Ovid’s phrase imminet orbi, intimating “the danger that the sky will pose to earth” in Phaethon’s ride.86 The palace image also looks back to Ovid’s cosmogony in Metamorphoses 1, which begins, less threateningly, “Before the sea and the land and the sky, which covers all” (1.5, ante mare et terras et quod tegit omnia caelum).87 The cosmogony, however, has by this point been shown to “establish a form of order that disintegrates into chaos,” the divine creator’s into the primal flood, Vulcan’s into Phaethon’s all-consuming fire.88 Ovid’s ecphrasis moves next to an account of sea deities, a vignette often commented upon for being considerably longer (eight lines) than the descriptions of the land and sky (two lines each) that follow, as well as for the particular gods—of “blended, unstable, and monstrous form”—to which our attention is drawn.89 Indeed, the imago mundi on the palace doors is framed by “Classical” 83

Wheeler 2000: 38. On the Homeric allusion see Lausberg 1982: 120–22; Brown 1987: 211; Wheeler 2000: 38. 85 For Homer’s shield as a cosmic icon, see Lausberg 1982: 120n1; Hardie 1986: 340–43; Brown 1987: 211. 86 Wheeler 2000: 38. 87 Such echoes continue, so as to suggest that Vulcan’s artistry imitates that of Ovid’s nameless divine creator, described as another artisan (“the artificer of the world” [1.57, mundi fabricator]; “that artisan of the universe” [1.79, ille opifex rerum]; Wheeler 2000: 38); cf. Hardie 2002: 177. 88 Wheeler 2000: 38. 89 Brown 1987: 218; see further Wheeler 2000: 38–39 with bibliography. 84

Heroic Objects: Ecphrasis in the Aeneid and Metamor phoses

163

order conveyed in the ecphrasis’s opening lines and in its conclusion, where we find a succinct account of “earth, men, and cities, woods and wild beasts and rivers, and nymphs and other country deities,” above which the “image of bright heaven,” displays a balanced tableau of “six constellations on the right-hand door and the same number on the left” (2.17–18). Within this setting, Ovid’s maritime world appears distinctly “Hellenistic,” the blending of the two aesthetics suggesting Ovid’s program in the Metamorphoses.90 So, after “sonorous Triton” (8, Tritona canorum), the passage continues, and Proteus of uncertain form, and Aegaeon pressing the huge backs of whales with his arms. Proteaque ambiguum ballenarumque prementem Aegaeona suis inmania terga lacertis. (Ovid, Metamorphoses 2.9–10) This section ends with the uncanny image of Doris and her sea-nymph sisters, some swimming, some sitting on rocks while drying their green hair, and others riding fish (11–13). “Their faces were not all the same,” we learn, “nor yet were they entirely different, as is fitting in sisters” (13–14, facies non omnibus una,/non diversa tamen, qualem decet esse sororum). Two-formed sea-deities such as Triton and elaborate aquatic imagery were staples of the Hellenistic art tradition.91 What has all this got to do with Phoebus? We sense here the intervention of Ovid’s narrator, with disproportionate emphasis on details that Vulcan would have had no reason to focus upon, an asymmetry whose effect is “subtly to undermine the metaphysical symbolism of the Sun’s palace” through allusion to “the more disorderly world of the Metamorphoses” itself.92 Out of rational order, that is, a universe neatly divided into discrete zones, emerges a far different picture dominated by the shape-changing Proteus, Aegaeon the Hundred-hander, and similar-looking nymphs. So Ovid’s metamorphic world seems to resist the imposition of an order like that envisioned on the palace, which is nevertheless also a part of Ovid’s total aesthetic in the poem. The palace’s imagery not only becomes available for narrative flashback and foreshadowing undermining its ostensible purpose, but also incites the narrator to visual “excesses” symbolizing this disorderly, ambiguous world in miniature. Forging a connection to his audience’s experience, Ovid invites it to reflect upon its own reactions to Augustus’s use of myth through recollections of 90

Cf. Brown 1987: 217 on the Hellenistic qualities of the ecphrasis and see below, pp. 165–66 on the blending of classical and Hellenistic as distinctive of Ovid’s program in the Metamorphoses. 91 See Pollitt 1986: 105, 148, 226. 92 Brown 1987: 218.

t h e c a p t o r’s i m a g e

164

the temple of Palatine Apollo. The chariot of Sol or Helios, closely associated with Apollo in Augustan poetry, appeared on the roof of the Palatine temple (cf. Propertius 2.31.11 and Chapter 4, p. 134), where it would have offered a clear material reference point for Ovid’s whole account of Phaethon’s ride.93 If not a palace itself, the Palatine temple was nevertheless joined to Augustus’s house, modest but still the home of a ruler. Like the two Vergilian passages recalling the Palatine temple, furthermore, the opening of Metamorphoses 2 moves from “a sketch of a splendid temple . . . to a detailed description of its artwork” glorifying its resident deity, a technique first appearing in the Vergilian ecphrases, discussed above.94 And marine motifs such as Tritons and dolphins would have been especially familiar to Ovid’s audience from Augustus’s use of them in iconography celebrating the sea-battle at Actium, which the Palatine temple also memorialized.95 The prominence of this battle and its watery setting is, as we have seen, a primary way in which Vergil’s Shield of Aeneas alters the pattern on the Shield of Achilles, to which Ovid also alludes before likewise diverging from this model. Finally, if the Hundred-hander Aegaeon, whom Vergil describes fighting “against Jove’s thunderbolts” at Aeneid 10.565–68 can recall the hints of Gigantomachy in Vergil’s Actium scene, which is itself closely linked to the Palatine temple within the Shield of Aeneas, another, more distant recollection of the monument emerges.96 As much as Ovid’s ecphrasis highlights metamorphosis (in Proteus) and its accompanying thought-world of monstrosity and uncanny similarity, its context is nevertheless a self-defining, though ultimately unsuccessful, quest: Phaethon’s struggle to prove his paternity, described at the end of Book 1 in distinctly Aeneid-like terms as a journey across Ethiopia and India to his “ancestral origins” (779, patrios ortus; cf. Verg. A. 3.129, “and let us seek our ancestors” [proavosque petamus]; 167–68, “from here Dardanus arose and father Iasus” [hinc Dardanus ortus/Iasusque pater]). The lurking dangers he faces in the heavens are likewise reminiscent of Aeneas’s Odyssean journey past monstrous adversaries such as the Harpies and Scylla, while also recalling the Ovidian narrator’s excursus on the metamorphic marine world depicted on the doors: “The way lies through traps and forms of beasts,” Sol warns (Ov. Met. 2.78, per insidias iter est formasque ferarum; that is, constellations such as the Bull, the Lion, the Scorpion, and so 93

On the close proximity of Sol and Apollo in the Augustan and other eras, see Bömer 1969– 1986: 1.244–45. 94 See Brown 1987: 212 and above, pp. 116–25, 148–52. 95 Wheeler 2000: 39n115 citing Zanker 1988: 82–85. 96 For the “possible undercurrent of violence” in Ovid’s Aegaeon, see Brown 1987: 218. Vergil departs from the Hesiodic account of the Hundred-handers as allies of the Olympian gods in defeating the rebellious Titans (WD 665–732; for the conflation of Titans and Giants in ancient accounts, see above, p. 16, on Naevius’s Bellum Punicum). On allusions to the Gigantomachy in the Shield of Aeneas, see above, pp. 155–56. Gigantomachy is also at Ov. Met. 1.151–62 and 10.150–51.

Heroic Objects: Ecphrasis in the Aeneid and Metamor phoses

165

on; cf. the “Serpent” at 138). Whereas Aeneas, however, becomes engrossed by images that signal and precipitate his reversion to troubling Greek models, Phaethon ignores—or at least apparently so—the images that forecast his father’s careful, repeated directions to him about which course to follow through the cosmos’s different regions and what dangers to avoid (63–89, 126–40). The contrast with Aeneas is borne out by his story’s conclusion. While Phaethon’s error ultimately brings on a disastrous mixing of elements and permanent transformations in the people and lands of the world (161–303), his own tale concludes not with metamorphosis (like that of Aeneas, whom Ovid will later show transformed into the god Indiges [14.605–08]), but rather with his death (2.319–43). If Phaethon’s efforts are to some extent redeemed in the grave inscription set up by the western Naiads, who laud his “great things dared” (328), he remains an anti-Aeneas, a hero crushed by paternal example rather than, like Aeneas, building upon it. That Phaethon and Augustus share a paternal figure in Sol/Apollo, Augustus’s patron deity, has not escaped the notice of scholars, some of whom have found in Ovid’s tale a thinly-veiled warning aimed at the Princeps.97 The Aeneid, as we have seen, certainly warns of large-scale destruction from unchecked violence, a threat emerging especially through the distortion of Aeneas’s martial ambitions. Yet the predominantly cautionary aspect of Ovid’s Phaethon nevertheless contrasts with another salient feature of Vergil’s Aeneas, whose relations with his own father, Anchises, suggest Augustus’s expressed pietas toward his adoptive father, Julius Caesar.98 Ovid’s ecphrasis of the Palace of the Sun thus both limns a trajectory Phaethon might have followed and one he could not possibility attain (not being Sol himself); it forecasts his disaster but suggests the underlying truth of the universe’s metamorphic nature as Ovid portrays it. Metamorphosis and self-definition again appear at odds, yet ecphrasis nevertheless presents us with heroic objects whose power itself is two-sided: generative and destructive, undergirding official authority and challenging it, wondrously intermediary between the human and the divine.

Imaging Captives: Greeks and Non-Greeks in the Tapestries of Minerva and Arachne The Metamorphoses’ most famous ecphrasis, the description of tapestries created by Minerva and Arachne in a primal weaving contest (6.70–128), has long struck readers for the programmatic nature of its central contrast. Minerva, challenged by the upstart Lydian girl Arachne, produces an image of “Classical”

97 98

See Doblhofer 1973; Schmitzer 1990: 89–107; Bretzigheimer 1993: 46–47, 72–74. See e.g., Tarrant 1997: 179.

166

t h e c a p t o r’s i m a g e

order celebrating the gods’ punishment of those who reject their rule. Arachne, in an account both reminiscent of Ovid’s Ars Amatoria and, like the sea-deities on the Palace of the Sun, suggestive of the Metamorphoses’ “Hellenistic” asymmetry and flux, weaves a seemingly irreverent picture of the gods’ lust for goddesses and mortal women.99 The whole passage can thus be seen to embody the Metamorphoses’ two-sided portrayal of the gods as both “a force of order” and “a force participating in the flux of nature,” and beyond this the poem’s fusion of “symmetrical and asymmetrical principles of organization and style.”100 Further, the episode’s generic differentiations help establish the elusive nature of Ovid’s epic and “locate that identity firmly within the Roman response to Alexandrian poetics.”101 At the same time Arachne’s tale is “a fable about the problematic relationship between artist and power” that becomes “a sinister omen for the fate of Ovid himself ”: the scandalized Minerva’s authoritarian punishment of Arachne by transforming her into a hapless spider eerily prefigures Augustus’s exile of the poet to the Black Sea region of Pontus in 8 ce, apparently a response, in part, to the transgressive nature of his work.102 Too little remarked by scholars, however (in spite of the connection with exile) is the ecphrasis’s keen interest in non-Greeks—those living at or beyond the borders of Hellas—and their experience and depiction of the Greco-Roman gods’ authority. The Arachne episode, that is, portrays non-Greek challengers to Minerva’s order, the Trojans among them, as subject to the same capricious exercise of divine force as the Greeks through the gods’ use of metamorphosis to impose their violent will. Thus Ovid suggests that such outsiders’ quest for self-distinction, which would also include Rome’s, must come up against the (in)ability of those in power to control images: both the “official” imagery of rule and a more generalized shape changing that utilizes a deceptive self-image to exploit the vulnerable. Ovid intimates further that destructive transformation may be the ultimate outcome of all such efforts, Rome’s included, to distinguish oneself from the rest of mankind. Yet Ovid is careful here to avoid any overt reference to the monuments or iconography of Augustan Rome; so careful, in fact, that one scholar declared the rival tapestries “not . . . windows looking towards a world outside of the poem,” but solely “mirrors of the poem itself.”103 Ovid’s 99 On the Arachne episode, see esp. Leach 1974a; Barkan 1986: 1–18; Harries 1990; Feeney 1991: 190–94; Smith 1997: 54–64; Rosati 1999; Oliensis 2004; Johnson 2008: 74–95; Vial 2010: 304–07 and passim. 100 Leach 1974a: 103. 101 Harries 1990: 64. 102 Rosati 1999: 251. On the circumstances of Ovid’s exile, its probable causes, and premonitions of it in the Arachne episode, see, recently, Johnson 2008: 3–21, 74–95. The poet cryptically describes his exile as due to “a poem and an error” (Tr. 2.207, carmen et error); these are often understood as, respectively, the Ars Amatoria and an affair with Augustus’s granddaughter, Julia. 103 Leach 1974a: 106.

Heroic Objects: Ecphrasis in the Aeneid and Metamor phoses

167

subtle play, however, with notions of Greek and non-Greek, power and vulnerability, change and self-definition, presents an allegory for the Augustan moment. Can Romans as non-Greeks, Ovid asks, wield the metamorphic power of art in their quest to outdo Greek authority, or will such efforts end in fatal transformation within a fluctuating universe where power over change rests ultimately with the gods themselves?104 Let us begin where Ovid does, with Minerva’s tapestry. Its central image depicts the twelve Olympian gods sitting in judgment while Minerva and Neptune compete for patronage of Athens. Surrounding this scene in the corners are four “contests” (85, certamina), really punishments of fabled transgressors. All are non-Greek, suggesting a divine authority reaching beyond territorial boundaries to master the edges of the known world. The tapestry itself embodies this world, its limits, significantly, coextensive on the whole with the horizons of Roman imperium in Ovid’s day. First there is Thrace, a difficult area for Augustus, but made a client kingdom well before Ovid’s exile. Minerva depicts its mountains Rhodope and Haemus, once mortals “who bestowed upon themselves the names of the highest gods” (89, nomina summorum sibi qui tribuere deorum). Almost nothing else is known of their story, though their outrageous ambition, literally self-defining, is clear from Ovid’s brief account.105 Next appears a Pygmy woman (and so a resident of northern Africa, Scythia, Thrace, or perhaps India). Her crime is to have vied somehow with Juno (again the details are unknown to us), who transforms her into a crane and induces her to “declare war upon her own people” (92, populisque suis indicere bellum). Vergil’s Aeneas, too, is a victim of Juno’s wrath whose transformation into the Achillean figure of the Aeneid’s second half involved him in war upon his own people, the Italians. He is distantly related to them through Dardanus, and this conflict is a prototype for Roman civil war. So Minerva’s illustrations, however arcane to us, are in their way familiar stories to an Augustan audience.106 Recollection of the Aeneid seems unavoidable, yet far from unambiguous, as the ecphrasis continues. Minerva next depicts Antigone, the Trojan princess and daughter of Laomedon who also dared vie with Juno, for which she was

104

For another interpretation of the Arachne episode as referring allegorically to Augustan visual self-representation (though with conclusions somewhat different from my own), see Oliensis 2004, esp. 295: “By having Minerva herself produce her self-aggrandizing representation, what Ovid discloses is the interestedness of Augustan (self)representations: their composite function as an expression, mystification, justification, and medium of power.” 105 Anderson 1972: 163 is perhaps right to see them as lovers “ecstatic in their mutual devotion,” who “considered themselves happier than the gods.” 106 For formal similarities between the Arachne episode and the Aeneid’s ecphrases, see Harries 1990: 67–69.

168

t h e c a p t o r’s i m a g e

changed into a bird. Ovid’s commentary on the tale sounds a dissonant note under a regime for which the Troy story was of central importance: Troy was of no help to her, or her father Laomedon, but that she took on white feathers and applauded herself, a stork, with rattling beak. nec profuit Ilion illi Laomedonve pater, sumptis quin candida pennis ipsa sibi plaudat crepitante ciconia rostro. (Ovid, Metamorphoses 6.95–97) Here a Trojan opposed by Juno ends, transformed, in futile self-applause, her illustrious Trojan ancestry (connected with Aeneas’s through Tros, grandson of Dardanus) simply “no help.” If the Julian line, descendants of Aeneas, had avoided an analogous fate, could the Rome that was Troy reborn nevertheless become helpless and unrecognizable through its ambitions, all its claims of stature rendered mere self-adulation? (Recall Vergil’s lines on the curse of Laomedon at Georgics 1.501–02).107 Minerva’s final picture juxtaposes the Trojan case with another transgressive easterner, Cinyras. His grief, which the image highlights, is the result of his daughters’ transformation into temple-steps (a further obscure reference).108 As though setting an imperial seal on the geographical regions limned by her handiwork, Minerva surrounds her tapestry with a border of “peaceful olive . . . her own tree” (6.101–02). Noting the echo of Augustus’s name at 6.73, where the gods sit “with august gravity” (augusta gravitate), Oliensis compares the golden “shield of virtue” (clupeus virtutis) awarded to Augustus in 27 bce and inscribed with a text recalling key Augustan values: similarly, Minerva’s tapestry embodies a “Minervan constitution, a self-authorizing display” of power.109 Minerva’s portrayal of these non-Greek transgressors and their punishments is all the more resonant for the fact that Arachne herself is soon to become one of their number: Ovid emphatically calls her “Maeonian,” that is, Lydian (6.5, Maeoniaeque; 103, Maeonis); her ambition to achieve a “name worthy of 107 “Now long since have we paid sufficiently with our blood for the perjuries of Laomedontean Troy” (satis iam pridem sanguine nostro/Laomedonteae luimus periuria Troiae). Laomedon incurred the ill-favor of Apollo and Neptune for failing to pay them for their help in building Troy. 108 The name Cinyras has Semitic roots; see Morris 1992: 8. This may or may not be the same person as the Cinyras who is king of Cyprus and, in Ovid, grandson of Pygmalion (whose name points to Phoenician origins: see Reed 2007: 75–77). Met. 10.298–518 recounts this Cinyras’s incestuous union with his daughter, Myrrha. Later Cinyras is associated with Assyria, and this tradition perhaps represents a different figure to whom the Arachne episode refers; see Bömer 1969–1986: 3.34. 109 Oliensis 2004: 286–87.

Heroic Objects: Ecphrasis in the Aeneid and Metamor phoses

169

mention” (12, nomen memorabile) precipitates a quest “through the Lydian cities” (11, Lydas . . . per urbes), where she boasts of a talent surpassing Minerva’s own (6–7). In the lines that introduce her, Ovid “flaunts his geographical knowledge of Arachne’s distant home” by inserting no fewer than seven references to Lydia and areas of Ionia that came under Lydian rule.110 Unlike the images on Minerva’s tapestry, moreover, hers feature Greek maidens (or Greco-Roman deities and their daughters) subjected to lustful gods.111 An opening exception, given the most space in Ovid’s account (five lines), nevertheless proves the rule: Europa, the Phoenician princess who embodies transition from East to West (Phoenicia to Crete) as well as from Minerva’s work to Arachne’s. Thus, in a whirlwind of concupiscence, Jupiter assumes eight different forms to achieve eight different conquests, from Asterie to his own daughter, Proserpina (108–14). Neptune is shown engaged in six similar acts of deception and sexual domination; Apollo, four; Bacchus and Saturn, one apiece (115–26). If Minerva shows the divine powers squelching the outsized ambitions of foreigners, the foreign Arachne reveals the gods’ sexual transgressions against the Greeks themselves in seemingly graphic detail that appears designed specially to offend the sensibilities of her virginal opponent.112 The challenge, like the object itself, is heroic; yet it will destroy the artist. Minerva’s vengeance on Arachne for her flawless work not only demonstrates the goddess’s might but also renders artistic creation itself as a process perversely consuming the creator.113 Arachne’s metamorphosis into a spider is not simply the end of her ambitions, but their dreadful twisting into the defining feature of her existence at the moment they also become permanently unfulfillable. Feeney aptly quotes the Younger Seneca’s insistence that the spider’s work is not art, since spider-webs are all the same, while “Whatever art bestows is uncertain and uneven” (Sen. Ep. 121.23, incertum est et inaequabile quidquid ars tradit)114—much like Arachne’s fateful tapestry, which Minerva viciously destroys (Ov. Met. 6.131). Doomed to “hang” (cf. 136, pende) forever in her “web of old” (145, antiquas . . . telas), which now doubles the suicide’s rope with which she attempted to take her life after Minerva cruelly struck her in the forehead with 110

Anderson 1972: 152. Greek women: Leda, Antiope, Alcmene, Danae, Aegina, Canace; divinities: Asterie, Mnemosyne, Proserpina, Iphidemeia. Only Admetus (6.122) is a male love object. 112 Johnson 2008: 83–88 suggests we are to imagine Arachne’s weaving as pornographically explicit. Worth remarking, however, is Leach’s observation that reading Arachne’s tapestry as a challenge to Olympian authority involves adopting the perspective of Minerva: “It is Minerva’s interpretation that makes the subject immoral” (Leach 1974a: 117; developed by Oliensis 2004: 290–91). 113 Envy, Ovid remarks in one of the passage’s more self-evident Callimachean allusions, could find no fault with it (Ov. Met. 6.129–30). On the suppression of Envy in Callimachus’s Hymn to Apollo, see above, p. 124, n47. 114 Feeney 1991: 193–94 (his translation). 111

t h e c a p t o r’s i m a g e

170

her shuttle (132–35), Arachne illustrates in “sickeningly appropriate” terms how metamorphosis and self-defining struggle may conflict, here due unequivocally to the overmastering, arbitrary power of the Olympian deities.115 If Augustus, Ovid intimates, has the power of a god, then Rome, and especially artists such as Ovid himself, have to fear similar capriciousness from their ruler. Yet if Trojan Rome, non-Greek Rome, is itself an upstart like the Lydian Arachne, Augustus’s own transformation of Rome’s power and self-image may be as vulnerable as her metamorphic art, failed rival of Minerva’s empire-in-miniature, her woven imperium.

Refiguring Vergilian Ecphrasis: The Cup of Anius With the Metamorphoses’ third and final ecphrasis, depicting a cup bestowed upon Aeneas by the Delian king Anius (13.681–701), this chapter comes full-circle. The “little Aeneid” is Ovid’s most pointed answer to Vergil’s poem, and the Anius passage “encapsulates the aspiring successor’s effort to emulate the famous pictorial texts in the Aeneid in terms of intricacy, density, and power of interpretation.”116 Here Ovid uses a Hellenic object not, as Vergil does, to suggest threats or impediments to Aeneas’s mission itself, but in such a way as to focus still greater attention than Vergil on the problems inherent in Greek mythical exempla as potential “inspiration” for Aeneas’s self-sacrificial journey, and so for Rome’s own journey of self-transformation into a society that would, in Augustus’s vision, both distinguish itself from the Greeks and outdo them in their own areas of cultural achievement.117 As before, Ovid engages with both of the attitudes toward Greece that we have seen structuring the Aeneid’s ecphrases, and by juxtaposing them afresh within an epic narrative, again complicates— here all but explicitly—Augustus’s public appropriation of Greek culture as underpinning for imperial rule. Unusually among the metamorphic stories embedded in the “little Aeneid,” the Anius ecphrasis bears a clear thematic relation to its surrounding narrative.118 115

Feeney 1991: 193: “Her metamorphosis into a spider is a sickeningly appropriate punishment for Minerva to devise.” 116 Papaioannou 2005: 24; on the ecphrasis see esp. her discussion at 19–42. I have also found helpful Guthmüller 1964: 89–90; Galinsky 1975: 221; Solodow 1988: 158–59; Myers 1994: 100; Hopkinson 2000: 201–04. 117 Alcon’s exact origins remains a mystery. The Theban scenes on the cup create the expectation that he is a Boeotian, but the epithet Hyleus transmitted at Met. 13.684 in some MSS is not the correct form of the adjective from Hyle, a town in Boeotia. See Hopkinson 2000: 201, who also rejects the reading Lidius “Lydian,” which, intriguingly, would make Alcon Arachne’s compatriot (this reading is defended by Papaioannou 2005: 20n2). 118 Galinksy finds the connection to the Aeneas story virtually unique among the stories embedded in the “little Aeneid”: “This attempt to create some thematic affinity with the situation of Aeneas contrasts effectively with the near total absence of any such endeavor in other stories, which are not inspired by the Aeneid” (Galinsky 1975: 221).

Heroic Objects: Ecphrasis in the Aeneid and Metamor phoses

171

As with the Aeneid’s ecphrases, therefore, an understanding of the passage’s narrative frame is crucial to grasping its larger import. The description forms the culminating moment of the “little Aeneid”’s first major episode: the Trojans’ visit to Delos after their initial flight from Troy. More specifically, it follows upon Anius’s account of the fate of his four daughters, who, descendents of both Apollo and Bacchus, were gifted with the divine power to transform anything they touched into grain, wine, and olive oil (13.652–54). Their miraculous abilities, Anius recounts, became known to Agamemnon, who, returning from the sack of Troy, demanded that they be handed over to him to provide sustenance for the Greek fleet. When the young women fled, two to Euboea and two to their brother’s eponymous island of Andros, the Greeks pursued them and threatened war if they were not surrendered. From this point on Anius provides only the story of the two daughters in Andros. Giving themselves up to save their brother, who lacked adequate forces to ward off a Greek attack, the pair, already placed in captives’ chains, called upon Bacchus and were turned into white doves (655–74). Although one might easily regard this as a liberation and a redemptive confirmation of the daughters’ noble self-sacrifice, Anius views their metamorphosis entirely in a negative light: it is the “the culmination of evil” (673, summa mali). After this, he makes no further mention of his son or expresses any feelings of gratitude and relief in the latter’s escape from imminent destruction. In fact, what he has already said about Andros suggests ambivalence about him: absent, even as a ruler, he offers his father no help (647–48, quod enim mihi filius absens/auxilium . . .?), his “loyalty was conquered by fear” (662, victa metu pietas), and “you might pardon the timid brother” (664, timido possis ignoscere fratri) since he lacked a hero such as Aeneas or Hector to defend him (665–66, non hic Aeneas, non, qui defenderet Andron,/Hector erat). Anius links his own suffering, furthermore, to that of the Trojans by prefacing his tale of Agamemnon’s threats with the words “lest you think that we, too, had not to some degree experienced the storm of troubles that you did” (656–57, ne non ex aliqua vestram sensisse procellam/nos quoque parte putes). Thus Ovid prepares his audience to draw a connection between Anius’s story and the very similar one depicted on his cup, as well as to regard Anius’s gift as a sympathetic offering meant to express his fellow-feeling for the Trojans. Scholars have long noted the parallels between the tale of the Aniads and that of the Orionids, as well as the details linking both stories to others in the Metamorphoses, especially those of Thebes in Books 3 and 4 and Rome’s revival after the visit of Aesculapius in Book 15.119 The cup of Anius encapsulates the Metamorphoses’ larger comparison of Thebes with Troy/Rome, a fascinating

119

See Papaioannou 2005: 20–42 for discussion and bibliography.

172

t h e c a p t o r’s i m a g e

parallelism that has recently attracted renewed critical attention.120 The essential link, moreover, between Aeneas’s story and the Orionids’ (and so also the Aniads’) has been found in the particular narrative depicted on the cup: its artist, Alcon, is said to show Thebes beset by plague (687–91), then delivered by the Orionids’ self-sacrifice “for their people” (695, pro populo . . . suo), then celebrating the miraculous appearance of a pair of young men from the girls’ ashes: then from the virgin ash, lest their line die out, [he makes] twin young men, whom wide report calls the Coronae, appear and lead the procession for their mothers’ ashes. tum de virginea geminos exire favilla, ne genus intereat, iuvenes, quos fama Coronas nominat, et cineri materno ducere pompam. (Ovid, Metamorphoses 13.697–99) Thebes’s story here anticipates Troy’s “regeneration as Rome through Aeneas and his descendents,”121 the tale Ovid will in fact go on to tell. Metamorphosis and the Trojan/Roman quest seem harmonized at last; yet anyone familiar with Thebes’ recurring history of disaster in ancient myth, particularly its problematic status in Greek tragedy as an alter-ego of Athens, cannot help but feel uneasy about the comparison. And indeed, as one enters closely into the details of the Anius ecphrasis, the positive interpretation of its Roman parallelisms becomes increasingly difficult to isolate from more doubtful associations. To begin with, there is the literary provenance of the cup itself. Fashioned by the sculptor Alcon, Ovid’s cup recalls not only the ecphrases of the Aeneid but also the cups of Alcimedon in Vergil’s Eclogue 3.122 The similarity of Alcon’s name to Alcimedon’s is already suggestive, but reinforcing this connection is the placement of each artist’s name at line-end, as well as the recurrence of the verb caelo (Ecl. 3.36–37, caelatum divini opus Alcimedontis [“the engraved work of the divine Alcimedon”]; Met. 13.683–84, fabricaverat Alcon/ . . . et . . . caelaverat

120 In his choice of Thebes as both “the symbol of a dysfunctional polis and metaphor for Rome,” Ovid follows “a pattern already established in Greek tragedy,” which situated Thebes in a similar relation to Athens (Papaioannou 2005: 27, building upon the insights of Zeitlin 1986); cf. Janan 2009: 7. The imposing shadow cast by Thebes over the whole of the Metamorphoses, particularly its closing episodes on Rome, is the subject of a seminal article by Philip Hardie (1990) and more recently a book by Micaela Janan (2009), whose Lacanian reading of Ovid’s Theban books interprets Ovid’s obsession with Thebes through the equally Thebes-obsessed discourse of psychoanalysis. 121 Papaioannou 2005: 24; cf. Guthmüller 1964: 89–90; Galinsky 1975: 221. 122 Hopkinson 2000: 201, 204.

Heroic Objects: Ecphrasis in the Aeneid and Metamor phoses

173

[“Alcon had made and chased”]). An even clearer allusion to Eclogue 3 concludes the Ovidian passage. There, the rim of Alcon’s cup is said to be “rough with gilded acanthus” (701, inaurato . . . erat asper acantho); that is, made rough by the engraved or embossed pattern disturbing its surface, words that playfully invert the “soft acanthus” with which Alcimedon had decorated Damoetas’s pair of cups (Ecl. 3.45, molli circum est ansas amplexus acantho).123 Doubling this inversion is that of Alcimedon’s otherwise unknown status in Ovid’s Alcon, who shares his name with a well-known Hellenistic craftsman of drinking-cups.124 In Chapter 3, we saw how the cups of Eclogue 3 instill doubt over the relation between artistic and social order, with Vergil’s rustic herdsmen failing to appreciate their highly cultivated Greek iconography. For an audience sensitive to this feature of Vergil’s cups, the opening of Ovid’s ecphrasis must cast doubt at once on the exact relation between imagery and context—both within the Metamorphoses and in Augustan Rome. What follows only enhances such uncertainty by calling attention, on the one hand, to the ambiguities specific to visual imagery and, on the other hand, to the artist’s role in shaping his audience’s response to such objects. Anius’s cup depicts a plague at Thebes. This city, however, is unnamed, although identified definitively, we are told, by its seven gates (Met. 13.685–86), which function “in place of a name” (686, pro nomine) and “showed which city it was” (quae foret illa docebant). Ovid’s self-conscious nod here to the “need for interpretation” imposed by silent artworks and the often-deceptive gap they create between themselves and the viewer finds a parallel in the way the cup’s images of mothers with unbound hair and bare breasts “signify grief ” (689, significant luctum) and the nymphs whose springs have dried up “seem to weep” (flere videntur). For a moment the imagery nearly takes on a life of its own, as a bare tree “stands stiff ” (691, riget) and goats “nibble” (691, rodunt) . But the power of the artist asserts itself immediately after this in the words “behold, he makes” (692, ecce facit) recalling through the vividness of the present tense the artistic process—startling in a description introduced in the pluperfect (fabricaverat . . . caelaverat).125 The interruption here in the flow of the ecphrastic narrative reminds us abruptly not only of Alcon’s creative presence but also of Ovid’s and his narrator’s, as though to put us on our guard as we reach the heart of the description. The Orionids’ death for the good of their city now comes to the fore. They are literally central, appearing “in the middle of Thebes” (692, mediis . . . Thebis), and their tale illustrates the centrality of female self-sacrifice throughout both the “little Iliad” and “little Aeneid” passages of the Metamorphoses.126 The exact 123

Hopkinson 2000: 204; as he notes, asper is “a technical term for the effect of embossing.” Hopkinson 2000: 201. 125 Cf. Hopkinson 2000: 202. 126 Keith 2000: 125–26. 124

t h e c a p t o r’s i m a g e

174

details of their deaths—each stabbing herself in the neck with a shuttle—appear in lines (693–94) whose authenticity has been doubted.127 Yet even if the characterization of the first daughter’s wound as “not womanly” (693, non femineum) is the observation of a commentator eager to display knowledge gleaned from later accounts, the epithet well reflects the way the women’s story becomes available as a precedent for Aeneas’s own.128 A more certain parallel with Aeneas is the description of the Orionids’ act as “for their own people” (695, pro populo . . . suo) and of the Coronae’s emergence from their ashes as occurring “lest their line die out” (698, ne genus intereat). Not only familial continuity but also glory attends the Orionids’ demise: their bodies are borne “through the city in a splendid funeral” (695–96, pulchrisque . . . per urbem / funeribus) and burnt, as a sign of honor, in its marketplace (696). Focusing on these details in isolation, an audience could perhaps read Anius’s cup simply as a token of encouragement for Aeneas and his Roman inheritors, yet Ovid emphatically disallows this. For the juxtaposition of the Aniads’ deaths, recalled by the ecphrasis, raises unanswered questions about the Orionids’ fate. Once again, we are faced with the impossibility of isolating a single mythic narrative from others to which it is related by contiguity or formal similarity. As one who uses a mythological narrative to encourage Aeneas in his mission, Anius stands in a position analogous to that of Augustus, whose visual implementation of the Aeneas myth itself was meant as encouragement toward a renewed and expanded Roman empire. Through the images he chose to publicize, Augustus sought to give the impression of, on the one hand, a transparent likeness between his own political re-organization of the Roman state and the founding acts of Aeneas and, on the other hand, of the unassailable singularity of Rome’s metamorphosis into a society to outdo the Greeks. Ovid, by contrast, presses his audience to ask why Anius chooses this cup for Aeneas and whether the parallels it creates between one mythic narrative and another are indeed as transparent as they seem, or whether Aeneas’s story, once seen in this light, can ever recapture an aura of uniqueness like that which Augustus hoped would adhere to his regime. The iconography of Anius’s cup suggests not only the rewards of heroic self-sacrifice but also, in looking back at Anius’s own painful loss of his daughters, implies that self-sacrifice for an ostensibly noble cause may, from a different perspective, appear to be only loss and “evil.” Anius has not realized any benefit from his daughters’ self-sacrifice, because his “timid” son’s consequent survival on Andros offers him no “help.” Thus what creates this ambiguity is the Greek mythical tradition itself, both mirror and abyss to those seeking a parallel for 127

See Hopkinson 2000: 203. On the lines’ authenticity and the Orionid myth as reported by the later imperial author Antoninus Liberalis (perhaps 2nd c. ce), see Hopkinson 2000: 32–33, 203. 128

Heroic Objects: Ecphrasis in the Aeneid and Metamor phoses

175

heroic action in the present. Vergil, although far more interested than Ovid in the reactions of his internal viewer, Aeneas, creates a similar vantage point on Aeneas’s journey by juxtaposing it with the story of Daedalus told in an ecphrasis, and Ovid’s use of such contrast between description and frame assuredly owes something to this precedent. Ovid thereby reworks still another Greek tradition, since both passages recall the contrastive Hellenistic technique typified by Moschus’s Europa and adapted to Latin verse, as we have seen, already in Catullus 64.129 But even more than Vergil, Ovid shines a spotlight on the act of juxtaposing mythic narratives for “inspirational” effect and exposes its difficulties. There is, of course, no explicit rejection of “official” ideology here: Aeneas still goes forward to success, and Rome still follows in his glorious wake. Yet an audience receptive to the “spirit of irony and ambivalence”130 expressed both here and in the Metamorphoses’ other ecphrases could perhaps never look upon Augustan public monuments in the same way again. Ovid’s heroic objects, like Vergil’s, present elusive allegories of an Augustan viewer’s experience, and, at the same time, of the Augustan epic poet’s when faced with the more restrictive ambitions of public iconography.

Conclusion: The Heroism of the Epic Image In conclusion, let us draw together this discussion by returning to the theme of heroism. In the Aeneid, images act heroically upon the hero Aeneas so as to further the course of his ambiguous trajectory toward and away from Greekness. They do not change him the way a god might (e.g., make him invisible as Venus does), but exert exceptional fascination by being both of his world and beyond it. Any great artwork might be said to possess a “spark of the divine,” but these objects possess an exceptional power competing with that of the hero himself, and on similar terms. They also confront, implicitly, Augustus’s own heroic appropriation of mythological iconography with a force both affirming and derailing, opening up associative expanses that Vergil does not ultimately delimit, indeed, whose lack of firm delimitation is precisely the point. In the Metamorphoses, ecphrastic images lack self-evident effects on their internal viewers, but nevertheless body forth the poem’s two-sided power to affirm and undermine traditional notions of heroism, to construct a new heroism centered in part on change itself. This heroic power once again extends to Augustus’s own image-world, which the poem refigures at least as ambiguously as the Aeneid. Both epics, moreover, amplify our sense of the image’s bivalent power in episodes lying outside this chapter’s main focus. This is easiest 129

Cf. Hopkinson 2000: 33. On the “characteristically Alexandrian” nature of the ecphrasis of Anius’s cup, see Myers 1994: 100. 130 Papaionannou 2005: 32.

t h e c a p t o r’s i m a g e

176

to see in the Metamorphoses, with its well-known stories of passage from body to object, from object to body, and the beautiful image’s transformative grip on the human psyche. Recall the tales of Pygmalion and Narcissus, the petrified, statue-like Niobe, and the statues into which Perseus, aided by the Gorgon’s head, changes his foes.131 Even in the Aeneid, artistry absorbs human beings into its own superhuman or nonhuman designs, as in the Trojan Horse, Daedalus’s wings, or, perversely, Pasiphae’s wooden cow. And the ecphrases treated here are of course the models—heroic again—for countless later examples, a prodigious mimesis already begun in Ovid’s allusions to Vergil. In their heroism, Augustan epic’s ecphrases can be understood to assert themselves, too, over the author who constructs them and the audience who visualizes them, by embodying something the author, narrator, text, and audience cannot finally control in full. The text that creates them also presents itself as ceding to their superhuman power, just as with the statues of divinities treated in Chapter 4. Their type of agency and heroism competes with those of the epic poet, also a heroic figure, and this is part of their whole raison d’être. They are simply too great, too wondrous, to tell of, even for a Vergil or an Ovid. Thus it is no wonder that they can be mobilized, through the audience’s visualizing imagination, to complicate even an emperor’s public visual program, specifically to create tension between two aspects of the epic ethos that Augustus sought to fuse: the transformative journeying of the epic hero and his encounter with visual wondrousness, itself a result of his quest and an occasion for reflection upon its very nature. Greece as unrivaled source of artistic wonders becomes a realm of significant tension in the hands of the Augustan epicist and, through him, of the Princeps as well. Yet not all imperial-age Roman responses to Greek ecphrasis were so heroic. Witness Petronius’s Satyricon.

131

For their attention to viewers, art, and transfixing (or transfixed) visual objects, all of these stories, though not involving ecphrasis proper, are often discussed in connection with the passages treated here. See, e.g., Hardie 2002; Elsner 2007: 67–176; Feldherr 2010: 243–341.

6

Sex, Satire, and the Hybrid Self in Petronian Ecphrasis

The act of describing an artwork often has an erotic dimension. The history of Roman ecphrasis traced thus far confirms this amply, but until now we have not remarked as such the striking array of both comic and serious elements embraced by the Latin authors in mobilizing the trope’s erotic associations. Already in Roman comedy, Plautus has Menaechmus I advertise his erotic appeal by parading, as it were, imagined paintings of Ganymede and Adonis that he insists are the image of his crossdressed self. Terence, by contrast, makes Chaerea find justification for raping Pamphila in a painting of Jupiter and Danae that he describes to Antipho. Catullus has the voyeuristic narrator of poem 64 envision an image of Ariadne in terms that recall “Catullus’s” tormented (though not uncomical) passion for Lesbia. Horace’s figwood Priapus describes himself, erect member and all, though only to deflate his sexual prowess via the expulsions from his split buttocks. Even in the loftier register of epic, Vergil can anticipate Aeneas’s fateful love affair with Dido through the image of Achilles and Penthesilea, while Ovid deploys possibly obscene images of gods coupling with mortals to convey Arachne’s challenge to divine authority. As the Horatian example especially suggests, the erotics of ecphrasis, in their recurrent association with doubtful or troubling male claims of dominance over supposedly passive, feminized visual objects, lend themselves to the serio-comic aims of satire. It fell to the author of the Satyricon, our sole example of a satiric, prosimetric novel in Latin, to give this truth its most complex expression in any surviving work from Greco-Roman antiquity. In the Satyricon’s gallery scene (83–90.1), its Greek-named, Latin-speaking protagonist Encolpius views and inquires about a series of paintings supposedly by Greek masters. The episode directs its satiric barbs not only at the attempt to produce a version of literary ecphrasis in Latin, but also at the attempt to characterize images from and of the Greek past as in need of such

177

t h e c a p t o r’s i m a g e

178

explication in the first place.1 The scene is thus ideal for inclusion in this study for the way it challenges assumptions about the Latin ecphrastic text’s dominance over the Hellenic image in still a different fashion from those outlined in other chapters. Here Petronius assimilates his characters’ efforts at ecphrasis to failed or comical pederastic conquest. Encolpius’s pederastic fantasies, which he expresses in relation to a series of mythological paintings, revolve around his love-object, the boy Giton, from whom he has been painfully separated in the scene immediately preceding this one in the Satyricon’s now fragmentary text. The fantasies of Eumolpus, another would-be ecphrasist who enters later in the scene, center on a beautiful boy he says he encountered in Pergamum, that city’s proximity to Troy lending humorously erotic content to his interest in a painting depicting the Sack of Troy. Such goings-on reveal how, in Petronius’s satiric world, seemingly no entity, whether text or image, Greek or Roman, can escape the decentering, destabilizing force of the satirist’s gaze. Indeed, the fusion and friction of competing Greek and Roman identities— both premise and product of Petronian ecphrasis—actually becomes a theme of this part of the narrative through the hybrid world of southern Italy that forms its backdrop. As much as it confirms my argument, however, the Satyricon occupies a problematic position within this book’s essentially chronological structure, insofar as the identity of its author and even its rough date remain disputed. Identification of the Petronius whose name appears in our manuscripts with the figure mentioned as a courtier of Nero in Tacitus’s Annals (16.18–19) is still an unsettled question. While many would accept a Neronian date for the Satyricon, and the text gives the appearance of being set in about the mid-first century ce, we cannot rule out definitively its composition at a later time.2 Thus while other chapters elaborate the connection between texts and their historical contexts at some length, I will limit discussion of the Satyricon’s hypothesized relation to Nero’s Rome to a few concluding remarks. Absent discussion of this kind, however, I will place more emphasis than elsewhere on ways this particular instance of ancient ecphrasis can be productively read with one of my modern theoretical touchstones, the work of W. J. T. Mitchell.3 Mitchell’s perspective, while especially valuable here for its insistence on affect as well as on the social implications of ecphrasis’s supposedly “neutral” stance, leaves itself open, I will argue, 1

I refer to Encolpius as “Greek-named, Latin-speaking” because we simply do not know what he and the Satyricon’s other Greek-named protagonists, Ascyltos, Giton, and Eumolpus, “really” are—Greek, Roman, or something in between: we simply are not told in what we have of the text. For reflections on Encolpius’s identity, see recently Laird 2007: 151–52. 2 Laird 2007 offers a helpful survey of the extensive speculation over the identity of the Satyricon’s author and its date. Cf., more recently, Schmeling 2011: xiii–xvii; Völker and Rohmann 2011. 3 For Mitchell’s views on ecphrasis, see above, pp. 3n5, 10, 74.

Sex , Satire, and the Hybrid Self in Petronian Ecphrasis

179

to appropriation by Petronian satiric discourse, whose terms in fact anticipate its own. This becomes especially clear in the gallery scene but we cannot understand fully the satiric use of ecphrasis there without first examining the only other, earlier fragment where it plays an important role, the Cena Trimalchionis, an account of an outrageous dinner party thrown at the house of the nouveau-riche freedman Trimalchio. Lacking, sadly, the whole novel within which to situate these two episodes, we can say that the Cena anticipates the gallery scene through two sets of images that, like the gallery’s paintings, collectively implicate author and narrator in a web of satiric associations involving both Greek culture and the erotic. The first is the Homeric and biographical murals near the entrance to Trimalchio’s house, together with the tomb that Trimalchio later instructs his friend Habinnas to build for him. The second is the series of preposterous food-artifacts crafted by a cook named after the primal Greek artist, Daedalus. In the first case, the image of Trimalchio’s “triumphal” entry, as a slave, into Rome foreshadows a similar blend of epic and quotidian themes in painted scenes from Homer’s epics and gladiatorial combats. It is capped in ring-composition by the tomb’s focus on Trimalchio’s achievements, including his amatory relations, and, again, gladiators. In the second case, the viewing and description of comically mock epic dishes, especially a great, circular zodiac dish reminiscent of the epic shields we have treated in Chapter 5, culminates in a pastry Priapus whose proffered delicacies end up defiling the guests. As we shall see, all of these images inform the Satyricon’s larger “exploration of aesthetic awareness”4 as it relates to both eros and Greek culture, a nexus of themes that is key to the gallery scene. What the Satyricon lacks in definitive connections to a historical milieu, it makes up for, so to speak, in its dense, kaleidoscopic engagement with the Greco-Roman literary tradition, as well as in its scrutiny of ecphrasis itself as both an aesthetic and cultural practice. Here epic allusions jostle side by side with recollections of verse satire, comedy, and mime (among other literary genres) and all contribute to the satirist’s paradoxically self-subverting aims. Through interaction with a wide field of Greek cultural associations inseparable from the text’s Roman milieu, ecphrasis in the Satyricon brings out and develops the satiric potential of the trope as it was already being used in Latin literature. Yet the satiric impulse itself can become, for us, a valuable way into the assumptions behind ecphrasis, not only in antiquity but also as the subject of theoretical inquiry today. Thus the brilliance of Petronian wit, when focused on the Hellenophile pretensions of the Satyricon’s bawdy, bathetic cast of characters, illuminates both its literary antecedents and its conceptual legacy. 4

Rimell 2002: 61.

t h e c a p t o r’s i m a g e

180

Chewing on Art: Ecphrasis, Greek Culture, and Eros in the Cena Trimalchionis Petronius’s account of the dinner party at Trimalchio’s will always be thought of in connection with the outrageous display of food and its overindulgent consumption, aspects of the scene brought home especially by Fellini’s memorable filmic rendition. Yet the themes of art, literature, and culture are never far off, even at the episode’s most ludicrous moments, and are in fact intrinsically related to its interest in gross misappropriation of social convention including sexual mores.5 All of these elements come together in the figure of Trimalchio, an extravagantly appetitive creature whose questionable interest in art virtually encompasses the entire Cena, since the description of his unusual wall paintings coincides with Encolpius’s entrance into his house and the account of Trimachio’s decorated tomb is among the wildly self-centered, death-obsessed performances that he inflicts upon his guests toward the end of the dinner-narrative as we have it. Let us begin our close analysis of ecphrasis in the Satyricon, therefore, by examining how Trimalchio’s gaffe-ridden, narcissistic attempts to display a knowledge of Greek culture—an important antecedent for those of Encolpius in the gallery scene—play out in these framing ecphrases, visual reminders of the whole novel’s amalgam of high and low, self and other, Greek and Roman. Trimalchio’s perceptions of art are of course inseparable from Encolpius’s, since the audience experiences the former through the latter (whichever voice is actually describing the object in question) and each bears upon the other while also implicating those of the “hidden author,” Petronius.6 In the course of the dinner, Trimalchio, as reported on by Encolpius, will make ludicrous missteps where basic ecphrastic conventions are concerned, as for example when he says (52.1) that he has drinkware illustrating Cassandra killing her sons, with the boys lying “dead” in a lifelike fashion!7 But Petronius puts his audience on guard about Encolpius himself as an interpreter of art at the very moment he enters Trimalchio’s house, when his misperception of a painted dog guarding the door nearly causes him to fall and break his leg (29.1). This inauspicious beginning encourages reflection on how the images that follow link Encolpius as ecphrasist to his host, particularly as imperfect manipulator of the startling, 5

My understanding of the Cena Trimalchionis derives especially from Sullivan 1968 passim; Walsh 1970: 111–40; Zeitlin 1971a: 659–66; Slater 1990: 50–86; Panayotakis 1995: 52–109; Conte 1996; Connors 1998: 51–62; Plaza 2000: 84–164; Rimell 2002: 49–59. 6 The term is Conte’s (1996); see above, p. 41, n5. 7 Cassandra’s sons by Agamemnon were murdered by Aegisthus (she is here confused with Medea), while Trimalchio makes a comical hash of the ecphrastic topos attributing life to visual images (see above, pp. 132–33). Cf. below, pp. 183–84 for further mistakes like this on Trimalchio’s part.

Sex , Satire, and the Hybrid Self in Petronian Ecphrasis

181

deceptive spectacles punctuating the Cena. Although Petronius distinguishes the more educated Encolpius as “an outsider to the milieu of the freedmen at the banquet,”8 he nevertheless slips in further encouragements toward an interpretation connecting the two. This is essential to my approach here, since it tightens the connection between the Cena and the gallery scene, where Encolpius and his view of art, together with those of another alter ego, Eumolpus, will reclaim the spotlight. Indeed, one of the ways that Petronius most unsettles ecphrasis’s authoritative regard upon its objects, as well as its supposed grounding in the stable identity of a speaking subject, is in the use of descriptions by one character that are reported by another, who nevertheless resembles the ecphrasist “himself ” to an uncanny degree.9 Like Encolpius at this point in the narrative, Trimalchio as portrayed in his paintings is in the midst of an entrance, and his servile status points to the subordinate position Encolpius has taken on vis-à-vis Trimalchio in his role as master of the house. Encolpius sees a slave market, complete with descriptive titles, where Trimalchio, long-haired and holding a caduceus, is being led into Rome by Minerva (29.3). Here the detail of Trimalchio’s long hair suggests the belief in his own sexual appeal he will later express in claiming to have served his owners as a sex object (69.3, 75.11; cf. Eumolpus’s verses concerning the erotic appeal of hair at 109.9–10). The vain Encolpius’s obsession with his own sexual appeal and abilities is, in turn, one of the novel’s perennial themes, as is his attraction to young boys, another trait he shares with Trimalchio, who presumably appears as a young man in the painting. More images of Trimalchio’s life, also explained by identifying inscriptions, follow: Trimalchio is shown learning how to keep accounts, then becoming “steward” (29.4, dispensator). At the end of the portico, Mercury lifts him by the chin onto a high platform, while Fortune holds a cornucopia and the three Parcae spin the thread of Fate. The whole pompous visual narrative of Trimalchio’s social advancement—its main cultural referent is the iconography of major achievements found in the atria of aristocratic Roman houses—likewise suggests the social ambitions of Encolpius himself, whose efforts to get ahead through rhetorical training are in evidence before this (1–5). Encolpius’s view of his host’s self-display turns our gaze back on Encolpius himself as both object and socially situated gazer. There seems at first to be little culture in Trimalchio’s attempts at visual self-promotion. After this beginning, however, the imagery of his dwelling takes 8

Zeitlin 1971a: 663. Cf. 665: “In the Cena, the pretentions to literature and to philosophy satirize the gaucherie of the freedmen, but the inadequacy of these standards as a sign of aristocracy and humanitas is also mocked.” 9 Cf. Rimell 2002: 46: “[W]hen Trimalchio, whose role throughout the Cena is to deceive through dramatic or visual representation, describes the pictures on his crockery, we cannot rely on Encolpius to confirm that the description is actually representative, or to decipher where description crosses over into interpretation.”

t h e c a p t o r’s i m a g e

182

on a more Hellenic—and more overtly literary—cast in the pictures illustrating the Iliad and the Odyssey that Trimalchio’s steward describes for the inquisitive Encolpius (29.9). As decoration situated near the entrance, these paintings would, like the biographical frieze, be on view to nearly every visitor to Trimalchio’s house, their contents meant to reflect the image of himself he was especially eager to present. Thus they are especially funny both in the way they follow upon the self-inflating biographical pictures and in the way Trimalchio has conjoined them with an altogether different kind of scene, the “gladiatorial games of Laenas” (29.9), presumably a local official of recent vintage. Satiric bite inheres in the misplaced collocation of the oldest, most august Greek literature and vulgar, contemporary entertainment, a gladiatorial spectacle. The implicit reduction of Homer to the level of a gladiatorial show, or the elevation of gladiators to Homeric status, throws Trimalchio’s claims to Hellenophile literary sophistication into doubt as soon as they are made. In this, he continues to mirror Encolpius, whose literary knowledge, if more ample than Trimalchio’s, is often exploited by Petronius to cast him in a highly satiric light. For an audience steeped in the ecphrastic tradition, moreover, both this and the biographical paintings resonate satirically with the kind of imagery appropriate to epic or possibly the Greek novel, where those who encounter programmatic paintings at the outset of the narrative may elicit the services of an interpreter, just as Encolpius depends on the steward.10 The burlesque of epic suggested by its visual proximity to arena spectacle resonates in general with the Satyricon’s mock epic aesthetic, and particularly with that of the Cena. But a specifically Vergilian allusion is at work here: just as Aeneas views Daedalus’s biographical and (Homeric) epic imagery before entering the underworld in Aeneid 6 (cf. chapter 5), Encolpius views absurd versions of such images, depicting another journeyer, Trimalchio, before entering the latter’s house, which is itself clearly assimilated to the Vergilian underworld through this and other aspects of the narrative.11 Petronius thus characterizes both Encolpius and Trimalchio as Daedalean wanderers of a satirical variety, while Encolpius becomes the quasi-Roman revision of the Greek artificer in the model of Vergil’s Aeneas. Encolpius is in this way, too, Trimalchio’s successor on a dubious path toward self-distinction. These literary foreshadowings will be taken up in the body of the Cena Trimalchionis and developed further in the gallery scene, but to turn, for a moment, to the framing recollections of them near the end of the Cena, let us consider Trimalchio’s description of his own tomb monument (71.5–12). It, too, will depict gladiators—“all the fights of Petraites” (71.6, Petraitis omnes pugnas), evidently a famous arena combatant (cf. 52.3)—while its outsized dimensions, one hundred 10 11

See below, pp. 195–96. See Courtney 1987; Slater 1990: 77–79; Bodel 1994; Connors 1998: 35–36.

Sex , Satire, and the Hybrid Self in Petronian Ecphrasis

183

feet in width and two hundred feet in depth, recall epic.12 The cultural satire effected by these details returns at the end of the ecphrasis in Trimalchio’s boast (part of the inscription for his tomb) that he “never listened to a philosopher” (71.12), in spite of the highbrow learning he is eager to show off elsewhere. Further enhancing the bathos of Trimalchio’s tomb is the runaway hodge-podge of personal symbols, sculpted and actual, it is to contain, and here the erotic dimensions of his self-display are still more apparent. The monument, Trimalchio specifies, is to include sculptures of his little dog, wreaths, and unguents, and its enclosure is to be planted with fruit trees and vines. There will also be sculpted ships, an image of Trimalchio in his magistrate’s attire distributing largess, a sculpted banquet scene and a statue of his wife, Fortunata, holding a dove and leading another little dog. While the dove, a symbol of Venus, and the dog suggest Fortunata’s sexual fidelity, other erotic interests of Trimalchio’s color the ecphrasis, as in the representations of his boy favorite and a broken urn with a boy grieving over it. Mention of a sundial and, finally, an appropriately self-aggrandizing inscription round off the passage, which like the account of the earlier paintings thus revolves around the themes of dubious personal achievement, cultural missteps, and eros. The presence of these themes in art representing both the beginnings and the end of Trimalchio’s life story indicates their importance in the dinner-narrative they enclose. Turning, therefore, to the Cena’s central sections, let us first follow up the epic allusion most clearly gathering together these themes within its frame: reference to the mythical Daedalus. If the Cena recalls the Aeneid’s underworld sequence in beginning with the biographical image of a Daedalus-figure and ending with a tomb and funeral (cf. those of Caieta at Verg. A. 7.1–4), Daedalus also makes his presence felt in the name of Trimalchio’s “most learned” (Petr. 74.5, doctissimo) cook.13 The mythological misprision implied by turning the great, though transgressive, Greek artisan into a culinary jack-of-all-trades (he received his name, Trimalchio informs us, because he could make foods look like other foods [70.1–2]) is very much in keeping with the hilarious errors Trimalchio perpetrates, for example, when he says he has drinkware showing the mythical Daedalus enclosing Niobe in the Trojan Horse (52.2, another brief ecphrasis) or in his interpretation of the show put on by the Homeric performers at his banquet (59).14 The banquet’s repeated exploitation of whole animal carcasses and their surprising contents (40.3–7; 49) even seems to echo 12

Yet another collocation of epic and gladiatorial imagery has occurred before this when Trimalchio says he has drinkware depicting Cassandra, Daedalus, Niobe, and “the fights of Hermeros and Petraites” (52.3, Hermerotis pugnas et Petraitis). On the identities of these gladiators, see Smith 1975: 139–40; Schmeling 2011: 214. 13 For more on the significance of Daedalus-the-cook’s name, see Connors 1998: 21–22, 36; Rimell 2002: 45–47, 54, 106, 155, 168, 178–79, 201. 14 Cf. the literary mistakes at Petr. 48.7 and the botched etiology of Corinthian bronze at 50.2–7.

t h e c a p t o r’s i m a g e

184

the erroneous Niobe reference in recalling the mythical Daedalus’s fabrication of an artificial cow for Pasiphae to enter so as to mate with the bull.15 And the culinary Daedalus’s delicacies at least once embody Trimalchio’s misunderstanding of Greek myth, as when a performer dressed as Ajax attacks a boiled calf wearing a helmet after Trimalchio has wrongly attributed Ajax’s madness to his having been deprived of Iphigenia (59.5–7). In transforming Daedalus into a cook, furthermore, Trimalchio also removes him from his mythical role as architect of the labyrinth designed to enclose the monstrous evidence of Pasiphae’s sexual transgression, which he in fact abetted. This in spite of the fact that Encolpius calls Trimalchio’s house a “labyrinth of a new kind” (73.1, novi generis labyrintho) when the steward informs him that no guest ever leaves by the same door through which he came in. The dinner party itself thus comes to embody a dizzyingly bathetic reception of Greek culture in its display of the reassigned Daedalus’s handiwork.16 Ecphrasis is at the heart of this satiric tour de force. Consider one of Daedalus-the-cook’s most impressive creations, a dish illustrating all the signs of the zodiac with food items corresponding to each one. This implement recalls the Vergilian description of art that is not in fact by Daedalus: the Shield of Aeneas.17 Like the Shield, the zodiac dish is a circular object (35.2, rotundum . . . repositorium), the arrangement of its images reflecting its rounded appearance (compare 35.2, “the twelve zodiac signs laid out in a circle” [duodecimo . . . signa in orbe disposita] with Verg. A. 8.673, where the dolphins on Aeneas’s shield run “around in a circle” [circum . . . in orbem]).18 It is linked to the epic Shield by the way Encolpius introduces it as “a tray plainly not as big as we thought it would be” (35.1, ferculum . . . plane non pro expectatione magnum), witty epic parody lurking in the internal audience’s disappointed expectations of a “big” plate bearing plentiful grub. The surface with the astrological signs actually turns out to be a lid covering a receptacle with more ample—and no less mock epic—fare: fowls and sows’ udders surrounding a

15

Cf. Rimell 2002: 71, who calls attention to the further similarity between “Daedalus”’s pig carcass, which spills out fake guts, and the Trojan Horse (as later described by Eumolpus), which spills out Greek soldiers. 16 Cf. also Trimalchio’s bestowal of extravagant rewards and privileges on his cook in excess of the latter’s servile status (50.1, 70.3, 12). 17 Cf. Grossardt 2009: 339–43, who argues for the zodiac dish as a recollection of the Homeric Shield of Achilles; see esp. 342n31 on the Shield of Aeneas. Additional literary precedents include elaborate dishes in comedy and the exploitation of astrological knowledge in mime; see Panayotakis 1995: 70–71. For more on the zodiac dish within the Cena and the Satyricon, see Rimell 2002: 52–55. 18 Further, the slave’s placement of food on top of each image (35, structor imposuerat cibum) suggests Vergil’s repeated references to Vulcan’s construction of the Shield with verbs in the pluperfect (cf. A. 8.628 fecerat ignipotens; 637, addiderat; 724–26, Mulciber . . . finxerat).

Sex , Satire, and the Hybrid Self in Petronian Ecphrasis

185

hare fitted with wings to resemble Pegasus (36.1–2). But what makes this Shield manqué especially apt as a send-up of Vergil’s lengthy ecphrasis is the fact that it elicits drawn-out description by not one but two characters, first Encolpius as narrator when he recounts the visual puns linking each food item to its corresponding zodiacal sign (35.3–5), then Trimalchio as astrologer when he predicts the types of people born under the different signs (39.5–15). Encolpius and Trimalchio are associated again through the act of visual description. This doubled account restores the memory of an over-familiar literary chestnut in refreshingly risible form. The details of the zodiac plate make its allusion to Vergil’s Shield all the clearer. The plate’s “novelty” (35.1, novitas), which attracts the attention of the guests, wittily recalls the wondering Aeneas’s ignorance over the contents of his prophetic shield (Verg. A. 8.730, rerumque ignarus). Like Vergil’s model, the Shield of Achilles, which was interpreted as a cosmic icon already in antiquity, the Shield of Aeneas is replete with cosmic significance.19 Yet while Aeneas’s armament conveys the cosmic dimensions of Roman history’s culmination in the Battle of Actium, Trimalchio’s plate finds cosmic meaning in a random assortment of eatables, whose connection, through the astrological signs, to a motley collection of characters presents a laughable parallel with the opposition of heroic figures and formidable enemies on the Vergilian shield. Vergil depicts momentous struggle to fulfill a glorious destiny, but Petronius throws together self-assertive or questionably successful figures (note especially the “domineering men” [39.9, imperiosi] born under Leo) and the subordinate (criminals, animals, slaves, the deformed, etc.) in a kind of dog-eat-dog race to the bottom.20 Such inversion of Vergil culminates in the fact that the center of the plate, which bears a piece of turf, represents the earth as the center of the universe (note in medio [35.5, 39.15]), while the Shield puts the sea battle of Actium at the center of the Vergilian cosmos (Verg. A. 8.675, in medio). Further evidence for this Vergilian connection is the fact that Trimalchio launches into his astrology just after quoting the Aeneid, notably at a moment of pointed allusion to Homeric precedent, Laocoon’s warning to the Trojans about accepting the Trojan Horse into Troy, “Is Ulysses known to be thus?” (Petr. 39.3, sic notus Ulixes?; cf. Verg. A. 2.44). He even goes on to declare his knowledge of “literary learning” (philologiam), and the zodiac dish, he announces, is illustration of the fact that “nothing new can be brought to me” (39.4, mihi nihil novi potest afferri), apparently another nod, on Petronius’s part, to the Shield of Aeneas’s familiarity.21 19 On the cosmic implications of the two Shields, see above pp. 154, 162 and cf. Grossardt 2009: 339–43. 20 Cf. Rimell 2002: 53. 21 Cf. Grossardt 2009: 340–41 on the suggestion here of Homeric philology on the Shield of Achilles.

t h e c a p t o r’s i m a g e

186

Finally, when Trimalchio wins the guests’ praise for surpassing Hipparchus and Aratus in astronomical knowledge (40.1), the learned audience recognizes currents of Hellenistic learning also found in Vergilian ecphrasis.22 While eros is not prominent in the zodiac plate, the trope’s erotic aspects are undeniably explored through a different food item, the pastry image of Priapus (60.4, Priapus a pistore factus) that appears just after a central event in the dinner spectacle, the parting of the ceiling to reveal another circular device bearing gifts for the diners (60.1–4). Here especially, in pointed anticipation of the gallery scene’s satiric blend of sex and artistic description, Encolpius’s ecphrastic account turns the described object’s erotic potential back upon the ecphrasist to call his dominance over the image into the utmost doubt. This Priapus appears in the middle of a plate filled with cakes, “and was holding up fruits and grapes of every sort in its quite ample bosom in the usual fashion” (60.4, gremioque satis amplo omnis generis poma et uvas sustinebat more vulgato). We are clearly to imagine an edible version of a statue something like the one depicted in figure 6.1. The Priapus shown (although its member has been obscured by post-classical carving; but cf. figure 3.2) suggests not only how the act of consumption would be obviously sexualized in the Petronian context, but how it would amount to an implicit admission of dependency on the deity’s powers, the proximity of phallus and fruit alluding his role as a fertility god responsible for vegetable as well as human fecundity. The textual image, however, is not just one delicacy among others, but occupies a special place in the Cena, since the Satyricon appears to take as an overarching theme the god’s torment of Encolpius, who would thus resemble Odysseus as pursued by Poseidon or Aeneas as tormented by Juno.23 Priapus’s appearance qua cake thus becomes a satiric epiphany, an effect enhanced by Encolpius’s insistence that he and the other guests thought that this must be a “sacred” dish (60.7, sacrum) because it was “overspread with such religious trappings” (tam religioso apparatu perfusum), as well as in the blessings for the Emperor that the guests subsequently invoke in a show of reverence (venerationem). As a god whose power over the Satyricon’s narrator is potentially manifest at every point in the novel, the edible Priapus bears satirically all the connotations of the religious icon explored in Chapter 4: such an object cannot definitively be said entirely to “submit” to verbal description, especially since the image that the ecphrasist—here, Encolpius—describes may be the product, through phantasia, of divine agency. But what is more, Petronius leaves no doubt about the sexually debased position of Encolpius here: as soon as he and the other diners touch Priapus’s cakes 22

See above, pp. 76–92. See Sullivan 1968: 216; Walsh 1970: 76–77; Conte 1996: 94; Connors 1998: 24–33; Rimell 2002: 98–112. 23

Sex , Satire, and the Hybrid Self in Petronian Ecphrasis

187

Figure 6.1 Priapus statue holding fruit. Vatican Museums, second c. ce. Photo Credit: Vanni/Art Resource, NY.

and fruits, they spurt liquid saffron, this “disagreeable fluid” (60.6, molestus umor) landing upon the guests themselves.24 Given the association between statues of Priapus and violent sexual humiliation that we have observed in the Carmina Priapea, the subordinate position of the defiled guests, stained by Priapus’s

24

Perhaps even on their faces or in their mouths. For MS H’s ad nos (“on us”) Buecheler 1922: 41 conjectured ad os (“on [our] faces”).

t h e c a p t o r’s i m a g e

188

ejaculatory emissions, could not be more clear.25 Yet in spite of the unpleasant outcome, the guests continue to grab at the fruit (cf. 60.7, poma rapientibus), filling their napkins with these sweets in a greedy show, especially Encolpius, “who was thinking that I could not load up Giton’s bosom with an ample enough gift” (qui nullo satis amplo munere putabam me onerare Gitonis sinum). Would-be worshippers, the guests end up as little more than over-excited thieves, while impotence and other sexual misfortunes, the effects of Priapus’s displeasure, continue to dog Encolpius in the novel’s subsequent fragments. As the Cena’s ludicrously “climactic” ecphrasis, the description of Priapus’s image both repeats and satirizes in sexual terms the ecphrastic convention whereby the wondering gazer, through careful scrutiny of an art object, intimates some deeper understanding of its form and meaning. Earlier chapters have offered clear instances of this approach to the trope. Propertius’s description, for example, of an Apollo statue in Augustus’s Palatine complex signals a poetic epiphany of which Elegies 2.31 itself is evidence. The Aeneid’s elaborate ecphrases are a means of conveying essential aspects of that poem’s relation to its social, cultural, and political contexts, and here the role of the internal viewer, as distinct from that of the author or narrator, is especially marked. Indeed, Encolpius, expressing doubts over the possible religious significance of the pastry Priapus at the same time as Petronius signals its broader import in the narrative, perhaps again recalls Aeneas’s ignorant wonder over his shield’s Roman imagery, which is Vergil’s domain. Encolpius, however, is not only ignorant here but also made ridiculous, a few stolen cakes and his saffron-stained person is all he has to show for his efforts. In this he is much more the heir of Horace’s self-descriptive Priapus in Satires 1.8, a figure comically undone at the conclusion of his ecphrastic performance. Encolpius, another Priapic antihero whose member fails him, meets his match in the god Priapus—or his pastry avatar. Petronius in fact departs from Horatian precedent, and many other previous ecphrases, in suggesting the visual object’s masculine, sexually aggressive powers over the feminized, passive figure who describes it. The Cena Trimalchionis is essential evidence for the Satyricon’s larger exploration of viewing, viewers, and their visual objects, yet its very density as satire, the exuberant, mad pleasures of its text, make generalization about its deployment of ecphrasis (or anything else) difficult. Among all the other bizarreries of its multifarious realm, curious art objects and quasi-art objects come and go, described by the unreliable narrator Encolpius and his counterpart as director of the literary feast, Trimalchio. These ecphrastic images often engage Greek culture as a status symbol, but as such it is merely one of the status markers over which Trimalchio possesses a laughably incomplete control, and the fluid 25

See above, pp. 95–96. On the violent sexual connotations here cf. Rimell 2002: 100 refuting the rather different reading of Connors 1998: 30.

Sex , Satire, and the Hybrid Self in Petronian Ecphrasis

189

rebelliousness of the Priapus pastry emblematizes such incompleteness where all of the episode’s “art” objects are concerned. As we will now see, such sexual satire is a still more prominent thematic component of the Petronian gallery scene, where it also becomes linked even more forcefully to characters’ Greek cultural pretentions.

Riding the Trojan Horse: Painting, Pretension, and Pederasty in Petronius’s Gallery Scene At section 83 of the Satyricon, not long after his visit to Trimalchio’s, Encolpius enters a wondrous art gallery containing a variety of paintings.26 Surveying its contents, he finds himself drawn to what he perceives to be the work of the great Greek artists Zeuxis, Protogenes, and Apelles, whom he praises in extravagant terms for their realism. He becomes fixated on images depicting the gods’ loves for young boys: On this side the eagle aloft was carrying the Idaean boy [Ganymede] to the heavens, on that fair Hylas was repelling the shameless Naiad. Apollo was condemning his guilty hands [for killing Hyacinthus] and was conferring honor upon his unstrung lyre with the newborn flower. hinc aquila ferebat caelo sublimis Idaeum, illinc candidus Hylas repellebat improbam Naida. damnabat Apollo noxias manus lyramque resolutam modo nato flore honorabat. (Petronius, Satyricon 83.3) Both of Encolpius’s preferences here—for great Greek artworks and for depictions of boy-love—fit in well with what we have already learned about him by this point in the Satyricon.27 They are especially topical because he has just lost the Greek-named Giton to his former companion Ascyltos, and has spent three days in agonized seclusion (79.9–81.6). The paintings Encolpius sees in the gallery, as he recalls the mythical narratives to which they refer, prompt him to cry out “as though alone” (83.4, tamquam in solitudine), 26

My interpretation of the gallery scene owes much to Walsh 1970: 93–97; Zeitlin 1971b: 56–67; Conte 1996: 14–21, 39–41, 90, 184; Slater 1990: 91–101, 186–90, 220–30; Connors 1998: 62–64, 84–99; Rimell 2002: 60–76; Elsner 2007: 177–99. 27 Encolpius shares the aspirations toward Greek culture displayed by numerous others in the “Greek city” (81.3, Graecae urbis) that forms the setting for this part of the text. Note for example the eulogy of Greek authors in his declamation at 2.3–8. His love for Giton has by this point already become a major theme.

t h e c a p t o r’s i m a g e

190

“So, then, love touches even the gods. Jupiter in his heavens did not find an object for his love, and though he intended to transgress on earth, nevertheless did no one injury. The nymph who preyed on Hylas would have controlled her love, if she had believed that Hercules would come to forbid it. Apollo called the shade of his boy back in the form of a flower, and all their myths, too, have told of embraces without a rival. But I admitted into fellowship a guest more cruel than Lycurgus.” “ergo amor etiam deos tangit. Iuppiter in caelo suo non invenit quod eligeret, et peccaturus in terris nemini tamen iniuriam fecit. Hylan Nympha praedata imperasset amori suo, si venturum ad interdictum Herculem credidisset. Apollo pueri umbram revocavit in florem, et omnes fabulae quoque habuerunt sine aemulo complexus. at ego in societatem recepi hospitem Lycurgo crudeliorem.” (Petronius, Satyricon 83.4–6) Now a gray-haired old man enters and approaches Encolpius. Petronius does not name him at first, but we learn later that he is Eumolpus, whose name in Greek means something like “good songster,” an epithet that will often resonate comically in what follows.28 After apologies for his shabby appearance, including some lines of verse, Eumolpus tells the story of his deceitful seduction of a young Greek boy in Pergamum. At Encolpius’s prompting, he then launches into a hackneyed diatribe on the decline of philosophy and the arts from their heights in ancient Greece. Next, seeing Encolpius’s attention fixed on a painting of the Sack of Troy, he embarks on a poem on this subject, a strange work in iambic trimeters which, after sixty-five lines, is cut off by a rain of stones hurled by other occupants of the gallery, and he and Encolpius run for cover. While the Cena Trimalchionis portrays in passing Trimalchio’s botched efforts to describe mythological imagery on his drinkware, the gallery scene goes much further in representing ecphrasis as a cultural practice open to satire for its assumptions about Latin-speakers’ ability to illuminate images from and of the Greek past. Trimalchio’s errors are patent: Niobe inside the Trojan Horse, Cassandra killing her sons, and so on. In the gallery scene, however, Petronius reveals far more subtle dangers and pitfalls inherent in the attempt by those who participate in Roman culture, however Greek their names or affiliations, to use ecphrasis to display their philhellenism and thereby reappropriate Greek culture as an aspect of their identity. Turning now to a close analysis of this episode, 28

The name belongs to several figures from the Greek world with connections to poetry, including a famous Eleusinian priest whose connection with the famous mystery cult enhances the bathetic religious associations of Eumolpus as artistic exegete. See Connors 1998: 62n34; Elsner 2007: 188.

Sex , Satire, and the Hybrid Self in Petronian Ecphrasis

191

I will show how the scene relates its aesthetic concern with the opposition of texts and images to a cultural concern with the complex interrelation of Roman and Greek by exposing its protagonists as neither comfortably Greek nor comfortably Roman in their cultural identifications, as well as hopelessly fallible in their erotic adventures. Encolpius’s strong identification with the work of the canonical Greek artists is apparent in his expressions of wonder, awe, and even worship for the Greek painters. His initial response to the gallery’s artworks (83.1–2) emphasizes their Greek provenance, before he goes on to describe their subject matter, which is, again, predominantly Greek (83.3–5). He calls the gallery itself mirabilis, and the first attempts of Protogenes, which rival, he says, nature itself, he views with a certain “awe” (83.1, horrore). He reserves the greatest praise, however, for Apelles, the outlines of whose figures are in his view “clear-cut with such great delicacy, that you would believe it was a picture of their souls as well” (83.2, tanta . . . subtilitate . . . ad similitudinem praecisae, ut crederes etiam animorum esse picturam). When he sees Apelles’ painting known as the “one-shinned” (monocnemon; perhaps the image of a goddess), “I even worshipped it,” he exclaims (etiam adoravi). But Petronius’s audience must already have its suspicions about Encolpius’s qualifications as a connoisseur of Greek art. In Trimalchio’s house, as we have seen, he seems to have had difficulty identifying pictures of the Iliad and Odyssey (29.9). He is also deceived by Trimalchio’s picture of a dog (29.1), and a series of thematic ties bind him closely to the risible Trimalchio as art-owner and viewer. Now, moreover, Petronius may be deploying deliberately flawed reminiscences of the Greek novel to suggest a still more imperfect identification with the cultural traditions of Greece.29 Like the narrators of Longus’s Daphnis and Chloe (1praef. 1) and Achilles Tatius’s Leucippe and Clitophon (1.1.2), Encolpius encounters art in a sacred setting (Petr. 90.1, templum) and, again like them, he is drawn to erotic images that mirror his own experience as a lover. In Leucippe and Clitophon, for example, the narrator specifies that it is as a lover himself (1.2.1, ὢν ἐρωτικὸς) that he is drawn to the image of Eros leading the bull, Zeus, who carries off Europa. Both Longus and Achilles Tatius, moreover, represent the texts of their novels as verbal responses to the images they describe in ecphrasis, and when Encolpius verbalizes his own thoughts upon seeing the 29

The unsettled dating of the Satyricon relative to the extant Greek novels makes caution warranted on this score. If the Satyricon is Neronian, the five extant, complete Greek novels likely all postdate it. See Laird 2007, however, for a useful survey of the chronological difficulties here as well as the possibility of an earlier date for Chariton’s Chaereas and Callirhoe and some surviving fragments of Greek fiction. Slater (1990: 91n9; 143n7, 221–23), Elsner (2007: 185, 189) and others read the extant Greek novels for traits of which Petronius would have been aware from earlier, lost examples. Jensson 2004 makes the case for a single Greek text that was the direct model of the Satyricon itself.

t h e c a p t o r’s i m a g e

192

paintings of boy-love, his resemblance to the narrator of Leucippe and Clitophon is especially close: like the Greek narrator, Encolpius is moved to a brief, sententious utterance, an emotional expression of a supposedly timeless truth. Achilles Tatius’s narrator, however, expresses a succinct, objective, and uncontroversial interpretation of the image of Eros and the bull: “See how a little child rules sky and land and sea” (1.2.1, οἷον . . . ἄρχει βρέφος οὐρανοῦ καὶ γῆς καὶ θαλάσσης). Encolpius’s interpretation, on the contrary, moves from a succinct axiom to a rather clumsy attempt to contrast what he sees in the paintings with his own experience. Encolpius, that is, dwells self-indulgently on his own predicament in a manner that distances him from the Greek novel. His departure from this tradition is underscored by his comical status as a failed lover in a Roman present gazing at images of erotic success from a Greek past, while his self-indulgence is heightened by the detail “as though alone.” Encolpius forgets that he has an audience. In addition to possible echoes of the Greek novel, there is a chronologically verifiable example of the way Petronius characterizes Encolpius’s identification with Greece as imperfect: the Aeneid’s influence on his self-indulgent response. Already in the Cena Trimalchionis, Petronius has made it clear that Encolpius knows and appreciates the Aeneid, since, for example, he reacts with displeasure when Habinnas’s slave proceeds to recite an error-ridden version of Book 5 with verses from Atellan farce mixed in (68.4–5).30 Now, as likewise in the Cena, Petronius again uses both narrative and characterization to depict Encolpius as an Aeneas figure. In the episode intervening between the Cena and the gallery scene, Petronius has made us especially conscious of Encolpius’s status as a stranger in a strange land, wracked by feelings of deracination and loneliness (to arrive in the south Italian town that is home to Trimalchio, he has journeyed from somewhere else, possibly Massilia).31 These traits emerge, for example, when Encolpius asks himself melodramatically whether all his previous adventures have amounted to becoming an exile (exsul) lying alone (desertus) in lodgings in a Greek city (Graeca urbs) (81.3). Encolpius proceeds from the shore where his lodgings are to the temple that houses the art gallery. Here in a moment of consolation tinged with a profound sense of loss, he identifies emotionally with the paintings, one of which (Ganymede) recalls

30

Cf. below, p. 193 on Petr. 132.11. From Serv. A. 3.57 we know of a lost episode from Petronius that may indicate Encolpius’s Massilian origins: in it, a poor man from Massilia, which is suffering under a plague, is fed at public expense for a year and then exiled in the hopes of expiating the pestilence. This could be Encolpius, his exile thus corresponding to that of Priapus himself from Lampascus as well as Aeneas’s from Troy, and the plague, coming perhaps at the opening of the Satyricon, echoing that of the opening of the Iliad. See Sullivan 1968: 40–42; Connors 1998: 27–29. On the Vergilian subtext of the whole of Petr. 79–98, see Paschalis 2011. 31

Sex , Satire, and the Hybrid Self in Petronian Ecphrasis

193

the Trojan War.32 And Encolpius eventually gives his full attention to a depiction of the Trojan War (“I see you are completely engrossed in that painting,” says Eumolpus [89.1, video te totum in illa haerere tabula]; cf. Verg. A. 1.495, “while he becomes astonished and engrossed, fi xed in one gaze” [dum stupet obtutuque haeret defixus in uno]).33 Aeneas, too, is an exile upon his arrival in Carthage in Book 1 and, if not alone, believes he has lost many of his comrades to the storm that opens the epic. After travelling from the shore to the temple of Juno and there discovering the Trojan War images, he becomes transfixed while literally identifying himself among the combatants, and pours out his heart in a mixture of pride and sadness. Encolpius, attempting to tie his personal difficulties directly to the images, further recalls Aeneas’s recognition of his own “labor” (Verg. A. 1.460, laboris), and his initial reaction, “So, then, love touches even the gods” (Petr. 83.4, ergo amor etiam deos tangit) may even echo Aeneas’s “mortality touches the heart” (Verg. A. 1.462, et mentem mortalia tangunt), since, like Vergil, Petronius has associated the verb tango with a noun indicating an emotional or psychological disposition (mentem in Vergil, amor in Petronius), while inverting Vergil’s emphasis on mortalia in the noun deos as well the syntactic roles of these semantic components between subject and direct object.34 Such bathos would be in keeping with that of Trimalchio’s zodiac plate, through which another one of Vergil’s purplest patches appears in a new light. While Aeneas’s identification with paintings is entirely appropriate in his case, since he is actually part of their subject matter, Petronius opens his protagonist to the charge of being solipsistically misguided in somehow aspiring, consciously or unconsciously, to Aeneas’s heroic self-recognition. Encolpius’s lament over his erotic misadventures with Giton is a bathetic counterpart to Aeneas’s lament over the destruction of Troy and the loss of his countrymen. This contrast becomes all the more pointed in that the images on Dido’s temple set the stage for Aeneas’s erotic conquest of her, while the paintings Encolpius sees signal only his erotic defeat. Bathetic citation of Vergil, moreover, in the context of erotic defeat seems to be a habit of Encolpius’s, as we realize later through the combination of Vergilian quotation and pastiche with which he describes his impotent penis (132.11).35 The first two verses of Encolpius’s poem here are a quotation of Aeneid 6.469–70, which refer to the ghost of Dido, while the third 32

Ganymede constitutes another link to the Cena, since one of Trimalchio’s guests is named Ganymede and Trimalchio mistakenly identifies Ganymede as Helen’s brother at 59 (Rimell 2001: 64). 33 Zeitlin 1971b: 60. 34 Cf. Zeitlin 1971b: 60; Conte 1996: 15; Habermehl 2006: 75; Paschalis 2011: 84. Conte 1996: 16 detects a further echo of Vergil in the rare technical use of manus “hand,” i.e., “style,” coupled with a word indicative of wonder: mirabilem in Petronius (83.1), miratur in Vergil (A. 1.455–56). 35 On the larger theme of Encolpius’s impotence within the Satyricon, see McMahon 1998.

t h e c a p t o r’s i m a g e

194

is a combination of Eclogues 5.16, where Menalcas says to Mopsus that Amyntas yields to him in song as surely as pliant willow yields to the olive, and Aeneid 9.436, on the death of Euryalus.36 Encolpius is no pure Hellenic connoisseur, but a belated, (proto-)Roman poseur. Encolpius’s response to the paintings reveals, further, an internalization of the Latin literary tradition that is open to satire insofar as he seems to have forgotten famous depictions in earlier Latin literature of the painted images he describes.37 In contrast, for example, to Encolpius’s assertion that Jupiter did no one any injury when descending to earth to transgress, Vergil’s description of the cloak awarded to Cloanthus for his victory in the boat race at Aeneid 5.252–57 depicts Ganymede’s aged guardians holding out their hands in vain to heaven and dogs barking at the air as the Trojan boy is carried off. Further, Ovid’s depiction of the rape of Ganymede includes Juno’s objections (Met. 10.161), and while Encolpius sees Apollo calling the shade of Hyacinthus back in the form of a flower, Ovid has Apollo fail to restrain Hyacinthus’s departing soul in spite of his healing herbs (Met. 10.188–89). We might have expected Encolpius, who boasts to Ascyltos, “both you and I know letters” (Petr. 10.5, et tu litteras scis et ego), to be aware of these other treatments of the myths, but he seems to have blocked them out of his literary memory, and this adds to the humor of his self-indulgent identification with the painted images. A humorous tension between Greek and Roman is even operative at the linguistic level if it is correct to hear interlinguistic puns in Encolpius’s quasi-art-historical descriptions.38 In general, Encolpius’s praise of the Greek artists and their talents perpetuates the terms of Roman art-historical discourse by echoing an account such as the Elder Pliny’s perhaps roughly contemporaneous catalogue of artists in Book 35 of the Naturalis historia. For example, perfect emulation of nature, which Encolpius applies to Protogenes, is also Pliny’s standard for artistic greatness, while Pliny, too, commends Apelles for his exact likenesses.39 Insofar as he recalls an encyclopedic work such as Pliny’s, Encolpius is to be associated with a distinctly Roman ambition toward universal knowledge accompanying Rome’s drive for empire (and the very inclusion of the names of known artists that makes Encolpius’s words suggestive of contemporary art history sets the scene apart from the Greek novels, which describe anonymous

36

Encolpius’s satiric self-identification with Aeneas recalls still other Vergilian echoes earlier in the Satyricon. Scholars including Collignon 1892: 121–22, Walsh 1970: 37, Zeitlin 1971b: 59, and Conte 1996: 3, 8–10, for example, have remarked a similar phenomenon in the scene immediately preceding the gallery scene (Petr. 81–82). See above, pp. 182–83 on the similarities between Trimalchio’s house and the Vergilian underworld. 37 Connors 1998: 85. 38 The puns described here are suggested by Elsner 2007: 183–84. 39 Elsner 2007: 183 citing Plin. Nat. 35.88.

Sex , Satire, and the Hybrid Self in Petronian Ecphrasis

195

works). But Encolpius seems to display an imperfect control of Latin as he attempts to describe Greek art in this Roman vein. He says, for example, that he sees a work by Apelles, which the Greeks “call” (appellant) the one-shinned (83.2). The word rudimenta (83.1, “first beginnings” or “first attempts”), used to describe the work of Protogenes, looks like a Latin translation of Protogenes’ own name (“first-born” or “primeval”). And if worshipping the painting of the “one-shinned” figure (83.2) implies going down on one knee, a pun may lurk here as well. In any case, it should now be clear that Encolpius’s eroticized response to the gallery’s artwork situates him uncomfortably between Greek and Roman. As he aspires to a cultural identification with Greece through a libidinous expression of philhellenism, Encolpius becomes enmeshed in the web of Petronian satire through echoes of a famous ecphrastic episode from Vergil’s Aeneid, canonical version of originary conflict between Greece and the forefathers of Rome.40 The satiric combination of Greek and Roman elements we have observed so far in the gallery scene continues into its next part: the entrance, monologue, and poetic performance of Eumolpus. He, too, professes an avid love of Greek culture, especially in his diatribe on the decline of philosophy and the arts from their heights in Greece (88.2–10). He is openly disdainful, moreover, of contemporary Roman culture, which he sums up simply as “wine and whores” (88.6, vino scortisque), the product of an age when the senate itself seems afflicted by greed for gold (88.9). As in Encolpius’s case, Petronius may be underscoring this philhellenism by adaptation from the Greek novel, and echoes of philosophical tradition possibly contribute to the same effect. Encolpius says that Eumolpus, when he first appears, “seems to promise some great thing” (83.7, nescio quid magnum promittere) and Eumolpus soon becomes Encolpius’s exegete as he seeks to understand the paintings in more depth. This is very similar to the situation in both Daphnis and Chloe and Leucippe and Clitophon.41 In Achilles Tatius, the narrator’s exclamation, “See how a little child rules sky and land and sea,” is overheard by a young man standing nearby, who, approaching the narrator, claims to have suffered much from love (1.2.1), and when the narrator asks him to say more, the young man, Clitophon, relates the story of the novel. Achilles Tatius has his narrator remark that the young man’s looks suggest that he is not far from being an initiate of the god (1.2.2): in the domain of erotic narrative, in other words, he promises much, just as Petronius’s Eumolpus does. Longus’s narrator, furthermore, says that he sought and found an “interpreter” 40 In Encolpius’s response to the paintings, Elsner finds a humorous allusion to the Stoic concept of phantasia as a vehicle of truth and the extension of this idea in literary and aesthetic theory (cf. the introduction, pp. 9–10, n25). Through Encolpius’s self-indulgent reactions, which belie the Stoic notion, phantasia becomes “a target that Petronius implicitly subverts” (Elsner 2007: 187). 41 Cf. Slater 1990: 91n9, 221–23; Elsner 2007: 189.

t h e c a p t o r’s i m a g e

196

(1praef. 3, ἐξηγητὴν) before writing the novel’s four books in response to the paintings he has discovered. And another exegete, here specifically philosophical, conveys the allegory of a picture in the Platonic Tabula of Cebes, possibly a work of the first century ce, while Eumolpus, like him an old man, will later be mistakenly identified as “one of the philosophers” above the sensual passions (Petr. 85.2).42 At last, Petronius hints playfully, true insight into visual imagery may be at hand. No sooner has Petronius introduced Eumolpus, however, than he begins to undermine his qualifications, too, as a philhellenic preceptor of art. After some initial remarks, Eumolpus launches into a “Milesian tale,” a sordid story about his seduction of the Pergamene boy through deceit and trickery.43 Illicit seduction is by no means a method shared by artistic exegetes in the Greek novels or the Tabula, and Eumolpus, as a duplicitous pederast, even seems to embody a satire of Platonic theory in becoming “a perfect living exemplar of the deceptive nature of the kinds of images he is being asked to interpret.”44 Petronius infuses this narrative, moreover, with a very particular Roman perspective on the events recounted, since Eumolpus’s story begins “When I was taken into Asia on the payroll of a quaestor, I received hospitality at Pergamum” (85.1, in Asiam cum a quaestore essem stipendio eductus, hospitium Pergami accepi). For all that he favors Greek culture over Roman, Eumolpus has been in the service of an arm of Roman provincial government and has had its privileges accorded to him. The locals over whom he helped rule were Greeks, but his view of them is in part that of the Roman administration. Of course, we need not imagine that Eumolpus, possibly a Greek himself, actually sympathized with the quaestor and his staff, or that he encountered a “Romans-go-home” mentality in Pergamum, which had by the mid-first century bce been under Roman rule for some two centuries. His interests are first and foremost in his own personal gratification, or at least in giving the impression that he benefited from belonging to the quaestor’s retinue. Here recall the Catullan complaints in poem 10 about Roman administrative abuses in Asia Minor (through which Catullus also alludes to Troy) together with the false braggadocio of that poem’s narrator over his supposed material enrichment in the provinces. Eumolpus, too, boasts that he took what he could get. Reminiscences of the Troy story again signal tension between Roman and Greek. The Romans routinely identified Pergamum with Troy. Indeed,

42

Elsner 2007: 185–86, 188–93. The telling of erotic tales in an art-gallery setting suggests the conventions of the Greek novel, although both Encolpius’s and Eumolpus’s erotic histories, as Elsner points out (2007: 185), represent departures from the Greek novels’ use of ecphrasis through their focus on same-sex love rather than love between the sexes. On conventions of the “Milesian tale” in the Satyricon, see Jensson 2004. 44 Elsner 2007: 192. 43

Sex , Satire, and the Hybrid Self in Petronian Ecphrasis

197

Eumolpus’s journey to Pergamum has the quality of a Trojan war in reverse, with an associate of the Trojans’ descendants, the Romans, realizing an erotic rather than a military conquest over the local Greeks.45 Eumolpus’s success, moreover, with the Pergamene boy, such as it is, brands him as one who breaks the bonds of trust inherent in ancient hospitality, just as Paris did by abducting Helen from Sparta: whenever the dinner conversation, he recalls, turned to amorous topics, he would feign great embarrassment, and because of this the mother of his young charge began to view him as “one of the philosophers” (85.2). He nevertheless achieves the boy’s seduction after three successive nights of promising ever more extravagant gifts, finally by the promise of a Macedonian horse, his story of conquest thus culminating in the ruse of a horse like the story of the Trojan War.46 In its emphasis on pederastic conquest, Eumolpus’s story takes up the Troy myth specifically from Encolpius’s earlier reference to Ganymede, as depicted in one of the gallery’s paintings.47 But again there is a significant inversion: Eumolpus, working for Rome, plays the part of the Greek god, Zeus, as seducer, while the boy who plays the role of the Trojan prince is Greek. This reversal, casting Eumolpus absurdly as Zeus, anticipates that of Eumolpus’s fortunes as his own sexual powers fail him. With no intention of actually following through on a gift that might arouse suspicion from the boy’s parents, Eumolpus makes the boy upset with him, but nevertheless persists in his wrongful ways: “Nothing is so difficult,” he declares, “that unscrupulousness cannot extort it” (87.3). Some nights later, after encountering resistance in further attempts at seduction, he finally succeeds in inflaming the boy so much that the latter’s appetite comically outstrips his own, and he is forced to turn the boy’s initial threats back against him. Importuned for more sex than he can handle, he exclaims, “Either go to sleep or I’ll tell your father” (87.10). Eumolpus winds up a conquered conqueror. As Encolpius’s erotic interests in boys reemerges in Eumolpus (just as it did before in Trimalchio), fissures continue to show up in both characters’ philhellenic facades. Encolpius’s response to Eumolpus’s story, which he himself has reported, develops his own characterization as one whose love of things Greek falls prey easily to satire. Rendered erectus (88.1, “bold” or simply “erect”) by Eumolpus’s words, Encolpius begins to ask him about art. Inquiring about the age of the pictures and aspects of their subject matter that he finds obscure, he also wishes to know why the arts have died in the present age, painting in particular. But what he really seems to want is someone who can tell him about the seduction of boys, the “art” in which he 45

Cf. Connors 1998: 93. Zeitlin 1971b: 61; Connors 1998: 93. 47 Rimell 2002: 63. 46

198

t h e c a p t o r’s i m a g e

has so recently failed.48 Eumolpus, whose qualifications as a philhellene have been rendered suspect by the deceit he used to seduce the Pergamene boy, responds with a hackneyed diatribe blaming love of money for the decline of philosophy and the arts. This would seem to strengthen his position as a Hellenophile, insofar as the passage amounts to an encomium of Greeks, among them Democritus, Eudoxus, and Chrysippus, as well as the artists Lysippus, Myron, Apelles, and Phidias, and a condemnation of present-day Roman pastimes (88.2–10). But here Eumolpus again trips himself up: his information about two of the artists he mentions, Lysippus and Myron, appears simply to be wrong.49 While he claims that hunger killed Lysippus, fi xated on the form of a single statue, and Myron, who almost captured the soul of men and beasts in bronze, left no heir (88.5), the Elder Pliny reports that Lysippus died a rich man after producing hundreds of statues (Nat. 34.37) and Myron failed precisely in giving expression to the “feelings of the soul” (34.58). Not only its hackneyed quality but also the presence of factual errors mars Eumolpus’s account of Greek greatness. Concluding his diatribe, Eumolpus notices that Encolpius has focused on a single painting of the Sack of Troy or, as Eumolpus identifies it using the Greek term, the Troiae halosis.50 He immediately says he will attempt “to disclose the work in verses” (89.1, opus versibus pandere), his words presumably the prelude to an interpretive description of its details. But what follows is hardly recognizable as an ecphrasis at all. Rather, Eumolpus begins a poem that simply recounts events of the Trojan War, without any apparent reference to the image itself, let alone exploitation of familiar ecphrastic devices such as references to the craft of the artist or the lifelike nature of his art. On the contrary, Eumolpus’s poem is much more like a messenger speech from Senecan tragedy, where a first-hand witness reports a series of fearful events. It is possible, of course, that this is a targeted literary satire: of Nero’s lost work on the same subject, perhaps, or of Lucan’s similarly lost juvenile effort, the Iliacon. But in the context of this scene and our limited knowledge, the poem strikes a reader first and foremost as a failure—at the very least a dramatic departure—from the perspective of generic expectations. Eumolpus is far less concerned with heightening a viewer’s appreciation of a painted scene than with milking a mythical subject’s possibilities to his own ends. His self-interested response to art recalls Encolpius’s earlier in the scene, a

48 Cf. Elsner 2007: 191: “The desire of Encolpius to understand art is in Petronius a sublimation of his desire to learn more about the techniques of homosexual seduction from a self-proclaimed expert.” 49 Walsh 1970: 96; Elsner 2007: 193. 50 On the “Troiae halosis” poem in the Satyricon, see esp. Sullivan 1968: 165–89; Zeitlin 1971b: 58–67; Slater 1990: 95–101, 186–90; Connors 1998: 84–99; Rimell 2002: 60–76.

Sex , Satire, and the Hybrid Self in Petronian Ecphrasis

199

further mise-en-abîme reminiscent of the Cena’s intermingling of Encolpius’s preferences with Trimalchio’s. Eumolpus’s “Troiae halosis” crowns his display of philhellenism vitiated by an immersion in things Roman, now especially by his status as an heir of Vergil. For it is in fact based heavily on the content of Aeneid 2 rather than, what we might expect, the Trojan War ecphrasis of Aeneid 1. Eumolpus’s own knowledge of Vergil comes through in his quotation, for example, of Aeneid 4.34 in the Widow of Ephesus story (111.12), and he also reflects on Vergil as a literary artist at Satyricon 118.5. The perspective he adopts in the first-person narrative of the “Troiae halosis” could well be that of one of Vergil’s Trojans, including Aeneas himself, since he recounts, in much shorter scope, the Vergilian events surrounding the Trojan Horse, from its construction through the Greek massacre.51 His poem assumes, moreover, audience knowledge of Vergil’s text, as in the verses “Laocoon fills the whole mob with uproar” (89.19–20, omne Laocoon replet/clamore vulgus), which are obscure without the Aeneid’s account of the arguments Laocoon addresses to the crowd.52 Details of Eumolpus’s language, such as the exclamation o Patria (89.11; cf. Verg. A. 2.241), also recall the Aeneid.53 Thus, following upon Encolpius’s earlier ecphrases in this scene and the Cena Trimachionis, the “Troiae halosis” represents still another satiric refraction of Aeneid 1 simply by excluding it as a model.54 While Vergil’s ecphrasis is highly conscious of itself as a description of art, its third-person narrative continually describing Aeneas in the act of looking and making the reader aware that the images seem to depict life (Α. 1.455–56, 464, 466, 485–93), Eumolpus becomes completely absorbed in the subject-matter of the painting, and this makes his use of a dramatic meter, iambic trimeter, all the funnier. Eumolpus wants to be a part of the action rather than attempt the aesthetic distance demanded by conventional ecphrasis.55 At another level, the “Troiae halosis” dramatizes the subject of Trojan responses to Greek artifice and the ultimate failure of these responses. In this way it becomes an allegory of the gallery-scene itself, in which Encolpius and Eumolpus attempt appropriate responses to Greek art and fail. The Trojans’ failure to identify the horse as a Greek ruse leaves them drifting between a deadly encounter with Greek might and their Roman future, and Petronius’s characters, similarly faltering in the face of Greek artisanship, are left to drift between Greek and Roman, especially as a sea voyage will soon deposit them in Croton, another Graeca urbs in southern Italy. Eumolpus’s poem 51

Connors 87; cf. Zeitlin 1971b: 62. Slater 1990: 97–98. 53 Cf. Connors 1998: 89. 54 Cf. Slater 1990: 188. 55 Elsner 2007: 195–96. 52

t h e c a p t o r’s i m a g e

200

actually includes language that portrays the building of the horse as a kind of figuration: The oaks of Ida, which will figure forth a threatening horse, are dragged along and, split, fall into a mass. Idae trahuntur scissaque in molem cadunt robora, minacem quae figurabunt equum. (Petronius, Satyricon 89.5–6) The artifice of the horse’s appearance is emphasized further by the words “[the Greeks] hide in their votive offering” (89.10, in suo voto latent) and the later line, “the young men went as captives [in the Trojan Horse] while they captured Troy” (89.27, ibat iuventus capta, dum Troiam capit) actually alludes to Horace’s dictum (Ep. 2.1.156–57) about Greek culture’s conquest of the Romans who were Greece’s conquerors in war.56 The captor’s image indeed.

The Hall of Mirrors: W. J. T. Mitchell and the Petronian Legacy Reading the Satyricon in the terms of W. J. T. Mitchell, we see that Petronius directs the barbs of his satire at attempts to render the text/image distinction an allegory “of power and value disguised as a neutral metalanguage.”57 By exposing the erotic investment of his characters in the supposedly neutral practice of artistic description, Petronius frames the problem of ecphrasis as one of self and other, just as Mitchell does many centuries later. Encolpius and Eumolpus both have strong reasons for valuing the silent images they encounter in the art gallery, and each feels a compulsion to supplement the paintings with spoken language, as if this act of verbalization supplied a need that the paintings, in their silence, somehow expressed, and which they, Encolpius and Eumolpus, were particularly qualified to understand and fulfill. Petronius’s protagonists seem to have found their own particular answers to the Mitchellesque question, “What do pictures want?”58 Encolpius takes solace in the paintings’ realism for its capacity to represent a lost world of pederastic gratification, lost both in the sense of its location in a mythological past and in the past of Encolpius’s own erotic life. His worship 56

Connors 1998: 94. Mitchell 1994: 157. 58 The title of Mitchell 2005. 57

Sex , Satire, and the Hybrid Self in Petronian Ecphrasis

201

of the Greek painters as realist artists shades quickly into praise for art’s ability to give persuasive visual form to the dictum amor etiam deos tangit, through which he seeks to reveal the paintings’ significance in relationship to his own experience. He wishes to identify with the paintings, to assimilate them, in all their Greekness, into his own sense of himself. Petronius, however, skewers his solipsistic self-importance in believing that his perception of such a relationship is significant, and that he has discovered the key that makes the paintings meaningful by contrasting their content with his own life. The images become merely a replacement for Giton.59 A humorous Greek parallel for Encolpius’s attitude actually exists in Herodas’s Mimiamb 4, where two women, looking at paintings by Apelles, respond erotically to the image of a naked boy, and we may also recall our discussion of the women excited by an image of Adonis in Theocritus’s Idylls 15.60 In this way, too, the Catullan narrator of poem 64 takes his place as one of Encolpius and Eumolpus’s literary forebears. Further, the connotations of Encolpius’s own name—“Mr. Crotch” or something of the kind—encourage us to read erotic content into all of his enthusiasms. Encolpius believes that his interpretation, like the behavior of Jupiter who, he says, “did no one injury” by taking mortal love-objects, is a beneficial act. But he actually does a kind of violence to the paintings by turning their beautiful, evocative silence to his own self-centered ends. Eumolpus, in his patronizing pity for Greek artists such as Apelles and Phidias, feels compelled to come to their defense through his own account of Greek art history and to “disclose” (pandere) one of the paintings in the gallery through his poem on the Sack of Troy.61 Yet his account of art history displays a Romanized perspective on Greece reminiscent of his attitude toward the Pergamene boy, while his non-ecphrastic poem on the Trojan War is the problematic outgrowth of this ambivalent attitude. Petronius underscores the parallel through the fact that Eumolpus’s third attempt to seduce the boy occurs while the latter is asleep, the likeness of sleeping love object and painted image having been ingrained in the Latin literary tradition at least since Propertius’s Elegies 1.3, with its comparison of the sleeping Cynthia to a series of mythical figures directly familiar from 59 Eumolpus seems to confirm this when he later praises Giton as Ganymede (92.3), the subject of one of the paintings in the gallery. Still later (126.16–17), the substitution of art-object for love-object becomes explicit when Encolpius compares the woman Circe, with whom he attempts a sexual union, to Praxiteles’ statue of the goddess Diana (cf. Slater 1990: 224–25). Just as his interpretation, so to speak, of Circe as a statue of a goddess reflects his wish to possess her sexually (a wish thwarted by his impotence), Encolpius’s approach to depictions of boy-love reflects a desire to possess them through the intellect and emotions. 60 See above, pp. 69–71. 61 Eumolpus at one point calls Apelles and Phidias “delirious little Greeks” (88.10, Graeculi delirantes), words echoing the disrespect he claims all great art now receives, though with what degree of irony it is not clear.

202

t h e c a p t o r’s i m a g e

Roman wall painting. Like Encolpius, Eumolpus is satirically deflated by errors, and above all by the shower of stones that cuts his poetic performance short. His “Troiae halosis” is shot through with a perspective propagated by the Aeneid, an epic to outdo Homer, but he winds up embodying a latter-day, Roman-world version of what Callimachus says about his own soul: that it is “deserving to be stoned and unlucky in love” (Epigr. 41.5–6, λιθόλευστος . . . καὶ δύσερως). His “reading” of the Trojan War painting is another failed attempt to conquer the Greek/Trojan past in its silent, seductive images, to wrest a Mitchellesque “gift” for the audience from this much-quarried source. But if Mitchell helps us understand Petronian ecphrasis as a problematically eroticized appropriation of Greek culture, Petronius also sheds light on implications of Mitchell’s theory not emphasized by Mitchell himself. As we have seen before, the text/image dyad of ecphrasis need not only stage a tension-ridden encounter between a self and an other figuring larger social entities, but can also represent the voluble present understood both with and against the alluringly silent past. As with Keat’s Grecian urn, the past can be a still, silent, foreign object of contemplation, which our various discourses invest with erotic significance as we attempt to recapture its lost worlds and personages. Greece in the Satyricon, however, embodies simultaneously an unattainable past ideal, vanishing before one’s eyes in the artworks that Encolpius says have not yet been conquered by the ravages of time (83.1), and a living, speaking, desirable present, elusive in a different way. The erotic desire of the Pergamene boy, made difficult of access, then outstrips that of his seducer. Ambiguously Hellenic art objects overwhelm their hybrid ecphrasists’ attempts to delimit their significance once and for all. This interplay of a cultural past and a cultural present points to Rome’s self-conscious status as a cultural intercessor, aware of its role in both appropriating and handing down a vanishing Greek artistic tradition prized by Romans and Greeks alike. The application of Mitchell’s theory leads us to position him in a dialogic relationship with the ancient ecphrastic tradition, a tradition he inherits in commenting upon it. For Petronius, finally, is like Mitchell in remarking ecphrasis’s broader social ramifications, and yet involves himself, even as a “hidden” author, in the self-questioning web he spins around its affective dimensions. Indeed, in the Satyricon, the affective underpinning of ecphrasis renders its practitioners, including Petronius, especially open to satire. Encolpius’s amor, for example, which he expresses both for a real Greek-named youth and, betraying Mitchell’s ecphrastic “hope,” for the painted Greek boys he describes for the audience, helps undo him as a reliable interpreter of art. Conversely, the horror he tries to convey suggests Mitchell’s ecphrastic “fear” insofar as it implies an awareness of/desire for the impossibility of his task: Apelles’ paintings are simply indescribable, too marvelous for words. If they were susceptible to description, something of their majesty would be lost. Similarly, like the Pergamene boy who bests Eumolpus through

Sex , Satire, and the Hybrid Self in Petronian Ecphrasis

203

an excess of erotic energy, the Greek traditions over which he attempts to assert interpretive control retain their vitality in the face of efforts to direct them to self-serving ends. Eumolpus is no Apollo, in spite of his name. These characters’ affective ties to ecphrasis are invitations to inquire about Petronius’s own (this could even apply to a certain “indifference” betrayed by his stone-throwing mob) and an implicit admission that he, too, cannot escape charges of affective investment in ecphrasis. The power of his text, in other words, derives from its deeply questioning stance toward ecphrastic “fascination” even in himself. Can Mitchell be said to exhibit this sort of Petronian self-awareness? Perhaps not, once his influential text on ecphrasis is enlisted in the service of a Petronian viewpoint. Mitchell, that is, actually seems anxious about the possibility of a counter-reading when he admits that a skeptic “indifferent to the magic of ekphrastic hope” might point out that his argument is “a kind of sham, creating difficulties at the level of representation and signs that do not exist.”62 His solution, however, that “from the semantic point of view . . . there is no essential difference between texts and images and thus no gap between the media to be overcome by any special ekphrastic strategies”63 represents a deferral of affect within a pose of distance (but not commonsense “indifference”). “The semantic,” like “ecphrasis,” is a construct with its own social history, but this goes unstated. The reduction of texts and images to pure semiosis is its own sort of “gift” to the reader and critic, both now freed to indulge ecphrastic fascination safe in the knowledge that its unacceptable ideological ramifications are founded upon an illusion. The Satyricon explodes even authorial gestures of this sort, when any gift it offers comes inscribed under the sign of that most deceptive of benefactions, the Trojan Horse. Mitchell, like all of us, must contend with the fact that ecphrasis begins as a supreme instance of Greek guile, firmly installed, through Homer, in the same tradition that brought the Horse to Troy and to us. This, however, is Petronius’s very point of departure, and to the extent that we are his heirs in the reception of this tradition from antiquity, we are all Encolpiuses now.

Conclusion: Neronian Ecphrasis? The subject of shams and deceptions leads us back to the issue of the Satyricon’s putative relation to a Neronian context, and in turn back to Trimalchio. Apart from the name “Petronius” on the manuscripts, the Neronian dating hangs in no small part on similarities between Trimalchio’s foibles and those ascribed to the emperor in Suetonius’s Life of Nero.64 Is Trimalchio’s love of Greek culture 62

Mitchell 1994: 158. Mitchell 1994: 160. 64 For a survey, see Walsh 1970: 137–38; Laird 2007: 160; cf. Walsh 1970: 70. 63

t h e c a p t o r’s i m a g e

204

a contemporaneous satire of Nero’s own? Although simply noting that both of these imperious characters kept the trimmings of their first beards in golden caskets or had bemedalled runners among their slaves cannot be the basis of a definitive argument, it is still worth considering the possibility that the Cena and gallery scene might have appealed to an original audience of elite, Hellenophile Romans under Nero by addressing anxieties over the effort to express appreciation of Greek cultural artifacts without giving up a secure place in the dominant Roman elite. Another unforgettable aspect of the historians’ Nero is his flamboyant philhellenism, which led him to establish Rome’s first set of Greek-style games, the Neronia, and to inflict his Hellenic literary efforts mercilessly upon captive audiences, whose long suffering was finally ended by the emperor’s suicide, even this act accompanied, if we can believe Suetonius, by recitation of a line from the Iliad (Suet. Nero 49). Love of Greece is potentially riskier under a philhellenic despot of waning popularity. Among confirmed Neronian authors, a suggestive anxiety emerges in Persius’s first satire, which expresses the hope that it will find a reader who appreciates the transfer (à la Horace) of the Greek old-comic tradition into Latin, not one who takes pleasure in mocking what he perceives to be merely Greek affectations, symbolized here by a kind of Greek footwear, the crepida (123–28).65 The Elder Pliny’s Naturalis historia, completed not long after Nero’s reign, is more than once dismissive of Nero’s use of the Greco-Roman artistic tradition for his own megalomaniac ends, as in his display of a 120-foot high painting of himself (35.51) or the painted decorations of the massive Golden House, the “prison” (carcer) of art by a certain Famulus, whose work virtually disappeared when this outrageous structure was destroyed (35.120). Trimalchio epitomizes similar missteps. As an upstart nouveau-riche freedman he is a monstrosity from a traditional Roman perspective, and this is part of what makes his pretensions to Greek culture available for outright ridicule. Reading the gallery scene with the Cena, however, reminds us that the Satyricon’s satire of philhellenic ecphrasis has a broader scope, not limited to the freedman class, absurd objects, or the laughably ignorant. If the Satyricon is Neronian, one way it would appear to be engaging with its historical context is precisely through an emphasis on contemporary “deception and decadence” that compounds our problems. For in this reading the text ends up as “a self-reflexive satire of [itself] . . . as a frame for the exposure of the Neronian era . . . a text which indulges lavishly and self-consciously in all the crimes of which it accuses its age.”66 The very terms of Encolpius’s praise 65

But cf., helpfully, Jensson 2004: 212 on “the well known anxiety of Roman aristocrats about the moral implications of a widespread study of Greek rhetoric and literature,” a theme found in a broad range of Greek and Roman authors, not just Persius. 66 Elsner 2007: 196–97.

Sex , Satire, and the Hybrid Self in Petronian Ecphrasis

205

for deceptively lifelike Greek art can be turned back against the Satyricon as an ecphrastic narrative, when “the success not only of the image but of the exegete himself . . . is measured in terms of his ability to deceive.”67 Our exegete, Petronius, succeeds if he deceives us into thinking that his text, art objects and all, is a self-subverting reflection of a deceptive age, and succeeds again if his deception can be shown up for what it is, since in that case it only “mirrors” its age more (which is to say less) successfully.68 Of course we cannot know exactly how the Satyricon’s original audience responded, or what they thought of the Greek cultural forms on display in its labyrinthine chambers. And what if, in the Satyricon, the Greek art isn’t really there in the first place? Let us close by remarking yet another way in which Petronius may be having the last laugh at the expense of his protagonists (as well as himself, his audience, and his intellectual heirs). It is likely that at least some Romans would have understood the paintings Encolpius and Eumolpus encounter to be Roman versions of Greek originals.69 To accept this caveat is to reinforce our sense of Petronius’s self-conscious implication in his own satire, especially if his model was indeed a lost Greek Satyrica.70 Encolpius and Eumolpus’s reactions to the gallery’s paintings, symbols of the Satyricon itself, would then be directly comparable to the external audience’s reactions to Petronius’s text, and their mistakes a pointed challenge to that audience to reflect upon whether it could do better (the very situation we encountered in Catullus 64).71 As we shall see in the next chapter, the nature of art connoisseurship, particularly of much-prized Greek works like those Encolpius believes he sees, would become an overt theme of Latin poetry under the rule of another avid philhellene, the emperor Domitian, whose predilections opened up still new possibilities for the Roman ecphrastic author.

67

Elsner 2007: 197. For an extreme form of skepticism over a Neronian date for the Satyricon, see Laird 2007, who reconsiders the idea that the text could be the work of a later author seeking to generate a Neronian “period backdrop” (158) and even writing under the pseudonym “Petronius” (161). 69 Cf. Slater 1990: 222n15. 70 As per Jensson 2004. 71 On the gallery scene as a “parody of audience reaction,” cf. Slater 1990: 222. 68

7

The Patron’s Image Philhellenism, Panegyric, and Ecphrasis in Statius and Martial

Lauding a patron by describing his art had been a tactic largely implicit in the Latin ecphrastic poetry thus far examined.1 But as a means of philhellenic display and of exploring the complex relations of both poets and patrons to a series of powerful, richly evocative images, the practice nevertheless emerged as a significant one in the Latin literary tradition. As we have seen, in the Eclogues and Georgics Vergil uses ecphrases of fictive art objects to gesture toward the achievements of Hellenophile patron figures, whether Asinius Pollio or Octavian/Augustus, and ally himself, as poet, with them. In Satires 1.8, Horace exploits a comically obscene statue of Priapus to suggest the Hellenistic refinement of Maecenas’s circle (himself included), while Propertius’s description of the art in Octavian’s Palatine sanctuary of Apollo (Elegies 2.31) accompanies a poetic epiphany together with muted commendation of Rome’s de facto ruler. With this chapter, the practice becomes explicit, and the patron’s mediating role comes into greater prominence. In the occasional poetry of Statius and Martial written under the Flavian emperor Domitian (ruled 81–96 ce), we find extant Latin verse for the first time using patrons’ actual art as the main focal point

1 In what follows I will use the terms patron and patronage in the sociological sense defined by Saller to describe a relationship characterized by asymmetry, reciprocity, and duration (Saller 1982: 1) and well attested in the ancient Greco-Roman world. This is the definition adopted recently by Nauta in his study of literary patronage in the age of Domitian and applied by him to Statius’s and Martial’s interactions with both the emperor and figures outside the imperial court (Nauta 2002). These terms have been controversial in Roman studies at least since White’s denial of their usefulness—though with somewhat different reference (White 1978: 75). The question is complicated by the fact that the Latin word patronus never bears the sense of “literary patron” in antiquity (White 1978: 78–79; cf. Saller 1982: 8–11; Nauta 2002: 14). Other influential studies of poets and patrons in the Domitianic period include White 1974, 1975, and 1993: 3–91 (which covers the early Empire); Garthwaite 1978 and 1993; Hardie 1983; Coleman 1986; Ahl 1984a and b; Leberl 2004; Zeiner 2005. For a fuller survey see Nauta 2002: 11.

206

The Patron’s Image

207

for extravagant panegyric.2 The Silvae and Epigrams thus offer new complications of the ecphrastic text’s supposed efforts to dominate its image, insofar as the image’s Greek otherness more clearly informs the identity of another Roman, to whom the Latin poet’s own subordination structures his philhellenic verbal performance in large part. As with descriptions of religious images in Chapter 4, the texts treated here illustrate a feature of Roman ecphrasis unfamiliar enough to go almost completely unrecognized in modern theory. I refer to the subtle way that relations between patron and poet, the asymmetrical bond linking ownership to authorship, may come to inform an ecphrastic text down to its smallest details, especially through allusion to the art object’s presence in particular physical surroundings. Unfamiliarity itself plausibly explains the omission in the theory, since this type of ancient poetry for patrons is perhaps one of the hardest to grasp with our modern sensibilities trained in post-Romantic notions of what “real” poetry is or does. In an age where private ownership and connoisseurship of art risks the stigma of isolated elitism as opposed to the public viewing practices afforded by museums open (at least putatively) to all, the influence of wealthy, powerful patrons on particular objects’ reception might well be a topic avoided by theorists wishing to understand the phenomenon of ecphrasis “in itself.” Consider, for example, the tone of exception adopted by Heffernan in his treatment of Browning’s “My Last Duchess,” a poem in his view “truly remarkable in the history of ekphrasis” for being spoken by the owner of the painting it describes, and the “most dramatically paragonal of all ekphrastic poems” for the conflict it appears to stage between the image and a narrator expressing a will “to sound controlling, to dominate the picture with his words.”3 The ancient poems discussed below, though never spoken entirely in a patron’s voice, nevertheless exploit a range of related techniques through which the influence of patron-poet relations on ecphrasis emerges as a far more complex and nuanced phenomenon than an interpretation such as Heffernan’s might lead us to believe. This chapter’s chief contribution to the book as a whole thus lies in the way its close readings—of poems describing Domitian’s public iconography and a Hercules statuette owned by a private patron, Novius Vindex—bring out a different sort of power dynamic at work within the staged Roman receptivity to Greek culture. What Mitchell has described as ecphrasis’s “tripartite” structure, wherein 2

I stress “extant” poetry, since a great deal of now-lost Latin verse, particularly epigrams, probably anticipated Statius and Martial’s techniques, and they could count on their audiences to be familiar with Latin poems of this kind. Occasional poetry, of which the vast bulk, again, no longer survives, may in fact have been the most common Latin verse form in the first century ce as in other periods. For backgrounds to Statius’s and Martial’s verse and precedents in extant Greek poetry, see below. 3 Heffernan 1993: 141, 142, 144.

208

t h e c a p t o r’s i m a g e

a speaking subject offers the ecphrasis as a “gift” to a third party, starts to look even less straightforward than in Chapter 3’s discussion of multiple audiences. For these poems express far more fully and directly the idea that the patron has already bestowed a welcome gift upon the poet by creating the whole environment in which the latter has the opportunity to say what he supposedly wishes to say about the object in question. Domitian accomplishes such stage-setting, as it were, simply by being the all-powerful ruler of Rome’s vast empire, and the guardian in particular of the city of Rome, while Vindex offers his home and art collection as such an environment. Through such means, the patron might provide an ineluctable set of hermeneutic data—indeed, in Vindex’s case possibly even verbal instructions—as to what he wished the object to mean in relation to himself and his social milieu. The ecphrasis, in turn, becomes a pointed and self-conscious counter-gift, elusively objectifying the process through which the patron’s self-articulation and the identity he forms for himself in all its detail become those of the poet and vice-versa. These poems thus especially reveal the overwhelming importance of another cultural dimension of Roman ecphrasis: its ties to a competitive philhellenism dependent upon real social interactions between patrons and clients, for all that ecphrasis may illustrate individual poets’ ideas about their poetry itself.4

Ecphrasis for Patrons in Flavian Rome The potential benefits of such verse for both poets and patrons are not far to seek, given the importance of certain art objects in Domitian’s Rome. Consider first the emperor’s “official” iconography. Domitian’s public art expressed a conception of rule informed strongly by Hellenistic notions of kingship, just one feature of his oft-remarked philhellenism.5 Although once-prevalent views of Domitian as a brutal, autocratic tyrant throughout his reign have been revised, there is no doubt that he used such objects in a manner reminiscent of the Hellenistic kings to project an image of absolute power backed by and even rivaling that of the gods.6 The two artworks treated in Part I of this chapter contribute to this impression: the first a massive equestrian statue of Domitian placed in the center of the Forum Romanum and holding, in its left hand, an image of the emperor’s patron deity, Minerva, with Medusa’s head; the second

4 For reasons of space, I will treat only the most salient parts of Statius’s ecphrases, while analyzing Martial’s much shorter poems in their entirety. 5 On Domitian’s philhellenism, see Jones 1992: 110, 112; Darwall-Smith 1996: 221–26, 249; Newlands 2002: 11–13; Leberl 2004: 74–79. 6 For the revision of views on Domitian’s despotism, see Jones 1992; Southern 1997.

The Patron’s Image

209

a statue of Hercules with Domitian’s features set up as a cult image in a new temple to Hercules on the Appian Way.7 To praise these sculptures in philhellenic terms while honoring Domitian’s power would not only have redounded to the emperor’s credit, but must also have seemed an obvious way for a poet to attract his attention and good favor. This is all the more true because of Domitian’s central role in conferring literary renown, another philhellenic trait.8 Under his reign, writers and intellectuals had access to the court and might on occasion receive benefits directly from the emperor: Domitian did not, like Augustus, enlist the informal services of a Maecenas to help generate a literature celebrating his reign. His main emphasis was on the cultivation of authors—both Greek and Latin—through institutional means.9 A poet himself, he established, like Nero, Greek-style games with literary competitions, the quadrennial Capitolia, as well as the annual Alban Games in honor of Minerva, which also included contests for poets and orators in Greek fashion.10 The Capitolia, over which Domitian presided wearing Greek dress, even took place in newly built Hellenic structures, the Odeon and Stadium, the latter, at least, impressively decorated with Hellenistic statues.11 One result of Domitian’s games policy was to institutionalize literary achievement itself, making imperial approval virtually a prerequisite for success.12 Competition for approval no doubt grew intense as poetry became “a mass activity in which all might participate, as poets or listeners.”13 And both Statius and Martial appear to 7

The Gorgon’s head, as noted in Chapter 4, was a symbol used by Hellenistic monarchs, and had been employed as such by Augustus himself, while Domitian’s coinage reveals a notion of a warlike Minerva (Minerva Propugnatrix/Athena Promachos) stemming ultimately from Hellenistic models. See above, pp. 115–16 and the following note. Statius’s Thebaid (12.606–10), moreover, portrays Minerva with the Gorgon’s head fighting on the side of Theseus and the Athenians, and the figure of Theseus here has been seen as evoking Domitian (Braund 1996; Ripoll 1998: 495–502) or at least an ideal, however problematized, of Augustan “clemency” (clementia) toward which there was reason to hope a Flavian emperor would aspire (McNelis 2007: 176). Self-representation as Hercules, furthermore, was one of the most enduring features of Alexander’s own propaganda. For more on Domitian’s iconography, including coinage and the remarkable Pamphili Obelisk with its inscriptions in Egyptian hieroglyphics, as reflecting Hellenistic ideas of the ruler, see Darwall-Smith 1996: 136; Newlands 2002: 11–13, 22–23; Leberl 2004: 55–56, 74–79. 8 On Domitian’s patronage of literature and its Greek background, see Hardie 1983; Coleman 1986; Jones 1992: 103–05; Darwall-Smith 1996: 223–26, 249; Leberl 2004: 113–42. 9 See Hardie 1983: 46–47 for Domitian’s games policy as the development of a form of imperial patronage. 10 The evidence is collected by Coleman 1986: 3097–100; Leberl 2004: 75–77. 11 Leberl 2004: 75–76. Darwall-Smith 1996: 249 calls these buildings “the final act in Rome’s hellenisation,” as “the only major types of Greek building which had yet to achieve permanent incarnation there.” Cf. Darwall-Smith 1996: 221–23. 12 Agoustakis and Newlands 2007: 120; cf. Newlands 2002: 29. 13 Hardie 1983: 49.

t h e c a p t o r’s i m a g e

210

take an acute interest in the emperor’s good opinion.14 If the cultivation of court poets itself had roots in Hellenistic practice, Domitian’s games, which spread the benefits of imperial patronage further, were an even more self-conscious adaptation of Greek tradition to a Roman context. While Domitian’s love of Greek culture found highly visible, public expression, Statius and Martial kept still other, private objects in view. With the emperor’s interests necessarily limited, the literary “boom” of the first century ce placed fresh importance on patronage by wealthy elites outside the court, some of them avid art collectors.15 In such circles, Greek masterpieces were especially prized, and poetry “commissioned” from a noted author might serve as a vehicle for advertising one’s aesthetic acumen and the impressiveness of one’s collection.16 In Part II of this chapter, we will see both poets praising not only the beauty of Vindex’s statuette but also its “pedigree” of previous owners stretching back through Sulla and Hannibal to Alexander the Great. Both, moreover, link their reactions to Vindex’s Hellenophilia, which they now characterize as the predilection of a private citizen rather than a ruler. Though centered on a character from Greek myth, the Vindex poems reflect pressing aspects of the poets’ social world from which the emperor is never entirely absent. For a writer such as Statius, himself the son of a Greek poet and grammaticus, maintaining a broad array of patrons such as Vindex must have been especially valuable. Although his father attracted the attention of Vespasian and may have been tutor to the young Domitian, Statius remained something of an outsider, and the great majority of the Silvae address persons other than the emperor himself.17 In turn, however, the formidable Greek credentials that the bilingual Statius gained from his father, a number of whose students apparently achieved 14

Cf. Silv. 3.5.31–33; 4.2.65–67; 5.3.225–33. Statius entered both the Capitoline and Alban Games, though apparently much to his disappointment won a victory only in the latter. Praise for Domitian is abundant in the Silvae and prominently placed in his epic poems, the Thebaid and the Achilleid, and accompanies at least one report of the emperor’s personal largess. See Geyssen 1996: 2, 8. At Silv. 3.1.62–63, Statius indicates that the running water in his Alban estate was a gift from Domitian. See further Nauta 2002: 335–36. Martial, who also lavishes praise on the living Domitian (though he would recant after the latter’s death and damnatio memoriae), speaks of being admitted to court and, though childless, received the “right of three children” (ius trium liberorum) at both Titus’s and Domitian’s hands. Admission to court: Mart. 6.10; ius trium liberorum: Mart. 2.91–92; 3.95.5; 7.97.5. On these and other marks of imperial favor enjoyed by Martial, see Nauta 2002: 336–37. For Martial’s recantation, see Mart. 10.72. 15 Cf. Hardie 1983: 53; Newlands 2002: 29–30; on Roman art collectors and collecting, see Zeiner 2005: 94–97; Bartman 2010. 16 See Zeiner 2005: 190–200 for sources and comparanda. The Younger Pliny’s prose letter to Annius Severus (Ep. 3.6) on a bronze statue of an old man that Pliny bought for the temple of Jupiter at Comum, combines, as Coleman notes (1988: 175) “ecphrasis . . . with the broader aim of self-advertisement.” Cf. esp. Henderson 2002: 155–71. 17 Cf. Coleman 1988: xv–xx; McNelis 2002: 72, 91; Newlands 2002: 27–32; cf. Stat. Silv. 5.3.

The Patron’s Image

211

positions of prominence in Roman government and the military, was part of the valuable “cultural capital” that association with him might provide.18 Statius’s list in Silvae 5.3.146–58 of sophisticated Greek authors whose work his father taught— including Pindar, Ibycus, Alcman, Sappho, Callimachus, and others—suggests the depth of learning that Statius himself could bring to his literary associations. Martial, too, as the offspring of Spanish provincials, required connections in Rome, and pursues a large number of patrons in his work, although the posture of mendicancy he adopts surely belies the truth.19 In spite of the fact that he and Statius mention few acquaintances in common, the two may have been direct rivals, a view supported by their co-presence as Vindex’s encomiasts.20 In any case, in the Vindex poems as elsewhere, Martial, like Statius, strives for intimacy with a wealthy amicus through the description of a treasured possession. This is only one aspect of a much larger poetic response to such figures’ personal material interests, whether in art, houses, slaves, or other belongings. But even here Domitian’s presence can be felt, since Vindex’s statuette represents a deity over whom the emperor himself claimed a kind of ownership. As we shall see, thematic, formal, and verbal links among the poems treated in the two halves of this chapter remind us again of the fluid boundaries between public and private, political and nonpolitical, where Roman ecphrasis is concerned. All this, finally, is to counter the diminishing effect of the pernicious label “occasional verse.” Ecphrasis is one way that the “occasions” from which the Silvae and Epigrams take their impetus especially reveal their larger social, cultural, and political significance.

I. A Ruler for the Ages: Greek Past, Roman Present, and Domitian’s Statues Hellenic Horsemanship: The Equus Domitiani and Silvae 1.1 The equestrian statue of Domitian erected in the Forum Romanum in or about 90 ce presented Statius with a signal opportunity to align himself with the 18 See McNelis 2002 on the “cultural capital” that Statius’s father would have been able to convey through the teaching of a sophisticated curriculum of Greek poets to up-and-coming orators among the Roman elite, and the evident success of his students, indicated by Silv. 5.3.185–90. Cf. Holford-Strevens 2000: 52: “Statius . . . implicitly presents himself as a perfect bilingual . . . that is to say, as we read his poem, a Roman who is perfect in Greek; but when we contemplate his origin, we should look rather for the Hellene who is perfect in Latin.” See further below, p. 229. 19 White 1978: 88–89; Sullivan 1991: 1–6. 20 For the poets’ common acquaintances, see White 1975. For the possibility of a rivalry, see Hardie 1983: 57; Sullivan 1991: 73n32, 113–14, 125–26; Dewar 1991: xxxv–xxxvi; Henriksén 1998; Bonadeo 2010: 43–47. Ecphrasis of another sort, we may note, informs their poems on the baths of Claudius Etruscus (Mart. 6.42; Stat. Silv. 1.5)

212

t h e c a p t o r’s i m a g e

emperor’s interests while announcing his own literary project in the Silvae.21 Although officially a gift to Domitian (following long-standing precedent, it was dedicated by the Roman senate and people in his honor), it was no doubt meant to function as a sign of Domitian’s greatness and hence to be understood as one of many benefactions furnished by his reign. The senate and people, that is, express their gratitude to the emperor through the statue, but part of the gratitude that they are supposed to feel is for having an emperor great enough to inspire such a massive undertaking in the first place. This, in any case, is how Statius presents the monument: a wondrous object bestowed upon Rome (note nobis [“for us”] in line 5), perhaps by Domitian’s fellow-divinities themselves. Preserved visually for us today only on a coin of 95/96 ce (figure 7.1), the statue as Statius describes it embodies Domitian’s sway over an empire extending its reach in both space and time, the monument’s huge form surpassing marvels of the mythical past and its iconography denoting victories over troublesome foreign peoples.22 And so the implicit understanding of Domitian’s statue as a stunning gift informs Statius’s proffered counter-gift of a startling poem. Indeed, Statius pushes so far the tensions and paradoxes of the small, elegantly wrought Roman Hellenistic poem encountering the overwhelming visual expression of the emperor’s power that Silvae 1.1 was at one time a central piece of evidence for the Silvae’s supposed program of “doublespeak” and subversiveness, a view that has now lost favor among scholars.23 In light of earlier chapters, the striking paradoxes of Silvae 1.1, particularly its simultaneous ascription of a Greek and a Trojan/Roman “lineage” to Domitian’s statue, can be seen to grow out of both Latin ecphrasis and imperial iconography itself from at least the time of Augustus. We benefit further from a general contrast with Chapter 5’s account of Augustan epic, which incorporates but does not limit itself to panegyric and uses tension between Greek, Trojan, and Roman as a part 21 On the dating of the statue and the poem, see Geyssen 1996: 21. Major discussions of the poem include Cancik 1965: 89–100; Ahl 1984a: 91–102; Geyssen 1996; Nauta 2002: 422–26; Newlands 2002: 46–73; Leberl 2004: 143–67; Dewar 2008. 22 On the coin (BMC 476† [Mattingly 1976: 406]), see Darwall-Smith 1996: 228–29; Geyssen 1996: 23–24; Thomas 2004: 28n34; Dewar 2008: 71n9. On the exact placement of the statue, part of whose base may survive (though this is speculative), see Darwall-Smith 1996: 229–30; Thomas 2004; Dewar 2008: 75–77. 23 See Zeiner 2005: 6–7 for discussion. This view found its strongest expression in Ahl 1984a: 91–102; contra, see Geyssen 1996; Nauta 2002: 424–26; Zeiner 2005: 6–9; Leberl 2004: 143–67; Dewar 2008: 79n20. Newlands seeks a middle ground in suggesting that the poem is ridden with “faultlines” where inconsistencies emerge, resulting in “a poem of anxiety as well as celebration” about a monument that “is itself ambiguous” (Newlands 2002: 24–25, 48–49). But questions of panegyric “sincerity,” easily overstated in the ancient context, need not preoccupy us in either Statius’s or Martial’s case, since my focus will be the ostensible posture each adopts in relation to his patron and Greek culture. On the “sincerity of the poet’s belief ” as a critical notion misplaced in the context of ancient panegyric, see Geyssen 1996: 7. For debates over the possibility of “hidden criticisms” of Domitian in Martial, see Fitzgerald 2007: 112 with n12.

The Patron’s Image

213

Figure 7.1 Coin depicting equestrian statue of Domitian. 95–96 ce. © The Trustees of the British Museum.

of much larger pattern of ambiguity. Conversely, Silvae 1.1 celebrates Domitian’s statue for its marvelous paradoxicality, and uses the layering of Greek, Trojan, and Roman elements to enhance this effect. This is nowhere better illustrated than in the interplay of Greek and Roman, and the paradoxical relation of Statius’s ecphrastic poetics to Domitian’s statue, that characterize the poem’s opening lines: What mass, twinned by the colossus placed on top of it, stands, having embraced the Latian forum? Has the work slipped, fully formed, from heaven? Or, fashioned in Sicilian forges, did the statue leave Steropes and Brontes weary? Quae superimposito moles geminata colosso stat Latium complexa forum? caelone peractum fluxit opus? Siculis an conformata caminis effigies lassum Steropen Brontenque reliquit? (Statius, Silvae 1.1.1–4)

t h e c a p t o r’s i m a g e

214

The rapid sequence of information conveyed by the imposing first sentence immediately imparts wonder over an object—or perhaps two (cf. 1, geminata)— that for now seems to defy definition: a “mass, twinned by the colossus placed on top of it, stands, having embraced the Latian forum” (1–2). The word colosso, given prominence at line-end, draws attention for its twofold, Greek-and-Roman resonance, while the contrast between its seemingly static grandeur and the “embrace” achieved by the unnamed moles thrusts paradox to the fore. “The Colossus” was, first and foremost, the wondrously tall sculpture of Helios in Rhodes, but the name had more recently been applied to the gargantuan statue of Nero, its head replaced with that of the sun-god, still standing next to the Flavian amphitheater in Statius’s day.24 Both statues become important later in Statius’s poem, since near the end of Silvae 1.1 he asserts that “fierce Rhodes, having scorned its Phoebus, would prefer your [i.e., Domitian’s] glance imitating starry radiance” (103–04), a veiled reference to the replacement of Nero’s visage with Sol’s on the Roman “colossus.”25 So already here the word colosso pays Domitian the compliment of having surpassed his reviled predecessor Nero while enveloping the former’s statue in the air of Greek culture that surrounded the latter’s.26 Statius’s specification of the “Latian” forum (i.e., the Forum Romanum) magnifies the Greek echo through contrast and the Roman one by association. Can a “mass” embrace a forum? This surprising image seems already to associate and subordinate Statius’s Hellenistic aesthetics to a more wondrous kind of mastery he ascribes to the statue, and beyond it to Domitian as patron and ruler. The conceit foregrounds the issue of containment at the very opening of this neo-Hellenistic poem daring to encompass a massive object. Indeed, Statius perhaps borrowed this idea directly from Hellenistic epigram.27 But whether or not Statius had such poetry in mind, the paradox of a diminutive verse form that manages to contain something much larger than itself (e.g., themes borrowed from the epic tradition) is a quintessential element of Hellenistic verse in general. Silvae 1.1, Statius suggests, does indeed possess an artistic analog in its object, but while the poem revolves around the contrast of small and great, the statue is a massive object containing the yet-more-massive. To this extent the statue’s reach subtly begins to exceed the poetry’s: although Statius will go on to describe the nearby buildings of the Forum Romanum and Forum Iulium (22–36, 84–90) the 24

See Geyssen 1996: 24. Geyssen 1996: 26. As he notes, the comparison between the statue of Nero/Sol and the Colossus of Rhodes was “obviously a commonplace” since it appears before this in Martial (Sp. 70.7–8; cf. 2.1). On the political ramifications of the Flavians’ transformation of the area, see Dewar 2008, esp. 74 and 78–80 for a rejection of Geyssen’s hypothesis (1996: 24–27) that Domitian’s statue was a “modest tribute [by Domitian] to himself ” (27). 26 Cf. Geyssen 1996: 25. 27 On AP 6.171 and APl. 82 on the Colossus of Rhodes, see Gow and Page 1965: 588–89; Hardie 1983: 131). The name of the creator of Domitian’s statue goes unmentioned in Statius’s text. 25

The Patron’s Image

215

statue’s “embrace” of its surroundings suggests Domitian’s power over all aspects of Rome and its empire, including, not least, Statius’s poetic effort itself.28 The words that come next, “Has the work slipped, fully formed, from heaven?” (2–3, caelone peractum/fluxit opus?) reinforce both the parallelism and the implicit hierarchy between poem and statue. Fluo, though in a somewhat different sense, is the same verb that Statius uses in Book 1’s preface to describe the genesis of the Silvae, which, he says, “flowed forth for me in a sudden enthusiasm and a certain pleasure in hastening” (1praef. 3–4, mihi subito calore et quadam festinandi voluptate fluxerunt).29 But the lexical similarity points up an important divergence between poet and emperor crystallized in the opposition caelo/mihi: Domitian’s statue is so impressive that it seems no artist’s hands could have fashioned it (thus its emergence “fully formed” from heaven), while Statius’s poem is the product of his inspiration, however impressive he may wish us to find the results of his supposedly extempore technique. The rush of four dazzled questions with which Silvae 1.1 begins (1–7) exemplify Statius’s inspired poetic “flow” at the outset, while also seeming to convey the immediacy of first-hand experience.30 The statue’s hypothesized passage from heaven is similar yet of another order: so unexpected (how could such an object “slip” or “fall gradually”?),31 so wondrous, that it can become an imagined prototype for Statius’s poetic process while nevertheless standing above it as an ideal. Statius’s counter-gift of a poem fantasizes its own inspiration from the statue, which is represented as a divine gift. The onslaught of marvelous alternatives continues. Another possible superhuman origin for the statue, one that returns the audience to the notion of Roman rivalry with Greek culture and Statius’s with the epic tradition, emerges in the next two lines. Statius’s second guess about the statue’s production is that it was forged by the Cyclopes Steropes and Brontes, who labored over it to the point of exhaustion (3–4, Siculis an conformata caminis/effigies lassum Steropen Brontenque reliquit?). The allusion here to Aeneid 8.424–25, on the creation of the arms of Aeneas, has long been recognized:32 The Cyclopes worked the iron in their vast cave, Brontes and Steropes and Pyragmon with naked limbs. ferrum exercebant vasto Cyclopes in antro, Brontesque Steropesque et nudus membra Pyragmon. (Vergil, Aeneid 8.424–25) 28

Cf. Ahl 1984a: 92. Geyssen 1996: 43–44. 30 Geyssen 1996: 35–38. 31 For these meanings of fluo see OLD 12a. Cf. Dewar 1991: 169 on Stat. Theb. 9.581. Fluo perhaps also plays here on the idea of flowing molten metal that eventually sets. 32 Vollmer 1898: 216; Geyssen 1996: 44–45; Newlands 2002: 53. 29

216

t h e c a p t o r’s i m a g e

Statius here sets his short poem against Vergil’s full-length epic and his ecphrastic object, Domitian’s statue, against Vergil’s, the Shield of Aeneas. Vergil himself in fact alludes later in the Cyclopes passage to Callimachus’s account of the Cyclopes’ labor in the Hymn to Artemis, making the Shield a paradoxically Callimachean object in part.33 Statius’s claim, however, to outdo the Vergilian Shield, as well as its Greek model, Homer’s Shield of Achilles (also Theocritus’s target in Idylls 1), rests on the suggestion that in spite of the statue’s impressive size his own poem will not be a drawn-out epic to “weary” his audience, but a work of actual Callimachean dimensions.34 In turn, Statius indicates that Domitian’s massive statue can be taken as the present-day answer to Aeneas’s mythical armament, and Domitian himself as a new Aeneas figure corresponding to the Greek Achilles—or perhaps something even greater. The Aeneid, as we have seen, repeatedly calls its Shield’s heroic import into question, making Aeneas’s acquisition of it into an ambiguous event within the scope of the poem. But Statius, who only offers the Shield of Aeneas as a potential comparandum before moving on, implies that Domitian’s monument is still more impressive than this object, since the statue would leave the Vergilian Cyclopes weary. Both Greek and Trojan echoes persist in Statius’s third question about the statue’s origins, which receives the greatest rhetorical weight as the final and longest in its series. Statius now asks whether Minerva herself crafted the image of Domitian, here addressed by his honorific title Germanicus (5), as conqueror of the Rhine and the Dacians (5–7), a reference to Domitian’s claimed victories.35 Of course, Minerva’s most famous creation was the Trojan Horse, to which Statius immediately turns.36 Though daring, the comparison that follows (8–15) between the statue and this infamous object, prototype of Greek artisanal treachery and downfall of Troy, is very much in keeping with the poem’s 33 A. 8.449–53; cf. Call. Dian. 46–61. In Callimachus, Artemis demands that the Cyclopes forge for her a bow, arrows, and quiver (81–83), an order they obey; she thus tames and redirects their fearful, noisy, epic-scale labor for her own purposes, just as Callimachus tames and redirects the epic tradition for his. Vergil alludes to Callimachus’s Cyclopes in order to signal that he is writing an epic poem of Alexandrian polish. The Aeneid passage reproduces G. 4.171–75, following which a simile comparing the bees with the Cyclopes is marked by the programmatic tag “if it is permitted to compare small things with great” (176, si parva licet componere magnis); on the Callimachean allusion see Thomas 1988: 2.179–80. 34 Pace McNelis 2007: 73: “The Cyclopes’ grandiose artistic programme clearly resembles that of the Telchines and is thus anti-Callimachean.” I prefer to emphasize Statius’s evocation, through Vergil, of the Callimachean Cyclopes themselves. 35 In 89 CE, Domitian celebrated a double triumph over the Dacians and the Germans. This followed upon two earlier triumphs over northern peoples, in 83 over the Chatti (for which he claimed the title Germanicus) and in 86, also over the Dacians. The actual outcomes of Domitian’s campaigns were often far less decisive than his record of triumphs suggests. See Jones 1992: 129, 139. 36 For Minerva as the Trojan Horse’s creator, see Hom. Od. 8.439; Verg. A. 2.15; cf. Ahl 1984a: 92.

The Patron’s Image

217

opening lines.37 Statius has thus far been concerned to situate his new poetic project in relation both to the Aeneid and to Greek culture in a manner whose very unexpectedness does homage to the emperor’s capacity, as both patron and ruler, to encompass paradox. This capacity sets Domitian apart from ordinary mortals, but signaling it in this way also allows Statius an opportunity for further ingenuity in his response to Vergil, since lines 11–13, which express the notion that Domitian’s horse is too big even to have been received into Troy, allude to Aeneid 2.234–39 by cleverly recasting their depiction of the horse’s entry into the doomed city.38 Statius’s reappropriation of Vergil’s Trojan Horse is one more instance of surprising containment in a poem that has by now indicated this as a defining theme. Just as Domitian’s monument has paradoxically “embraced” the forum in which it stands, and surpasses even the Shield of Aeneas in grandeur, so Statius now shows that his poem will embrace aspects of the epic tradition that might seem to lie beyond its acceptable field of reference. Statius’s compliment to the emperor hinges on the idea that Domitian’s command over past and present, over empire and cultural tradition, is so great that even such seemingly negative comparisons can redound to his credit. In the poem’s last twenty-four lines, a cluster of references to Greek artists and art objects, envisioned through the frame of Domitian’s Rome, encapsulates the entire poem’s panegyric strategy where Greece and Rome are concerned. Statius had earlier illustrated his claim that the statue’s “location is equal to the work” (22, par operi sedes) with an account of buildings in the Forum Romanum, beginning with the temple of Divus Iulius (22–31). Comparison with Julius Caesar there sets Domitian’s clemency in relief: Caesar learns from Domitian’s sculpted visage how much “milder” he is in arms, as illustrated by his treaties with the Chatti and Dacians (25–27). With Domitian carrying the standards, Statius asserts, Caesar’s enemies Cato and Pompey would have surrendered (27–28), a backhanded deprecation of Caesar for instigating civil violence like the violence the Flavians were proud to have suppressed.39 Now Statius turns to the Forum Iulium, with its equestrian statue of Caesar placed near the temple of Venus Genetrix, here called “Latian Dione” (84, Latiae . . . Diones) in honor of her status as mother of Aeneas and, through him, ancestor of both Caesar and the whole Roman race.40 Statius bids the 37

The comparison especially bothered Ahl (1984a: 92). Geyssen, by contrast, emphasizes “the good the Trojan Horse produced for Rome” (in assuring the palladium’s transfer and Rome’s rise) rather than “the destruction it brought to Troy” (1996: 51–52), while Newlands sees this and the coming comparison of Domitian with Orion as two of the poem’s “faultlines” revealing “a gap between the noble myth of military success and containment and the lurking threat of indiscriminate power and violence” (2002: 59). 38 On the allusion, see Wagenvoort 1955: 195–96; Courtney 1984: 331; Geyssen 1996: 49. 39 Dewar 2008: 72–74. 40 On the statue and its location, see Geyssen 1996: 86n45.

t h e c a p t o r’s i m a g e

218

statue “yield” (84, cedat) to Domitian’s, then, referring to the fact this statue was supposedly an image of Alexander the Great sculpted by Lysippus but refitted with Caesar’s head (86–87, “Soon it bore the face of Caesar upon its wondering neck” [mox Caesaris ora/mirata cervice tulit]), he describes it as the one “which you, Lysippus, having dared it, gave to the Pellaean leader” (85–86, quem traderis ausus/Pellaeo, Lysippe, duci). The connotations of the verb tradere, which can also mean “surrender” in a military context (OLD 3), amplify the idea of Statius’s own subjection to Domitian, as well as the latter’s martial prowess.41 Itself bidden to “yield” to Domitian’s monument, Caesar’s horse becomes at once the Greek and the Roman precedent for it. Lysippus’s offering, in this way and others, anticipates Statius’s. The Roman poet’s identification with the Alexandrian sculptor emerges plainly through this apostrophe, its effect possibly enhanced by a further echo of the preface to Silvae, book 1, where Statius describes the delivery to Domitian of Silvae 1.1 with the words “I dared deliver” (1praef. 19, tradere ausus sum).42 But again, the parallelism is asymmetrical and complex. Both statue and poem are offerings to a ruler, but while Caesar expressed simultaneously his emulation of Alexander and his Hellenophilia through appropriation of the latter’s monument, Statius’s Hellenistic-inspired poem for Domitian affirms the emperor’s superiority to both Caesar and Alexander and thus his supreme embrace of the Greek legacy. Statius forces this issue of Domitian’s superiority back upon the audience in the next two lines. An observer, Statius remarks, would scarcely be able to measure the “downward view” (88, despectus) from Domitian’s statue to Caesar’s. No one, he suggests, would be so “unsophisticated” (89, rudis) as not to equate the difference in scale between the two horses with the difference in stature between the two rulers (89–90). So Statius presents his own sophistication as a vehicle for conveying Domitian’s greatness. After lines asserting the statue’s permanence and imagining it as a meeting place for the deified spirits of Domitian’s family (91–98), Statius returns for a final time to the themes of Greek art and artists, now to figure the statue as a more impressive successor to a series of famous and large-scale Greek works. Apelles, he insists, would have wished to paint Domitian (100, “Apelles’ wax would desire to represent you” [Apelleae cuperent te scribere cerae]).43 Given Statius’s earlier reference to Alexander the Great (86), the mention of Apelles in this context is likely to have recalled his most famous painting of Alexander housed in the temple of Artemis at Ephesus: here Alexander was portrayed in 41

Again I thank Mira Seo for this observation. Geyssen 1996: 79–80, who is careful to note the difference in voice and meaning between the two uses of tradere. On the emended reading ausus sum in the preface, see Geyssen 1996: 28–30. 43 This refers to the encaustic wax used in painting. See OLD scribo 1b; cera 3e. 42

The Patron’s Image

219

the image of Zeus Keraunophoros or “Thunder-Bearer,” the statue’s iconography one among many possible models for Domitian’s own identification with Jupiter in art.44 Two major paintings of Alexander by Apelles, moreover, were on display in the Forum Augustum, providing an even more familiar reference point for Statius’s remark, which would then also mobilize Augustan precedent in Domitian’s favor.45 Statius saves the biggest for last. Phidias, he goes on, would have wished to fashion a sculpture of Domitian to place “in a new temple of Elean Jove” (101–02); that is, a successor to the temple at Olympia housing the colossal statue of Zeus, like the Colossus of Rhodes one of the seven wonders of the ancient world. Tarentum, too, he adds, would prefer Domitian’s image (102–03)—presumably a reference to its colossal bronze statue of Zeus, another work of Lysippus46—and Rhodes’ image of Helios, as noted, completes the list of Greek masterworks (104). In a culminating paradox, Statius’s small-scale poem on the Equus Domitiani becomes a stand-in for the large-scale images that the Greek masters of old would have created of Domitian if they had had the opportunity. As literary heir to both Homer and Vergil, Statius daringly places himself in a similar relation to these artists: his concluding catalogue even adumbrates artistic filiation by including both the work of Lysippus (at Tarentum) and of his pupil, Chares of Lindos, sculptor of the Colossus of Rhodes. But Silvae 1.1 ends with Domitian’s imagined surveillance of his own familial lineage, not Statius’s artistic lineage. With a sentiment familiar from still other poetic forefathers, Horace and Ovid, Statius begs Domitian to remain on earth for many years before he decides to dwell as a god in heaven (105–07).47 This will allow him to see offerings brought to the statue by his own “grandsons” (107, nepotes), the poem’s final word.

The Hut of Molorchus: Martial’s Epigrams on a Statue of Hercules as Domitian However much Statius vaunts his poetry as the appropriate medium in which to celebrate Domitian’s statue, the pair’s relation to the Greek past and to each other remains asymmetrical, the difference in scale between poem and artwork serving as a reminder that no mere mortal, even the most gifted poet, can rival the emperor’s achievements. For its emphasis on the

44

On Apelles’ painting see Pollitt 1986: 22. See above p. 137; cf. Pollitt 1986: 22–23. For a different view of Alexander’s role in Silvae 1.1 and 4.6, see Spencer 2002: 147–48, 151–54, 185–87. 46 Tarentum was also home to a colossal bronze of Hercules, also by Lysippus. On both sculptures, see Pollitt 1986: 49. 47 Cf. Hor. Carm. 1.2.45–52; Ov. Trist. 5.2.47–54; Geyssen 1996: 126–31. 45

t h e c a p t o r’s i m a g e

220

paradoxical containment of the Greek past within the Roman present, and of the emperor’s imposing image by the small-scale poem, Martial’s series of epigrams on a statue of Hercules with Domitian’s features (9.64, 65, 101) is every bit the rival of Silvae 1.1.48 These poems were in fact published later than Silvae, Books 1–3, in an epigram book that stands out within Martial’s oeuvre for frequency of imperial panegyric, due possibly to a desire to compete with Statius as a court poet.49 The Hercules poems, moreover, have a major Statian precedent in Silvae 3.1, on a temple of Hercules built by Pollius Felix at Surrentum. While thus clearly engaged in an analogous endeavor, Martial here takes to still greater extremes than Statius the compression of epic theme into ecphrastic poetry of Callimachean size and polish, and in a less capacious meter, the elegiac couplet. Whereas Statius, furthermore, uses comparison between Greek and Roman as one among a range of panegyric techniques, Martial focuses continuously on a stark contrast between the Hercules of Greek myth and Hercules-as-Domitian, whom he portrays as having outdone his predecessor. The juxtaposition of the “lesser” Greek and “greater” Roman Hercules is facilitated in formal terms by the unequal lines of the elegiac couplet itself. It accompanies and underwrites Martial’s self-fashioning as a Roman poet who, though working 48

Leberl discusses these three poems together as a “Herkuleszyklus” in Book 9 (2004: 73, 272, 310–17). Mart. 9.101, as the longest and nearly the last epigram in book 9 (it is followed by only two relatively short poems) clearly serves as “a concluding summary of the imperial theme in Book 9” while 64 and 65 go together as a pair (Henriksén 1998–1999: 1.19). Both Barwick 1958 and Garthwaite 1993 argue for all three as part of a large “imperial cycle” in the book, while Henriksén 1998–1999: 1.16–20, following the definition of Grewing 1997: 30–31, adopts a more limited view of “cycles” in the book. For more on these epigrams as a unit and as part of Book 9, see Lorenz 2002: 203–08; 2003. On the statue and the temple that housed it, see Darwall-Smith 1996: 133–36. Representations of Hercules as Domitian survive in a massive gilded Hercules statue on display in the Musei Capitolini (Museo nazionale Romano, inv. no. 8573) and a giant black marble Hercules found on the Palatine and now housed in the Galleria Nazionale at Parma. See Prioux 2008: 291. 49 Henriksén 1998–1999: 1.18; cf. the previous note. On the dating of Epigrams, Book 9 (late 94 or early 95 CE) in relation to Silvae 1–3 and 4, see Coleman 1988: xvi–xvii, xix; Henriksén 1998–1999: 1.11–13; Bonadeo 2010: 51–56. Martial’s book actually suggests a broader interest in statues: the epigram attached to the book’s dedicatory epistle mentions a bust of Martial himself in Toranius’s possession (3), 9.50.5 refers to a statuette owned by the tyrannicide Brutus, the pair of poems on Vindex’s Hercules (43–44) also comes from this collection, and still another pair from the same book (23–24) refers to a bust of Domitian owned by the poet’s friend, Carus. Lorenz 2003 argues for the structural importance of these poems within the book and their varied relation to the theme of praise for the emperor. Cf. 9.74 on a picture of the deceased child, Camonius. For more on references to statues in Martial, see below, pp. 240–41. On ecphrasis (of buildings) as a broader technique in Martial’s poetry for patron figures, see Fabbrini 2007. Martial, Book 14, the Apophoreta, on gifts given at the Saturnalia, contains his most concentrated series of epigrams on a diverse series of objects, including some statues.

The Patron’s Image

221

in a small-scale form, is nevertheless worthy to sing the praises of a mighty emperor. Even more overtly than Statius in Silvae 1.1, Martial uses ecphrasis to play upon the notion of a Roman identity that is ambiguously Greek—or a Greek identity now made ambiguously Roman—his own poetry’s amalgam of Greek and Roman finding a supreme exemplar in the imagery’s combination of these elements. The first epigram in the series, 9.64, in fact begins with a couplet remarkably suggestive of the opening of Silvae 1.1, where the gift of Domitian’s equestrian statue has “slipped” down from heaven to embrace the “Latian forum” (1–3): Caesar, having deigned to descend into the features of great Hercules, gives a new temple to the Latian road [i.e., the Via Appia] . . . Herculis in magni voltus descendere Caesar dignatus Latiae dat nova templa viae. . . (Martial, Epigrams 9.64.1–2) Martial’s emphasis on godlike descent as well as the “Latian” circumstances of the envisioned art object recalls Statius’s poem, although unlike Statius, Martial leaves no doubt about the image’s subject matter. The word “Latian,” given prominence before the pentameter’s caesura, suggests a contrast between Greek and Roman even if Hercules is not yet associated emphatically with Greece (as he soon will be). The first line, framed by the names “Hercules” and “Caesar,” the former in the genitive, the latter in the nominative, at once sets “the order of precedence . . . it is Domitian who deigns to lend his features to Hercules, not vice versa.”50 Martial’s poem presents itself as a grateful gift in return. As in Statius, Domitian embodies magnitude surpassing that which is already great: Hercules’ greatness (1, magni) is less than the emperor’s, a conceit to which the poem will return at its conclusion. The contrast between the hexameter and the pentameter applies this idea to the relation between the statue and its surroundings: great Caesar, who is still greater than Hercules (1) is also greater than the temple and road (2) to which he descends, his downward motion suggested by the decline from the longer line to the shorter. In some sense, the Domitianic Hercules has “deigned,” too, to become the subject of Martial’s brief poem, which here announces its intention to contain, however improbably, its great object.

50

Henriksén 1998–1999: 2.66.

t h e c a p t o r’s i m a g e

222

Containment of the large both within the small and within the still larger likewise informs the poem’s second couplet on the location of the new temple: [Caesar establishes the temple] where the wayfarer, while he seeks the wooded realm of Trivia, reads the eighth milestone from the imperial city. qua Triviae nemorosa petit dum regna, viator octavum domina marmor ab urbe legit. (Martial, Epigrams 9.64.3–4) The sacred precinct of Diana Nemorensis was on the shore of the Lacus Nemorensis south-east of Rome. But the area where Diana exerts her sacred power—her “realm” (3, regna)—is only one region within the vast expanse over which Domitian’s Rome, “the imperial city” (4, domina . . . urbe), extends its power.51 This containment, in a manner again reminiscent of Silvae 1.1, has a paradoxical quality since, although the Roman empire contains Diana’s regna, the city of Rome contains within its grasp the much larger empire. Martial’s subject matter again cleverly tropes his poetry’s formal and thematic characteristics, since the small epigram has now broadened its own reach to encompass a statue, a temple, a city, a realm, and an empire, all by line 4. Martial even wittily places the word “eighth” (octavum) at the beginning of his fourth line, as though to suggest the multiplicative difference in magnitude between his verse and what it depicts. Similarly, the hexameter’s inclusion of Diana’s regna (3) finds its counterpart in the pentameter’s focus on the “milestone” (4, marmor), the small object that nevertheless marks Rome’s expansive reach. But Martial quickly returns to the contrast between the traditional Hercules and Hercules/Domitian. The next couplet refocuses attention on the appearance of the statue and its surroundings, while suggesting Martial’s own poem as an act of reverence: Before, [Hercules] was worshipped with vows and copious blood [i.e., of sacrificial victims]; now the lesser Alcides himself worships the greater. ante colebatur votis et sanguine largo, maiorem Alciden nunc minor ipse colit. (Martial, Epigrams 9.64.5–6) Line 5 sets up the expectation of contrast between earlier means of worship and those employed at present, but this does not seem at first to be the contrast that 51

For dominus as “imperial,” see OLD 2c.

The Patron’s Image

223

emerges with line 6, where the worshipper and object of worship, the “lesser” and “greater” Hercules, take center stage. The couplet’s paradoxical logic only becomes apparent upon reconsideration of the statue, which, because it introduces Domitian in Hercules’ place while still indicating the traditional Hercules’ presence, itself becomes the latter’s means of worshipping the emperor. The statue’s face is Domitian’s, while the body, clothed, no doubt, in traditional Herculean accoutrements, both is and is not his. Along with this reemphasis on ambiguous identification between container and contained (the body as containing the face, the clothing as containing the body) comes a more overt suggestion of Hercules’ Hellenic origins through the Greek patronymic, Alcides. The Roman Domitian, Martial suggests, even appropriates the lineage of the Greek Hercules and makes it, too, “greater,” an implication underscored by the syntactic placement of Alciden with maiorem rather than minorem (6). So Martial’s own “smaller” Latin verse brings its Greek-inspired aesthetic to the task of venerating a philhellene emperor “greater” than itself—and one who makes it “greater” than itself. With more clever self-reference, Martial’s final couplet hints at the additional benefits that his own poem might elicit from Domitian, now by disassociating the lesser Hercules from the greater according to gifts each can bestow: One man asks this [Hercules] for great wealth, another for honors; to that [Hercules] a man perfunctorily makes lesser vows.52 hunc magnas rogat alter opes, rogat alter honores; illi securus vota minora facit. (Martial, Epigrams 9.64.7–8) The benefits envisioned by anonymous worshippers of the greater Hercules, by suggesting Domitian’s actual power to bestow “wealth” and “honors,” point to Martial’s hopes for a favorable reaction from the emperor. In turn, the lines allude to the effort that Martial has put into his epigram: far from a careless or perfunctory offering (cf. 8, securus), this poem has benefitted from Martial’s own cura, which sets it apart from “lesser” works (cf. 8, minora) undeserving of Domitian’s attention. Here, again, we sense the paradoxes of Martial’s Callimachean epigram, which, though small, is still greater than less polished verse: cura, if not exactly the same as the labor indicating Callimachean efforts in Latin, can nonetheless connote the attention lavished upon a literary work.53 And the fact that Hercules, as a topic of the Victoria Berenices, was among the 52

I owe my translation of securus vota minora facit to Henriksén 1998–1999: 2.68; for securus as “careless” or “perfunctory” see OLD 4. 53 OLD 3b; cf. esp. Coleman 1988: 185 citing Stat. Silv. 1.3.104, 1.5.64, 4.6.1, 5.2.71, 5.3.34. For the Callimachean associations of labor, see, e.g., Clausen 1994: 293 on Verg. Ecl. 10.1.

t h e c a p t o r’s i m a g e

224

touchstones of Roman Callimacheanism, only enhances the suggestion that in Domitian’s Hercules statue, Martial has found an ideal subject through which to articulate epigram’s self-distinction from epic in terms set forth by the Alexandrian poet himself.54 Reconsidering the epigram’s opening, we now find a parallel between Domitian’s deigning to descend into Hercules’ features, and Hercules’ own figurative descent into the humble circumstances of the poor Molorchus’s hut in Callimachus’s “epyllion.” Like the statue, Martial’s Latin poem becomes the Greek pauper’s hut that receives the Roman Hercules. The next poem in the series, 9.65, begins with a still more explicit suggestion of the Greek past’s containment in the Roman present, now through apostrophe of the traditional Hercules, who is deemed fortunate for possessing Domitian’s features: Alcides, now to be acknowledged as his own by the Latian Thunderer, now that you bear the handsome face of the god Caesar . . . Alcide, Latio nunc agnoscende Tonanti, postquam pulchra dei Caesaris ora geris . . . (Martial, Epigrams 9.65.1–2) This opening contrasts with that of the previous poem insofar as Hercules, rather than Domitian, is made the grammatical subject of the first finite verb (geris). Yet the shift in perspective is nevertheless an extension of Martial’s technique in the earlier epigram, which moved from Domitian’s assumption of Hercules’ appearance and gift of the temple, to the viator’s reading of the milestone, to the “lesser” Hercules’ worship of the “greater,” to the anonymous worshippers’ requests. Martial also introduces a new onlooker: Jupiter Capitolinus (“the Latian Thunderer”), who, as a Roman god, could not properly “acknowledge” the Greek Hercules as his son, even if, as the counterpart of the Greek Zeus, Jupiter was in fact Hercules’ father.55 He can do so now because Hercules bears the features of “the god Caesar.” While Domitian, that is, became in the previous poem a Roman Hercules both distinct from and still identified with his Greek counterpart, the Greek Hercules now seems to become a Roman, though only because he looks like Domitian. Following close upon the heels of Epigrams 9.64, 9.65 thus lends Domitian’s statue a still more ambiguous identity between Greek and Roman. Having pointed to Hercules’ new Romanness, Martial goes on imagine how much better this Hercules, in possession of isti vultus habitusque (3), would have fared in the world of Greek myth. These words themselves, however, foster ambiguity by their possible translation as either “that face and bearing” or “that face 54 55

On the Victoria Berenices and Latin poetry, see Thomas 1983b. Cf. Henriksén 1998–1999: 2.69.

The Patron’s Image

225

and style of dress”; that is, Hercules’ customary attributes.56 The former translation makes Hercules’ improved fate entirely the result of looking like Domitian, while the latter suggests that the combination of Roman and Greek characteristics would have constituted Hercules’ advantage in the Greek past. The changes, moreover, that Martial imagines in Hercules’ story retroject aspects of Domitian’s rule into the myth, so as to make the Greek narrative appear quasi-Roman. Thus Martial asserts that the peoples of the world (gentes) would not have seen Hercules serve “the Argive tyrant” (Argolico . . . tyranno, i.e., Eurystheus), but Hercules would have “given orders to” (iussisses) Eurystheus (5–7). Martial here implies Domitian’s prowess as a ruler, as well as his worldwide renown (note the introduction of the gentes as another onlooker to follow Jupiter Capitolinus), while being careful to distinguish his reign from the arbitrary rule of a tyrannus. “Nor would deceitful Lichas,” Martial goes on, “have brought you the treacherous gift of Nessus” (7–8, nec tibi fallax/portasset Nessi perfida dona Lichas). The emphasis on avoided treachery and deceit, somewhat exaggerated since the herald Lichas was unaware of the poison infusing the cloak of Nessus he brought to Hercules, suggests Domitian’s own experience of treachery (alleged or otherwise) and his efforts against it.57 Hercules/Domitian, Martial continues, would have approached “the stars of his father supreme” (10, astra patris summi) without needing to have been burnt in Hercules’ funeral pyre on Oeta (9). The words allude to Domitian’s own apotheosis and return to the company of his deified father, Vespasian (also intimated at Silvae 1.1.106). Finally, after rejecting the ludicrous idea of Hercules/Domitian’s subservience to the Lydian queen, Omphale (11), Martial wraps up his hypothetical tale with another hint at the emperor’s immortality: the new Hercules “would not have seen Styx and the dog of Tartarus,” Cerberus (12, nec Styga vidisses Tartareumque canem). For Domitian, no descent into death. We now move from a Romanized Greek past to a Hellenized Roman present. The poem concludes by listing a series of advantages the transfigured Hercules enjoys in his new guise: Now Juno is propitious to you, now your Hebe loves you, now if the Nymph sees you, she will return Hylas. nunc tibi Iuno favet, nunc te tua diligit Hebe; nunc te si videat Nympha, remittet Hylan. (Martial, Epigrams 9.65.13–14)

56

For the ambiguity and the translation, see Henriksén 1998–1999: 2.69. In 93 ce, Domitian apparently brought seven important accusations of maiestas leading to three executions and four exiles. See Newlands 2002: 9. 57

t h e c a p t o r’s i m a g e

226

These assertions map even more easily onto Domitian. Given the associations between Domitian and Aeneas in, for example, Statius’s Silvae 1.1, Juno’s reconciliation with Hercules/Domitian and his marriage to her daughter Hebe suggest an elegiac, erotic version of Juno’s famous reconciliation with the Trojans at Aeneid 12.841. More directly, “your Hebe” and Hylas point to the emperor’s wife, Domitia Longina, and his boy-favorite, Earinus, the subject of his own series of epigrams in Book 9 (11–13, 16–17) as well as of Statius’s Silvae 3.4.58 The Nymph who, in myth, stole Hylas from Hercules will, Martial declares, now give him back, a humorous nod to Domitian’s universal authority. But more than simply offering a mythical parallel, Martial here suggests that Domitian and his court are somehow the harmonious revision and fulfillment of the myth itself: the Roman present realizes, outdoes, and even replaces, Greek myth by improving specifically upon it.59 Martial’s emphasis on Hylan, the poem’s final word, is perhaps programmatic, since the tale of Hercules’ love for the boy was a ubiquitous theme in Hellenistic and Latin literature, decried as such by Vergil at Georgics 3.6, as we have seen, and by Martial himself at Epigrams 10.4.3.60 Martial ends his elegiac epigram, that is, with the Greek hero’s diminutive love-object, exploited by earlier poets to embody “the traditionally conflicting themes of heroism and pederastic love.”61 Yet insofar as Domitian’s statue stands for a Roman improvement upon Greek myth, it imbues the literary Hylas with a larger cultural significance here, while at the same time allowing for an implicit claim to have found a novel use for this overfamiliar figure. The climax of Martial’s extant eulogies of Domitian and the longest poem in book 9, Epigrams 9.101 is nevertheless a tour de force of compression, recounting the mythical Hercules’ Twelve Labors and one of his “byworks” or πάρεργα in a mere seven lines (4–10), only to follow this with a catalogue of Domitian’s achievements, especially his foreign wars, in ten (13–22). The slight difference in length between the two accounts is enough to underscore the prowess of the “greater” Hercules over that of the “lesser” (11), of the present Alcides over the “earlier” one (3, prioris), while Martial again frames the whole poem as inspired by the statue of Hercules/Domitian in its Roman surroundings. Addressing the Appian Way, which “Caesar hallows, to be

58

While a rift between Domitian and Domitia appears to have occurred in 83 ce, she was apparently reinstated the following year, although her reputation has suffered from the strongly anti-Domitianic bias of our sources. See Jones 1992: 34–36; Southern 1997: 41–42. 59 For this idea, I am indebted to Fitzgerald’s development (2007: 48–52) of an argument by Pailler (1990: 180–81) concerning Martial’s characterization of Flavian arena spectacles that were presented via the iconography of Greek myth. 60 See above, p. 119 and Henriksén 1998–1999: 1.138. 61 Thomas 1988: 2.38.

The Patron’s Image

227

worshipped in an image of Hercules bearing his features”62 (1, simili venerandus in Hercule Caesar/consecrat), Martial refers to the statue as “the greatest glory of the Ausonian [i.e., Italian] road” (2, Ausoniae maxima fama viae). As with the adjective “Latian” used in the earlier epigrams and Silvae 1.1, “Ausonian” here implies a contrast between the statue’s Greek associations and its Roman milieu, while the word Appia with which the poem begins likewise pits Roman tradition against Greek myth by recalling the venerable Roman aristocrat, Appius Claudius Caecus, who was the road’s builder. Domitian has surpassed both traditions, Greek and Roman, in that his statue is now the Appian Way’s “greatest glory.” The notion of Hercules/Domitian’s Roman superiority to the mythical Greek Hercules is further underscored by suggestive details of the ensuing narratives. While Martial shapes no exact correspondence between Hercules’ feats and Domitian’s, a series of thematic and geographical parallels creates the impression that, once again, Domitian’s reign is the fulfillment of the mythological events. For example, the “Scythian girdle” (5) that Hercules took from the Amazonian queen Hippolyta prefigures Domitian’s campaigns against the Sarmatians (17) and perhaps also the nearby Dacians, while the “treacherous horns of the Sarmatian Ister” (17, cornua Sarmatici . . . perfida . . . Histri), which Domitian is said to have “crushed” three times, suggests Hercules’ breaking off of the river god Achelous’s horn in combat to win Deianeira.63 Domitian’s having “washed” (18, lavit) his sweating horse in Getic snow recalls Hercules’ having “washed” (10, lavit) the cattle of Geryon in the Tiber; that is, his crossing of the Tiber at the time of his encounter with Cacus, an event that already links Hercules to Rome. The mythological characterization of Domitian’s battles culminates in his bearing off a “name” from “the Hyperborean region” (20, Hyperboreo nomen ab orbe tulit), a reference to the title Germanicus that he acquired after his triumph over the northerly Chatti in 83 ce. Here again there is a possible reminiscence of the Hercules myth, since Pseudo-Apollodorus (Bibliotheca 2.114, 119) places Atlas and the golden apples of the Hesperides (cf. 4, aurea poma) in Hyperborea rather than Libya, as Martial seems to initially. Martial’s account of Domitian’s battles would then conclude, in Alexandrian fashion, with an allusion to a scholarly disagreement over the first locale to which he refers in the Hercules section, Libya (4). Yet Martial also praises Domitian for accomplishments that seem to go far beyond what Hercules achieved, especially in the rapid list of deeds at the very end of the comparison (21–22). Sheer numbers bear out the poem’s final conceit that “Hercules’ godhead does not suffice for such great deeds” (23, Herculeum 62

On the translation of simili here, see Henriksén 1998–1999: 2.171. For this reading of line 17 see Henriksén 1998–1999: 2.182. This is not among the deeds of Hercules to which Martial refers. 63

228

t h e c a p t o r’s i m a g e

tantis numen non sufficit actis): while Hercules may have founded one religious structure at Rome, the Ara Maxima, Domitian’s building of multiple “temples for the gods” (21, templa deis) outdoes this.64 Likewise, Hercules’ own translation to divinity cannot match the apotheosis and catasterization of a number of Domitian’s family members (22, astra suis, caelo sidera), together with his own presumed deification.65 Hercules’ destruction of unjust, violent enemies (e.g., Antaeus and Cacus) pales before Domitian’s moral reforms and pacification of the empire (21, mores populis dedit, otia ferro); and Hercules’ connection with the Muses, enshrined in the Roman temple of Hercules Musarum founded by M. Fulvius Nobilior in 189 bce, is but a prelude to Domitian’s Capitoline games (22, serta Iovi).66 “Let this god,” Martial concludes, “lend his features to his Tarpeian father” (24, Tarpeio deus hic commodet ora patri), that is, Jupiter. Martial here reactivates not only the notion behind 65.1–2—that Jupiter Capitolinus can now recognize the new Hercules as his son—but implies further that the son has become the father: Jupiter’s statues, too, should now bear Domitian’s appearance. The end of Epigrams 9.101 thereby takes us once more into a realm already traversed in Silvae 1.1, which culminates in parallels between Domitian’s equestrian statue and images of Zeus/Jupiter, and which likewise situates this artwork within a broader field of cultural achievements. Martial, too, with his concluding flourish, dares to present his poem as a monument of Domitian’s Rome, and an offering to the Roman Jupiter. In their ecphrastic poems on statues of Domitian, we see Statius and Martial gravitating toward a common set of themes and techniques that they nevertheless handle in subtly different ways. Here Greece is made to cede its authority to that of an unsurpassable Roman ruler, who sheds the aura of his Hellenophile majesty upon the poet’s acts of literary emulation and rivalry, so as to become their ideal. Both poets celebrate the Roman present as a paradoxical fulfillment of Greek myth, and Domitian’s art objects as embodying the fusion of Greek and Roman tradition. But Statius, introducing the Troy story as a further, startling antecedent, uses simultaneous, layered allusion to various precedents to create the impression of a seamless whole, while Martial, exploiting echoes of Silvae 1.1 for his own purposes, rings changes on the old and the new, the present as a greater version of the past. As we shall now see, these poets’ work for a non-imperial patron reveals still further ingenuity in making ecphrasis channel the dynamics of the patron/client relationship.

64 According to one version of the myth Hercules himself founded the Ara Maxima (Liv. 1.7.10–11; Ov. Fast. 1.581; Prop. 4.9.67; Verg. A. 8.271), while according to another Evander did (D.H. 1.40.6.; Tac. Ann. 15.41; Str. 5.3.3; Macr. 3.11.7, 12.4). 65 See Henriksén 1998–1999: 2.185. 66 For more on the comparable aspects of Hercules, see Bonadeo 2010: 97–128.

The Patron’s Image

229

II. Statius and Martial on the Hercules Epitrapezios of Novius Vindex Hercules, My Centerpiece: The Convivial Connoisseurship of Silvae 4.6 With Silvae 4.6, ecphrasis comes (back) to dinner.67 Focused on an image of Hercules “Epitrapezios” (“on a table” or perhaps “at table”), and reworking themes prominent in Silvae 3.1, this poem describes a Lysippan statuette that Statius claims to have encountered at a dinner party given by another wealthy friend, Novius Vindex.68 Whether actually by Lysippus or a copy, Vindex’s image may have been a reduced version of a massive prototype, of which a fullsize copy/adaptation has possibly survived in the seated Hercules from Alba Fucens (figure 7.2).69 The text’s convivial setting, an occasion perhaps arranged by Vindex for the specific purpose of having Statius view a prized statue and compose a poem on it, returns us to circumstances of power and dependency surrounding the display of Hellenophile sophistication.70 As is also implicit in Petronius’s Cena Trimalchionis (cf. Chapter 6), the dinner party was a common form of largess bestowed by patrons upon their clients: a potentially generous act, but nevertheless a site where the pair’s unequal status was made clear. While the host incurred the responsibility of a satisfying meal and diverting entertainment, the guest in turn was expected to express appreciation, which in Statius’s case takes poetic form. Statius’s own Greek origins and qualifications, combined with his reputation as a poet, may have made him an especially attractive choice for Vindex’s purposes. In any event, Statius uses Silvae 4.6 to profess his profound sensitivity to the particular version of Greek culture he asserts that

67 Scholarship on Silvae 4.6 has recently been placed on a new footing by the publication of Bonadeo’s book-length study of this single poem (2010). I have also learned much from the earlier analyses by Coleman 1988: 173–94; Kershaw 1997; Newlands 2002: 73–87; Nauta 2002: 228–29, 256–57, 321–23; Lorenz 2003; Zeiner 2005: 190–200; Chinn 2005; and McNelis 2008. 68 Coleman 1988: 174 hypothesizes that the title ἐπιτραπέζιος meaning “on a table” was “an epithet acquired by the miniature, either Lysippus’ own or a copy, alluding to its potential as a table decoration,” while any large-scale prototype (see below) was “probably not originally known as ἐπιτραπέζιος,” since it could not have rested “on a table” nor depicted the god in the accubatio position, i.e., reclining “at table,” but in the sitting posture Statius describes. See further Bonadeo 2010: 24–42. On Vindex’s identity, see Coleman 1988: 173; Bonadeo 2010: 15–23. 69 For discussion, see Coleman 1988: 173; Bonadeo 2010: 24–42 (with further parallels in extant art). Lysippus is reported to have made small-scale versions of his own works. The statue from Alba Fucens largely corresponds in pose to the one Statius describes, but holds the wine-cup and club in opposite hands, if we take Mart. 9.43.4 as our guide. 70 On the possible motives for Vindex’s dinner (and the further possibility that the whole occasion has been invented by Statius), see Coleman 1988: 176; Bonadeo 2010: 49–50.

t h e c a p t o r’s i m a g e

230

Figure 7.2 Seated Hercules from Alba Fucens. Second-first c. bce. Su concessione della Soprintendenza per i Beni Archeologici dell’Abruzzo—Chieti.

Vindex has supplied him in abundance. The Roman poet’s Hellenophilia finds an ideal ambiance in his patron’s house, and his ecphrastic poem becomes a form of repayment to an artwork’s actual owner. This conceit is at the fore almost from the outset of Statius’s poem. After a breezy opening reminiscent of Horace’s Satires, Silvae 4.6 moves directly to the dinner’s powerful effect on the poet, especially for its refined discussion of literary topics.71 Statius imagines the Greek Muses infusing the entire occasion with their inspirational power, of which his poem itself becomes evidence. “This [dinner],” he insists, “having sunk deep into my inmost soul, remains unconsumed” (4–5, haec [cena] imos animi, perlapsa recessus/inconsumpta manet). “A true passion,” he continues, “and conversation fetched from the midst of Helicon, and cheerful jests” (12–13, verus amor medioque Helicone petitus/sermo hilaresque ioci) made up the nightlong revels (13–16); would that they could have lasted as long as the night Hercules’ own father, Zeus/Jupiter, lengthened 71

On the Horatian echoes, especially of the opening of S. 1.9, see Coleman 1988: 176; Bonadeo 2010: 171–76.

The Patron’s Image

231

to make love to his mother, Alcmene (17). The night of Vindex’s dinner even possesses its own “enduring genius” (19, genium . . . perennem), a “quality of divinity derived from the divinely inspired literary conversation.”72 No true poet, Statius suggests, could have been more at home than he was on this occasion. Turning to the theme of visual art, Statius expresses the utmost pleasure in Vindex’s wondrous collection, a feeling he links to the latter’s unfailing discernment where Greek masterworks are concerned. He begins: Then, in that place, I got to know thoroughly innumerable images of bronze and ancient ivory and waxes of deceptive substance on the point of speaking. mille ibi tunc species aerisque eborisque vetusti atque locuturas mentito corpore ceras edidici. (Statius, Silvae 4.6.20–22) The detail locuturas, recalling the trope of the lifelike, “speaking” artwork familiar from Greek epigram, underscores Statius’s sensitivity to Vindex’s sculptures.73 The narrator of Catullus 64, as we have seen, uses direct speech to convey the words of figures in an imaginary tapestry; Statius all but hears Vindex’s mute artworks speak, and he will go on to address one of them, the Hercules statuette, directly (89–108).74 Again, Vindex’s own connoisseurship has afforded Statius this rare opportunity for communion with the finest art, since no one, he suggests, can compete with Vindex in power of observation (22, oculis) and the ability to recognize the work of old masters even without an identifying inscription (24, non inscriptis . . . signis). Indeed, Vindex “will show you” (tibi . . . monstrabit) the distinguishing traits of works by Myron, Praxiteles, Phidias, and Apelles (25–30). For art is his true passion—when not writing poetry (30–31). With the mention of Vindex’s poetry in this context, Statius makes his own conjunction of text and image still a further reflection of Vindex’s interests. Statius now turns to the statuette itself, whose inspired creation by Lysippus he links to a viewer’s response through implicit reference to the notion of phantasia.75 Statius first relates his own strong reactions to the image. Amid all of 72

Coleman 1988: 180. On connections between Silvae 4.6 and Greek epigram, see esp. Hardie 1983: 130–31; Coleman 1988: 175; Chinn 2005; Bonadeo 2010: 129–48 and passim. 74 “Epyllion” is in fact among the generic precedents posited for Silvae 4.6. See Bonadeo 2010: 133, 135–36. 75 On the connection between phantasia and ecphrasis see the introduction, pp. 9–10, n25; on the deployment of this idea in the present passage, see Coleman 1988: 184; Bonadeo 2010: 68–70. 73

t h e c a p t o r’s i m a g e

232

Vindex’s artworks, he asserts, the Hercules sculpture “seized my soul with much love and did not satisfy my eyes even with long viewing” (33–34, multo mea cepit amore/pectora nec longo satiavit lumina visu). This Statius ascribes to its “dignity” (honos) and the “majesty” (maiestas) contained within its diminutive form (finesque inclusa per artos) (35–36). Lysippus’s work, he goes on, is the result of the god Hercules’ self-revelation to the artist: “A god he is, a god,” Statius exclaims, “and to you, Lysippus, bestowed himself to look upon” (36–37, deus ille, deus, seseque videndum/indulsit, Lysippe, tibi). More specifically, Hercules, though appearing small, left an impression of greatness upon Lysippus’s mind (37–38, parvusque videri/sentirique ingens). In turn, the image will compel a viewer to pronounce it the real Hercules and exclaim, “By this breast the ravager of Nemea [i.e., the Nemean lion] was crushed; these arms bore the lethal oak club and broke the oars of the Argo.” “hoc pectore pressus vastator Nemees; haec exitiale ferebant robur et Argoos frangebant bracchia remos.” (Statius, Silvae 4.6.40–42) Such sentiments reinforce the general perception of Lysippus’s work as especially lifelike, as in Propertius’s claim that “it is Lysippus’s glory to fashion lifelike statues” (Prop. 3.9.9, gloria Lysippo est animosa effingere signa).76 Silvae 4.6 thus posits a two-part sequence of inspired visions—φαντασίαι—beginning with Hercules’ initial appearance to Lysippus.77 The ventriloquism of an enthusiastic viewer’s words implies that Statius’s own poem is one such reaction to a divinely inspired phantasia, and so generalizes the statue’s power beyond this particular occasion in a compliment both to Lysippus for his artistry and Vindex for his discerning acquisition of the work. Statius encourages his audience to imagine that the mental image evoked by his poem is yet another version of Hercules’ self-revelation, the god’s own form realized once more. To this extent, we can draw a parallel with Propertius’s cult statue of Apollo (Chapter 4), which, as the god “himself,” seems to inspire the poetry. Statius’s stress on the aptness of the Roman Vindex’s ownership of this Greek statuette—especially salient for my argument—comes through especially in his account of its previous owners. From the opening of his account, Statius emphasizes the sculpture’s good fortune in arriving in Vindex’s house, as opposed to 76 77

For this meaning of animosus here, see Richardson 1977: 350. Coleman 1988: 184. On phantasia see further the introduction, pp. 9–10, n25.

The Patron’s Image

233

a royal court or some other official setting: “Its fortune is worthy of the sacred work” (59, digna operi fortuna sacro), he insists.78 The statuette, Statius asserts, was Alexander’s constant “companion” (61, comitem), always present on his table and touched “by the right hand with which he had taken away and bestowed diadems [i.e., of Persian rulers] and overturned great cities” (62–63, qua diademata dextra/abstulerat dederatque et magnas verterat urbes). Here the idea of Alexander’s subjection of eastern potentates, although a precedent for Roman ambitions, nevertheless acquires an air of danger in Statius’s reference to great cities destroyed. Likewise, Alexander’s close relations with Hercules while at war in eastern and Greek lands—Statius imagines him seeking courage from the statuette and even recounting to it his victories (65, huic . . . narrabat)— accompany his regret at the destruction of Thebes, Hercules’ birthplace (64–70). Such intimacy gains special point from this statue’s imitation of Hercules’ image itself through phantasia as well as from Lysippus’s broader reputation for lifelike work, noted above. In wrapping up the Alexander passage with that monarch’s supposed death from poisoned wine (72, letale merum), Statius leaves no doubt that Hercules has now found a better home. Lysippus’s handiwork even reacts to Alexander’s death by changing its expression and sweating (73–74). Subjection to Hannibal, whose recognized bond with the Phoenician deity Melqart makes him a fitting successor to Alexander in this context, is alien both to the statuette’s Hellenic origins and to its present, Roman milieu.79 If Hannibal’s very name suggests barbarism (cf. 106–07, barbarus . . ./Hannibal), Statius magnifies this impression by referring to him as Nasamoniaco . . . regi (75, “king of the Nasamones”) an epithet associating him not with the Hellenophile Carthaginians but with a more remote North African people, most familiar in Statius’s day for having been crushed by Domitian.80 Hannibal here possesses “the status of a barbarian king,”81 his non-Hellenic qualifications further stressed in Statius’s emphasis on the destruction of Saguntum (82–84), believed in antiquity to be a Greek settlement founded by Hercules.82 The striking enjambment that puts ipsius (“his own”) at the beginning of line 83 makes this point eloquently. Not only this outrage but also Hannibal’s hostility to Rome provokes Hercules’ negative response. Hercules “hated [Hannibal] spattered with the 78 Lines that recall Silvae 1.1.22, “its location is equal to the work” (par operi sedes), where Statius introduces the very different, public circumstances of Domitian’s monument. Cf. Bonadeo 2010: 87. 79 Melqart was identified with Hercules already in antiquity, while Hannibal’s ties to Melqart (evidently in imitation of Alexander) were also recognized in Greco-Latin literature; see Bonadeo 2010: 78–83. 80 In 85 or 86 ce. Nasamoniacus became a general term for “African,” as Coleman 1988: 189 points out. Cf. Bonadeo 2010: 248–49. 81 Coleman 1988: 189. 82 Its name falsely derived from Ζάκυνθος. See Coleman 1988: 190; Bonadeo 2010: 255–56.

t h e c a p t o r’s i m a g e

234

blood of the Italian race and bringing with him dire flames for Roman houses” (78–79, Italicae perfusum sanguine gentis/diraque Romuleis portantem incendia tectis/oderat). Hannibal himself is “savage and exultant in his perjured sword” (atrox . . . periuroque ense superbus), a reference to proverbial Carthaginian bad faith. By contrast, Hercules’ own hatred and grief (81, maerens) find their counterpart in the “honorable fury” (84, furias . . . honestas) Hannibal inspired in the Saguntines and Romans.83 Statius’s brief account of Sulla (85–88), another ruler who identified with Hercules, seems at first to present a more favorable picture of the statuette’s subsequent history, but later in the poem we find an echo of the conventional portrait of a brutal dictator: “the harsh voice of savage Sulla” (107, saevi . . . vox horrida Sullae), Statius insists, could not have celebrated the statue as well as Vindex in his own poetry.84 Here, however, Statius emphasizes Sulla’s patrician rank by referring to his house as “not plebeian” (85–86, nec . . . plebeia domus) and his household gods as “illustrious” (87, claros . . . penates). The statue itself, moreover, is described in this passage as “happy in the lineage of its masters” (88, felix dominorum stemmate), with felix an allusion to Sulla’s cognomen.85 In this way, the “snob appeal” of owning a work that belonged to a famous Roman aristocrat seems momentarily to trump Sulla’s associations with civil war and butchery.86 Statius’s overall lack of emphasis, however, on Sulla’s ownership may constitute a tacit acknowledgment of his unsuitability as a comparandum.87 It may owe something as well to the fact that his campaigns in the Greek East and Asia were thought to be the first moment that Roman soldiers acquired an interest in the looting of artworks and other types of depravity, while Sulla himself was known to have robbed the sanctuary of Zeus Eleutherios at Athens, from which his thefts included a painting of a Hippocentaur by Zeuxis.88 Given Sulla’s reputation, an implicit contrast becomes inevitable between his traits, whether brutality or a crass attitude toward Greek art, and the companionable Vindex’s refined connoisseurship. Like the entire passage on the statuette’s owners, the poem’s final movement (89–109) begins by summarizing how Vindex blends Greek culture with Roman values in a way that is completely appropriate to the statuette’s presence in his house as opposed to the earlier owners.’ The passage opens with a suggestion of 83

For this interpretation of populis (84, lit. “in the peoples”), see Coleman 1988: 190–91. Cf. Coleman 1988: 191; Bonadeo 2010: 279–80. For Sulla’s identification with Hercules, see Bonadeo 2010: 83–84. 85 The allusion is perhaps ironic. See Coleman 1988: 191; Bonadeo 2010: 259–60. 86 See Coleman 1988: 175 for other passages in Latin literature illustrating the snob value of art that had belonged to distinguished owners. 87 Bonadeo 2010: 85–86 proposes an implicit contrast between Sulla’s “absolute lack of fides” and Vindex’s paradigmatic embrace of fides (cf. Silv. 4.6.91–93). 88 See Coleman 1988: 191 for references. 84

The Patron’s Image

235

Epicurean belief, insofar as Statius voices hesitation over the gods’ actual concern with knowing men’s ways and souls (89–90). Epicureanism comes through more strongly in the “happy tranquility” (96, laeta quies) that Hercules now enjoys.89 The Greek cast of such language combines, however, with an emphasis on “the obligations of amici towards each other which form the basis of Roman social relations.”90 Addressing the statuette itself, Statius declares that “no royal hall . . . nor regal honor surrounds” it (90, non aula quidem . . . nec . . . /regius ambit honos), but rather the chaste and blameless mind of a master to whom belongs ancient good-faith and the unending bond of friendship once begun. casta ignaraque culpae mens domini, cui prisca fides coeptaeque perenne foedus amicitiae. (Statius, Silvae 4.6.91–93) Here prisca fides is especially evocative of long-standing Roman custom, while the ensuing idea that the ghost (95, umbrae) of Vestinus, apparently a close friend or benefactor inserted abruptly in the text, holds Vindex in its “embrace” (complexibus) introduces traditional Roman notions about the afterlife into context where they might not at first appear to belong.91 Vindex’s house improves upon the Greek world precisely in its suffusion by Roman values and beliefs. Statius’s poem both celebrates and depends upon this state of affairs, and Statius ends by suggestively merging his voice with his patron’s, whose own poetry on Hercules’ exploits he describes in some detail (99–105). We are not told whether the latter work was extant or planned, but Vindex’s role as poet means that the statuette does “not look upon war and ferocious battles but lyre and fillets and laurels that love song” (97–98, nec bella vides pugnasque feroces/sed chelyn et vittas et amantes carmina laurus). Here the Greek word chelys points once 89 Coleman 1988: 191–92; she well describes how Statius’s Vindex evinces traits opposite to Alexander’s and Hannibal’s and embodies while also exhibiting a generally Epicurean cast of mind. 90 Coleman 1988: 191; cf. Bonadeo 2010: 107. 91 Note especially the fact that Vestinus is called “equal to his great ancestors” (94, par magnis . . . avis). On his identity, see Coleman 1988: 192; Bonadeo 2010: 267–68. Epicureanism rejected the idea of any afterlife for the deceased. With this passage cf. Silv. 2.7.120–23, where the ghost of the Roman poet Lucan may be allowed, on the model of the mythical Greek couple Protesilaus and Laodamia, to return for one day of life with his widow, Polla. Even without such contact, Polla “worships and cherishes you yourself [i.e., the dead Lucan], fixed in her more deeply than her deepest marrow” (126–27, ipsum . . . colit et frequentat ipsum/imis altius insitum medullis).

t h e c a p t o r’s i m a g e

236

again to the philhellenic basis of Vindex’s cultivation, and perhaps to Statius’s Greek origins as well (we are also not told whether Vindex composes in Latin or Greek). In the summary of Vindex’s possibly hymnic poetry that follows, something of Martial’s epigrammatic penchant for condensing heroic narrative into a very few lines contributes to the prevailing effect of greatness-in-smallness: note especially the triple repetition of quantus [“how great”] in verses 99–101.92 Vindex, we are told, “will relate in solemn song” (99, sollemni memorabit carmine) Hercules’ exploits at Troy and capture of Diomedes’ horses from Thrace, the labors of the Stymphalian birds and Erymanthian boar, the capture of Geryon’s cattle and killing of Busiris, the descent to the underworld, the seizure of the apples of the Hesperides, and the taking of Hippolyta’s girdle, here compressed into seven lines (99–105). A miniature version of Vindex’s work, Statius’s summary is still another figure for the miniature sculpture, a further containment of this paradoxical object by the poem that describes it. As Silvae 4.6 concludes, poet and patron both sing and see across the social divide separating them, which nevertheless remains in place. None of the statue’s previous owners, Statius declares, could have celebrated it “in these measures” (108, his . . . modis), nor would Lysippus, addressee of the poem’s last line-and-a-half, prefer the approval of “other eyes” (109, aliis . . . oculis).93 Read within the context of the previous lines, modis and oculis naturally refer to Vindex’s poem and eyes other than his. Nothing, however, prevents an audience from hearing them as referring to Silvae 4.6 and eyes other than Statius’s.94 The ambiguity conjoins the two in sympathy of word and vision, yet the more specific referent of “other eyes,” is of course Lysippus’s own patron, Alexander. With “a final allusion to the question of patronage,” Statius thus concludes by elevating Vindex’s private beneficence toward him above Alexander’s toward his “official” court sculptor.95 The last line of Silvae 4.6 reworks still another technique familiar from Silvae 1.1 to underscore the opposition between private and public life animating the whole poem. Identification with Lysippus (note the prominent vocative, Lysippe) once again makes his art an icon of Statius’s own poetry. But just as with the Lysippan statue of Alexander appropriated by Julius Caesar, we are dealing with Greek art inextricable from its Roman context, indeed, both shaping and 92

For Vindex’s poem as possibly a hymn (cf. 99, sollemni carmine), see Coleman 1988: 192; Bonadeo 2010: 272–73. 93 For the possible play on his . . . modis as also meaning “within these limits” see Bonadeo 2010: 280. 94 Cf. Newlands 2002: 85; Bonadeo 2010: 280–82. 95 Newlands 2002: 85–86. As she notes, Horace at Ep. 2.1.232–4 relates that Alexander had good taste in visual art but not in poetry; Statius can thus claim to offer Vindex “far better poetry than that received by Alexander, according to Horace, from the worthless Choerilus” (Newlands 2002: 86).

The Patron’s Image

237

shaped by this context in such a way as to render its Greekness an indelible part of Romanness. Statius himself, like his poem and Vindex’s statue, embodies the paradoxes of the Greek-turned-Roman. The hybrid world of Silvae 4.6 encompasses a realm of intimacy between him and his Hellenophile patron, the Roman-turned-Greek.

“Don’t You, a Poet, Know Greek?”: Martial’s Epigrams on Vindex’s Statue Martial’s two epigrams on the same statue (9.43–44) confirm that intimacy such as this was a valued asset and, like Silvae 4.6, emphasize the ways in which the bond between the Roman patron and his client informs the poetry’s entire posture vis-à-vis the Greek image it describes.96 The first of the two poems bears striking similarities to Silvae 4.6, similarities so close, in fact, that they have been thought to reflect a set of instructions issued by Vindex and dutifully followed by both poets.97 Thus Martial refers, though in the opposite order from Statius, to the sculpture’s seated posture on a lion skin covering rocks, its upturned gaze, and the objects it holds in its hands, before turning to its three earlier, illustrious owners and its delight in inhabiting a private home just as the mythical Hercules once became Molorchus’s guest. But once again Martial exploits the epigrammatic form and the elegiac meter, tagged as Callimachean by the presence of Molorchus, to compress his subject matter, now in a manner especially suggestive of the statuette’s Hellenistic aesthetic. The meter in particular allows him to situate the work’s latent paradoxes within a literary dialogue between epic and elegy, a technique informing Roman ecphrasis at least since Propertius 2.31 (cf. Chapter 4). These effects begin at once. The first poem’s expansive opening hexameter, which describes the lion skin “spread out” (43.1, porrecto) on the rock whose hardness (dura) likewise belongs to an epic ethos, immediately confronts elegy’s softer associations in the first word of the pentameter, mitigat (2, “softens”). Just as the sculpted Hercules softens the rock with the skin of the savage lion, the poem’s second line mitigates the epic theme by calling the statue “a great god in a small work of bronze” (exiguo magnus in aere deus). Here the conceit seems even more at home than in Silvae 4.6, since a similar one structures Greek ecphrastic epigrams such as AP 9.776 and APl. 120.98 Similarly, the next couplet begins with reference to one of Hercules’ most impressive labors, his hefting of the universe until Atlas could obtain for him the apples of the Hesperides (3), 96

Mart. 9.43–44 are very often discussed in connection with Silvae 4.6. In addition to the studies cited in n67 above, see Schneider 2001. 97 White 1975: 287. 98 Henriksén 1998–1999: 1.208; Coleman 1988: 183.

t h e c a p t o r’s i m a g e

238

while the pentameter marks a turn from epic to the sympotic realm of elegy, noting first the club in the statuette’s left hand, then the wine in its right (4). The third couplet situates the great/small dichotomy in direct relation to the opposition between Rome and Greece: the hexameter declares, “He is no recent object of fame nor the glory of our [i.e., a Roman] chisel” (5, non est fama recens nec nostri gloria caeli), with the pentameter asserting, “You see the noble product and handiwork of Lysippus” (6, nobile Lysippi munus opusque vides).99 Rome is the large-scale setting for the Greek Lysippus’s noble, small-scale art. The remaining couplets enact related juxtapositions of subject and theme. In the next hexameter, the “Pellaean tyrant” Alexander’s table is said to have “held” the statue’s “godhead” (7, hoc habuit numen Pellaei mensa tyranni), while, in elegiac vein, the pentameter suggests that Alexander’s swift conquests could not prevent his untimely death (8, qui cito perdomito victor in orbe iacet). Here the noun victor is isolated and diminished by its position in the middle of the shorter line. The statue’s passive role as the object by which the young Hannibal swears at Libyan altars (9) gives way to its active role in commanding “savage Sulla” (10, Sullam . . . trucem) to lay down his rule. Because Sulla did in fact retire, this line is especially striking for the power it imaginatively grants the statuette within the pentameter’s shorter compass. The idea of retirement from politics sets up the next two couplets, contrasting the fearful, haughty life of the aula (11) with that of Vindex’s household (12) and assimilating the hospitality of “kindly Molorchus” (13, placidi . . . Molorchi) to that of “learned Vindex” (14, docti Vindicis). The single word doctus, the regular Latin term to indicate Alexandrian learning, does the work of many lines of Silvae 4.6 in the way it relates Vindex’s own philhellenic cultivation to Martial’s. And although the poor Molorchus thereby becomes Vindex’s mythic counterpart, Molorchus’s role as host to the hero Hercules—that is, Hercules before his ascent to divinity—emerges as less impressive than Vindex’s possession of Hercules the “god” (14, deus). The very structure of the final line succinctly confirms Vindex’s improvement upon earlier conditions through its emphasis on the divine Hercules’ choice of Vindex: sic voluit docti Vindicis esse deus. The second epigram on the statue revisits such themes in a more lighthearted vein—now perhaps in direct parody of Silvae 4.6—yet nevertheless mobilizes connotations of grandeur with a serious bearing on the statue and its context.100 Composed in hendecasyllables, a meter with jocular associations, the poem begins with Martial’s reported question about the statue’s creator, “whose work and happy labor?” (2, cuius opus laborque felix). Scholars are divided over 99

Caelum here may also be heard as “heaven,” i.e., the statue is “not made under a Roman sky”; see Henriksén 1998–1999: 1.208. 100 For parody of Silvae 4.6 in Mart. 9.44, see Kershaw 1997; cf. Henriksén 1998: 108–11; 1998–1999: 1.212.

The Patron’s Image

239

whether the text of line 1 indicates the statue or Vindex as addressee, and so also over the speaker of lines 4–5.101 If the statue speaks, Epigrams 9.44 uses a convention more familiar from Greek poetry than Latin to complement the likewise Greek technique of beginning with a question of this kind.102 In either case, however, the laughter and “light nod” (3, levi . . . nutu) with which the addressee responds undermine the weighty theme suggested by the poem’s first word, Alciden (or Alcides). Following upon the previous poem, moreover, this epigram elicits amusement by temporally preceding it, since there the sculptor’s identity is already known.103 Martial infuses the imagined discussion with suggestive play between Greek and Roman: “Don’t you,” [the addressee] said, “a poet, know Greek? The base is inscribed and indicates the name.” “Graece numquid,” ait “poeta nescis? inscripta est basis indicatque nomen.”104 (Martial, Epigrams 9.44.4–5) No one, the verses imply, could legitimately be called a Roman “poet” without a knowledge of Greek (poeta itself being in origin a Greek word) and the special pertinence of Greek for the epigrammatist emerges in an allusion to the genre’s association with inscription (inscripta est basis). Greek epigrams, particularly those inscribed on objects, are models from which Martial announces an ambivalent distance. The statuette, Martial suggests further, carries with it a Greek pedigree that would make its celebration in Greek verses especially appropriate, and in this light Martial’s Latin offering risks being seen as something of a disappointment. Indeed, before the reference to the inscription in line 5, one might hear in the previous line’s “Don’t you, a poet, know Greek?” the implied question, “Can’t you, a poet, compose in Greek?” or perhaps, “Don’t you, a poet, know that this kind of question is a worn-out trope of Greek epigram?” With 101 Henriksén 1998–1999: 1.212 follows Lindsay 1929 and Heraeus-Borovskij 1976 in preferring the reading of the β-group, Alciden . . . Vindicis, which would make the statue the addressee; Schneidewin 1842 and Friedländer 1886 prefer the γ–group’s Alciden . . . Vindicem, which would make Vindex the addressee; Shackleton Bailey 1990 accepts Gilbert’s (1901) Alcides . . . Vindicem, which would also mean that Martial addresses Vindex. 102 Kershaw 1997: 272; McNelis 2008: 269–70. 103 McNelis 2008: 268. 104 The meaning changes somewhat if a comma is introduced after poeta so as to make it vocative: “Poet, don’t you know Greek?” See Henriksén 1998–1999: 1.213 for the preferability of omitting the comma here. Kershaw 1997: 272n10 notes the possibility that Martial, not the interlocutor, speaks the words inscripta . . . nomen.

t h e c a p t o r’s i m a g e

240

characteristic self-deprecation, Martial sets himself up for a fall, and his laughable ignorance only offsets Vindex’s artistic acumen, especially since the ability to identify an artist’s work without an inscription is exactly what Statius praises Vindex for at Silvae 4.6.22–24.105 Similarly, the poem’s conclusion is all the wittier for Statius’s allusion to phantasia. “I read the name of Lysippus,” Martial avers, “I thought it was a work of Phidias” (6, Lysippum lego, Phidiae putavi).106 The line turns on a contrast between the statue’s diminutive appearance and the grandeur of the image it creates in the viewer, a grandeur worthy of colossal Phidian works like those discussed above. Martial emphasizes his own internal state in the contrast between lego and putavi, and thereby suggests that a similar contrast ought to exist between the internal state of viewer or audience and the poem he or she reads in Martial’s book. Ignorance suddenly becomes the clever vehicle for a compliment to Vindex. A further witty turn of events emerges if we are right to hear a pun on Phidias’s name and the Greek verb φείδομαι (“spare”) to the effect that in creating Vindex’s statuette, a “sparing” Phidias was, contrary to expectation, living up to his name.107 Martial, too, has, as it were, defeated expectation in knowing Greek very well and so earning the title poeta of which he was nearly deprived. Just as identification with Lysippus provides Statius with an opportunity to articulate a poetic program, so Martial uses Lysippus for programmatic ends, but unlike Statius suggests that the key to his own programmatics lies in a simultaneity of Lysippan refinement and a kind of Phidian grandeur represented not so much by the individual epigram as by the entire, varied corpus of his poetry.108 When assessed against a sculptor such as Phidias, the embodiment of classical order and monumental proportion, Lysippus (although he, too, fashioned colossal works) stands easily for Martialian epigram’s anticlassical, Hellenistic-inspired aesthetic, “the novel artistic style that corresponds to innovative approaches of Hellenistic poets.”109 This is confirmed, for example, by a programmatic passage of the Greek epigrammatist Posidippus only recently brought to light in the Milan Papyrus: here Lysippus’s novel works (νεάρ’) stand in contrast with the efforts of the “old style” (παλαιοτέχνης) sculptor Polyclitus (62.4–6 105

Cf. Kershaw 1997: 271–72; Henriksén 1998–1999: 1.212; Chinn 2005: 253. Shackleton Bailey 1993 follows Housman 1907: 247 in preferring the MSS’ reading Lysippum instead of the Greek Λυσίππου, which although printed in the editio Aldina of 1501 and preferred by most modern editors lacks MS support. As Henriksén (1998–1999: 1.213–14), offering parallels, points out, a Greek inscription on the statue is more likely to have read Λύσιππος ἐποίει (or ἐποίησεν), while Λυσίππου “would perhaps rather suggest that the statuette was a copy”; Lysippum gives the preferable meaning, “the name of Lysippus.” 107 Schneider 2001: 709; McNelis 2008: 268–69. 108 McNelis 2008: 269–72. 109 McNelis 2008: 269. 106

The Patron’s Image

241

Austin-Bastianini), an artist many years Phidias’s junior.110 In Epigrams 9.43–44, Martial implicitly claims this aspect of Lysippus’s Hercules as a precursor to his own poetic innovations. And yet Martial’s Phidias is more than a symbol of classicism. Elsewhere in the Epigrams, Phidias, who appears with relative frequency, does not merely symbolize idealized, old-style grandeur, but combines a reputation for works of cosmic import with more realist, less august associations, just as Martial treats the high and the low together to create his own, tendentiously veristic regard on Rome’s universal empire.111 In Epigrams 7.52, for example, the cosmic grandeur with which Domitian’s architect, Rabirius, has fashioned the emperor’s palace makes him worthy to create a temple for the Phidian Zeus. Similarly, in an earlier poem of Book 9, Martial has insisted that the statue of Domitian in Carus’s possession outdoes this same Phidian image in depicting “the face of the heavens, the features of a serene Jove” (24.3, mundi facies . . . Iovis ora sereni). Contrast, however, the glib, two-line Epigram 3.35, on a relief-work depiction of fish identified as Phidias’s for being so lifelike: “add water,” Martial remarks, “and they’ll swim” (2, adde aquam, natabunt).112 Hardly a grand topic, Martial’s theme here finds a still more bathetic counterpart in 6.73, on a Priapus statue whose erect member is said, with obscene double-entendre, to be “worthy of Phidias’s hand” (8, Phidiaca . . . digna manu).113 Within Martial’s work, Phidias stands for a range of scale and subject matter both exceeding those attributed to Lysippus and bearing a different sort of resemblance to Martial’s own concerns. In purportedly mistaking Vindex’s Lysippus for a Phidias, Martial inscribes this object more deeply within his own poetic cosmos. Yet this, too, is ultimately a way of paying homage to Vindex’s taste and judgment and deprecating himself, since the Phidian statue in Martial’s imagination is emphatically not the one Vindex actually owns. While evoking the notion of phantasia, Martial nevertheless playfully undermines the claims of ecphrasis to appropriate the visual object “itself ” for a verbal artist’s own designs. Martial even seems to admit that he first 110 McNelis 2008: 269, who notes (n56) that Lysippus also appears elsewhere in Greek epigram as an analogue for Hellenistic poetic style. Phidias appears in Greek epigram in, e.g., APl. 81 on his statue of Zeus at Olympia. 111 Cf. McNelis 2008: 271–72. 112 Cf. Mart. 3.40 on a bowl by Mentor decorated with a lizard. Phidias was known to have created small-scale works as well as large-scale ones. See Grewing 1997: 139 and cf. Mart. 6.13, where the poseur Charinus claims to have relief work by Phidias (4); in 10.87 the same kind of object is given as a gift by an “aged admirer of the ancient forefathers” (15, mirator veterum senex avorum). 113 On the double-entendre, cf. above, p. 103. Grandeur and the lifelike, however, come together in Ep. 6.13 on a portrait of Domitian’s niece, Julia, as Venus that one would believe to have been “formed by Phidias’s chisel” (1, Phidiaco formatam . . . caelo) for its similarity to the real goddess.

t h e c a p t o r’s i m a g e

242

thought of the statue as another Phidian icon of his own poetry before being instructed, whether by Vindex or the statue itself, to regard it as Lysippus’s handiwork. We are reminded once again of Catullus 64, where the narrator’s apparent mistakes in reacting to (imaginary) art from the Greek past prod an audience to reconsider its own attitudes toward Greek culture, and measure these against the postures on display in the verse itself. Such a ploy would, further, resemble Vergil’s in Eclogue 3 or Horace’s in Satires 1.8 if its target is indeed Silvae 4.6, and Martial’s poem a bid to divert favor away from a poetic rival. Thus the brilliance of Phidiae putavi: with these two words Martial activates a compliment, a programmatic statement couched, typically, in self-deprecation, and a potential jab at literary foe. Getting Greek culture “wrong” can, it turns out, have its advantages, too.

Conclusion: Ceding Power Martial’s “mistake” is the culminating example of how the poetry treated in this chapter casts yet another form of doubt upon the idea of an ecphrastic text’s habitual expression of dominance over the image it describes, an ongoing concern of this book. The circumstances of occasional, praise-centered poems for patron-figures, so unfamiliar to our modern sensibilities, require that these texts serve the interests of their addressees. The image that ecphrasis makes its audience “see” reflects direct acquiescence to another’s wishes, which thus become inseparable, too, from the articulation and pursuit of aesthetic goals. Domitian’s mastery over the statues bearing his image is a notion implicit in both Statius and Martial; Statius calls Vindex his artwork’s dominus (Silv. 4.6.92, “owner” but also “master”),114 while Martial says the statuette is “learned Vindex’s god” (Mart. 9.43.14). The Greekness of these objects must in a sense belong to Domitian or Vindex first, then Statius or Martial after, however much this conceit reverses what is arguably the case, that “the visual artist seeking perpetual fame, as well as the ruler or collector, is dependent on the poet’s text.”115 Statius purports to see well-known Greek sculptures, even one in the Forum Iulium, in a different light following Rome’s reorientation around the Equus Domitiani; Martial sees a “greater,” Roman Hercules because Domitian’s face appears on the traditional hero’s body; Statius professes “love” for the Hercules statuette that Vindex’s ownership has delivered into ideal surroundings; the dazzled Martial’s judgment, as it were, fails before “Vindex’s god.” That the Greco-Roman gods could be thought to inhabit their images has already prompted us to broader reflection on the nature of ecphrasis in Chapter 4. This chapter complements the earlier discussion by illustrating how, 114 115

Cf. Bonadeo 2010: 263–64. Newlands 2002: 85.

The Patron’s Image

243

at Rome, the numinous power within statues of divinities might become subject to a rival power much less explicit in the Augustan period, the divinity of the emperor himself. While Propertius stages the epiphany of an Apollo for whom Octavian has constructed a new temple, the Flavian poets’ Domitian has attained to the godhead of Hercules and even Jupiter. In these conditions, the ecphrastic image, too, may become permeated by the emperor’s might, even when the object in question is neither official nor publically displayed. Just as Martial’s Book 9 accompanies the Vindex poems (43–44) with epigrams on Hercules as Domitian (64–65, 101), so the opening poems of Silvae 4 contain prominent comparisons of Domitian with Hercules (2.50–51, 3.155), the first showing Hercules rejoicing “to rest his side on the spread lion-skin” (51, strato latus adclinare leoni; cf. 4.6.58) after fulfilling Eurystheus’s “harsh commands” (50, horrida iussa; cf. 4.6.107, vox horrida Sullae). Numerous echoes of Silvae 1.1 in Silvae 4.6 recall, as we have seen, the emperor and his power even if Statius’s introductory poem makes no mention of Hercules per se. Art objects may change hands, but Domitian’s ownership of Hercules forms an ultimate horizon for Vindex’s possession of the god’s diminutive image, in spite of contrasts drawn with regal uses to which the statuette had supposedly been put. For all that a poet may exercise “his virtuoso and controlling gaze”116 on the Hellenic imagery of Domitian’s Rome, his efforts must finally render unto Caesar that which is Caesar’s—which is to say, a god’s.

116

Newlands 2002: 72.

Epilogue Captives and Captors: Apuleius and Philostratus

A book on classical Latin ecphrasis ends naturally with Apuleius’s second-century ce novel, the Metamorphoses or “Golden Ass,” a work traditionally regarded as among the last surviving examples of pagan Latin literature, and one adapting for its own purposes a series of literary texts and genres treated in earlier parts of this study. The work is doubly appropriate given the specific argument I have advanced here. For although not as prominent in this text as, apparently, in the Greek novelistic tradition to which it is strongly connected, ecphrasis of art objects is still a significant feature of Apuleius’s multilayered receptivity to Greek culture.1 And here, more than three centuries after Naevius, this crucial dimension of the trope, the subtle cultural competition infusing Apuleius’s whole text from beginning to end, still outweighs in importance ecphrasis’s narrowly programmatic qualities. At the opening of Book 2, the novel’s protagonist, Lucius, discovers a wondrous statue of Diana and Actaeon in the home of his aunt, Byrrhena, a resident of the Thessalian town of Hypata. Unrivaled by any other extant Roman ecphrasis for the way it suggests Greek art’s relation to threatening though perhaps salvific changes of identity, the description of this statue forges a dense interrelation of author, narrator, audience, and sculpted figure around this issue.2 Actaeon’s sculpted metamorphosis into a stag prefigures Lucius’s real transformation into an ass, in which degraded form he will spend most of the narrative before being changed back again by the goddess Isis, whose initiate he subsequently becomes. Further changes await Lucius, since at Isis’s command he will travel to Rome, where he will take up a legal career in Latin. A “Grecian story” (1.1, fabulam Graecanicam) in which a Greek, as it were, 1

On ecphrasis of art objects in the Greek novels, see above, pp. 191–92, 195–96. For my understanding of this passage I am indebted especially to Peden 1985; Winkler 1985: 168–70; Schlam 1992: 49, 71, 139n6; Shumate 1996: 67–71; Slater 1998; and 2008; Harrison 2000: 221; van Mal-Maeder 2001: 98–113; Paschalis 2002; Freudenburg 2007. 2

244

Epilog ue

245

becomes Roman via an Egyptian cult; a story, moreover, by a Latin author from the Roman province of Africa Proconsularis and one that defines its subject through ecphrasis of a Greek myth as told in particular by a previous Latin author (Ovid in Met. 3.138–252), Apuleius’s tale caps our discussion in obvious ways. The Apuleian Lucius’s intense receptivity to the sculpture tropes and is troped by Actaeon’s fateful gazing—and this in an episode not found in our closest Greek parallel for Apuleius’s novel as a whole, Λούκιος ἢ ὄνος (Lucius or the Ass, ascribed to the second-century sophist Lucian).3 Beginning his account with the exclamation “Look!” (2.4, ecce), Lucius packs his description with words indicative of the statue’s beauty, artistry, and verisimilitude, and concludes with the image of himself “examining [it] closely again and again” (2.5, identidem rimabundus). The voyeuristic Actaeon, “waiting” (2.4, opperiens) for Diana to step into her bath, is his mythical counterpart, especially if it is correct to read curioso optutu (2.4, “with an inquisitive gaze”) or curiosum optutum as a “proleptic mise-en-abyme”4 linking Actaeon’s behavior to that of the overcurious Lucius, whose insatiable desire to experience the ways of magic will soon plunge him into asinine misadventures.5 The Romanized Greek Isis-worshipper Lucius (whose mother, Salvia [2.2], would seem to be a Roman citizen)6 figures in turn Apuleius, the Hellenized Roman African and possibly native speaker of Punic. We have in fact returned to the territory of Naevius’s Bellum Punicum, though now the trope that linked Rome’s Punic foe to defeated mythical rebels may point instead to the Punic subject’s appropriation of Hellenic sophistication prized by the Roman conqueror.7 The external audience, moreover, drawn into a scene of erotic viewing evocative of the rest of the novel’s far more racy content, finds its own experience represented here as well. 3 Though now commonly regarded as an epitome of the Metamorphoses of Lucius of Patras and as sharing this source with Apuleius’s novel. See Harrison 2000: 218–19 for discussion. 4 Van Mal-Maeder 2001: 111. Textual difficulties problematize interpretation here. F transmits curioso optutum in deam tu proiectus, with tu corrected to sum and the abbreviation ~ of optutum scratched out by a later hand (see van Mal-Maeder 2001: 110). Proiectus would seem to be one in a series of nominative singular adjectives applied to Actaeon (cf. ferinus, opperiens), while sum proiectus would seem to make Lucius, rather than Actaeon, the person who looks inquisitively. For a survey of the numerous scholarly conjectures see van Mal-Maeder 2001: 110–11. 5 A further refinement of this point involves the distinction, developed in Winkler’s foundational narratological study of the Metamorphoses, between Lucius-auctor, the narrator looking back on his actions, and Lucius-actor, the protagonist as acting within the narrative. In Paschalis’s view (2002: 139), the Lucius who describes the statue is the former rather than the latter. 6 Cf. Bowersock 1965: 289. 7 Apuleius may himself have been an initiate of the Isis cult (cf. Apol. 55.8, where he claims to have been initiated into “several” cults in Greece). Lucius shares, further, his elite education and interest in “sophistic” display with Apuleius. For discussion of the extensive resemblances between Lucius and Apuleius, see Harrison 2000: 215–20.

246

epilogue

The statue can be seen as symbolic of Apuleius’s text, perhaps especially in the way the text infuses it with life and vibrant motion, situating it simultaneously within its mythical narrative and that of the Metamorphoses. Diana’s statue is “perfectly splendid, its clothes billowing, vigorously running forward to meet those entering” (2.4, perfecte luculentum, veste reflatum, procursu vegetum, introeuntibus obvium). The sculpted dogs are so lifelike that “if from any place close at hand the sound of barking threatens, you will think it comes from the stone jaws” (sicunde de proximo latratus ingruerit, eum putabis de faucibus lapidis exire).8 With their chests lifted off the ground, “their back feet stand fixed, their front feet run” (pedes imi resistunt, currunt priores), a detail self-consciously indicative of their status as miraculously veristic sculpture, and one by which the artist has revealed “the highest indication of his skilled handiwork” (summum specimen operae fabrilis). “Art emulous of nature” (ars aemula naturae) has produced grapes so realistic, “you would think that some of them could be plucked from there for food” (putes ad cibum inde quaedam . . . posse decerpi); reflected in the water that “running about, undulates in a soft wave” (discurrens in lenem vibratur undam) near Diana’s feet, they, too, seem to move. Actaeon appears “already animal in the midst of becoming a deer” (iam in cervum ferinus).9 The statue that seems, via Apuleius’s Latin prose, to burst into motion in its depiction of a Greek myth signals and stands for the remobilization of Greek narratives within/as the Metamorphoses. But as much as one might be tempted to find, at last, firm evidence of the Latin ecphrastic text’s dominance over its Hellenic image (an imaginary sculpture in a parodic, make-believe Greek world served up for the viewer’s titillation and delight10), here, too, a religious image is at issue, and the notion of divine power it implies cannot be written off as a mere fiction given the whole novel’s much-noted play with the idea of religious salvation and its “display of authorial elite religious knowledge.”11 During his prayer to the “Queen of heaven” (regina 8 In such emphasis on “the lifelike realism of visual mimesis” within a text (the Metamorphoses as a whole) that often alludes to mime performances, Kirichenko 2010: 43 finds a reliance on “more or less the same aesthetic principles as the women of Theocritus’s and Herodas’s mimes” that include ecphrasis of art objects (cf. above pp. 69–71 on Theocritus, Idyll 15 and Herodas, Mimiamb 4). 9 Or perhaps “about to be animal through his transformation into a deer”; cf. van Mal-Maeder 2001: 112. 10 This sculpture in fact corresponds to no extant artwork, but rather represents “a composite of famed Hellenistic images” (Schlam 1992: 71; cf. 139n6). 11 Harrison 2000: 248. The question of whether or not the Metamorphoses should be read as a proselytizing text recommending the worship of Isis as a path to salvation is one of the oldest in the scholarly literature. See Harrison 2000: 238–52 for discussion and bibliography. In formulating his own view of Lucius’s initiation into the Isis cult as fundamentally parodic of “the narrator’s youthful derangement and credulity” (248), Harrison responds primarily to Winkler 1985: 223–27, who views the text as offering the reader a choice between proreligious and anti-Isiac interpretations but finally endorsing neither. My own position does not depend on acceptance of either Harrison’s or Winkler’s view, since the fact remains that Apuleius, perhaps an initiate of the

Epilog ue

247

caeli) in Book 11, Lucius-ass in fact suggests Diana as one of her possible identities (11.2), and when Isis appears to him she names “Dictynna Diana” (i.e., Diana as identified with an ancient goddess of Crete) as one of the numerous forms in which she is worshipped the world over (11.5).12 The sculpture of Book 2, which depicts Diana’s vengeance upon a transgressor, is “venerable with the majesty of divinity” (2.4, maiestate numinis venerabile) and calls particular attention to Diana’s power when considered in light of previous literary and artistic versions. Here the dogs that will soon tear Actaeon apart belong not to him (the traditional story) but to her: they “protect the goddess’s flanks in both sides” (utrimquesecus deae latera muniunt);13 their eyes seem to “threaten” (minantur) not Diana but her onlooker(s).14 This image of the spied-upon goddess, for all its secondary status as a mute, feminine, Greek image subject to the voluble, masculine, Latin text, indicates nonetheless the larger, ultimately ungovernable power accorded by the ancients to divine self-revelation. The inseparability of Greco-Roman deities from their plastic representations, noted especially in Chapters 4 and 7, suggests Diana’s elusive presence to the gazing Lucius already in Metamorphoses, Book 2. So here at the “end” of classical Latin literature, ecphrasis remains a richly evocative way of staging the Greek influence on Roman identity, while in this case subsuming both within the widely pervasive phenomenon of Isis-worship. Apuleius’s novel, then, makes an appropriate conclusion; yet our theme of Roman receptivity to Greek culture finds, paradoxically, another sort of end-point in an actual Greek text from near the finale of its own literary tradition (probably the early third century ce), the Eikones (Paintings) ascribed to the Elder Philostratus.15 A characteristic text of the Second Sophistic, the vastly popular literary and philosophical movement that sought to redefine Hellenism and the Greek legacy for the inhabitants of the Roman Empire, Eikones is anticipated by such works as Lucian’s ecphrastic De Domo and Essay on Images, the descriptive passages of Pausanias’s Description of Greece, and the numerous ecphrases of the Greek novels. Like Philostratus’s, such texts look back in turn to earlier Greek Isis cult himself (see above, n5) appeals to similar religious interests on the part of his audience (cf. Harrison 2000: 217 on cult initiation as a “by-product of elite educational travel”). For more on visuality and religious experience in Apuleius’s Metamorphoses, see Elsner 2007: 290–302. 12 A still closer connection exists between the Diana sculpture and the revelation of Isis at the end of the Metamorphoses if, as Peden 1985: 382 suggests, the decoration surrounding Diana in Byrrhena’s house has Isiac connotations. 13 Cf. van Mal-Maeder 2001: 101. 14 Slater 2008: 246. 15 Most modern scholars agree in viewing this work (Eikones or Imagines) as by the same Philostratus who wrote The Lives of the Sophists, the Life of Apollonius of Tyana, and Heroicus. Another book of Eikones is ascribed to the Younger Philostratus, who claims to be this author’s grandson. See Newby 2009: 322 for discussion and bibliography.

epilogue

248

literature, so that Eikones indexes both a Greek tradition running parallel to the Roman one and an innovative Greek style of ecphrasis emerging into special prominence at this time—even coming to overshadow its Roman counterpart in the eyes of posterity. For while the influence of the Latin ecphrases, albeit considerable, is limited chiefly to subsequent Western literature, Eikones, a collection of forty-eight self-standing prose ecphrases of paintings supposedly viewed in a picture-gallery near Naples, can be said to have furnished an essential impetus to the modern discipline of Art History and its related writing practices.16 And yet its pairing with Apuleius, a “Roman sophist” and author of a “sophist’s novel,” who conceived of his literary output in direct response to the activities of the Greek sophists, is apt.17 In some way a Roman phenomenon itself, though never mentioning Rome or the Romans, Eikones embodies an extreme, totalizing, and explicitly didactic version of the receptivity to Greek culture that informs the Latin texts studied in this book, and so will help set some of their characteristic features in high relief. Assuming the work’s ascription to the Elder Philostratus to be correct, we might have expected this author’s experience at Rome to have formed part of the narrative frame of Eikones, which situates itself almost immediately in Italy (1praef. 4). According to the Suda (s.v. Φιλόστρατος 421), Philostratus enjoyed a sophistic career in Athens and then in Rome. His arrival there (perhaps around 203–207 ce) is the most likely moment for his introduction to the emperor Septimius Severus, his wife Julia Domna, and the Hellenophile intellectual coterie that the Syrian-born Julia gathered around her at the Severan court. Philostratus refers to this milieu near the opening of his longest work, On Apollonius of Tyana.18 His connection with the Severans continued into the reign of Caracalla, in whose presence he witnessed the sophist Heliodorus plead for his country, an event about which we learn in another of his major works, the Lives of the Sophists.19 In this latter text, moreover, activity in both Athens and Rome is already an accepted feature of the sophistic life, as in the cases of Philagrus of Cilicia and Hadrianus of Tyre.20 By no means a figure confined 16

Until recently the scholarship on Eikones was focused largely on the question of whether or not it described real paintings. See Newby 2009: 322 with n5. For more recent approaches sensitive to the text as a rhetorical performance and as an ecphrasis, see Webb 2006; Newby 2009; Squire 2009: 339–56, 416–28; Platt 2011 passim (with emphasis on religious epiphany in the ecphrases; cf. above on Apuleius). Inaugurating “the tradition of independent prose descriptions of works of art,” Eikones stands as “a fundamental development of Kunstbeschreibung, in which the description of art finally became the supreme exemplar of ekphrasis as defined by the writers on rhetoric” (Elsner 2002: 14). 17 Although he apparently choose not to compete with them in their own language. See Harrison 2000, esp. 6, 210–59. 18 Philostr. VA 1.3. My reconstruction of Philostratus’s life at Rome depends on Bowie 2009: 20. 19 Philostr. VS 2.32.625–26. Heliodorus’s provenance is unknown; see Bowie 2009: 20n11. 20 Cf. Bowie 2009: 19 on the similarities with what we know of Philostratus’s own life.

Epilog ue

249

within the Hellenic world, Philostratus the star performer won the attentions of the Roman imperial court itself. Eikones, however, stages ecphrasis as a didactic performance by a Greek for Greeks—and this in an Italian city, Naples, described simply as home to men of “the Greek race,” who, as “people of refinement,” are “Greek in their zeal for discourse” (1praef. 4, ἡ δὲ πόλις ἐν Ἰταλία ᾤκισται γένος Ἕλληνες καὶ ἀστικοί, ὅθεν καὶ τὰς σπουδὰς τῶν λόγων Ἑλληνικοί εἰσι), as though the city’s population had remained essentially unchanged since its days as a Greek colony. Indeed, as a result, Philostratus says, of his reluctance to give addresses (μελέτας) in public, the young men of Naples seek him out in the home of his host. A suitable topic on which to speak presents itself in a portico housing a series of paintings set in its walls. But further impetus for this performance comes from his host’s ten-year-old son, whose curiosity about the images leads Philostratus to propose a display-speech (1praef. 5, ἐπίδειξιν) on them when the young men arrive. The boy, he says, will be placed in front, and the discourse will be addressed to him, with the young men following the speech as well and asking questions if anything isn’t clear. Thus the transmission of Greek culture from master to initiates proceeds, through “instructional speeches” [1praef. 3, ὁμιλίας] that are composed for the young and that will allow them to “interpret” paintings “and give heed to what is excellent” in them (1praef. 3, ἀφ’ ὧν ἑρμηνεύσουσί τε καὶ τοῦ δοκίμου ἐπιμελήσονται). Here as in all his major works, Philostratus seeks “to bring forward a more exclusive model of Hellenic culture than had been accepted before and to present this as the natural culture of his elite peers.”21 Out of all the descriptions in Eikones, many of whose paintings depict well known Greek myths and often stories found in Homer, we need look no further than the first, on an image of the river-god Scamander at Troy, to gain an appreciation of the collection’s rarefied Hellenism, and particularly the way its Roman-exclusive stance is allowed to inform Greek ecphrasis within a movement popular in the Roman world.22 Addressing his pupil, the boy, Philostratus first asks whether he has noticed that the image comes from Homer—or whether he has not recognized this through sheer wonder at the existence of fire in water (1.1.1, a reference to the flames Hephaestus used to combat Scamander when the latter threatened to engulf Achilles; cf. Hom. Il. 21.328–84). In pursuit of “what it means” (ὅ τι νοεῖ), Philostratus instructs the boy to look away from the image itself, so as “to see the things from which the painting comes” (ἐκεῖνα ἰδεῖν, ἀφ’ ὧν ἡ γραφή). Phantasia allows the boy to envision the mythical events themselves, which of course are only the product of, in this case, Homer’s text. Knowledge 21

Swain 2009: 34. On the Scamander image, I have benefitted especially from the discussions of Elsner 1995: 29–30; Webb 2006: 118–19; Newby 2009: 326–28. For more on the subtle use of Homer and other Greek texts in the Eikones, see Squire 2009: 339–56, 419–27. 22

epilogue

250

of Homer, the central Greek cultural text, is the index of correct appreciation for the painting, so much so that the image itself becomes a kind of distraction from which Philostratus’s young charge must momentarily avert his gaze.23 At issue is the Greek perception of Troy—not Vergil’s, not Rome’s. No detail could be more telling in this regard than the very first one on which Philostratus has the boy concentrate after returning his eyes to the painting: the lofty city of Troy with its great plain, an area sufficing “to range Asia in battle against Europe” (1.1.2, τὴν Ἀσίαν πρὸς τὴν Εὐρώπην ἀντιτάξαι). Troy as proto-Rome, a perpetual concern, as we have seen, of Roman authors, is here entirely absent, in its place Asian Troy, other to the European self. And at the core of this self is Greekness, its purity guaranteed by Homer. Tellingly again, Philostratus ends this opening ecphrasis of the Eikones with further insistence on the Homeric paradigm. Scamander, he notes, has had his hair burnt off, while Hephaestus is not lame but running, and the flames are not yellow (Ξανθός is the gods’ name for the Scamander river in Homer) nor “of the accustomed appearance” (τῇ εἰθισμένῃ ὄψει) but golden and beaming like the sun. “These things,” he concludes, “are no longer Homer’s” (ταῦτα οὐκέτι Ὁμήρου). The painting’s deviation from the text of the Iliad is duly noted even (and especially) as it is given pride of place at the opening of Philostratus’s text. Looking back from Philostratus’s Eikones through Apuleius’s Metamorphoses, we gain an especially clear perspective on two themes latent in this book’s seven chapters. On the one hand—the more obvious point—Eikones makes us see how far the Latin examples are from being exclusive appropriations of Greek culture for predominantly Hellenist purposes. At every moment in the descriptions of art by Naevius, Plautus, Catullus, and their Roman successors, Romanness is somehow at issue, albeit at times so mediated through Greekness as to make any dividing line between the two indeterminable. Roman ecphrasis represents an ongoing, never fully successful endeavor to articulate this relation, though lack of success itself takes on thematic import within the trope, so that one can see its subtle, compelling reiteration in author after author as its own form of success. After centuries, these descriptions still draw us seductively into their own hybrid world, one replete with mystery and fascination for the very fact of its being indeterminate. Indeed, my own decision to end with a Latin and a Greek author together can be seen as a large-scale acknowledgment of the ineluctable two-sidedness of my subject matter. Even settling upon a concluding view of the phenomenon involves sliding between its two poles. On the other hand, Philostratus’s emphasis on the labor of interpretation, the need to teach it and learn it, and the work of becoming receptive to “what 23

See esp. Newby 2009: 326–28 on this detail as indicative of the text’s larger oscillation between “absorption and erudition,” and with her argument cf. in particular Chapter 1 on the ambivalent narrator of Catullus 64.

Epilog ue

251

is excellent” in art, calls attention to Roman ecphrasis as acculturation into an ambiguously Greco-Roman identity. In the Latin authors such didacticism is implicit rather than explicit, and reflects the uses to which ecphrastic literature is put by a politically dominant society captivated by a colonized subject. As we have seen, these uses change greatly over time. Yet already in early Roman comedy, there is no need to say explicitly what Greek art is “good” for, or “what is excellent” within it. You know it or you don’t, and an individual Roman’s response is understood to reflect both the fact of Roman dominance over Greece and Greek culture’s transformative effects on Roman identity, as measured both by elite interests and popular perception. Even as received notions about Greek artists become more frequent in Roman discourse, the trick is still to mobilize them in novel, unexpected ways, a technique we have seen operative from Vergil’s Eclogues to the poetry of Statius and Martial. The “excellence” that such texts implicitly instruct their audiences to perceive in the Hellenic art object does not reach back to some putative set of unchanging Greek values, but rather serves both to guarantee and interrogate a larger Roman philhellenism that is also constantly in flux. I hope to have shown that Roman ecphrasis greatly problematizes the opposition between active and passive poles of reception, as it complicates the social, cultural, and political stakes of receptivity to Greek culture. It thus has the potential to make us, as modern readers, newly sensitive to the way in which we ourselves express our attitudes toward ancient Greece and to the implications of so doing—though this would be the topic of a different sort of book. Already in antiquity, “captive Greece” has a long and intriguing history as the object of its Roman captor’s gaze, never more so than when the positions of captor and captive are reversed or occluded, and the Hellenized Roman subject gazes upon a Greece somehow already Roman, or a Rome simply unthinkable without Greek images, stories, texts, ideas, emotions, habits, and practices. This is the situation with which Roman ecphrasis, through its interest in pleasure, understanding, and epiphany as much as tension, conflict, and dominance, frequently presents us. If it has become a truism to say that somewhere at the heart of the Roman self there was a Greek, more true to the Romans’ own experience, we have seen, is the ongoing effort to discern there the captor’s image.

This page intentionally left blank

BIBLIOGRAPHY

Acosta-Hughes, Benjamin. 2002. Polyeideia: The Iambi of Callimachus and the Archaic Iambic Tradition. Berkeley: University of California Press. Ahl, Frederick M. 1984a. “The Rider and the Horse: Politics and Power in Roman Poetry from Horace to Statius.” Aufstieg und Niedergang der römischen Welt 2.32.1: 40–110. ———. 1984b. “The Art of Safe Criticism in Greece and Rome.” American Journal of Philology 105: 174–208. Anderson, William S. 1969. The Art of the Aeneid. Englewood Cliffs, NJ: Prentice-Hall. ———, ed. 1972. Ovid’s Metamorphoses, Books 6–10. Norman: University of Oklahoma Press. ———. 1982. Essays on Roman Satire. Princeton: Princeton University Press. ———. 1993. Barbarian Play: Plautus’ Roman Comedy. Toronto: University of Toronto Press. André, Jacques. 1949. La vie et l’oeuvre d’Asinius Pollion. Paris: Klincksieck. Arkins, Brian. 1986. “New Approaches to Virgil.” Latomus 45: 33–42. Augoustakis, Antony, and Carole E. Newlands. 2007. “Introduction: Statius’s Silvae and the Poetics of Intimacy.” Arethusa 40: 117–25. Austin, R. G., ed. 1971. P. Vergili Maronis Aeneidos liber primus. Oxford: Clarendon Press. Balensiefen, Lilian. 1995. “Überlegungen zu Aufbau und Lage der Danaidenhalle auf dem Palatin.” Mitteilungen des deutschen archaeologischen Instituts, Roemische Abteilung 102: 189–209. Balot, Ryan K. 1998. “Pindar, Virgil, and the Proem to Georgic 3.” Phoenix 52: 83–94. Barchiesi, Alessandro. 1984. La traccia del modello. Effetti omerici nella narrazione virgiliana. Pisa: Giardini. ———. 1994. “Rappresentazioni del dolore e interpretatzione nell’Eneide.” Antike une Abendland 40: 109–24. ———. 1997. “Virgilian Narrative: Ecphrasis.” In Martindale, 271–81. Cambridge: Cambridge University Press. ———. 2005. “Learned Eyes: Poets, Viewers, Image Makers.” In Karl Galinsky, ed., The Cambridge Companion to the Age of Augustus, 281–305. Cambridge: Cambridge University Press. Barkan, Leonard. 1986. The Gods Made Flesh: Metamorphosis & The Pursuit of Paganism. New Haven: Yale University Press. Baratte, François. 1986. Le Trésor d’orfèvrerie romaine de Boscoreale. Paris: Éditions de la Réunion des musées nationaux. Barsby, John. 1999. Terence: Eunuchus. Cambridge: Cambridge University Press. Bartman, Elizabeth. 2010. “Sculptural Collecting and Display in the Private Realm.” In Elaine K. Gazda, ed., Roman Art in the Private Sphere: New Perspectives on the Architecture and Decor of the Domus, Villa, and Insula, 71–88. 2nd ed. Ann Arbor: University of Michigan Press.

253

254

bibliography

Bartsch, Shadi. 1998. “Ars and the Man: The Politics of Art in Virgil’s Aeneid.” Classical Philology 93: 322–42. ———. 2006. The Mirror of the Self: Sexuality, Self-Knowledge, and the Gaze in the Early Roman Empire. Chicago: University of Chicago Press. Barwick, Karl. 1958. “Zyklen bei Martial und in den kleinen Gedichten des Catull.” Philologus 102: 284–318. Bass, R. C. 1977. “Some Aspects of the Structure of the Phaethon Episode in Ovid’s Metamorphoses.” Classical Quarterly, n.s., 27: 402–08. Baxandall, Michael. 1985. Patterns of Intention: On the Historical Explanation of Pictures. New Haven: Yale University Press. Beard, Mary. 2007. The Roman Triumph. Cambridge: Harvard University Press. Beck, Deborah. 2007. “Ecphrasis, Interpretation, and Audience in Aeneid 1 and Odyssey 8.” American Journal of Philology 128: 533–49. Becker, Andrew Sprague. 1995. The Shield of Achilles and the Poetics of Ekphrasis. Lanham: Rowman & Littlefield. ———. 2003. “Contest or Concert? A Speculative Essay on Ecphrasis and the Rivalry between the Arts.” Classical and Modern Literature 23: 1–14. Becker, Carl. 1964. “Der Schild des Aeneas.” Wiener Studien 77: 111–27. Belloni, Luigi, Alice Bonandini, Giorgio Ieranò, and Gabriella Moretti, eds. 2010. Le Immagini nel testo, il testo nelle immagini: rapporti fra parola e visualità nella tradizione greco-latina. Labirinti, 128. Trento: Università degli Studi di Trento. Berg, William J. 1974. Early Virgil. London: Athlone Press. Bettini, Maurizio. 1999. The Portrait of the Lover. Translated by Laura Gibbs. Berkeley: University of California Press. Blänsdorf, Jürgen, ed. 2011. Fragmenta poetarum Latinorum, post W. Morel et K. Büchner. Berlin: Walter de Gruyter. Bloom, Harold. 1997. The Anxiety of Influence: A Theory of Poetry. 2nd ed. New York: Oxford University Press. Bodel, John. 1994. “Trimalchio’s Underworld.” In James Tatum, ed., The Search for the Ancient Novel, 237–59. Baltimore: The Johns Hopkins University Press. Boitani, Piero. 1989. “Antiquity and Beyond: The Death of Troilus.” In Piero Boitani, ed., The European Tragedy of Troilus. Oxford: Clarendon Press. Bömer, Franz. 1969–1986. P. Ovidius Naso: Metamorphosen. 7 vols. Heidelberg: Carl Winter. Bonadeo, Alessia. 2010. L’Hercules Epitrapezios Novi Vindicis: Introduzione e commento a Stat. silv. 4,6. Naples: Loffredo. Bosworth, A. B. 1972. “Asinius Pollio and Augustus.” Historia 21: 441–73. Boucher, Jean-Paul. 1980. Études sur Properce. Problèmes d’inspiration et d’art. 2nd ed. Paris: Éditions de Boccard. Bowditch, Lowell. 2009. “Palatine Apollo and the Imperial Gaze: Propertius 2.31 and 2.32.” American Journal of Philology 130: 401–38. Bowersock, Glen W. 1965. “Zur Geschichte des römischen Thessaliens.” Rheinisches Museum für Philologie 108: 277–89. ———. 1971. “A Date in the Eighth Eclogue.” Harvard Studies in Classical Philology 75: 73–80. Bowie, Ewen. 2009. “Philostratus: The Life of a Sophist.” In Bowie and Elsner, 19–32. Bowie, Ewen, and Jaś Elsner, eds. 2009. Philostratus. Cambridge: Cambridge University Press. Boyd, Barbara W. 1995. “Non enarrabile textum: Ecphrastic Trespass and Narrative Ambiguity in the Aeneid.” Vergilius 41: 71–90. Boyle, Anthony J. 1972. “The Meaning of the Aeneid: A Critical Inquiry, Part II. Homo Immemor: Book VI and its Thematic Ramifications.” Ramus 1: 113–51. ———. 1975. “A Reading of Virgil’s Eclogues.” Ramus 4: 187–203. ———. 1986. The Chaonian Dove: Studies in the Eclogues, Georgics, and Aeneid of Virgil. Leiden: Brill.

Bibli og raphy

255

Boyle, Anthony J., and J. L. Penwill, eds. 2004. Rethinking Terence. Ramus 33. Braun, Maximilian. 2000. “Mos maiorum und humanitas bei Terenz.” In Maximilian Braun, Andreas Haltenhoff, and Fritz-Heiner Mutschler, eds., Moribus antiquis res stat Romana, 205–15. Beiträge zur Altertumskunde 134. Munich: K. G. Saur. Braund, Susanna M. 1996. “Ending Epic: Statius, Theseus and a Merciful Release.” Proceedings of the Cambridge Philological Society 42: 1–23. Breed, Brian W. 2006. Pastoral Inscriptions: Reading and Writing Virgil’s Eclogues. London: Duckworth. Breen, Carolyn C. 1986. “The Shield of Turnus, the Swordbelt of Pallas and the Wolf.” Vergilius 32: 63–71. Bretzigheimer, Gerlinde. 1993. “Jupiter Tonans in Ovids Metamorphosen.” Gymnasium 100: 19–74. Brown, P. Michael, ed. and trans. 1993. Horace, Satires I. Warminster: Aris & Phillips. Brown, Robert. 1987. “The Palace of the Sun in Ovid’s Metamorphoses.” In Michael Whitby, Philip Hardie, and Mary Whitby, eds., Homo Viator: Classical Essays for John Bramble, 211–20. Bristol: Bristol Classical Press. Browning, Robert. 2002. “Greeks and Others: From Antiquity to the Renaissance.” In Thomas Harrison, ed., Greeks and Barbarians, 257–77. New York: Routledge. Bryson, Norman. 1981. Word and Image: French Painting of the Ancien Régime. Cambridge: Cambridge University Press. Buchheit, Vinzenz. 1972. Der Anspruch des Dichters in Vergils Georgika: Dichtertum und Heilsweg. Darmstadt: Wissenschaftliche Buchgesellschaft. Buecheler, Franz. 1922. Petronii saturae et liber priapeorum. Sixth edition revised by Wilhelm Heraeus. Berlin: Weidmann. Cairns, Francis. 2006. Sextus Propertius: The Augustan Elegist. Cambridge: Cambridge University Press. Cancik, Hubert. 1965. Untersuchungen zur lyrischen Kunst des P. Papinius Statius. Spudasmata 13. Hildesheim: Georg Olms. Candilio, Daniela. 1989. “Nero antico.” In Maxwell L. Anderson and Leila Nista, eds., Radiance in Stone: Sculptures in Colored Marble from the Museo Nazionale Romano, 85–91. Rome: De Luca Edizioni d’Arte. Carey, Sorcha. 2003. Pliny’s Catalogue of Culture: Art and Empire in the Natural History. Oxford: Oxford University Press. Casali, Sergio. 1995. “Aeneas and the Doors of the Temple of Apollo.” Classical Journal 91: 1–9. Cassio, A. C. 1973. “L’incipit della Cioma Callimachea in Virgilio.” Rivista di Filologia e di Istruzione Classica 101: 329–32. Caston, Ruth R. 2003. “Rivaling the Shield: Propertius 4.6.” In Thibodeau and Haskell, 145–62. Afton: Afton Historical Society Press. Chinn, Christopher. 2005. “Statius Silv. 4.6 and the Epigrammatic Origins of Ekphrasis.” Classical Journal 100: 247–63. Christenson, David M. 2000. Plautus: Amphitruo. Cambridge: Cambridge University Press. Clare, Ray J. 1996. “Catullus 64 and the Argonautica of Apollonius Rhodius: Allusion and Exemplarity.” Proceedings of the Cambridge Philological Society 42: 60–88. Clarke, John R. 1991. The Houses of Roman Italy, 100 B.C.—A.D. 250: Ritual, Space, and Decoration. Berkeley: University of California Press. Clausen, Wendell V. 1982. “The New Direction in Poetry.” In E. J. Kenney and W. V. Clausen, eds., The Cambridge History of Classical Literature. Vol. 2: Latin Literature, 178–206. Cambridge: Cambridge University Press. ———. 1994. A Commentary on Virgil, Eclogues. Oxford: Clarendon Press. Clay, Diskin. 1988. “The Archaeology of the Temple to Juno in Carthage (Aen. 1.446–93).” Classical Phililogy 83: 195–205. ———. 1998. Paradosis and Survival: Three Chapters in the History of Epicurean Philosophy. Ann Arbor: University of Michigan Press.

256

bibliography

Coleman, Kathleen M. 1986. “The Emperor Domitian and Literature.” Aufstieg und Niedergang der römischen Welt 2.32.5: 3087–3115. ———. 1988. Statius: Silvae IV. Oxford: Clarendon Press. Coleman, Robert, ed. 1977. Vergil: Eclogues. Cambridge: Cambridge University Press. Collignon, Albert. 1892. Étude sur Pétrone. Paris: Hachette. Connors, Catherine. 1998. Petronius the Poet: Verse and Literary Tradition in the Satyricon. Cambridge: Cambridge University Press. Conte, Gian Biagio. 1986. The Rhetoric of Imitation: Genre and Poetic Memory in Virgil and Other Latin Poets. Translated and edited by Charles Segal. Ithaca: Cornell University Press. ———. 1996. The Hidden Author: An Interpretation of Petronius’ Satyricon. Translated by Elaine Fantham. Berkeley: University of California Press. Courtney, Edward. 1966. “On the Silvae of Statius.” Bulletin of the Institute of Classical Studies 13: 94–100. ———. 1984. “Criticisms and Elucidations of the Silvae of Statius.” Transactions of the American Philological Association 114: 327–41. ———. 1987. “Petronius and the Underworld.” American Journal of Philology 108: 408–10. ———, ed. 1990. P. Papini Stati Silvae. Oxford: Clarendon Press. ———, ed. 1993. The Fragmentary Latin Poets. Oxford: Clarendon Press. Culler, Jonathan. 1977. “Apostrophe.” Diacritics 7.4: 59–69. Curran, Leo C. 1969. “Catullus 64 and the Heroic Age.” Yale Classical Studies 21: 171–92. Currie, Harry MacL. 1976. “The Third Eclogue and the Roman Comic Spirit.” Mnemosyne 29: 411–20. Cuvigny, Hélène. 2002. “Silver Celestial Globe from Antiquity.” In Alexis Kugel, ed., with the collaboration of K. Van Cleempoel and J.-C. Sabrier, Spheres: The Art of the Celestial Mechanic, 22–26. Paris: J. Kugel. Darwall-Smith, Robin H. 1996. Emperors and Architecture: A Study of Flavian Rome. Collection Latomus 231. Brussels: Latomus. DeBrohun, Jeri Blair. 1999. “Ariadne and the Whirlwind of Fate: Figures of Confusion in Catullus 64.149–57.” Classical Philology 94: 419–30. Dekel, Edan. 2012. Virgil’s Homeric Lens. New York: Routledge. Dessen, Cynthia S. 1995. “The Figure of the Eunuch in Terence’s Eunuchus.” Helios 22: 123–39. Dewar, Michael, ed. and tr. 1991. Statius: Thebaid IX. Oxford: Clarendon Press. ———. 2008. “The Equine Cuckoo: Statius’ Ecus Maximus Domitiani Imperatoris and the Flavian Forum.” In Johannes J. L. Smolenaars, Harm-Jan van Dam, and Ruurd R. Nauta, eds., The Poetry of Statius, 65–83. Mnemosyne Supplementum 306. Leiden: Brill. Dickie, Matthew W. 2001. Magic and Magicians in the Greco-Roman World. London: Routledge. Dix, T. Keith. 1995. “Vergil in the Grynean Grove: Two Riddles in the Third Eclogue.” Classical Philology 90: 256–62. Doblhofer, Ernst. 1973. “Ovid—ein ‘Urvater der Résistance’?” In Wilhelm Danhofer, ed., 400 Jahre Akademischen Gymnasiums in Graz, 1573–1973. Festschift. 143–54. Graz: Verlag des Akademisches Gymnasium in Graz. Downey, Glanville. 1959. “Ekphrasis.” In Theodor Klauser, ed., Reallexikon für Antike und Christentum 4.921–44. Anton Hiersemann: Stuttgart. duBois, Page. 1982. History, Rhetorical Description, and the Epic: From Homer to Spencer. Cambridge: D. S. Brewer. Dufallo, Basil. 2000. “Satis/satura: Reconsidering the ‘Programmatic Intent’ of Horace’s Satires 1.1.” Classical World 93: 579–90. ———. 2007. The Ghosts of the Past: Latin Literature, the Dead, and Rome’s Transition to a Principate. Columbus: The Ohio State University Press. DuQuesnay, I. M. Le M. 1984. “Horace and Maecenas: The Propaganda Value of Sermones I.” In T. Woodman and D. West, eds., Poetry and Politics in the Age of Augustus, 19–58. Cambridge: Cambridge University Press.

Bibli og raphy

257

Edmunds, Lowell. 2002. “Mars as Hellenistic Lover: Lucretius, De rerum natura 1.29–40 and Its Subtexts.” International Journal of the Classical Tradition 8: 343–58. ———. 2009. “Horace’s Priapus: A Life on the Esquiline (Sat. 1.8).” Classical Quarterly 59: 125–31. Elsner, Jaś. 1995. Art and the Roman Viewer: The Transformation of Art from the Pagan World to Christianity. Cambridge: Cambridge University Press. ———, ed. 1996. Art and Text in Roman Culture. Cambridge: Cambridge University Press. ———. 2002. “Introduction: The Genres of Ekphrasis.” In Jaś Elsner, ed., The Verbal and the Visual: Cultures of Ekphrasis in Antiquity, 1–18. Ramus 31. ———. 2004. “Seeing and Saying: A Psychoanalytic Account of Ekphrasis.” Helios 31: 157–85. ———. 2007. Roman Eyes: Visuality and Subjectivity in Art and Text. Princeton: Princeton University Press. ———. 2009. “A Protean Corpus.” In Bowie and Elsner, 3–18. Cambridge: Cambridge University Press. Erskine, Andrew. 2001. Troy between Greece and Rome: Local Tradition and Imperial Power. Oxford: Oxford University Press. Evans, Jane DeRose. 2009. “Prostitutes in the Portico of Pompey? A Reconsideration.” Transactions of the American Philological Association 139: 123–45. Evenepoel, Willy. 1990. “Maecenas: A Survey of Recent Literature.” Ancient Society 21: 99–117. Faber, Riemer. 1995. “Vergil Eclogue 3.37, Theocritus 1 and Hellenistic Ekphrasis.” American Journal of Philology 116: 411–17. ———. 2000. “Vergil’s ‘Shield of Aeneas’ (Aeneid 8. 617–731) and the Shield of Heracles.” Mnemosyne 53: 49–57. Fabbrini, Delphina. 2007. Il migliore dei mondi possibili: gli epigrammi ecfrastici di Marziale per amici e protettori. Florence: Università degli studi di Firenze. Fantham, Elaine. 1997. “Images of the City: Propertius’ New-Old Rome.” In Thomas Habinek and Alessandro Schiesaro, eds., The Roman Cultural Revolution, 122–35. Cambridge: Cambridge University Press. ———. 2004. “Terence and the Familiarisation of Comedy.” In Boyle and Penwill, 20–34. Fantuzzi, Marco, and Richard Hunter. 2004. Tradition and Innovation in Hellenistic Poetry. Cambridge: Cambridge University Press. Farrell, Joseph. 1991. Vergil’s Georgics and the Traditions of Ancient Epic: The Art of Allusion in Literary History. New York: Oxford University Press. ———. 1992. “Literary Allusion and Cultural Poetics in Vergil’s Third Eclogue.” Vergilius 38: 64–71. ———. 1997. “The Virgilian Intertext.” In Martindale, 222–38. Cambridge: Cambridge University Press. Fedeli, Paulo. 2005. Properzio: Elegie Libro II. Cambridge: Francis Cairns. Feeney, Dennis C. 1988. Literature and Religion at Rome: Cultures, Contexts, and Beliefs. Cambridge: Cambridge University Press. ———. 1991. The Gods in Epic: Poets and Critics of the Classical Tradition. Oxford: Clarendon Press. Feldherr, Andrew. 2010. Playing Gods: Ovid’s Metamorphoses and the Politics of Fiction. Princeton: Princeton University Press. Fitzgerald, William. 1984. “Aeneas, Daedalus and the Labyrinth.” Arethusa 17: 51–65. ———. 1995. Catullan Provocations: Lyric Poetry and the Drama of Position. Berkeley: University of California Press. ———. 2007. Martial: The World of the Epigram. Chicago: University of Chicago Press. Flower, Harriet I. 1996. Ancestor Masks and Aristocratic Power in Roman Culture. Oxford: Clarendon Press. Fowler, Don P. 1991. “Narrate and Describe: The Problem of Ekphrasis.” Journal of Roman Studies 81: 25–35.

258

bibliography

———. 2002. Lucretius on Atomic Motion: A Commentary on De rerum natura, Book Two, Lines 1–332. Oxford: Oxford University Press. Fraenkel, Eduard. 1954. “The Giants in the Poem of Naevius.” Journal of Roman Studies 44: 14–17. ———. 1957. Horace. Oxford: Clarendon Press. Francis, James A. 2009. “Metal Maidens, Achilles’ Shield, and Pandora: The Beginnings of ‘Ekphrasis.’” American Journal of Philology 130: 1–23. Fränkel, Hermann. 1935. “Griechische Bildung in altrömischen Epen II.” Hermes 70: 59–72. Fredrick, David. 1995. “Beyond the Atrium to Ariadne: Erotic Painting and Visual Pleasure in the Roman House.” Classical Antiquity 14: 266–87. ———, ed. 2002. The Roman Gaze: Vision, Power, and the Body. Baltimore: Johns Hopkins. Freudenburg, Kirk. 1993. The Walking Muse: Horace on the Theory of Satire. Princeton: Princeton University Press. ———. 2001. Satires of Rome: Threatening Poses from Lucilius to Juvenal. Cambridge: Cambridge University Press. ———. 2007. “Leering for the Plot: Visual Curiosity in Apuleius and Others.” In Paschalis, Frangoulidis, Harrison, and Zimmerman, 238–62. Groningen: Barkhuis & Groningen University Library. Friedländer, Ludwig, ed. 1886. M. Valerii Martialis epigrammaton libri. 2 vols. Leipzig: Hirzel. Friedländer, Paul. 1912. Johannes von Gaza und Paulus Silentiarius: Kunstbeschreibungen justinianischer Zeit. Leipzig: Teubner. Frischer, Bernard. 1982. The Sculpted Word: Epicureanism and Philosophical Recruitment in Ancient Greece. Berkeley: University of California Press. Gaisser, Julia H. 1995. “Threads in the Labyrinth: Competing Views and Voices in Catullus 64.” American Journal of Philology 116: 579–616. Gale, Monica R. 1994. Myth and Poetry in Lucretius. Cambridge: Cambridge University Press. ———. 1997. “The Shield of Turnus (Aeneid 7.783–92).” Greece & Rome 44: 176–96. ———. 2000. Virgil on the Nature of Things: The Georgics, Lucretius and the Didactic Tradition. Cambridge: Cambridge University Press. Galinsky, G. Karl. 1972. The Herakles Theme: The Adaptations of the Hero in Literature from Homer to the Twentieth Century. Oxford: Basil Blackwell. ———. 1975. Ovid’s Metamorphoses: An Introduction to the Basic Aspects. Berkeley: University of California Press. ———. 1996. Augustan Culture: An Interpretive Introduction. Princeton: Princeton University Press. Gardner, Hunter H. 2007. “Ariadne’s Lament: The Semiotic Impulse of Catullus 64.” Transactions of the American Philological Association 137: 147–79. Garthwaite, John. 1978. Domitian and the Court Poets Martial and Statius. PhD Diss. Cornell. ———. 1993. “The Panegyrics of Domitian in Martial Book 9.” Ramus 22: 78–102. Garvie, A. F., ed. 1994. Homer: Odyssey, Books VI–VIII. Cambridge: Cambridge University Press. Gazda, Elaine K., ed. 2000. The Villa of the Mysteries in Pompeii: Ancient Ritual, Modern Muse. Ann Arbor: Kelsey Museum of Archaeology, University of Michigan Museum of Art. Germany, Robert. 2008. Mimetic Contagion in Terence’s Eunuchus. PhD Diss. University of Chicago. Geyssen, John W. 1996. Imperial Panegyric in Statius: A Literary Commentary on Silvae 1.1. Studies on Themes and Motifs in Literature 24. New York: Peter Lang. Gilbert, Walther, ed. 1901. M. Valerii Martialis epigrammaton libri. Editio stereotypa emendatior. Leipzig: Teubner. Goldberg, Sander M. 1986. Understanding Terence. Princeton: Princeton University Press. ———. 1995. Epic in Republican Rome. Oxford: Oxford University Press.

Bibli og raphy

259

———. 2005. Constructing Literature in the Roman Republic: Poetry and Its Reception. Cambridge: Cambridge University Press. Goldhill, Simon. 1994. “The Naive and Knowing Eye: Ecphrasis and the Culture of Viewing in the Hellenistic World.” In Goldhill and Osborne, 197–223. ———. 2007. “What Is Ekphrasis For?” Classical Philology 102: 1–19. ———. 2009. Review of Webb, Bryn Mawr Classical Review 10.03. http://bmcr.brynmawr. edu/2009/2009-10-03.html. Goldhill, Simon, and Robin Osborne, eds. 1994. Art and Text in Ancient Greek Culture. Cambridge: Cambridge University Press. Gow, A. S. F. 1952. Theocritus. 2nd ed. 2 vols. Cambridge: Cambridge University Press. Gow, A. S. F., and D. L. Page, eds. 1965. The Greek Anthology: Hellenistic Epigrams. Vol. 2. Cambridge: Cambridge University Press. ———, eds. 1968. The Greek Anthology: The Garland of Philip and Some Contemporary Epigrams. 2 vols. Cambridge: Cambridge University Press. Gowers, Emily. 1993. “Horace, Satires 1.5: An Inconsequential Journey.” Proceedings of the Cambridge Philological Society 39: 48–66. ———. 2003. “Fragments of Autobiography in Horace, Satires 1.” Classical Antiquity 22: 55–91. Gransden, K. W. 1984. Virgil’s Iliad: An Essay on Epic Narrative. Cambridge: Cambridge University Press. Gratwick, A. S., ed. 1993. Plautus: Menaechmi. Cambridge: Cambridge University Press. Grewing, Farouk. 1997. Martial, Buch VI: Ein Kommentar. Hypomnemata 115. Göttingen: Vandenhoeck & Ruprecht. Graziani, Françoise. 2003. “Materiam superabat opus: Un art poétique ovidien.” In Emmanuel Bury and Mireille Néraudau, eds., Lectures d’Ovide publiées à la mémoire de Jean-Pierre Néraudau, 339–55. Paris: Les Belles Lettres. Griffiths, Frederick T. 1979. Theocritus at Court. Mnemosyne Supplementum 55. Leiden: Brill. Grossardt, Peter. 2009. “Die ‘Cena Trimalchionis’ gelesen als Parodie auf die ‘Ilias.’” Hermes 137: 335–55. Gruen, Erich S. 1992. Culture and National Identity in Republican Rome. Ithaca: Cornell University Press. ———. 2011. Rethinking the Other in Antiquity. Princeton: Princeton University Press. Grüner, Andreas. 2004. Venus ordinis: Der Wandel von Malerei und Literatur im Zeitalter der römischen Bürgerkriege. Paderborn: Schöningh. Gurd, Sean Alexander. 2007. “Meaning and Material Presence: Four Epigrams on Timomachus’s Unfinished Medea.” Transactions of the American Philological Association 137: 305–31. Gurval, Robert A. 1995. Actium and Augustus: The Politics and Emotions of Civil War. Ann Arbor: University of Michigan Press. Guthmüller, Hans-Bodo. 1964. Beobachtungen zum Aufbau der Metamorphosen Ovids. Diss. Marburg. Gutzwiller, Kathryn J. 2004. “Seeing Thought: Timomachus’ Medea and Ecphrastic Epigram.” American Journal of Philology 125: 339–86. Habash, Martha. 1999. “Priapus: Horace in Disguise?” Classical Journal 94: 285–97. Habermehl, Peter. 2006. Petronius, Satyrica 79–141. Ein philologisch-literarischer Kommentar. Vol. 1: Sat. 79–110. Berlin: de Gruyter. Habinek, Thomas N. 1998. The Politics of Latin Literature: Writing, Identity, and Empire in Ancient Rome. Princeton: Princeton University Press. Håkanson, Lennart. 1969. Statius’ Silvae: Critical and Exegetical Remarks with Some Notes on the Thebaid. Lund: Gleerup. Hallett, Judith P. 1981. “Pepedi/diffissa nate ficus: Priapic Revenge in Horace Satires I.8.” Rheinisches Museum für Philologie 124: 341–47.

260

bibliography

Hardie, Alex. 1983. Statius and the Silvae: Poets, Patrons and Epideixis in the Graeco-Roman World. ARCA Classical and Medieval Texts, Papers and Monographs 9. Liverpool: Francis Cairns. Hardie, Philip R. 1986. Virgil’s Aeneid: Cosmos and Imperium. Oxford: Clarendon Press. ———. 1990. “Ovid’s Theban History: The First ‘anti-Aeneid’?” Classical Quarterly 40: 224–35. ———. 1992. “Augustan Poets and the Mutability of Rome.” In Powell, 59–82. ———. 1993. The Epic Successors of Virgil: A Study in the Dynamics of a Tradition. Cambridge: Cambridge University Press. ———. 2002. Ovid’s Poetics of Illusion. Cambridge: Cambridge University Press. Hardie, Philip, Alessandro Barchiesi, and Stephen Hinds, eds. 1999. Ovidian Transformations: Essays on the Metamorphoses and Its Reception. Cambridge: Cambridge Philological Society. Hardwick, Lorna, and Christopher Stray. 2008. “Introduction: Making Connections.” In Lorna Hardwick and Christopher Stray, eds., A Companion to Classical Receptions, 1–9. Malden: Blackwell. Harries, Byron. 1990. “The Spinner and the Poet: Arachne in Ovid’s Metamorphoses.” Proceedings of the Cambridge Philological Society, n.s., 36: 64–82. Harrison, Steven J., ed. 1990. Oxford Readings in Vergil’s Aeneid. Oxford: Oxford University Press. ———. 2000. Apuleius: A Latin Sophist. Oxford: Oxford University Press. Häuber, Ruth Christine. 1991. Die Horti Maecenatis und die Horti Lamiani auf dem Esquilin. Geschichte, Topographie, Statuenfunde. Diss. Köln. Heffernan, James A. W. 1993. Museum of Words: The Poetics of Ekphrasis from Homer to Ashbery. Chicago: University of Chicago Press. Henderson, John. 1998. “Virgil’s Third Eclogue: How Do You Keep an Idiot in Suspense?” Classical Quarterly 48: 213–28. ———. 2002. Pliny’s Statue: The Letters, Self-Portraiture & Classical Art. Exeter: University of Exeter Press. Henriksén, Christer. 1998. “Martial und Statius.” In Farouk Grewing, ed., Toto Notus in Orbe: Perspektiven der Martial-Interpretation. Palingenesia 65, 77–118. Stuttgart: Franz Steiner. ———. 1998–1999. Martial, Book IX: A Commentary. 2 vols. Studia Latina Upsaliensia 24. Uppsala: Uppsala University Library. Heraeus, W., ed. 1976. M. Valerii Martialis epigrammaton libri. Corr. ed., Iacobus Borovskij. Leipzig: Teubner. Heyworth, S. J., 1994. “Some Allusions to Callimachus in Latin Poetry.” Materiali e discussioni per l’analisi dei testi classici 33: 51–79. ———, ed. 2007a. Sexti Properti elegos. Oxford: Clarendon Press. ———. 2007b. Cynthia: A Companion to the Text of Propertius. Oxford: Oxford University Press. Holford-Strevens, Leofranc. 2000. “In Search of Poplios Papinios Statios.” Hermathena 168: 39–54. Hollander, John. 1995. The Gazer’s Spirit: Poems Speaking to Silent Works of Art. Chicago: University of Chicago Press. Holliday, Peter J. 1997. “Roman Triumphal Painting: Its Function, Development, and Reception.” Art Bulletin 79: 130–47. Hollis, Adrian S., ed. and trans. 2007. Fragments of Roman Poetry c.60 BC—AD 20. Oxford: Oxford University Press. Hopkinson, Neil, ed. 2000. Ovid: Metamorphoses, Book XIII. Cambridge: Cambridge University Press. Horsfall, Nicholas M. 1990 (1973–74). “Dido in the Light of History.” In Harrison, 127–44. Oxford: Oxford University Press. Housman, A. E. 1907. “Corrections and Explanations of Martial.” Journal of Philology 30: 229–65.

Bibli og raphy

261

Hubbard, Thomas K. 1984. “Art and Vision in Propertius 2.31/32.” Transactions of the American Philological Association 114: 281–97. ———. 1998. The Pipes of Pan: Intertextuality and Literary Filiation in the Pastoral Tradition from Theocritus to Milton. Ann Arbor: University of Michigan Press. Hunter, Richard. 1996. Theocritus and the Archaeology of Greek Poetry. Cambridge: Cambridge University Press. ———, ed. 1999. Theocritus: A Selection. Cambridge: Cambridge University Press. ———. 2006. The Shadow of Callimachus: Studies in the Reception of Hellenistic Poetry at Rome. Cambridge: Cambridge University Press. Hutchinson, G. O. 2003. “The Catullan Corpus, Greek Epigram, and the Poetry of Objects.” Classical Quarterly 53: 206–21. Isager, Jacob. 1991. Pliny on Art and Society: The Elder Pliny’s Chapters on the History of Art. London: Routledge. Jaeger, Mary. 2008. Archimedes and the Roman Imagination. Ann Arbor: University of Michigan Press. James, Sharon L. 1998. “From Boys to Men: Rape and Developing Masculinity in Terence’s Hecyra and Eunuchus.” Helios 25: 31–47. Janan, Micaela. 1994. “When the Lamp Is Shattered”: Desire and Narrative in Catullus. Carbondale: Southern Illinois University Press. ———. 2009. Reflections in a Serpent’s Eye: Thebes in Ovid’s Metamorphoses. Oxford: Oxford University Press. Jensson, Gottskálk. 2004. The Recollections of Encolpius: The Satyrica of Petronius as Milesian Fiction. Ancient Narrative, Supplementum 2. Groningen: Barkuis Publishing & Groningen University Library. Johnson, Patricia J. 2008. Ovid before Exile: Art and Punishment in the Metamorphoses. Madison: University of Wisconsin Press. Johnson, William R. 1976. Darkness Visible: A Study of Vergil’s Aeneid. Berkeley: University of California Press. Jones, Brian W. 1992. The Emperor Domitian. London: Routledge. Karakasis, Evangelos. 2005. Terence and the Language of Roman Comedy. Cambridge: Cambridge University Press. ———. 2011. Song Exchange in Roman Pastoral. Trends in Classics—Supplementary Volumes 5. Berlin: De Gruyter. Karanika, Andromache. 2006. “Agonistic Poetics in Virgil’s Third Eclogue.” In M. Skoie and S. Bjørnstad Velázquez, eds., Pastoral and the Humanities: Arcadia Re-inscribed, 107–14. Exeter: Bristol Phoenix Press. Keane, Catherine. 2006. Figuring Genre in Roman Satire. American Classical Studies 50. Oxford: Oxford University Press. Keith, Alison M. 2000. Engendering Rome: Women in Latin Epic. Cambridge: Cambridge University Press. ———. 2008. Propertius: Poet of Love and Leisure. London: Duckworth. Kellum, Barbara. 1993 (1986). “Sculptural Programs and Propaganda in Augustan Rome: The Temple of Apollo on the Palatine.” In Eve D’Ambra, ed., Roman Art in Context: An Anthology, 75–83. Upper Saddle River: Prentice Hall. Kennedy, Duncan F. 1992. “‘Augustan’ and ‘Anti-Augustan’: Reflections on Terms of Reference.” In Powell, 26–58. London: Bristol Classical Press. Kershaw, Allan. 1997. “Martial 9.44 and Statius.” Classical Philology 92: 269–72. Ketterer, Robert C. 1986. “Stage Properties in Plautine Comedy III.” Semiotica 60: 29–72. Kirichenko, Alexander. 2010. A Comedy of Storytelling: Theatricality and Narrative in Apuleius’ Golden Ass. Bibliothek der klassischen Altertumswissenschaften, Neue Folge 2.127. Heidelberg: Universitätsverlag Winter.

262

bibliography

Klingner, Friedrich. 1956. Catulls Peleus-Epos. Munich: Verlag der Bayerischen Akademie der Wissenschaften. Klodt, Claudia. 1998. “Platzanlagen der Kaiser in der Beschreibung der Dichter.” Gymnasium 105: 1–38. Knapp, Charles. 1917. “References to Painting in Plautus and Terence.” Classical Philology 12: 143–57. Knauer, Georg N. 1964. Die Aeneis und Homer. Göttingen: Vandenhoeck & Ruprecht. Knorr, Ortwin. 2007. “Metatheatrical Humor in the Comedies of Terence.” In Kruschwitz, Ehlers, and Felgentreu, 167–74. Konstan, David. 1977. Catullus’ Indictment of Rome: The Meaning of Catullus 64. Amsterdam: Hakkert. ———. 1986. “Love in Terence’s Eunuch: The Origins of Erotic Subjectivity.” American Journal of Philology 107: 369–93. Kraggerud, Egil. 1998. “Vergil Announcing the Aeneid: On Georg. 3.1–48.” In Hans-Peter Stahl, ed., Vergil’s Aeneid: Augustan Epic and Political Context, 1–20. London: Duckworth. Krieger, Murray. 1992. Ekphrasis: The Illusion of the Natural Sign. Baltimore: Johns Hopkins University Press. Kroll, Wilhelm, ed. 1959. C. Valerius Catullus. 3rd ed. Stuttgart: Teubner. Kruschwitz, Peter. 2002. Carmina Saturnia Epigraphica. Hermes Einzelschriften 84. Stuttgart: Franz Steiner. Kruschwitz, Peter, Widu-Wolfgang Ehlers, and Fritz Felgentreu, eds. 2007. Terentius Poeta. Zetemata 127. Munich: C. H. Beck. Künzl, Ernst, Maiken Fecht, and Susanne Greiff. 2000. “Ein Römischer Himmelsglobus der Mittleren Kaiserzeit.” Jarbuch des Römisch-Germanischen Zentralmuseums Mainz 47: 495–594. Kustas, George L. 1973. Studies in Byzantine Rhetoric. Thessaloniki: Patriarchal Institute for Patristic Studies. Kuttner, Ann L. 1999. “Culture and History at Pompey’s Museum.” Transactions of the American Philological Association 129: 343–73. Laird, Andrew. 1993. “Sounding out Ecphrasis: Art and Text in Catullus 64.” Journal of Roman Studies 83: 18–30. ———. 1996. “Ut figura poesis: Writing Art and the Art of Writing in Augustan Poetry.” In Elsner 1996, 75–102. Cambridge: Cambridge University Press. ———. 2007. “The True Nature of the Satyricon?” In Paschalis et al., 151–67. Groningen: Barkhuis & Groningen University Library. Lausberg, Marion. 1982. “Ἀρχέτυπον τῆς ἰδίας ποιήσεως. Zur Bildbeschreibung bei Ovid.” Boreas 5: 112–23. Leach, Eleanor W. 1969. “Meam quom formam noscito: Language and Characterization in the Menaechmi.” Arethusa 2: 30–45. ———. 1974a. “Ekphrasis and the Theme of Artistic Failure in Ovid’s Metamorphoses.” Ramus 3: 102–42. ———. 1974b. Vergil’s Eclogues: Landscapes of Experience. Ithaca: Cornell University Press. ———. 1988. The Rhetoric of Space: Literary and Artistic Representations of Landscape in Republican and Augustan Rome. Princeton: Princeton University Press. ———. 2004. The Social Life of Painting in Ancient Rome and on the Bay of Naples. Cambridge: Cambridge University Press. Leberl, Jens. 2004. Domitian und die Dichter: Poesie als Medium der Herrschaftsdarstellung. Hypomnemata 154. Göttingen: Vandenhoeck & Ruprecht. Leclerq, R. 1996. Le divin loisir: Essai sur les Bucoliques de Virgile. Collection Latomus 229. Brussels: Latomus. Lefèvre, Eckard. 1989. Das Bild-Programm des Apollo-Tempels auf dem Palatin. Xenia 24. Konstanz: Universitätsverlag Konstanz.

Bibli og raphy

263

———. 2003. Terenz’ und Menanders Eunuchus. Zetemata 117. Munich: Beck. Leigh, Matthew. 2004. Comedy and the Rise of Rome. Oxford: Oxford University Press. Lessing, Gotthold E. 2005 (1898). Laocoon: An Essay upon the Limits of Painting and Poetry. Translated by Ellen Frothingham. Mineola: Dover Publications. Lindsay, W. M. 1929. M. Val. Martialis epigrammata. 2nd ed. Oxford: Clarendon Press. Linfert, Andreas. 1984. “Die Tochter—nicht die Mutter. Nochmals zur ‘Afrika’-Schale von Boscoreale.” In Nicola Bonacasa and Antonino di Vita, eds., Alessandria e il mondo Ellenistico-Romano: Studi in onore di Achille Adriani, 2.351–58. Rome: “L’Erma” di Bretschneider. Lipka, Michael. 2001. Language in Vergil’s Eclogues. Untersuchungen zur antiken Literatur und Geschichte 60. Berlin: Walter de Gruyter. ———. 2002. “Notes on Fagus in Vergil’s Eclogues.” Philologus 146: 133–38. Longfellow, Brenda. 2000. “Liber and Venus in the Villa of the Mysteries.” In Gazda, 116–28. Lorenz, Sven. 2002. Erotik und Panegyrik: Martials epigrammatische Kaiser. Classica Moncensia 23. Tübingen: Gunter Naar. ———. 2003. “Martial, Herkules und Domitian: Büsten, Statuetten und Statuen im Epigrammaton liber nonus.” Mnemosyne 56: 566–84. Lowenstam, Steven. 1993. “The Pictures on Juno’s Temple in the Aeneid.” Classical World 87: 37–49. Lowrie, Michèle. 2009. Writing, Performance, and Authority in Augustan Rome. Oxford: Oxford University Press. Manakidou, Flora. 1993. Beschreibung von Kunstwerken in der hellenistischen Dichtung: Ein Beitrag zur hellenistischen Poetik. Stuttgart: Teubner. Männlein-Robert, Irmgard. 2007. Stimme, Schrift und Bild: Zum Verhältnis der Künste in der hellenistischen Dichtung. Bibliothek der klassischen Altertumswissenschaften, Neue Folge 2.119. Heidelberg: Winter. Marshall, C. W. 2006. The Stagecraft and Performance of Roman Comedy. Cambridge: Cambridge University Press. Martin, Charles. 1992. Catullus. New Haven: Yale University Press. Martin, R. H., ed. 1976. Terence: Adelphoe. Cambridge: Cambridge University Press. Martindale, Charles. 1993. Redeeming the Text: Latin Poetry and the Hermeneutics of Reception. Cambridge: Cambridge University Press. ———, ed. 1997. The Cambridge Companion to Virgil. Cambridge: Cambridge University Press. ———. 2004. Latin Poetry and the Judgment of Taste: An Essay in Aesthetics. Oxford: Oxford University Press. ———. 2006. “Introduction: Thinking Through Reception.” In Charles Martindale and Richard F. Thomas, eds., Classics and the Uses of Reception, 1–13. Malden: Blackwell. Marx, Friedrich, ed. 1928. Plautus, Rudens. Abhandlungen der sächsischen Akademie der Wissenschaften zu Leipzig, Philologisch-Historische Klasse 38.5. Leipzig: Hirzel. Mattingly, Harold. 1976. Coins of the Roman Empire in the British Museum. Vol. 2. 2nd ed. Prepared by R. A. G. Carson. London: British Museum Publications. Maurice, Lisa. 2005. “A Calculated Comedy of Errors: The Structure of Plautus’ Menaechmi.” Syllecta Classica 16: 31–59. McCarthy, Kathleen. 2000. Slaves, Masters, and the Art of Authority in Plautine Comedy. Princeton: Princeton University Press. ———. 2004. “The Joker in the Pack: Slaves in Terence.” In Boyle and Penwill, 100–19. McDonnell, Myles. 2006. “Roman Aesthetics and the Spoils of Syracuse.” In Sheila Dillon and Katherine E. Welch, eds., Representations of War in Ancient Rome, 68–90. Cambridge: Cambridge University Press. McMahon, John M. 1998. Paralysin Cave: Impotence, Perception, and Text in the Satyrica of Petronius. Mnemosyne Supplementum 176. Leiden: Brill. McNelis, Charles. 2002. “Greek Grammarians and Roman Society during the Early Empire: Statius’ Father and his Contemporaries.” Classical Antiquity 21: 67–94.

264

bibliography

———. 2007. Statius’ Thebaid and the Poetics of Civil War. Cambridge: Cambridge University Press. ———. 2008. “Ut Sculptura Poesis: Statius, Martial, and the Hercules Epitrapezios of Novius Vindex.” American Journal of Philology 129: 255–76. Meban, David. 2008. “Temple Building, Primus Language, and the Proem to Virgil’s Third Georgic.” Classical Philology 103: 150–74. Miles, Gary B. 1980. Virgil’s Georgics: A New Interpretation. Berkeley: University of California Press. Miles, Margaret M. 2008. Art as Plunder: The Ancient Origins of Debate about Cultural Property. Cambridge: Cambridge University Press. Miller, John F. 2009. Apollo, Augustus, and the Poets. Cambridge: Cambridge University Press. Miller, Paul Allen. 1994. Lyric Texts and Lyric Consciousness: The Birth of a Genre from Archaic Greece to Augustan Rome. London: Routledge. Mitchell, W. J. T. 1986. Iconology: Image, Text, Ideology. Chicago: University of Chicago Press. ———. 1994. Picture Theory: Essays on Verbal and Visual Representation. Chicago: University of Chicago Press. ———. 2005. What do Pictures Want? The Lives and Loves of Images. Chicago: University of Chicago Press. Morgan, Llewelyn. 1999. Patterns of Redemption in Virgil’s Georgics. Cambridge: Cambridge University Press. Morris, Pam. 2003. Realism. London: Routledge. Morris, Sarah P. 1992. Daidalos and the Origins of Greek Art. Princeton: Princeton University Press. Moseley, Nicholas, and Mason Hammond, eds. 1953. T. Macci Plauti Menaechmi. Cambridge: Harvard University Press. Müller, Roman. 2007. “Pura oratio und puri sermonis amator.” In Kruschwitz, Ehlers, and Felgentreu, 111–25. Munich, Matthew. 2003. “The Texture of the Past: Nostalgia and Catullus 64.” In Thibodeau and Haskell, 43–65. Myers, K. Sara. 1994. Ovid’s Causes: Cosmogony and Aetiology in the Metamorphoses. Ann Arbor: University of Michigan Press. Mynors, Roger A. B. 1990. Virgil: Georgics. Oxford: Clarendon Press. Nappa, Christopher. 2005. Reading After Actium: Vergil’s Georgics, Octavian, and Rome. Ann Arbor: University of Michigan Press. Nauta, Ruurd R. 2002. Poetry for Patrons: Literary Communication in the Age of Domitian. Mnemosyne Supplementum 206. Leiden: Brill. Nava, Maria Luisa, Rita Paris, and Rosanna Friggeri, eds. 2007. Rosso Pompeiano: La decorazione pittorica nelle collezioni del Museo di Napoli e a Pompei. Milan: Electa. Nelis, Damien. 2001. Vergil’s Aeneid and the Argonautica of Apollonius Rhodius. ARCA Classical and Medieval Texts, Papers and Monographs 39. Leeds: Francis Cairns. Nelson, Stephanie. 1998. God and the Land: The Metaphysics of Farming in Hesiod and Vergil. Oxford: Oxford University Press. Newby, Zahra. 2009. “Absorption and Erudition in Philostratus’ Imagines.” In Bowie and Elsner, 322–42. Newlands, Carole E. 2002. Statius’ Silvae and the Poetics of Empire. Cambridge: Cambridge University Press. Nisbet, Gideon. 2003. Greek Epigram in the Roman Empire: Martial’s Forgotten Rivals. Oxford: Oxford University Press. Nisbet, R. G. M., and Margaret Hubbard. 1978. A Commentary on Horace: Odes, Book II. Oxford: Clarendon Press. Norden, Eduard, ed. 1957. P. Vergilius Maro: Aeneis Buch VI. 4th ed. Stuttgart: Teubner.

Bibli og raphy

265

O’Connell, Michael. 1977. “Pictorialism and Meaning in Catullus 64.” Latomus 36: 746–56. O’Connor, Eugene M. 1989. Symbolum Salacitatis: A Study of the God Priapus as a Literary Character. Studien zur klassischen Philologie 40. Frankfurt am Main: Peter Lang. O’Hara, James J. 1990a. Death and the Optimistic Prophecy in Vergil’s Aeneid. Princeton: Princeton University Press. ———. 1990b. “The Significance of Vergil’s Acidalia mater and Venus Erycina in Catullus and Ovid.” Harvard Studies in Classical Philology 93: 335–42. ———. 1994. “They Might Be Giants: Inconsistency and Indeterminacy in Vergil’s War in Italy.” Colby Quarterly 30: 206–26. ———. 1996. True Names: Vergil and the Alexandrian Tradition of Etymological Wordplay. Ann Arbor: University of Michigan Press. ———. 2007. Inconsistency in Roman Epic: Studies in Catullus, Lucretius, Vergil, Ovid and Lucan. Cambridge: Cambridge University Press. Oliensis, Ellen. 1998. Horace and the Rhetoric of Authority. Cambridge: Cambridge University Press. ———. 2004. “The Power of Image-Makers: Representation and Revenge in Ovid Metamorphoses 6 and Tristia 4.” Classical Antiquity 23: 285–321. Ong, Walter J. 1982. Orality and Literacy: The Technologizing of the Word. London: Methuen. Osgood, Josiah. 2006. Caesar’s Legacy: Civil War and the Emergence of the Roman Empire. Cambridge: Cambridge University Press. O’Sullivan, Patrick. 2000. “Satyr and Image in Aeschylus’ Theoroi.” Classical Quarterly 50: 353–66. O’Sullivan, Timothy M. 2007. “Walking with Odysseus: The Portico Frame of the Odyssey Landscapes.” American Journal of Philology 128: 497–532. Owens, William M. 2000. “Plautus’ Stichus and the Political Crisis of 200 B.C.” American Journal of Philology 121: 385–407. Pagán, Victoria E. 2006. Rome and the Literature of Gardens. London: Duckworth. Pailler, Jean-Marie. 1990. “Le poète, le prince et l’arène: À propos du “Livre des Spectacles” de Martial.” In Claude Domergue, Christian Landes, and Jean-Marie Pailler, eds., Spectacula I: Gladiateurs et amphithéatres, 179–83. Lattes: Musée archéologique Henri Prades. Panayotakis, Costas. 1995. Theatrum Arbitri: Theatrical Elements in the Satyrica of Petronius. Mnemosyne Supplementum 146. Leiden: Brill. Papaioannou, Sophia. 2005. Epic Succession and Dissension: Ovid, Metamorphoses 13.623–14.582, and the Reinvention of the Aeneid. Untersuchungen zur antiken Literatur und Geschichte 73. Berlin: Walter de Gruyter. Parker, W. H. 1988, ed. and trans. Priapea: Poems for a Phallic God. London: Croom Helm. Paschalis, Michael. 2002. “Reading Space: A Re-examination of Apuleian ekphrasis.” In Michael Paschalis and Stavros Frangoulidis, eds., Space in the Ancient Novel, 132–42. Ancient Narrative, Supplementum 1. Groningen: Barkhuis Publishing & The University Library Groningen. ———. 2011. “Petronius and Virgil: Contextual and Intertextual Readings.” In Konstantin Doulamis, ed., Echoing Narratives: Studies of Intertextuality in Greek and Roman Prose Fiction, 73–98. Ancient Narrative, Supplementum 13. Groningen: Barkhuis & Groningen University Library. Paschalis, Michael, Stavros Frangoulidis, Stephen Harrison, and Maaike Zimmerman, eds. 2007. The Greek and Roman Novel: Parallel Readings. Ancient Narrative, Supplementum 8. Groningen: Barkhuis & Groningen University Library. Pavlock, Barbara. 1990. Eros, Imitation, and the Epic Tradition. Ithaca: Cornell University Press. Peden, R. G. 1985. “The Statues in Apuleius Metamorphoses 2.4.” Phoenix 39: 380–83. Philippides, Katerina. 1995. “Terence’s Eunuchus: Elements of the Marriage Ritual in the Rape Scene.” Mnemosyne 48: 272–84. Platt, Verity. 2010. “Art History in the Temple.” Arethusa 43: 197–213.

266

bibliography

———. 2011. Facing the Gods: Epiphany and Representation in Graeco-Roman Art, Literature and Religion. Cambridge: Cambridge University Press. Plaza, Maria. 2000. Laughter and Derision in Petronius’ Satyrica: A Literary Study. Acta Universitatis Stockholmiensis. Studia Latina Stockholmiensia 46. Stockholm: Almquist & Wiksell International. ———. 2006. The Function of Humour in Roman Verse Satire: Laughing and Lying. Oxford: Oxford University Press. Pollitt, Jerome J. 1978. “The Impact of Greek Art on Rome.” Transactions of the American Philological Association 108: 155–74. ———. 1986. Art in the Hellenistic Age. Cambridge: Cambridge University Press. Porter, James I. 2010. The Origins of Aesthetic Thought in Ancient Greece: Matter, Sensation, and Experience. Cambridge: Cambridge University Press. Pöschl, Viktor. 1975. “Die Tempeltüren des Dädalus in der Aeneis (VI 14–33).” Würzburger Jahrbücher für die Altertumswissenschaft, Neue Folge 1: 119–23. Powell, Anton, ed. 1992. Roman Poetry and Propaganda in the Age of Augustus. London: Bristol Classical Press. Powell, Barry B. 1976. “Poeta Ludens: Thrust and Counter-Thrust in Eclogue 3.” Illinois Classical Studies 1: 113–21. Prioux, Évelyne. 2007. Regards alexandrins: Histoire et théorie des arts dans l’épigramme hellénistique. Leuven: Peeters. ———. 2008. Petits musées en vers: Épigramme et discours sur les collections antiques. L’art & l’essai 5. Paris: CTHS-INHA. Putnam, Michael C. J. 1961. “The Art of Catullus 64.” Harvard Studies in Classical Philology 65: 165–205. ———. 1970. Virgil’s Pastoral Art: Studies in the Eclogues. Princeton: Princeton University Press. ———. 1979. Virgil’s Poem of the Earth: Studies in the Georgics. Princeton: Princeton University Press. ———. 1998. Virgil’s Epic Designs: Ekphrasis in the Aeneid. New Haven: Yale University Press. Quenemoen, Caroline K. 2000. “The Architectural Significance of the House of Augustus.” PhD Diss. Yale. ———. 2006. “The Portico of the Danaids: A New Reconstruction.” American Journal of Archaeology 110: 229–50. Ravenna, Giovanni. 1974. “L’ekphrasis poetica di opere d’arte in latino: temi e problemi.” Quaderni dell’Istituto di filologia latina, Università di Padova 3: 1–52. Reckford, Kenneth J. 1958. “Some Appearances of the Golden Age.” Classical Journal 54: 79–87. Reed, Joseph D. 1995. “The Sexuality of Adonis.” Classical Antiquity 14: 317–47. ———. 2000. “Arsinoe’s Adonis and the Poetics of Ptolemaic Imperialism.” Transactions of the American Philological Association 130: 319–51. ———. 2007. Virgil’s Gaze: Nation and Poetry in the Aeneid. Princeton: Princeton University Press. Richardson, Lawrence, Jr., ed. 1977. Propertius: Elegies I–IV. Norman: University of Oklahoma Press. ———. 1992. A New Topographical Dictionary of Ancient Rome. Baltimore: Johns Hopkins University Press. Richlin, Amy. 1992. The Garden of Priapus: Sexuality and Aggression in Roman Humor. Revised edition. New Haven: Yale University Press. Ricoeur, Paul. 1981. Hermeneutics and the Human Sciences: Essays on Language, Action, and Interpretation. Ed. and trans. John B. Thompson. Cambridge: Cambridge University Press. Ridgway, Brunilde S. 2000. Hellenistic Sculpture II: The Styles of ca. 200–100 B.C. Madison: University of Wisconsin Press.

Bibli og raphy

267

Rimell, Victoria. 2002. Petronius and the Anatomy of Fiction. Cambridge: Cambridge University Press. ———. 2008. Martial’s Rome: Empire and the Ideology of Epigram. Cambridge: Cambridge University Press. Ripoll, François. 1998. La morale héroïque dans les épopées latines d’époque flavienne: tradition et innovation. Bibliothèque d’études classiques 14. Louvain: Editions Peeters. Roller, Duane W. 2010. Cleopatra: A Biography. Oxford: Oxford University Press. Rosati, Gianpiero. 1999. “Form in Motion: Weaving the Text in the Metamorphoses.” In Hardie, Barchiesi, and Hinds, 240–53. Rosen, Ralph M. 2007. Making Mockery: The Poetics of Ancient Satire. Oxford: Oxford University Press. Rosivach, Vincent J. 1998. When a Young Man Falls in Love: The Sexual Exploitation of Women in New Comedy. London: Routledge. Ross, David O., Jr. 1975. Backgrounds to Augustan Poetry: Gallus, Elegy and Rome. Cambridge: Cambridge University Press. Rouveret, Agnès. 2007. “From Alexandria to Rome: Confrontation of Roman Painting with Augustan Poetics.” Paper presented at seminar on hellenistic aesthetics. University of Cincinnati, September 29. Saller, Richard P. 1982. Personal Patronage under the Early Empire. Cambridge: Cambridge University Press. Sandström, C. E. 1878. Studia Critica in Papinium Statium. Upsala: Upsala Universitets Årsskrift. Saunders, Timothy. 2008. Bucolic Ecology: Virgil’s Eclogues and the Environmental Literary Tradition. London: Duckworth. Schäfer, Sabine. 1996. Das Weltbild der Vergilischen Georgika in seinem Verhältnis zu De rerum natura des Lukrez. Frankfurt am Main: Peter Lang. Schlam, Carl C. 1992. The Metamorphoses of Apuleius: On Making an Ass of Oneself. Chapel Hill: University of North Carolina Press. Schlegel, Catherine M. 2005. Satire and the Threat of Speech: Horace’s Satires, Book 1. Madison: The University of Wisconsin Press. Schmale, Michaela. 2004. Bilderreigen und Erzähllabyrinth: Catulls Carmen 64. Munich: Saur. Schmeling, Gareth. 2011. A Commentary on the Satyrica of Petronius. With the collaboration of Aldo Setaioli. Oxford: Oxford University Press. Schmitzer, Ulrich. 1990. Zeitgeschichte in Ovids Metamorphosen: Mythologische Dichtung unter politischem Anspruch. Stuttgart: Teubner. Schneider, Werner J. 2001. “Phidiae putavi: Martial und der Hercules Epitrapezios des Novius Vindex.” Mnemosyne 54: 697–720. Schneidewin, F. G., ed. 1842. M. Val. Martialis epigrammaton libri. 2 vols. Grimae: Gebhardt. Schultz, Celia E. 2003. “Latet anguis in herba: A Reading of Vergil’s Third Eclogue.” American Journal of Philology 124: 199–224. Scivoletto, Nino. 1981. “La città di Roma nella poesia di Properzio.” In Francesco Santucci and Salvatore Vivona, eds., Colloquium Propertianum (Secundum), 27–38. Assisi: Accademia properziana del Subasio. Scott, Grant F. 1994. The Sculpted Word: Keats, Ekphrasis, and the Visual Arts. Hanover: University Press of New England. Sebesta, Judith L. 1994. “Mantles of the Gods and Catullus 64.” Syllecta Classica 5: 35–41. Segal, Charles P. 1965. “Aeternum per saecula nomen, the Golden Bough and the Tragedy of History: Part I.” Arion 4: 617–57. ———. 1967. “Vergil’s Caelatum Opus: An Interpretation of the Third Eclogue.” American Journal of Philology 88: 279–308. ———. 1981. “Art and the Hero: Participation, Detachment, and Narrative Point of View in Aeneid 1.” Arethusa 14: 67–83.

268

bibliography

Segal, Erich. 1987. Roman Laughter: The Comedy of Plautus. 2nd edition. Oxford: Oxford University Press. Selden, Daniel. 1998. “Alibis.” Classical Antiquity 17: 289–420. Shackleton Bailey, D. R., ed. 1990. M. Valerii Martialis epigrammata post W. Heraeum. Stuttgart: Teubner. Sharland, Suzanne. 2003. “Priapus’ Magic Marker: Literary Aspects of Horace, Satire 1.8.” Acta Classica 46: 97–109. Sharrock, Alison. 2009. Reading Roman Comedy: Poetics and Playfulness in Plautus and Terence. Cambridge: Cambridge University Press. Shumate, Nancy. 1996. Crisis and Conversion in Apuleius’ Metamorphoses. Ann Arbor: University of Michigan Press. Skinner, Marilyn B. 2001. “Ladies’ Day at the Art Institute: Theocritus, Herodas, and the Gendered Gaze.” In André Lardinois and Laura McClure, eds., Making Silence Speak: Women’s Voices in Greek Literature and Society, 201–22. Princeton: Princeton University Press. Skutsch, Otto. 1968. Studia Enniana. London: Athlone Press. ———. 1985. The Annals of Q. Ennius. Oxford: Clarendon Press. Slater, Niall W. 1985. Plautus in Performance: The Theatre of the Mind. Princeton: Princeton University Press. ———. 1990. Reading Petronius. Baltimore: Johns Hopkins University Press. ———. 1998. “Passion and Petrifaction: The Gaze in Apuleius.” Classical Philology 93: 18–48. ———. 2008. “Apuleian Ecphraseis: Depiction at Play.” In Werner Riess, ed., Paideia at Play: Learning and Wit in Apuleius, 235–50. Ancient Narrative, Supplementum 11. Groningen: Barkhuis Publishing & Groningen University Library. Smith, Louise Pearson. 1994. “Audience Response to Rape: Chaerea in Terence’s Eunuchus.” Helios 21: 21–38. Smith, Martin S., ed. 1975. Petronii Arbitri Cena Trimalchionis. Oxford: Clarendon Press. Smith, R. A. 1997. Poetic Allusion and Poetic Embrace in Ovid and Vergil. Ann Arbor: University of Michigan Press. Solodow, Joseph B. 1988. The World of Ovid’s Metamorphoses. Chapel Hill: University of North Carolina Press. Southern, Pat. 1997. Domitian: Tragic Tyrant. London: Routledge. Spence, Sarah. 1991. “Clinching the Text: The Danaids and the End of the Aeneid.” Vergilius 37: 11–19. Spencer, Diana. 2002. The Roman Alexander: Reading a Cultural Myth. Exeter: University of Exeter Press. Squire, Michael. 2009. Image and Text in Graeco-Roman Antiquity. Cambridge: Cambridge University Press. ———. 2010. “Making Myron’s Cow Moo? Ecphrastic Epigram and the Poetics of Simulation.” American Journal of Philology 131: 589–634. Stanley, Keith. 1965. “Irony and Foreshadowing in Aeneid, I, 462.” American Journal of Philology 86: 267–77. Steiner, Deborah T. 2001. Images in Mind: Statues in Archaic and Classical Greek Literature and Thought. Princeton: Princeton University Press. Stephens, Susan A. 2003. Seeing Double: Intercultural Poetics in Ptolemaic Alexandria. Berkeley: University of California Press. Stewart, Andrew F. 1990. Greek Sculpture: An Exploration. Vol. 1. New Haven: Yale University Press. Stewart, Peter C. N. 1997. “Fine Art and Coarse Art: The Image of Roman Priapus.” Art History 20: 575–88. ———. 2003. Statues in Roman Society: Representation and Response. Oxford: Oxford University Press.

Bibli og raphy

269

———. 2004. Roman Art. Greece & Rome: New Surveys in the Classics 34. Oxford: Oxford University Press. ———. 2008. The Social History of Roman Art. Cambridge: Cambridge University Press. Strazzulla, Maria J. 1990. Il Principato di Apollo: Mito e propaganda nelle lastre “Campana” dal tempio di Apollo Palatino. Studia Archaeologica 57. Rome: “L’Erma” di Bretschneider. Strzelecki, Wladyslaw, ed. 1964. Cn. Naevii Belli Punici carminis quae supersunt. Leipzig: Teubner. Strocka, Volker M. 2008. “Vergil und die Danaiden.” In Stefan Freund and Meinolf Vielberg, eds., Vergil und das antike Epos: Festschrift Hans Jürgen Tschiedel, 41–66. Stuttgart: Franz Steiner Verlag. Strong, Donald E. 1963. “Some Observations on Early Roman Corinthian.” Journal of Roman Studies 53: 73–84. ———. 1966. Greek and Roman Gold and Silver Plate. London: Methuen. Stuart Jones, H., ed. 1926. A Catalogue of the Ancient Sculptures Preserved in the Municipal Collections of Rome. The Sculptures of the Palazzo dei Conservatori. Oxford: Clarendon Press. Reprinted, Rome: “L’Erma” di Bretschneider, 1968. Sullivan, J. P. 1968. The Satyricon of Petronius: A Literary Study. London: Faber and Faber. ———. 1991. Martial: The Unexpected Classic: A Literary and Historical Study. Cambridge: Cambridge University Press. Swain, Simon. 2009. “Culture and Nature in Philostratus.” In Bowie and Elsner, 33–46. Syndikus, Hans Peter. 1990. Catull: Eine Interpretation. Vol. 2. Darmstadt: Wissenschaftliche Buchgesellschaft. Szelest, Hanna. 1966. “Die Originalität der Sog. Beschreibenden Silvae des Statius.” Eos 56: 186–97. Taplin, Oliver. 1980. “The Shield of Achilles within the Iliad.” Greece and Rome 27: 1–21. Tarrant, Richard J. 1997. “Poetry and Power: Virgil’s Poetry in Contemporary Context.” In Martindale, 169–87. Thalmann, William G. 1984. Conventions of Form and Thought in Early Greek Epic Poetry. Baltimore: Johns Hopkins University Press. Thibodeau, Philip, and Harry Haskell, eds. 2003. Being There Together: Essays in Honor of Michael C. J. Putnam on the Occasion of His Seventieth Birthday. Afton: Afton Historical Society Press. Thomas, Michael L. 2004. “(Re)locating Domitian’s Horse of Glory: The Equus Domitiani and Flavian Urban Design.” Memoirs of the American Academy in Rome 49: 21–46. Thomas, Richard F. 1982. “Catullus and the Polemics of Poetic Reference (Poem 64.1–18).” American Journal of Philology 103: 144–64. ———. 1983a. “Virgil’s Ecphrastic Centrepieces.” Harvard Studies in Classical Philology 87: 175–84. ———. 1983b. “Callimachus, the Victoria Berenices, and Roman Poetry.” Classical Quarterly 33: 92–113. ———, ed. 1988. Virgil: Georgics. 2 vols. Cambridge: Cambridge University Press. ———. 1998. “Virgil’s Pindar?” In P. E. Knox and C. Foss, eds., Style and Tradition: Studies in Honor of Wendell Clausen, 99–120. Stuttgart: Teubner. ———. 2001. Virgil and the Augustan Reception. Cambridge: Cambridge University Press. Thomson, Douglas F. S., ed. 1997. Catullus. Toronto: University of Toronto Press. Tomei, Maria A. 1990. “Le tre ‘Danaidi’ in nero antico dal Palatino.” Bollettino di archeologia 5–6: 35–48. Tompkins, Jane P., ed. 1980. Reader-Response Criticism: From Formalism to Post-Structuralism. Baltimore: Johns Hopkins University Press. Trappes-Lomax, John M. 2007. Catullus: A Textual Reappraisal. Swansea: Classical Press of Wales. van Mal-Maeder, Danielle, ed. 2001. Apuleius Madaurensis: Metamorphoses, Livre II. Groningen: Egbert Forsten.

270

bibliography

Vial, Hélène. 2010. La métamorphose dans les Métamorphoses d’Ovide. Paris: Les Belles Lettres. Völker, Thomas, and Dirk Rohmann. 2011. “Praenomen Petronii: The Date and Author of the Satyricon Reconsidered.” Classical Quarterly 61: 660–76. Vollmer, Friedrich, ed. 1898. P. Papinii Statii Silvarum libri. Leipzig: Teubner. Vout, Caroline. 2007. Power and Eroticism in Imperial Rome. Cambridge: Cambridge University Press. Wagenvoort, H. 1955. “Ad Statii Silv. I 1 adnotationes.” In Pieter de Jonge et al., eds., Ut Pictura Poesis: Studia Latina Petro Iohanni Enk septuagenario oblata, 195–203. Leiden: Brill. Wallace-Hadrill, Andrew. 1988. “The Social Structure of the Roman House.” Papers of the British School at Rome 56: 43–97. ———. 1998. “Horti and Hellenization.” In M. Cima and E. La Rocca, eds., Horti Romani. Atti del Convegno Internazionale, Roma, 4–6 maggio 1995, 1–22. Bullettino della Commissione Archeologica Comunale di Roma, Supplementi 6. Rome: “L’Erma” di Bretschneider. ———. 2008. Rome’s Cultural Revolution. Cambridge: Cambridge University Press. Walker, Susan, and Peter Higgs, eds. 2001. Cleopatra of Egypt: From History to Myth. Princeton: Princeton University Press. Walsh, P. G. 1970. The Roman Novel. Cambridge: Cambridge University Press. Wandhoff, Haiko. 2003. Ekphrasis: Kunstbeschreibungen und virtuelle Räume in der Literatur des Mittelalters. Berlin: Walter de Gruyter. Watson, Gerard. 1988. Phantasia in Classical Thought. Galway: Galway University Press. Webb, Ruth. 1999. “Ekphrasis Ancient and Modern: The Invention of a Genre.” Word & Image 15: 7–18. ———. 2006. “The Imagines as a Fictional Text: Ekphrasis, Apatê, and Illusion.” In Michel Costantini, Françoise Graziani, and Stéphane Rolet, eds., Le Défi de l’art: Philostrate, Callistrate et l’image sophistique, 112–36. Rennes: Presses Universitaires de Rennes. ———. 2009. Ekphrasis, Imagination and Persuasion in Ancient Rhetorical Theory and Practice. Farnham: Ashgate. Weber, Clifford. 1983. “Two Chronological Contradictions in Catullus 64.” Transactions of the American Philological Association 113: 263–71. Welch, Tara S. 2001. “Est locus uni cuique suus: City and Status in Horace’s Satires 1.8 and 1.9.” Classical Antiquity 20: 165–92. ———. 2005. The Elegiac Cityscape: Propertius and the Meaning of Roman Monuments. Columbus: The Ohio State University Press. Wheeler, Arthur L. 1964. Catullus and the Traditions of Ancient Poetry. Berkeley: University of California Press. Wheeler, Stephen M. 2000. Narrative Dynamics in Ovid’s Metamorphoses. Classica Monacensia: Münchener Studien zur klassischen Philologie 20. Tübingen: Narr. White, Peter. 1974. “The Presentation and Dedication of the Silvae and the Epigrams.” Journal of Roman Studies 64: 40–61. ———. 1975. “The Friends of Martial, Statius, and Pliny, and the Dispersal of Patronage.” Harvard Studies in Classical Philology 79: 265–300. ———. 1978. “Amicitia and the Profession of Poetry in Early Imperial Rome.” Journal of Roman Studies 68: 74–92. ———. 1993. Promised Verse: Poets in the Society of Augustan Rome. Cambridge: Harvard University Press. Wilkinson, L. P. 1969. The Georgics of Virgil: A Critical Survey. Cambridge: Cambridge University Press. Williams, Gordon. 1980. Figures of Thought in Roman Poetry. New Haven: Yale University Press. Williams, R. D. 1981. “The Shield of Aeneas.” Vergilius 27: 8–11. ———. 1990 (1960). “The Pictures on Dido’s Temple (Aeneid I. 450–93).” In Harrison, 37–45. Oxford: Oxford University Press.

Bibli og raphy

271

Wimmel, Walter. 1970. “Vergil und das Atlantenfragment des Naevius.” Wiener Studien 83: 84–100. Winkler, John J. 1985. Auctor & Actor: A Narratological Reading of Apuleius’s Golden Ass. Berkeley: University of California Press. Wiseman, T. P. 1985. Catullus and His World: A Reappraisal. Cambridge: Cambridge University Press. Wormell, D. E. W. 1960. “The Riddles in Virgil’s Third Eclogue.” Classical Quarterly 54: 29–32. Zanker, Graham. 2003. “New Light on the Literary Category of ‘Ekphrastic Epigram’ in Antiquity: The New Posidippus (col. X 7-XI 19 P. Mil. Vogl. VIII 309).” Zeitschrift für Papyrologie und Epigraphik 143: 59–62. ———. 2004. Modes of Viewing in Hellenistic Poetry and Art. Madison: University of Wisconsin Press. Zanker, Paul. 1983. “Der Apollontempel auf dem Palatin: Ausstattung und politische Sinnbezüge nach der Schlacht von Actium.” In Kjeld de Fine Licht, ed., Città e Architettura nella Roma Imperiale. Analecta Romana Instituti Danici, Supplementum 10, 21–40. Odense: Odense University Press. ———. 1988. The Power of Images in the Age of Augustus. Translated by Alan Shapiro. Ann Arbor: University of Michigan Press. Zeiner, Noelle K. 2005. Nothing Ordinary Here: Statius as Creator of Distinction in the Silvae. New York: Routledge. Zeitlin, Froma I. 1971a. “Petronius as Paradox: Anarchy and Artistic Integrity.” Transactions of the American Philological Association 102: 631–84. ———. 1971b. “Romanus Petronius: A Study of the Troiae Halosis and the Bellum Civile.” Latomus 30: 56–82. ———. 1986. “Thebes: Theater of Self and Society in Athenian Drama.” In J. Peter Euben, ed., Greek Tragedy and Political Theory, 101–41. Berkeley: University of California Press. ———. 1994. “The Artful Eye: Vision, Ecphrasis and Spectacle in Euripidean Theatre.” In Goldhill and Osborne, 138–96. Cambridge: Cambridge University Press. Zetzel, James E. G. 1983. “Catullus, Ennius, and the Poetics of Allusion.” Illinois Classical Studies 8: 251–66. ———. 2002. “Dreaming about Quirinus: Horace’s Satires and the Development of Augustan Poetry.” In Tony Woodman and Dennis Feeney, eds., Traditions and Contexts in the Poetry of Horace, 38–52. Cambridge: Cambridge University Press.

This page intentionally left blank

INDEX

Accius, 42 Acculturation, 13, 251 Achates, 144, 147, 152 Achilles, 130, 142, 177, 249 military prowess of, 41, 146 Shield of. See Homer, Iliad wrath of, 140, 144–147, 153, 158, 167 Achilles Tatius Leucippe and Clitophon, 191–192, 195 Actaeon, 244–247 Actian games, 120 Adelphoe, 34 Admetus, 118, 169 n.111 Adonis painting of, 21–23, 26, 28 woven image of, 69–71 Aegaeon, 163–164 Aegeus, 40, 59, 64–66, 149 Aegyptus, 111, 115 Aemilius Paullus, L., 6 Aeneas Shield of. See Vergil, Aeneid as viewer, 142–160 Aesculapius, 171 Africa dish, 88–89 Agamemnon, 144, 171, 180 n.7 Ageladas, 103 Agrigentum, 17–19 Agrippa, 102 Alba Fucens, 229–230 Alban Games, 209–210 Alcimedon, 81–86, 107, 172–173 Alcman, 211 Alcon, 170, 172–173 Alexander the Great, 7, 88 n.48, 137, 218–219 as owner of art, 209–210, 233, 235–236, 238 Alexandrianism, 12, 40 n.4, 42, 69, 119, 127

Ambracia, 6 Amphrysus, 117–118 Anchises, 5, 123, 130, 143, 148, 151, 158, 165 Androgeus, 149 Andros, 171, 174 Aniads, 171–172, 174 Anius, 140–141, 160, 170–175 Anti-art, 102–104 Antigone, 167 Antikythera mechanism, 90 n.51 Antiochus, 35 Antipater of Sidon, 60 Antipater of Thessalonica, 88–89 Antonius, L., 76 Antony (M. Antonius) and Horace’s Satires, 92–93, 104 and Palatine temple of Apollo, 109, 111, 113–115 and Vergil’s Aeneid, 155–156 and Vergil’s Eclogues, 76–77 and Vergil’s Georgics, 119–120, 122 Apelles, 189, 191, 194–195, 198, 201–202, 231 paintings of Alexander by, 137, 218–219 Aphrodite, 70, 144 Apollo Cumaean temple of, 148–152 Palatine temple of, 5, 109–136, 154, 161, 164 Apollonius Rhodius Argonautica, 16 n.5, 42–44, 46, 59, 68–69, 156 n.60 Appian Way, 13, 209, 226–227 Apuleius Metamorphoses, 244–250 Arachne, 140–141, 165–170, 177 Aratus, 81, 186 Phaenomena, 88–89 Arcadia, 79, 85

273

274

Arcesilaus, 78 Archimedes, 81, 90 Argo, 40–44, 46, 58, 60, 232 Argos, 115 Ariadne, 5, 39–73, 126, 148, 150–152, 177 statue of, 60 Aristophanes, 11, 75, 94, 98–99, 106 Aristotle, 34, 99 Arsinoe II, 66 Artemis, 113, 131, 216, 218 Ascyltos. See Petronius, Satyricon Asterie, 169 Ataraxia, 100–101 Athena, 43, 115, 143 Athenis, 61, 111 Attis, 67 Auditorium of Maecenas, 93 Augustus. See also Octavian Forum of. See Forum Augustum and Ovid’s Metamorphoses, 160–175 and Vergil’s Aeneid, 142–160 Bacchante, 53, 60, 63, 68 Bacchus, 40, 53, 60–61, 68, 169, 171 cult of, 28 Baldric of Pallas. See Vergil, Aeneid Battle of Actium, 77, 92, 111, 114, 124, 135 in ecphrasis, 122, 140, 154–156, 158, 164, 185 Bavius, 82, 91, 107 Baxandall, M., 3 nn.5–6 Becker, A., 1, 3, 152 Belus, 115, 130 Berenice II, 66–67, 80 Lock of. See Callimachus, Aetia Lock of (constellation), 80 Bilingualism, 210–211 Bloom, H., 4 n.8 “Bore,” 91, 93, 105 Boscoreale, 50, 53, 88–89 Britons, 117, 121 Brontes, 213, 215 Browning, R., 25, 207 Brundisium, 77, 93, 104 Bryson, N., 3 n.5 Bupalus, 111 Busiris, 117, 119, 236 Byzantine ecphrasis, 108 n.1 Cacus, 156, 227–228 Caesar (C. Julius), 76, 78, 114, 124, 129, 159 and Anchises, 165 British campaigns of, 121 divinity of, 122 Forum of. See Forum Iulium and Pompey, 112 statue of, 217–218, 236

Ind e x Callimachus Aetia, 42, 80, 97, 119, 123 Lock of Berenice in, 66, 80 Victoria Berenices in, 69, 223–224 Epigrams, 119, 124 n.47, 133 Hecale, 59, 68 Hymn to Apollo, 124, 130, 133–134, 169 Hymn to Artemis, 216 Hymn to Delos, 134, 148 Iambi, 97 Camilla, 147 Campana reliefs, 115–116 Campus Martius, 7, 114, 120 Canidia, 95, 99, 105 Capitolia, 209 Capitoline Hill, 210, 228 Caprius, 98–99 Casa Romuli, 114 Cassandra, 180, 183, 190 Cato the Elder, 8, 25, 29 Cato the Younger, 217 Catullus Carmina Ariadne coverlet in, 39–73, 79, 149–150, 177, 201 Cebes Tabula, 196 Cena Trimalchionis. See Petronius Cephisodotus, 111, 113 Chalcis, 149 Chares of Lindos, 219 Chariton Chaereas and Callirhoe, 191 Chrysippus, 198 Cicero, 7, 77, 79, 90, 103, 133 n.64 Cinyras, 23, 168 Circe, 54–55, 201 Circus Maximus, 111 Cleopatra, 11, 88 n.48, 92, 109, 115, 119, 159 in ecphrasis, 155–156 Cloanthus, 194 Clonus, 157 Coins, 209, 212–213 Colonization, 3, 6, 149 Colossus of Rhodes, 214, 219 Columella De re rustica, 96 n.65, 103 Comatas, 85–86 Comedy, 10, 14, 79, 177–179, 184 Old, 98–99, 106, 204 Roman, 21–38, 177, 251 Connoisseurship, 110, 135, 205, 207, 229, 231, 234 Conon, 80–83, 85, 90 Consus, 35 Contaminatio, 34 Corinth, 7–8, 70–71, 183

Ind e x Coronae, 172, 174 Crassus (M. Licinius), 122 Cratinus, 98 Crete, 40, 59, 61, 148, 150–151, 169, 247 Culler, J., 45, 73 Cultural competition ecphrasis as, 4, 21, 40, 76, 106, 244 Cumae. See Apollo, Cumaean temple of Cup of Anius. See Ovid, Metamorphoses Cupid, 61 Cybele, 67 Cyclopes, 215–216 Cynicism, 100–101 Cynthia, 126, 128–130, 201 Dacians, 216–217, 227 Daedalus, 140, 161, 176, 182 and Aeneas, 1–2, 5, 10, 148–153, 157, 175 name of Trimalchio’s cook, 179, 183–184 Damoetas, 82–88, 90–91, 173 Damon, 87 Danae, 30, 33, 169, 177 Danaids, 55, 111–112, 115–116, 129–130, 157, 159 Danaus, 115, 126, 129–130 Daphnis, 83–84, 87, 195 Dardanus, 159, 164, 167–168 Deduco, 119–120 Delos, 117, 127, 130, 133–136, 148, 170–171 Delphi, 17, 115, 134 Democritus, 198 Demodocus, 143–145, 147 Diana, 111, 113, 152, 201, 222, 244–247 Dido, 5, 18–19, 129, 140, 177, 193 and Ariadne, 150–151 and Cleopatra, 159 and temple of Juno. See Vergil, Aeneid Dio Cassius, 93 Dionysius of Halicarnassus, 137 n.2 Dionysus, 53, 78, 103 Divus Iulius, 122, 217 See also Caesar Domitian, 13, 205–228, 233, 241–243 Donatus, 31 Doris, 163 Drinkware, 9, 180, 183, 190 duBois, P., 3 n.5 Earinus, 226 Empedocles, 156 n.60 Enargeia, 9 n.25 Encolpius. See Petronius, Satyricon Ennius, 31–32, 42, 120, 123 Annales, 32, 120 Medea exsul, 42 Envy, 12, 77, 110, 124–125, 169 Ephesus, 199, 218

Epicureanism, 11, 39, 75, 93–94, 100–102, 235 Epigram Greek, 94, 231, 239–241 Hellenistic, 96, 103, 214 See also Callimachus; Martial Epiphany, 12, 36, 248 n.16, 251 in Petronius, 186, 188 in Propertius, 109, 125–135 Epitrapezios, 229 “Epyllion,” 39, 69, 231 Equus Domitiani, 211–219 Eretria, 6 Esquiline Hill, 54, 93 Etruscans, 6, 18, 24 Euboea, 171 Eudoxus, 81, 90, 198 Eumenides, 65, 68 Eumolpus. See Petronius, Satyricon Euphrates, 122 Eupolis, 98 Euripides, 32 n.56, 41–42, 60 n.63, 67 Ion, 17 n.10 Medea, 42, 44, 67 Europa, 69, 169, 191 Euryalus, 194 Eurystheus, 117, 225, 243 Eurytus, 157 Evander, 157–158, 228 Exempla, 29, 34–35, 170 Fabius Maximus, Q., 6 Fabius Pictor, C., 8 Fabius Pictor, Q., 8 Famulus, 204 Fannius Synistor, P., 50, 53 Farnese Atlas, 90 n.52 Farnese Bull, 78 Fates, 40, 72 Fellini, F., 180 Fetish, 4 Flaminius (L. Quinctius), 6 Flavian amphitheater, 214 Forum Augustum, 137–138, 142, 219 Forum Gallorum, 77 Forum Iulium, 214, 217, 242 Forum Romanum, 13, 208, 211, 214, 217 Fulvius Flaccus, M., 35 Fulvius Nobilior, M., 6, 120, 228 Gallia Transpadana, 77 Gallus (C. Cornelius), 77 Ganymede in Petronius, 189, 192–194, 197, 201 n.59 in Plautus, 21–37 Gardens, 94, 97, 100, 102, 105 Giants, 16, 19, 164

275

276

Gift ecphrasis as, 206–243 Gigantomachy, 4, 16–19, 34, 155–156, 164 Giton. See Petronius, Satyricon Gladiators, 179, 182–183 Globes, 88–90 Golden Age, 46–47, 77, 87, 90, 92, 107 Gorgo, 70 Gorgon, 115–116, 155, 176, 209 Granicus Monument, 7 Habinnas. See Petronius, Satyricon Haemus, 167 Hall of Liberty, 78, 85 Hannibal, 210, 233–235, 238 Harpies, 164 Hecate, 105 Hector, 147, 158, 171 Helen, 193, 197 Helenus, 143 Helicon, 119, 230 Helios, 164, 214, 219 Hephaestus, 144, 150, 153, 162, 249–250 Heracles, 83, 85, 119, 140, 153, 155–156. See also Hercules Shield of. See Pseudo-Hesiod, Scutum Herculaneum, 53 Hercules Musarum, 120, 228 statues of, 229–243 Hermes, 97 Hermodorus of Salamis, 7 Herms, 96–97, 103 Herodas Mimiambs art objects in, 70 n.84, 201, 246 n.8 Heroic objects in ecphrasis, 137–176 Hesiod, 118 n.28, 119 Theogony, 16 Works and Days, 72, 164 n.96 Hipparchus, 186 Hollander, J., 3 Homer Iliad Shield of Achilles in, 16, 150, 152–157, 162 Odyssey Latin translation of, 4. See also Livius Andronicus paintings of, 182, 191. See also Odyssey Landscapes Phaeacian palace in, 39 n.1, 143 Horace Epistles, 4, 200 Epodes, 91, 99

Ind e x Satires Priapus statue in, 92–106 Horti Maecenatis, 93, 102. See also Gardens House of Augustus, 51 House of Julius Polybius, 103 House of the Colored Capitals, 103–104 House of the Cryptoportico, 52 Hundred-handers, 163–164 Hyacinthus, 189, 194 Hylas, 117, 119, 189–190, 225–226 Ibycus, 211 Icarus, 5, 148, 151–152, 161 Ida, 200 Idumaea, 119 Illyria, 26, 77–78, 92, 114 Imagines maiorum, 34 Inachus, 159 Inconsistency, 19, 40, 47–48, 68 Io, 69, 156, 159 Ionia, 169 Iphigenia, 184 Iser, W., 74 n.2 Isis, 115, 156, 244–247 Jason, 16, 67–69, 156 Jerome, 91 n.54 Julia Domna, 248 Juno temple of. See Vergil, Aeneid Jupiter painting of, 29–36 Krieger, M., 3 n.5, 20 n.17, 73 labyrinth, 62, 64, 150, 152, 160, 184 Lacon, 85 Laocoon, 185, 199 Laodamia, 65, 235 Laomedon, 167–168 Latona, 111, 113 Leonardo da Vinci, 4 n.6 Leonidas of Tarentum, 95–96 Lepidus (M. Aemilius), 77, 92 Lesbia, 47–48, 63–67, 177 Lessing, G., 108 Leto, 113 Libraries, 78, 107, 109, 111, 113 Libya, 126–127, 134, 227, 238 Livius Andronicus Odussia, 16 plays of, 14 n.1 Livy, 6–7 Longus Daphnis and Chloe, 191, 195

Ind e x Lucan Iliacon, 198 Lucian, 245, 247 Lucilius, 92, 94, 98–99, 101, 106 Lucretius, 13, 39 n.1, 100–102, 118 n.28 Lucullus (L. Licinius), 7 Ludi scaenici, 121 Luxuria, 7 Lycurgus, 190 Lydia, 165, 168–170, 225 Lysippus, 7, 198, 218–219 Hercules statuette of, 229–242 Macedonia, 6, 29, 197 Maecenas, C., 75, 91–94, 97, 100, 102, 104– 107, 206, 209 Maevius, 82, 91, 107 Magna Graecia, 6, 25, 35 Mantua, 117, 119–120 Marcellus (M. Claudius), 6, 8, 24, 90 Marcellus (M. Claudius; heir of Augustus), 151 Martial Epigrams statues of Hercules in, 219–229, 236–243 Massilia, 192 Medea, 41–42, 44, 46, 52, 67–68, 180 n.7 Medieval ecphrasis, 108 n.1 Medusa, 115–116, 208 Memnon, 147 Menalcas, 80–91, 194 Menander, 31–33, 34, 130 Mercury, 181 Metellus Macedonicus, Q. Caecilius, 6–7 “Milesian tale,” 196 Mincius, 117, 120, 122 Minerva, 140–141, 156, 165–170, 181, 208–209, 216 Minos, 149 Minotaur, 40, 62, 149–150, 152 Mise-en-abîme, 66, 199, 245 Mitchell, W. J. T., 3 n.5, 74, 178, 200–203, 207 Molorchus, 117, 219, 224, 237–238 Mopsus, 84, 194 Moschus, 69, 175 Mulciber, 162. See also Vulcan Mummius Achaicus, L., 7 Murals, 8, 140, 142, 144–146, 179 Muses, 17, 87–88, 97, 116–117, 119–120, 228, 230 Museums, 53, 56–57, 187, 207, 213 Mutina, 77 Myron, 111, 126, 132–133, 198, 231 Naevius Bellum Punicum Gigantomachy ecphrasis in, 16–20, 36–38, 245

277

Naiad, 189 Naples, 53, 99, 248–249 Narcissus, 176 Naulochus, 111, 114 Necromancy, 94, 102 Neptune, 156, 167–169 Nereids, 43–44, 46, 53, 60–61 Nero, 49, 178, 198, 203–205, 214 Neronia, 204 Nile, 121–122, 156 Niobe, 111, 131–135, 176, 183–184, 190 Novel Greek, 182, 191–192, 194–196, 244, 247 Roman. See Apuleius; Petronius Nymph, 163, 190, 225–226 Octavia, 77 Octavian and Horace’s Satires, 92–93, 102, 104, 106–107 and Propertius’s Elegies, 125–136 and Vergil’s Eclogues, 76–79, 107 and Vergil’s Georgics, 109–125 See also Augustus Odeon, 209 Odysseus, 54–55, 59, 142–145, 147–148, 186 Odyssey Landscapes, 54, 56–57, 59 Old Comedy. See Comedy Olympia, 121, 219, 241 Oracle, 149 Orion, 217 Orionids, 141, 171–174 Orpheus, 81–82, 86–87 Ortygia, 126–127, 133 Ovid Ars Amatoria, 130, 166 Metamorphoses, 137, 140–141, 160–176 Cup of Anius in, 170–175 Palace of the Sun in, 160–165 tapestries of Arachne and Minerva in, 165–170 Tristia, 129–130 Pacuvius, 8 Palace of the Sun. See Ovid, Metamorphoses Palaemon, 81, 84, 87 Palatine Hill. See Apollo, Palatine temple of Pallas, 140, 153, 157–159 Pamphili Obelisk, 209 n.7 Panegyric, 110, 124, 132, 206–243 Panhellenic Games, 120 Papirius, 35 Parcae, 181 Parma, 77, 220 Parnassus, 126, 134 Pasiphae, 150–151, 176, 184

278

Ind e x

Patronage, 75, 78, 82, 91, 167, 206–243 Pausanias, 247 Pederasty, 178, 189, 197, 200, 226 Peleus, 39–40, 43–46, 48, 60–61, 71 Penthesilea, 147, 177 Pergamene boy, 196–198, 201–202 Pergamum, 18, 109, 178, 190, 196–197 Perseus, 6, 115, 155, 176 Persia, 7, 233 Persius, 204 Perusia, 76 Petronius Satyricon, 177–205 Cena Trimalchionis in, 180–189 gallery scene in, 189–200 Phaeacia, 39, 143 Phaethon, 141, 160–162, 164–165 Phantasia, 9–10 n.25, 195 n.40, 231, 240–241 Pharsalus, 46, 76 Phidias, 103, 198, 201, 219, 231, 240–241 Philippi, 93, 104 Philostratus, 13, 244, 247–250 Eikones, 248–249 Life of Apollonius of Tyana, 248 Lives of the Sophists, 247–248 Phoenicia, 19, 169 Phradmon, 103 Pinakes, 53 Pindar, 18, 119, 124 n.47, 211 Piso Frugi (L. Calpurnius), 88–89 Planetaria, 90 Plato, 10, 99, 100, 196 Plautus Menaechmi painting of Ganymede and Adonis in, 21–28 Pliny the Elder, 8–9, 78, 102, 132 n.62, 194, 198, 204 Plunder art as, 2, 6–7, 14–15, 27 Plutarch, 6, 8, 155 Pollio (C. Asinius), 75–82, 85, 87–93, 107, 111, 206 Pollius Felix, 220 Polybius, 7, 34, 103 Polyclitus, 103, 240 Pompeii, 5, 32, 52, 54, 103–104 Pompey (Cn. Pompeius), 76, 78, 92, 102 Portico, 54, 129, 130, 181, 249 of the Forum Augustum, 137–138 of the Palatine temple of Apollo, 111–115, 126–129, 154, 159 of Pompey, 78, 129 Porticus, 7, 50, 55. See also Portico Poseidon, 186 Posidippus, 240 Posidonius, 90

Praxinoa, 70 Praxiteles, 78, 85, 103, 201, 231 Priam, 146–147, 158 Priapea, 95, 103, 187 Priscian, 17 n.8 Progymnasmata, 1 n.1, 10 Propertius Elegies Palatine temple of Apollo in, 125–135 Proserpina, 169 Protesilaus, 65, 235 Proteus, 118, 163–164 Protogenes, 189, 191, 194–195 Pseudo-Apollodorus, 23, 227 Pseudo-Hesiod Scutum, 83, 153, 155–156 Psychoanalysis, 4 n.8, 172 n.120 Ptolemy II Philadelphus, 66, 69–71 Ptolemy III Euergetes, 66, 80 Punic Wars, 6, 14, 16–17, 19, 24 Pygmalion, 140, 168, 176 Pythagoras, 102 Rabirius, 241 Religious iconography and theories of ecphrasis, 107–108 Remus, 153 Rhesus, 146 Rhine, 216 Rhodes. See Colossus of Rhodes Rhodope, 167 Romulus, 114, 137, 153 Room of the Masks, 51 Room of the Mysteries, 53–54 Rubicon, 76 Sagana, 95, 99 Sappho, 47, 211 Saturn, 169 Saturnalia, 220 Saturnian verse, 16 Scamander, 249–250 Scheria, 147 Scopas, 78, 103, 111, 113, 127, 132, 135 Second Sophistic, 13, 247 Secular Games, 148 Seneca, 169 Septimius Severus, 248 Servius, 19, 77, 81 Shield of Achilles. See Homer, Iliad of Aeneas. See Vergil, Aeneid of Heracles. See Pseudo-Hesiod, Scutum Sibyl, 113, 135, 148–149, 152 Sibylline Books, 148–149 Sicily, 17, 22–26, 79, 102, 142, 213

Ind e x Silence, 5, 69, 104, 130–133, 152, 200–201 Silenus, 100 Social structure of ecphrasis, 74, 76, 107 Socrates, 11, 75, 99–100 Sol, 141, 161–162, 164–165, 214. See also Sun Sorrento base, 113, 135, 148 Spain, 76, 77, 211 Sparta, 197 Spheres, 88–90 Stadium, 209 Staging ecphrasis as, 1, 2 n.3, 14–38, 69, 76, 142 Statius Achilleid, 210 n.14 Silvae equestrian statue of Domitian in, 211–219 statuette of Hercules in, 229–237 Thebaid, 209 n.7 Stephanus, 78 Steropes, 213, 215 Stoicism, 7, 10 n.25, 195 n.40 Suda, 248 Suetonius, 29, 203–204 Sulcius, 98–99 Sulla Felix (L. Cornelius), 72, 112, 125, 210, 234, 238 Sun Palace of. See Ovid, Metamorphoses Syracuse, 6–7, 18, 22, 24, 27, 35, 69–71, 90 Tacitus Annals, 178 Tantalus, 126, 134 Tarentum, 6, 93, 95, 219 Tauriscus, 78 Temple building, 110, 124 Terence Adelphoe, 34–35 Eunuchus painting of Danae in, 6, 14, 29–31, 35–36 Terracina, 112 Theocritus Idylls festival of Adonis in, 69–71, 201 Priapus statue in, 96 n.68 and Vergil’s Eclogues, 79–92 vessels in, 83–85, 119, 216 Timotheus, 111, 113 Tiresias, 148 Titans, 16, 164

Titus, 210 Tityrus, 77 Trimalchio. See Petronius, Satyricon Triton, 163, 164 Troiae halosis, 198–199, 202 Troilus, 146 Trojan Horse in Petronius’s Satyricon, 183–185, 190, 199–200, 203 in Statius’s Silvae, 216–217 and Vergil’s Aeneid, 144, 159, 176 Trojan War, 65, 142–143, 193, 197–199, 201–202 Trompe-l’oeil, 48–51, 55 Tros, 123, 168 Turnus, 140, 147, 150, 156–159 Valerius Messalla, M’. 17–18, 20, 35 Varius, 93 Varro, 7, 85, 103 Venus paintings of, 21, 26, 53 Vergil Aeneid baldric of Pallas in, 157–159 Cumaean temple of Apollo in, 148–152 Shield of Aeneas in, 152–160 temple of Juno in, 142–147 Eclogues herdsmen’s cups in, 76–92 Georgics temple in, 116–125 Vespasian, 210, 225 Villa of the Mysteries, 50, 53–54, 60, 103 Villa Oplontis, 50–53 Vindex (Novius), 207–208, 210–211, 220, 229–232, 234–243 Vortumnus, 35 Vulcan, 153, 156–157, 162–163, 184 Wall Painting First-Style, 50 Second-Style, 9, 11, 41–42, 48–60, 62, 67 Widow of Ephesus, 199 Xenophon, 99 Zeus colossal statue of, 219 temple of, 17, 20 Zeuxis, 189, 234 Zodiac dish, 179, 184–185

279