goldberg epic in republican rome

Epic in Republican Rome

This page intentionally left blank

in Republican Rome

Sander M. Goldberg

New York Oxford OXFORD UNIVERSITY PRESS

1995

Oxford University Press Oxford New York

Athens Auckland Bangkok Bombay Calcutta Cape Town Dar es Salaam Delhi

Florence Hong Kong Istanbul Karachi Kuala Lumpur Madras Madrid Melbourne

Mexico City Nairobi Paris Singapore Taipei Tokyo Toronto

and associated companies in Berlin Ibadan

Copyright © 1995 by Sander M. Goldberg

Published by Oxford University Press» Inc., 200 Madison Avenue, New York, New York 10016

Oxford is a registered trademark of Oxford University Press

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,

electronic, mechanical, photocopying, recording, or otherwise, without the prior permission of Oxford University Press

Library of Congress Cataloging-in-Publication Data Goldberg, Sander M.

Epic in Republican Rome / Sander M. Goldberg, p. cm.

Includes bibliographical references (p. ) and indexes. ISBN 0-19 -509372-0

1. Epic poetry, Latin—History and criticism, 2. Rome—History—Republic , 510-30 B.C. 3. Literature and history—Rome. 4. R o m e — I n literature. I. Title.

PA6054.G65 1995 8 7 3 \ 0 1 0 9 — d c 2 0 94-18379

1 3 5 7 9 8 6 4 2

Printed in the United States of America on acid-free paper

Preface

This book began as an exercise in reading with a fairly simple motive. The fragments of early Roman epic, though repeatedly edit-ed and annotated as texts, are rarely treated as poetry, and they have never been subjected as a group to a literary critic's close and me-thodical scrutiny. 1 found this an irresistible challenge, and the fragments have, I think, justified my New Critical impulse. Yet even a devotee of practical criticism must acknowledge the urgency of larger questions that close reading may raise but cannot answer. What did talented poets like Naevius and Ennius hope to accomplish by writing narrative poems on Roman themes? What encouraged them to do so, and what idea of epic emerged from their efforts? These are questions for literary history, but histories of Latin litera-ture have been reluctant to address them. The tendency has instead been to slight the artistry of early epics, to attribute mundane or even base motives (on rather scanty evidence) to their creators, and to treat their achievement as but a preamble to Vergil's still greater achievement. I do not find this an adequate response. While argu-ments built from fragmentary evidence are necessarily fragile, and a literary history based on such arguments may never fully satisfy, the effort must be made. To understand what epic meant to the Romans— and what poets like Vergil and Lucan would eventually do to that meaning—we need to know all we can about the works that estab-lished its norms. We cannot afford to hurry by the earliest landmarks on the epic scene simply because they are now ruins. Even ruins are witnesses of a kind, and these witnesses have an important story to tell. Their testimony may nevertheless be a little quirky, and a book that takes them seriously is bound to have a few quirks of its own.

I have allowed the fragments of epic to set their own agenda, which has meant tolerating the incomplete, erratic, and occasionally peculiar lines of argument they suggest. I do not consider this a fault. An uneven argument seems preferable to none at all. My only resis-tance has been to imposing preconceived notions on the fragments or judging their poetry by the standards of a later time. I have therefore limited my primary discussion to fragmentary texts. Some of Catullus' hexameter poetry and large sections of Lucretius' De rerum natura can doubtless claim a place in the history of Republican epic, but not in this particular history. Whole texts are bullies. They

vi Preface

threaten to dominate and bias discussion by diverting attention from their less fortunate associates and by tempting critics to privilege their more coherent and engaging testimony. Let this be one place where the victims of time may safely claim their due. When I have looked beyond the fragments of poetry, it is generally to those frag-ments of history that hold clues to the values and concerns of the society that encouraged these poets and supplied their first audi-ences. I am not a historian, but it does not take a historian to see that the historical record has things to teach and that students of litera-ture should heed its lessons. I therefore find myself less fearful of writing bad history than hopeful of writing useful criticism.

Where this study goes may also raise the question of where it does not go. It does not go to Alexandria. Though I remain sympathetic to claims of an Alexandrian influence on early Latin poetry, nothing in the texts I have read has sent me back to the groves of Callimachus and Philitas. What I find serious and significant in the extant poetry of Andronicus and Naevius does not seem to be the result of Hellen-istic learning. Even Ennius, as doctus a pioneer as we are likely to meet, has left surprisingly little evidence to support the oft-repeated assertion that he read Greek literature with Alexandrian eyes and sought to reproduce Alexandrian effects. He certainly claimed in the Annals to be dicti studiosus, a phrase that may recall the philologia we associate with Alexandrian learning, but that recollection does not justify importing an Alexandrian aesthetic, itself a somewhat shaky reconstruction, to Rome of the early second century. We have no basis for distinguishing, as surely we must, the veneer of learning from the genuine article. Recent attempts along these lines have proven unsatisfactory. The fragments of Ennius most often identified as Alexandrian "experiments'' are all tainted by suspicions of forgery and false attribution, while a quite plausible allusion to Calli-machus' Aitia prologue lacks the detail required to confirm a signifi-cant influence. What Ennius may have learned from Alexandria remains beyond our ken, and I will frame no hypotheses on such inadequate evidence.

Two other decisions deserve mention. First, I have deliberately abbreviated the customary apparatus. The scholarship on individual fragments runs deep and long, and it has been elegantly collated and clarified in the relevant editions of Karl Büchner and Otto Skutsch. Readers interested in pursuing it should consult their bibliographic notes. My own annotation aims simply to acknowledge explicit debts, identify ancient sources, and point the way to contrary and comple-mentary material of value. Its inevitable omissions seem a small price to pay for the enhanced readability of an uncluttered page. Second, the translations provided are neither cribs nor poetry. Except

Preface vii

where noted, they are my own creations, designed to render the general sense and to capture the specific features of the passages discussed. Translations are lacking when the point at issue is untrans-latable or when its interest is largely technical. This is particularly true in the notes. Readers deserve help with the peculiarities of early Latin syntax and morphology, but I will not insult the lin-guistically insecure by shielding them from the sight and sound of Latin words. There are limits to even a Latinist's pedantry,

I would like to acknowledge some pleasant and important debts. My interest in early epic developed from a review of Otto Skutsch's edition of the Annals that I was then scarcely qualified to write. Only an editor's insistence and, I suppose, a certain naive curiosity overcame my initial resistance to the task. I did my best, and from that rather halting effort came a series of exchanges with a truly extraordinary scholar. Professor Skutsch, as those who knew him will attest, was a prompt, generous, and meticulous correspondent, but not a man quick to change his mind. And not a literary man. My dis-agreements with him were therefore numerous, and many have found their way into these pages. Our points of agreement—and my many points of dependence—are inevitably less conspicuous, but no less important. This work rests on his, and my only real regret upon com-pleting it is that I cannot learn from his response.

Another great debt is to Erich Gruen, who first showed me how history can illuminate literature. He, too, has taught me much, and the good history in what follows is probably his. The weak history doubtless marks places where I ignored his advice. Other debts are equally important. Barbara and James Halporn patiently bore the many impositions that friendship can bring. Denis Feeney, Robert Kaster, and James Zetzel proved, as always, to be diligent and opin-ionated readers, enemies of cant, and lovers of a good fight. They will know what I owe them. I must also thank Alan Cameron, Edward Courtney, and Robert Kaster for sharing drafts of their own works in progress. I have tried not to plunder without acknowledgment what were then only defenseless manuscripts. Over the years, many UCLA students watched tolerantly and responded indulgently when I rode my archaic hobbyhorse. John Kearns and Steve Reece even held the reins and claimed it was fun. I thank them all.

Olli cernebant magnis de rebus agentes. Finally, some collective and institutional debts: to audiences at

UCLA, the University of Southern California, and the American Philological Association, who responded with vigor to some early versions of these chapters; to UCLA's Academic Senate Committee on Research for a steady stream of grants-in-aid; and to the University of California Office of the President for a Faculty Research Fellow-

vi i i Preface

ship in the Humanities, which provided leave to complete the bulk of the writing. Preliminary accounts of material now found in chapters 3 and 5 have appeared in, respectively, Roman Epic, edited by A. J. Boyle (London 1993), and TAPA 119 (1989).

Los Angeles October 1994

S.M.G.

Contents

Texts and Abbreviations, xi

1. Ruins, 3

2. Reconstructions, 28

3. Saturnian Aesthetics, 58

4. Hexameter Aesthetics, 83

5. Poetry and Patronage, 111

6. Ciceronian Sirens, 135

7. Envoi, 158

Works Cited, 172

Concordances, 182

Index of Passages, 190

General Index, 193

This page intentionally left blank

Texts and Abbreviations

In 1982, Willy Morel's long-standard Fragmenta poetarum latinorum epicorum et lyricorum praeter Ennium et Lucilium (Leipzig 1927) was reissued with revisions and bibliographic notes by Karl Büchner. These notes for each fragment are a very great convenience, and I therefore cite whenever possible from Büchner's edition. The one major exception is Naevius' Bellum Punicum, for which the superi-ority of Ladislas Strzelecki's text cannot be ignored. As for Ennius, though Vahlen's edition remains valuable for its introduction and complete text of minor works, the fruits of more recent scholarship must claim priority. The tragedies and epic are therefore cited here from the editions of, respectively, Jocelyn and Skutsch. Minor works are cited from Courtney. Other editions of Andronicus, Naevius, and Cicero, each with its own virtues (and all too often with its own numbering scheme) are referred to occasionally in the notes and appear in the bibliography. Concordances to Morel's unrevised edi-tion and to Vahlen's Ennius may be found at the back.

The major texts used are as follows:

Ennius Ennianae poesis reliquiae, 2nd ed. Ed. J. Vahlen. Leipzig: Teubner,

1928. The Annals of Quintus Ennius. Ed. with introduction and commentary

by Otto Skutsch. Oxford: Clarendon, 1985. The Tragedies of Ennius. Ed. with introduction by H. D. Jocelyn.

Cambridge: Cambridge University Press, 1967. The Fragmentary Latin Poets. Ed. with commentary by Edward

Courtney. Oxford: Clarendon, 1993.

Livius Andronicus Fragmenta poetarum latinorum epicorum et lyricorum praeter Ennium

et Lucilium. Ed. Carolus Buechner. Leipzig: Teubner, 1982.

Naevius Cm Naevii Belli Punici carmen. Ed. W. Strzelecki. Leipzig: Teubner, 1964.

xii Texts and Abbreviations

Ancient literary works are generally cited from the Oxford editions. Other major sources used include:

Festus Sextus Pompeius Festus, De verborum significatu quae supersunt cum

Pauli Epitome. Ed. W. M. Lindsay. Leipzig: Teubner, 1913. Macrobius

Ambrosii Theodosii Macrobii Saturnalia . Ed. J. Willis. Leipzig: Teubner, 1963.

Nonius Nonius Marcellus, De compendiosa doctrina. Ed. W. M. Lindsay. 3

vols. Leipzig: Teubner, 1903. Servius

Servil Grammatici qui feruntur in Vergilii carmina commentant. Ed. G. Thilo and H. Hägen. 3 vols. 1881. Reprint, Hildesheim: Olms, 1961.

Authors and titles of ancient works follow the abbreviations of the Oxford Latin Dictionary. The titles of periodicals are abbreviated according to the recommendations of the American Philological Asso-ciation and the Modern Language Association. The titles of other standard works are abbreviated as follows:

CA = Collectanea Alexandrina. Ed. J. U. Powell. 1925. Reprint, Oxford: Clarendon, 1970.

CHLL = The Cambridge History of Classical Literature II: Latin Literature. Ed. E. J. Kenney and W. V. Clausen. Cambridge: Cam-bridge University Press, 1982.

FGH = Fragmente der griechischen Historiker. Ed. F. Jacoby. Leiden: E. J. Brill, 1954-,

GLK ~ Grammatici Latini. Ed. H. Keil. 7 vols, and suppl. Leipzig: Teubner, 1855-80.

HRR = Veterum historicorum romanorum reliquiae. Ed. H. Peter. 2 vols. Leipzig: Teubner, 1870.

ILLRP = Inscriptiones latinae liberae rei publicae, vol. 1. Ed. A. Degrassi, Florence: La nuova Italia, 1957.

MRR = The Magistrates of the Roman Republic, by T. R. S. Broughton. 3 vols. 1951-52. Reprint, Atlanta: Scholars Press, 1968-86.

ORF3 = Oratorum romanorum fragmenta liberae rei publicae, 3rd ed. Ed. H. Malcovati. Turin: Paravia, 1953.

R - Scaenicae romanorum poesis fragmenta, 3rd ed. Ed. Ο. Ribbeck. 1887-98. Reprint, Hildesheim: Olms, 1962.

SH ~ Supplementum Hellenisticum. Ed. H. Lloyd-Jones and P. Parsons. Berlin: De Gruyter, 1983.

Epic in Republican Rome

Yet the writers of my generation were natural assimilators. We knew the literature of Empires, Greek, Roman, British, through their essential classics; and both the patois of the street and the language of the classroom hid the elation of discovery. If there was nothing, there was everything to be made. With this prodigious ambition one began.

Derek Walcott, "What the Twilight Says: An Overture"

1 Ruins

Theory is the use of language to talk about language. (J. Hillis Miller) What is sillier than to talk about talking? (M. Tullius Cicero)

Latinists as a group remain wary of theory, and especially wary of literary theory. So many ancient works beg for study and present so many specific problems of language, style, and interpretation that we often think it distracting or even self-indulgent to divert attention from the literature we serve to our method of serving it. Yet the introspective exercise that theory entails is not entirely frivolous, nor is it necessarily a bad thing. Karl Lehrs warned long ago against unthinking invocations of method, and one good check on that recurring temptation is a clear awareness of the assumptions on which our scholarly method is based.1 The need for such introspection is especially keen as we confront a subject as problematic and demanding as early Roman epic. The relevant poems are in fragments. Little is known of their creators, and much of what is said about them is fiction. The conditions that encouraged their work are poorly at-tested and even more poorly understood. Historical, textual, and aesthetic problems are thus inextricably bound, challenging our technical skill and critical acumen even as they tax our patience. What can we realistically hope to learn from such difficult and often uncongenial material?

The very complexity of the evidence—not to mention the scholarly arguments it engenders—has discouraged the detailed treatment of

1 Thus Lehrs' fourth commandment for classical philologists: "Du sollst den Namen Methode nicht unnütz im Munde führen/' See Calder, CW 74 (1980/81) 228. The headnote references are to Miller, PMLA 102 (1987) 283 and de., de Or. 1.112; "nam quid est ineptius quam de dicendo dicere, cum ipsum dicere numquam sit non ineptum, nisi cum est necessarium?" For Latinists' reluctance to theorize, see Galinsky, The Interpretation of Roman Poetry, 1-40.

3

4 Epic in Republican Rome

early epic in our standard histories. It is simpler to record the received "facts" of authors and works and a few received opinions concerning them than to attempt a coherent account of how and why the genre developed at Rome. The epics of Andronicus and Naevius therefore attract scant attention, while Ennius' poem stands apart as a respected but vague precedent for the much clearer achievements of Lucretius and Vergil.2 Yet a good deal more can and should be said about these pioneers. Philologists and historians have produced reliable texts and helpful commentaries. The biographical testimony has been carefully sifted, and increased attention is now being paid to the cultural climate of their age. If literary scholars have failed to provide a clear sense of their artistry, it is more for want of trying than for lack of evidence. A sympathetic reading of the relevant fragments will soon identify the favored techniques and deliberate effects that gave early Latin epic a distinct and often effective style.

The present study offers the results of such a reading, but the practical criticism of pre-Vergilian epic is only one of its aims. Though the fragments are interesting in their own right, they are also important witnesses to the literary climate of Republican Rome. This fact raises a set of larger questions. Why and how did the epic genre develop in the third and second centuries, and what did its rise mean for other, contemporary genres? How are we to understand the dynamics, as well as the details, of Roman literary activity? Such issues, which are fundamental for the development of Roman litera-ture, move us from the aesthetic analysis of poetic fragments to the interpretation of those fragments as literary and cultural documents. This interpretive process entails not simply finding more space in Roman literary history for Rome's earliest poets, but redefining the material on which we base that history and reconsidering what such a history should aim to accomplish. How do we identify the growth of a literary form and the complex relationships that developed be-tween and among its authors and their audiences? How can we assemble a coherent account of these developments using the evidence of fragments?

Though the principles that should guide the writing of literary history are at present hotly contested by literary theorists, the modern debate involves a set of problems with significant and revealing ancient precedents. Roman authors themselves began the writing of Roman literary history, and they could not entirely escape the conceptual difficulties this activity invariably poses. Because their accounts have both set the parameters and furnished much of

2 Gratwick's essay on Ennius, CHLL, 60-76, is a welcome exception. Con-trast his inspired treatment to the stolid reliability of Schanz-Hosius, Geschichte der römischen Literatur, 90-99.

Ruins 5

the evidence for many subsequent works, it is worth beginning this new history with a look at some of these ancient endeavors.

I

Accius, the last great tragedian of the Roman stage, was also its first historian. He devoted nine books of a treatise called Didascalica to the history of Greek and Roman theater, cataloguing dramatists and genres from the beginning down to his own day.3 When that history reached Roman times, it presumably drew upon the performance records of the poets' guild, but however official Accius' sources may have been, they did not secure him universal acceptance on all points. There must have been other evidence available, for other conclusions were certainly reached. Rome's first dramatist, for example, was widely thought to have been a Tarentine Greek named Livius An-dronicus. Accius associated his arrival in Rome with the capture of Tarentum by Fabius Maximus in 209 and dated his first dramatic production to the ludi luventutis of 197. Yet that chronology was im-possible, as Cicero would demonstrate with some heat in the Brutus, his own attempt to write a history of Roman oratory. Cicero's anti-quarian friend Atticus had established, and his own research in antiquis commentants confirmed, the year 240 for Andronicus' first dramatic production. This is a documented fact, says Cicero, and a logical one. How else could Andronicus have preceded Naevius and Plautus, who were themselves active in the third century? Dating Andronicus' career to the 190s would make him younger than his successors. This counterargument has prevailed, and not only because it is probably right.4 It has an internal logic and its own intellectual appeal.

Cicero's concern with chronology is the necessary consequence of an implicit theoretical perspective. His quarrel with Accius was not simply over facts and the sequence of authors and works those facts

3 Accius was born in 170 (Jer. Chron. a. 1878, Cic. Brut. 229) and became a prominent member of the collegium poetarum (Val. Max. 3.7.11). Since he lived to discuss literature with Cicero (Brut. 107), his death is usually set ca. 85. For the fragments of the Didascalica, see DeglTnnocenti Pierini, Studi su Accio, 58-67; Leo, Geschichte der römischen Literatur, 386-91, provides a good general appraisal. It is unclear if the work was in verse or prose.

4 Brut. 72-73. Accius' reference to the ludi luventutis is also incorrect as it stands: these games, vowed by Livius Salinator at the Metaurus in 204, were conducted not in 197, but in 191 by the consul's son (Liv. 36.36.5-6). Atticus' likely source was Varro's De Poetis (cf. Gell. 17.21.44-45); the authority of Cicero's antiqui commentant is unknown. See Mette, Lustrum 9 (1964) 41^6, and Gruen, Studies, 80-82. D'Anna, Ciceroniana 5 (1984) 81-90, shows how Cicero champions the chronology best suited to his own argument.


dictate. Such a sequence by itself produces the kind of chronological narrative that modern theorists call an intrinsic history, and this may well have been Accius' purpose in the Didascalica. Cicero, how-ever, was certainly not writing so naive an account. The literary history of the Brutus has a distinctly extrinsic aim: its goal was not simply the presentation of a factual record but an interpretation of the facts it records.5

The sequence of authors Cicero defends is central to this greater purpose and to a notion of literary history that is both evolutionary and teleological. The Brutus explains Roman oratory by presenting its history within a general scheme of literary progress that applies Aristotelian concepts of inherent nature (physis) and goal (telos) to literary phenomena. Aristotle himself had employed a similar idea of progress in the Poetics when he described the history of tragedy in terms of technical advances. Poets, he said, put tragedy through a sequence of changes until it achieved its natural form, at which point its development necessarily ceased.6 Because literary forms, like organic ones, have a telos toward which their nature leads, an inter-nal impetus for change is inherent in his idea of literature. Evolution then becomes the central issue in literary history. Nothing, says Cicero, picking up this argument in his Brutus, is simultaneously in-vented and perfected.7 Artistic progress is linear: sculpture develops from the rigidity of Canachus to the fluid realism of Polyclitus, painting from the limited palette of Zeuxis and Polygnotus to the brilliance of Apelles. Literary development follows a similar pat-tern. Andronicus' work was crude, and that very roughness declares its priority. The more refined poets necessarily succeeded him (Brut. 70-71). Naevius is the Myron to Ennius' Polyclitus, just as Cato in this scheme plays Lysias to Cicero's own Demosthenes. Cicero does not develop these parallels; the evolutionary scheme simply lends credi-bility to the claim that Roman oratory had found its Demosthenes, and thus its telos, in Cicero himself. The limitations and consequences of such an argument are worth tracing, for our own histories of Latin literature tend to share them.

5 For the concepts of intrinsic and extrinsic history, see Patterson, "Lit-erary History," 250-62.

6 Poet. 1449al5: και πολλάς μεταβολας μεταβαλουσα ή τραγωδία έπαύσατο, έπε! εσχε την αυτής φύσιν. Physis for Aristotle implies inherent principles of motion and rest (Phys. 2.1.192bl5). Change is thus axiomatic in his literary scheme, although completion of the process is not (Poet. 49a7-9).

7 "Nihil est enim simul et inventum et perfectum," Brut. 71, cf. de Or. 1.13. Similarly, Dionysius of Halicarnassus (Din. 1) introduced his own essay on Attic orators by categorizing them as either inventors (εύρεταί) or perfectors (τελειωταί) of a style.

Ruins 7

The teleological aspect of Cicero's history has lost favor with literary theorists. Its underlying biological metaphor—that of generation and decay—has been widely discredited, and we are far less willing than our nineteenth-century predecessors to allow the "final" form of a genre to shape our understanding of its antecedents.8

Teleology presupposes closed sets of texts, while we more naturally define open ones, and it posits absolute values instead of the more relative, impermanent ones we have learned to accept. Because we assume that new works can always be added to a literary sequence and new beginnings defined, we now regard literary systems as subject to a continual process not of generation and decay but of redefinition and réévaluation. T. S. Eliot put the matter well in his 1919 essay "Tradition and the Individual Talent":

The existing monuments form an ideal order among themselves, which is modified by the introduction of the new (the really new) work of art among them. The existing order is complete before the new work arrives; for order to persist after the supervention of novelty, the whole existing order must be, if ever so slightly, altered; and so the relations, proportions, values of each work of art toward the whole are readjusted; and this is conformity between the old and the new.9

Any telos claimed for such a set will be apparent only in long retrospect and only to a particular observer at a particular time. It is inevitably a matter more of perception than fact. Cicero's own teleological schemes in the Brutus are strikingly incomplete and ten-dentious. He ignored Lysippus, was necessarily ignorant of Vergil, and simply shrugged off the contemporary criticism of his own achievement. Yet Roman oratory did not cease developing after his Philippics any more than epic came to a halt with Ennius' Annals.

The evolutionary aspect of Cicero's argument is more congenial and more familiar. Shorn of teleological claims, evolution refers simply to the process of change in literature, and literary history must surely account for the alterations we observe in literature over time. This more limited definition avoids the disillusionment with evolu-tionary schemes that so frustrated René Wellek in his later years, when he abandoned his lifelong dedication to the evolutionist cause.10 Wellek had originally sought to explain the direction of change. We should now be content merely to describe its vagaries

8 Wellek, Theory of Literature, 255-57; Perkins, Is Literary History Pos-sibleP, 37-38.

9 Eliot, Selected Essays, 5; cf. Uhlig, CL 42 (1990) 193-207. A similar point is made somewhat more obscurely by Slawinski, NLH 19 (1988) 526.

10 "There is no progress, no development, no history of art except a history of writers, institutions, and techniques. This is, at least for me, the end of an illusion, the fall of literary history." Wellek, The Attack on Literature, 77.


accurately, though even that is no easy task. The notion of change includes not only the process by which variations develop in a genre as authors expand and extend its conventions but also those striking innovations that, at least in retrospect, can be seen to have redefined a genre's conventions and reshaped its message.11 Attention to such distinctions enables literary history to transcend the sequential record of individual achievements and become a coherent account of the relationships and processes which explain that record. A literary history of this kind can, for example, help us understand in cultural—as distinct from purely aesthetic—terms the significance of Cicero's failure as an epic poet and the uniqueness of Vergil's success. Identifying the patterns of change created in the course of epic's development is an essential step in constructing such an argument. This means tracing the genre's evolution, but in doing so we must take care not to mistake evolution for some things it is not.

Evolution is not synonymous with improvement. Cicero himself worked hard, though somewhat disingenuously, to keep the evolu-tionary argument of Brutus on a technical level. When the "Atticus" of his dialogue eventually points to the absurdity of equating Cato and Lysias in absolute terms, Cicero acknowledges the objection only to set it aside. He knows that the introduction of aesthetic values would confuse, if not derail, the evolutionary scheme.12 We too need to avoid comparative judgments of value. Not only are they too subjective, too limited by our own expectations, but they have too little explanatory power. The awareness of Vergil's technical superi-ority to Ennius in the writing of Latin hexameters, for example, neither diminishes nor explains the significance of Ennius' achieve-ment, though there is obvious value in tracing the stylistic inno-vations that enabled Vergil to set new standards for hexameter prosody. Changing styles and changing tastes may introduce improve-ments, but those changes are not in themselves the same as improvements. As Eliot rightly claimed in that same famous essay, "Art never improves, but . . . the material of art is never quite the same" (6). We should not confuse aesthetic and technical judgments. Nor should we assume that development is strictly linear. Evolution

11 I have borrowed this formulation from Cohen, "Innovation and Varia-tion/' 3-6, a discussion with roots in genre theory. The Rezeptionsästhetik of H. R. Jauss treats a similar problem in terms of changing readerships. Todorov, in Ducrot and Todorov, Encyclopedic Dictionary, 144-49, sees change as the central problem for literary history,

12 Brut. 293-94, cf. 68-69. The aesthetic question is deferred indefinitely as a res nova disputatione digna (297). Quintilian (12.10.7-9) should probably have done the same. When he tried adapting Cicero's artistic analogy to aesthetics, it of course broke down. See Douglas, ANRW 1.3 (1973) 108-15.

Ruins 9

is not the same as chronology. Even the Russian formalists had to modify their distinction between synchronic and diachronic as they came to recognize that no literary development is entirely indepen-dent of time. Synchronic systems can also be said to evolve, for they draw on works outside the present.1 3 Evolution, as a force for change, is independent of the chronological periods that impose a superficial order on literary history.

A n d finally, evolution must not become an excuse for ignorance or inattention. Eliot as poet comments on this tendency in his Four Quartets.

It seems, as one becomes older, That the past has another pattern, and ceases to be a mere

sequence— Or even development: the latter a partial fallacy Encouraged by superficial notions of evolution, Which becomes, in the popular mind, a means of disowning

the past. ("The Dry Salvages" II)

Classicists should be particularly wary of disowning the past, yet the more fragmentary its record, the more easily it seems to slip from our consciousness. L. P. Wilkinson, for example, begins a discussion of what he calls the "architectonics of verse/' the way meter is used to enrich sense in Latin poetry, with the following statement:

Early Latin verse, as represented by Ennius, seems innocent of ar-chitectonic structure, whether in the management of long sentences or in the disposition of short ones (though few of the fragments are long enough to enable us to judge). The poet gives the impression of versifying what he has to say just as it occurs to him.14

13 See Tynjanov and Jakobson, "Problems in the Study of Literature," 79-80: "Pure synchronism now proves to be an illusion: every synchronic system has its past and future as inseparable elements of the system. . . The opposition between synchrony and diachrony was an opposition between the concept of system and the concept of evolution: thus it loses its importance in principle as soon as we recognize that every system necessarily exists as an evolution, whereas, on the other hand, evolution is inescapably of a systemic nature." Wellek offers a lucid survey of attitudes toward literary evolution in his Concepts of Criticism, 37-53. See also Striedter, "Zur forma-listischen Theorie der Prosa," lx-lxx, and Jauss, Toward an Aesthetic of Reception, 16-18.

1 4 Wilkinson, Golden Latin Artistry, 189. Note that Wilkinson's paren-thetic disclaimer does not keep him from making this judgment. By confining himself to description of the phenomena, his predecessor Norden avoided such judgments entirely in his discussion of meter and meaning: cf. the presentation of data appended to his edition of Aeneid 6, 391^404.


Wilkinson was a particularly keen and sensitive reader of Latin, and yet this opinion is contradicted by many fragments of the Annals. One obvious example:

pellitur e medio sapientia, vi geritur res; spernitur orator bonus, horridus miles amatur

good sense is driven from the field, by force is business done; the honest orator is spurned, the uncouth soldier loved; (248-49)

Ennius says nothing "just as it occurs to him," though he may some-times create the illusion of spontaneity. This particular fragment shows his careful attention to word choice and his effective use of parallelism and chiasmus, not to mention his balancing of dactylic cola. This is not Vergilian architecture, but it is certainly archi-tecture of a kind, and Wilkinson was wrong to ignore it. Fascinated as he was by the beauties of "golden Latin artistry," he simply ne-glected to read early Latin authors with the same care he gave their successors. He does not argue his view of Ennius as he will go on to argue the case for Vergil. He simply assumes it. What he has to say about Augustan verse is no less true for this omission, but his discussion is less complete and his understanding of the phenomena less full than the available evidence would allow.

A different attitude, but a similar result, characterizes Gian Biagio Conte's more conscious appeal to the literary past. Vergil's problematic relation to what Conte calls "the epic code" is central to his reading of the Aeneid, and his critical instincts are often acute. Few would doubt his claim that the content of pre-Vergilian epic, with its tendency to cast Roman achievements in Homeric terms, "should, in particular, be identified with the supremacy of the state as an embodiment of the public good, with the acceptance of divine will as providential guidance, and with the historical ratification of heroic acts."15 The very titles Bellum Punicum and Annals suggest as much. If there is little in this claim to doubt, there is also little here on which to build. Conte's preoccupation with critical constructs keeps his discussion at some distance from the texts needed to define the old epic code to which he appeals. The two fragments of Naevius and five lines of Ennius that follow this generalization elicit only a theoretical discussion of the "gnomic terseness" that is really just the natural consequence of the conditions that preserved them. Neither the Bellum Punicum nor Annals is discussed as a poem or its fragments

1 5 Conte, The Rhetoric of Imitation, 144. The original passage occurs in Virgilio, 58-59. Conte's approach is beginning to stimulate productive debate, especially in Britain. See, for example, Fowler, PCPS 36 (1990) 42-63.

Ruins 11

as poetry. Conte's explicit refusal to define either the code or its norm reflects his tacit refusal to analyze pre-Vergilian poetry. He is more sensitive than Wilkinson to the abiding influence of early Roman precedents, but he too reads later texts with only presuppositions about the earlier ones. While his allusions to the past are welcome, they are still only allusions. His discussion remains less explicit, and therefore less enlightening, than the available evidence actually allows.

This tendency to hurry over the evidence of fragments is in part a function of the interpretive difficulties they present. These can be considerable, and also discouraging. The seven books of Naevius' Bellum Punicum, for example, had the power to inspire Vergil and annoy Horace, but only some sixty awkward and unattractive frag-ments of those books survive.16 Quotations range from three lines down to single words; the indirect tradition provides only a few al-lusions to the poem's content. The remains of Andronicus' Odussia are even scantier and offer no indication of when or why the work was undertaken. The ancient scholars and antiquarians who cite the old epics are rarely interested in the questions that concern modern critics. Yet the technical problems inherent in the evidence as pre-served are not the chief obstacles to understanding these poems. Their reputations suffer most from the persistent habit of treating them in retrospect, of making them ancillary to larger, later interpretations of someone or something else.

This habit in turn introduces a serious distortion because we cannot understand something as extensive and long-lived as the Roman epic tradition through the practice of Vergil, its most original member. Change may attract our attention, but commonality, not innovation, defines a tradition. The past, as Eliot says, is not merely a sequence of changes; it is also a set of continuities. "Evolution," rather than simply "change," is a more appropriate term for the alterations that occur in literature over time, not for evolution's biological associ-ations—I do not intend to cast Andronicus as the Neanderthal to Ennius' sapiens—but because an evolutionary scheme acknowledges the underlying pattern of continuity, as opposed to disjunction, in the phenomena. The Roman epic poets were, for all their differences in subjects and forms, united in a common enterprise by their dedication to a discernible set of values and techniques. The literary history I want to write is concerned with the emergence of those continuities and their appeal to poets and readers alike. It therefore requires not just the record of individual achievements but also the discovery of patterns in the growth and development of the literary form we

16 For the Augustans' reception of Naevius, see White, TAP A 117 (1987) 227-34, and Wigodsky, Vergil and Early Latin Poetry, 22-39.


recognize as Roman epic. To perceive these patterns, we need a close reading of the fragments, and this in turn means reconsidering both our approach to them and the context we build around them.

II

Fragments deserve to be read with sympathy, but too much sympathy may introduce dangers of its own. Fragmentary poems can be beguiling and treacherous allies, for they easily generate a meaning and appeal independent of their original context. F. W. Bateson once illustrated this phenomenon by culling a "fragment" from the opening sentence of Shakespeare's seventy-third sonnet:

That time of year thou mayst in me behold, When yellow leaves, or none, or few, do hang Upon those boughs which shake against the cold, Bare ruined choirs, where late the sweet birds sang.

Bateson claimed that the beauty of the famous fourth line actually grows with detachment from its context. "Grammatically," he observed, "the ruined choirs are simply a metaphoric extension of the trees' leafless boughs; it was on such boughs—not in the choirs—that the birds once sang. The line once detached from its linguistic context derives its pathos, however, from the birds' exclusion from a church or chapel that is now a ruin. This is not what the sentence says; it is what the line says, defying what the earlier lines want it to say." He follows this attack on context with a swipe at historicism: "And whether the choirs' ruin are an after-effect of the dissolution of the monasteries or of capitalist sheep-farming is wholly immaterial."17

Bateson is surely right to observe that the isolated line may suggest a different grammar—and so a different sense—from what the complete sentence requires. Whether we assume that the birds and/or the choirs are metaphorical or we wonder why real birds left their perch in a real ruin, we cannot determine the line's grammatical structure or restore the larger sense of which it was a part. If we take the whole sentence for our "fragment," the grammar at least becomes clear and we can admire how the old age-as-winter conceit develops into the poignant, physical image of a bleak monastic ruin. Yet the thought remains incomplete. We still cannot see where this meta-phor is heading, even if we know these are lines from a sonnet and are

17 Bateson, NLH 2 (1970) 121-22. Bateson here means to illustrate his belief, which I do not share, that literary history is "merely . . . a disreputable though not entirely useless by-product" produced by confusing historical and literary studies. The illustration is more compelling than the contention that engendered it.

Ruins 13

familiar with that form's conventions. A leap to the poem's actual conclusion—"to love that well which thou must leave ere long"— remains beyond our power. Only some philological observations are certain; that the line is indeed an iambic pentameter, that these choirs are not choruses, and that "sweet" describes the birds' song, not their succulence on a plate.

The Latinist facing true fragments must often wrestle with even greater uncertainties. Not only are grammar and syntax likely to be obscure, but the words may themselves challenge our vocabulary. Architectural images, whether factual or metaphoric, may be hard to recognize or may require special knowledge to understand. After we have gathered what philological help we can, the process of interpretation may still lead with disconcerting speed from fact to conjecture. Naevius' epic, for example, included the following description:

inerant signa expressa, quo modo Titani, bicorpores Gigantes magnique Atlantes Runcus ac Purpureus, filii Terras

Figures were represented on it, how the Titans, double-bodied Giants and great Atlases, Runcus and Purpureus, sons of Earth (fr. 4)

A work of art is apparently being described in a series of phrases that, as Eduard Fraenkel demonstrated, seem to stand here in apposition rather than sequence. Two figures named Runcus and Pur-pureus are portrayed in some action, presumably the Gigantomachy. That much is very likely, but what kind of work displayed this Gigantomachy, where it was located, and how its description fit into Naevius' poem are not questions to be answered with comparable certainty. They can only be subjects of speculation, though the specu-lation may still be worth making. The odd word "Atlantes," for example, reminded Hermann Fränkel of the stone figures called ατλαντες which were a famous architectural feature of the Temple of Zeus at Agrigentum. The building had a Gigantomachy on its east pediment, and Fränkel therefore thought that Naevius was de-scribing this sculpture.18 This identification, as we shall later see, has important ramifications for understanding the poem's structure. It is a fine but also typical suggestion: learned, appealing, and ulti-mately unverifiable. No Shakespearean scholar faces quite this level of difficulty: our choirs are in every sense a ruin. They and their

1 8 Fraenkel, J RS 44 (1954) 14-17; Fränkel, Hermes 70 (1935) 59-61. The lines are cited without context by the grammarian Priscian (GL Κ 2.198) to illustrate the archaic genitive in -as.


vanished singers require especially close attention and great care. What principles should guide our approach to the alluring but potentially deceptive remains of their music?

Editors and commentators generally adopt a technical approach, as did Bateson when he wrote on the history of English poetry. Even while demonstrating that poetry's appeal can transcend and even subvert context, Bateson maintained that philological understanding provides a necessary check on the errant subjectivity of literary criticism. He was an active participant in the controversy between "scholars" and "critics" that polarized midcentury students of litera-ture, and his spirited exchanges with F. R. Leavis over the relevance to literary analysis of such facts as the eighteenth-century conno-tations of "vassal" in Alexander Pope's Ditnciad remain landmarks of a kind.19 Classicists have never really confronted the issue that engaged Bateson and Leavis so passionately. Our philological train-ing ensures the priority of technical over literary concerns, but we do need to realize that neither philology nor history suffices to answer all the questions worth a reader's attention.

Consider, for example, the following line of Ennius, which describes the allocation of consular provinces in 200 B.C.:

Graecia Sulpicio sorti data, Gallia Cottae Greece went by lot to Sulpicius, Gaul to Cotta (324)

In annotating the line, Otto Skutsch pointed out that the provinces in question were more correctly styled Macedonia and Italia. Ennius, he suggests, altered the names to boost the political standing of Aurelius Cotta, who had claimed credit for a campaign that year in Gaul. Skutsch devotes a long note to arguing the point in detail by tracing a complex scheme of dynastic alliances and obligations involving Ennius and his "patron," Fulvius Nobilior. To the literary critic more interested in the alliterated parallelism this change of names created, Skutsch the philologist makes a technical response: "People of the old school who still compose verse know that practically the only places where a dactylic word ending in a vowel can be placed are the first and fifth foot, though there are a few exceptions in Ennius.

19 A representative exchange between Bateson and Leavis is reprinted in Leavis, Λ Selection from Scrutiny, vol. 2, 280-316; the debate between Rosamund Tuve and William Empson over George Herbert's Christian imagery is even more famous. Cf. Kenyon Review 12 (1950) 51-75 and 735-38. For the general context of these debates, see Graff, Professing Literature, 183-94, and for their relevance to literary history, Weimann, Structure and Society, 33-46. Reviews and occasional essays in the early volumes of Arion are as close as classicists have come to articulating similar problems.

Ruins 15

So where is Ennius' art in Graecia (1) and Gallia (5)?"20 The art, we may reply, is in the word choice and the repeated sounds and balanced cola it produces. The metrical pattern, as consequence rather than cause of that choice, creates a sense of inevitability or naturalness that enhances rather than diminishes the art. Aesthetic, historical, and technical considerations all support a single conclusion: the construction of the line is no accident of its meter.

By focusing on individual words, Skutsch demonstrates that Ennius' effects are the result of a calculated choice, but his interest as a philologist does not extend to the literary implications of that choice. While philology's attention to detail may clarify the ele-ments of poetic construction, it neither encourages nor equips its practitioners to pass judgment on their effects. Other kinds of know-ledge must then be supplied. We routinely import such external knowledge in explicating complete texts. When the reluctant Anchises of Aeneid 6 finally explains, "tu Marcellus eris," readers are not satisfied with the identification of this mournful shade as Augustus' ill-fated nephew, nor are they distracted from further queries by the poignant tale of Octavia's faint on hearing Vergil read these lines. Once the literal meaning is established, we naturally want to consider its wider significance: we start to wonder why Vergil caused this contemporary disappointment to intrude on his mytho-logical scene. The historical detail only raises a larger interpretive question that technical knowledge alone cannot answer.21 We simultaneously incorporate the insights of philology and history and move beyond them.

Though the study of fragments should be no less demanding in its expectations, it remains a singularly conservative and limited specialty. Philology has proven to be an intimidating, as well as enabling influence. When evidence is scattered, the pieces small and the lacunae great, it is often thought safer to focus on narrow problems of meaning. Ennius' Annals have escaped the worst effects of this restricted vision. Though the historical explication of his poem may sometimes fail to satisfy, its many points of contact with Homer and Vergil have always invited serious literary analysis. Its prede-

2 0 Skutsch, letter dated 29 March 1990. The observation came with a foot-note: "Here is some detail: in the 300 lines of Verg. Aen. I, IV and VI 1-100 there are 51 dactylic words in the first foot, one (I 96) in the second foot. In the same lines there are 91 dactylic words in the fifth foot, two in the fourth foot (I 40; IV 37)."

2 1 Scholarly agreement extends only to the common interest in these larger questions of significance. Styles of argument and standards of proof vary. On Marcellus and the procession of Aeneid 6, contrast Johnson, Dark-ness Visible, 105-11, and Feeney, PCP S 212 (1986) 1-24.


cessors in the epic tradition, however, have not been even this fortunate. While many scholars puzzle over the details of Naevius' life, the structure of his poem, and the oddities of its meter and morphology, few have attempted to read the fragments of the Bellum Punicum as literature or to consider their significance in the context of Rome's cultural and literary development. Important ques-tions therefore remain unexplored. No one seems willing to ask why Naevius wrote an epic, why Roman readers supported his effort, and what later writers made of the attempt. These and many similar questions about Naevius and his successors are well worth asking, but answers will only come by putting the fragmentary evidence to further uses. A literary history focused on the process of continuity and change that determined the genre's evolution should be able to get beyond the facts these fragments represent to the aesthetic and cultural conditions that produced them.

Widening the field of inquiry to consider problems of cultural context means taking a few risks. There remains practical criticism to be done, and also something more. We must add to our heritage of precise reading a kind of imaginative sympathy that has grown increasingly remote from the modern application of philological method. To give old poetry its due, to recapture something of the appeal it once generated, we must begin thinking more seriously and more innovatively about the conditions that produced it. Our grasp of details should encourage us to trust our instincts and our feelings for their significance.

There are certainly pitfalls in doing so, especially when we move beyond the reading of purely literary texts. The historical evidence we may need to consider is itself fragmentary, and these fragments are at least as challenging as their literary counterparts. Take, for example, the case of Ennius' portrait bust. According to Cicero, a marble statue representing Ennius may have adorned the tomb of the Scipios, and among Livy's comments on the confused traditions sur-rounding Africanus' death is passing mention of a monument there to the poet. Valerius Maximus and the elder Pliny report that a statue of Ennius was placed by the tomb on Africanus' instructions.22 Cicero, in quite a different context, also preserves an epitaph of the type that would have graced such a statue.

22 Cic. Arch. 22: "carus fuit Africano superiori noster Ennius: itaque etiam in sepulcro Scipionum putatur is esse constitutus ex marmore." Liv. 38.56.4: "Romae extra portam Capenam in Scipionum monumento très statuae sunt, quarum duae P. et L. Scipionum dicuntur, tertia poetae Q. Ennii." The other main sources are Val. Max. 8,14.1 and Plin. NH 7.114. For full texts and discussion, see Suerbaum, Untersuchungen, 210-12.

Ruins 17

Aspicite, ο cives, senis Enni imaginis formam: Hie vestrum panxit maxima facta patrum.

Gaze, fellow citizens, on aged Ennius' form and image: He set to verse your fathers' greatest deeds. (TD 1.34)

Uniting these reports of statue and epitaph is very tempting, es-pecially when we recall that Ennius wrote a trochaic poem called Scipio, with features akin to a triumphal song. In addition to extolling Scipio's military successes (fr. 32, 33 Courtney), the work also reflected his carefully cultivated public image. Thus, as Livy reports in that same context, Africanus made a point of refusing permission for statues to be erected in his honor (38.56.12), and a fragment of Ennius, restored and assigned to the Scipio, apparently alludes to this fact:

<o tum> quantam statuam faciet populus Romanus <tibi, Scipio>, quantam columnam, <claro> quae <praeconio> res tuas gestas loquatur.

Then what a statue the Roman people will erect for you, Scipio, what a column, to speak with renowned celebration of your deeds. (Scipio, no. 30 Courtney)

The poem may also have included a sympathetic account of Afri-canus' speech during the tumultuous trials o f 187.23 Small wonder, then, that the beleaguered aristocrat might raise a statue to Ennius, a rare distinction. He was linking his own honor with that of Rome's leading poet and, as it happened, thereby leaving to posterity strik-ing and provocative evidence both for poetry's prestige by the 180s and for its ties to the highest levels of Roman society.

Closer inspection of the evidence, however, does not support this reading. An analysis of this kind is reconstructing as well as inter-preting history, and the reconstruction here is particularly dubious. Cicero and Livy, our earliest sources for the story, are only reporting hearsay (putatur; dicuntur). They are not themselves certain that the statue in question was of Ennius, which strongly suggests that it did not display the epitaph Cicero quotes in his Tusculans. Valerius and Pliny may have linked the two reports and identified Ennius' reference to maxima facta patrum with Scipio's particular achieve-ments, but we lack justification for following their example. We cannot even be sure that a portrait bust ever stood in the tomb.

2 3 Cf. no. 34 and the discussion in Courtney, Fragmentary Poets, 26-30. For the form and purpose of the Scipio, see Scholz, Hermes 112 (1984) 183-99. An elegiac elogium for Africanus is also ascribed to Ennius (Courtney, Fragmentary Poets, 39-40).


Neither Cicero nor Livy actually saw it, though Livy personally inspected a rival monument at Liternum. Nor was such a bust, if it existed, necessarily a rare honor. Privately erected statues were becoming increasingly common in second-century Rome, and not just in tombs. The whole city was growing crowded with such monuments. In 179 the censors had to remove statuary cluttering Jupiter's temple on the Capitol, and in 158 they took from the forum ail the statues of magistrates that were not authorized by the senate or the people.24

Both the story of the bust and its interpretation may thus be based on fiction and misunderstanding, and with that fiction goes our witness to the mutual self-congratulation of Ennius and Scipio. We find our-selves, like Aeneas in the underworld, embracing an appealing shadow that only dissipates at our touch.

The moral, however, is not to avoid historical evidence, but to concentrate instead on other aspects of the historical record. Incidents like the curious affair of Numa's books in 181 and Anicius Gallus' triumphal games of 167 provide a different kind of insight into the cultural dynamics of Roman society. Events of this type will offer broader scope for interpretation without requiring such precarious reconstruction of the necessary facts. The interpretive enterprise may still require the play of imagination that rhetoricians called phantasia—the power of creative visualization—but the resulting arguments will not lack controls of their own.25 The historical record so essential to philology continues to be our ally, but we can aspire to produce something more than a set of chronological and prosopo-graphical notes. Taking account of history will mean looking up from our texts and considering the dynamics of the culture that fostered their creation. This requires a partnership between literary and historical inquiry that is, if not entirely new, at least a little dif-ferent from the orientation that has come to characterize the philological study of early Latin literature.

Grounds for this partnership lie in the imaginative faculty our two disciplines share. Historians have long appreciated the explanatory

24 Liv. 40.51.3; Plin. NH 34.30. For the growing popularity of honorific statues in this period, see Gruen, Culture and National Identity, 118-23. The proemium of Annals 16 apparently disparaged the fame associated with statues and tombs (404-05). A contrast with verse's power to immortalize its subjects probably followed.

2 5 For φαντασία and its cognates, which include ένάργεια, demonstratio, and repraesentatio, see Kroll, "Rhetorik," RE Suppl. 7 (1940) 111-12. I do not mean by this borrowing either the engaged fictions of Lembke, Bronze and Iron, 3-78, or the enargeia turned by a sophistic twist into the "social energy" of Greenblatt's Shakespearean Negotiations, 5-7, but something closer to the imaginative power introduced to historiography by Vico. See Berlin, Vico and Herder, 107-14.

Ruins 19

power of imagination. R. G. Collingwood once analyzed its role by making a distinction between what he called the outside of an event, by which he meant the physical action (Caesar crossing the Rubicon), and its inside, which he called "thought" (Caesar's reasons for doing so). The historian's work, he wrote, "may begin by discovering the outside of an event, but it can never end there; he must always re-member that the event was an action, and that his main task is to think himself into this action, to discern the thought of its agent."26

Collingwood went on to separate the historian's exercise of imag-ination from the novelist's by appealing to separate realms of fact and fiction, but that distinction has become increasingly difficult to maintain. Modern historians are now more inclined to acknowledge the subjective qualities of their discipline. Hayden White therefore puts the interpretive act that Collingwood's "thought" implies at the very center of the historian's task. "A specifically historical in-quiry," he writes, "is born less of the necessity to establish that certain events occurred than of the desire to determine what certain events might mean for a given group, society, or culture's conception of its present tasks and future prospects."27 In practice, this can mean stretching the historian's traditional generic boundaries to an extra-ordinary degree. Thus Simon Schama, who had previously likened history written without the play of imagination to digging in "an intellectual graveyard," recently confounded (and sometimes an-noyed) reviewers by following his lively narrative history of the French Revolution with a book that is at times very like a historical novel.28

If these developments make history start to read once again like literature, a complementary trend has been making literary study seem increasingly like history. Stephen Greenblatt has made much of what he calls the "permeability" of the boundaries between history and literature and in the process launched a New Historicism that subjects historical material to the techniques of literary criticism.29

This approach, which originated in the study of Renaissance drama and now ranges far beyond it, is not without certain difficulties of

2 6 Collingwood, The Idea of History, 213; cf. his remarks on the historical imagination, 240-49.

2 7 White, CI 12 (1985/86) 487 (his emphasis). He treats history throughout as a form of literature, a position congenial to classicists but rather less so to many historians. Novick, That Noble Dream, 598-607, places White within the context of the historians' debate.

2 8 I allude to Schama, Citizens, from whose preface I quote, and Schama, Dead Certainties, which recreates the 1849 murder of Dr. George Parkman by John W. Webster, Erving Professor of Chemistry at Harvard.

2 9 Greenblatt, Shakespearean Negotiations, 95.


conception and execution. The method currently in vogue is easily parodied and not above a kind of antiquarian voyeurism; its fondness for the telling anecdote and the obscure document can become, as a recent critic complained, "wayward, eccentric, and unhistorical." Nor is what Greenblatt calls "social energy" really so different from what earlier (and now less fashionable) critics called Geist.30 There is also an unfortunate inclination to politicize the critical process, which sometimes leads to tendentious and jejune declarations well beyond the traditional sphere of literary discourse: literary critics do not become competent historians simply by writing about historical texts, nor do they become social theorists because they talk about literature in cultural terms.31 The most interesting challenge the method raises, however, arises from the developing awareness among both historians and literary critics that history is itself a critical construct. It does not represent an external, objective reality which literary works reflect or against which they can be measured. New Historicists necessarily invent the history they interpret. "History," notes a tolerant, but doubting observer of recent developments, "speaks in our voice. History does not tell us what the text is, because we decide what history is, and then put history into the text, rather than the other way around."32

Classicists, who were never really estranged from the old historicism, have been understandably skeptical of this new incar-nation. Its ideological fits and idiosyncratic jerks are not in our usual style. The current reawakening of interest in historicism as a critical orientation nevertheless has important things to teach us. The oc-casional excesses of New Historicism aside, there is something to be said not just for treating history and literature as part of a larger cultural record, but for reading the events of history as we read texts: to analyze, to imagine, and above all to wonder at the significance of recurring themes in history as we already do in literature. It is precisely because history is itself an imaginative construct that we should not hesitate to follow Greenblatt's lead in applying the techniques of literary criticism to historical evidence. Doing so does

30 The critical quote is from Romany, Essays in Criticism 39 (1989) 287; cf. Porter, NEH 21 (1990) 253-72, and Perkins, Is Literary History Possible?, 148-52. For Geistesgeschichte, see Wellek, Theory of Literature, 119-24, and for its reincarnation among New Historicists, Liu, ELH 56 (1989) 740-45.

3 1 The idiosyncratic arguments, if not the self-confident tones, of the essays in Veeser, The New Historicism, suggest as much. Pechter, PMLA 102 (1987) 300-301 acquits the New Historicists of playing at politics, but the role of academic politics in their work is something else again. See the general comments of Fromm, Academic Capitalism and Literary Value, 1-13.

32 Pechter, PMLA 102 (1987) 298. Cf. Perkins, Is Literary History Possible?, 136: "When used to interpret literature, context is itself an interpretation."

Ruins 21

not make us historians, but there is much to be gained from seeking answers in the historicai record to questions posed by literary works.

This is hardly a new idea. Historical analysis has never been foreign to the study of ancient literature, nor is it employed simply to curb subjective and anachronistic readings. Vergilian criticism again provides a case in point. History and literature quickly find common ground in Vergil because of the way he echoed contemporary concerns in his poetry. Not only can historians furnish such necessary details as the identification of young Marcellus, but their interpretation of historical events inevitably influences readings of the poetry itself. Sir Ronald Syme, for example, considered Vergil's poetry a glorifi-cation of the new Principate; more recent critics, less enamored of the new order but equally fond of Vergil, find indications of the same tur-moil and doubt in his poetry that they see beneath the tranquil surface of the pax Augusta. Where Syme saw values confirmed, others now see them subverted. The Marcellus episode is increasingly read this way, and even purely legendary events can be fitted to the new, less hopeful pattern.33

The nighttime adventure of Nisus and Euryalus in Aeneid 9 is one of those programmatically dark episodes that so fascinate modern critics. Events take an ugly turn when Ascanius, playing at kingship in his father's absence, sends the pair off on their willful, bloody, and useless mission. Vergil's reworking of the Homeric Doloneia gives the episode a new element of pathos, and also a suggestion of horror as initial success leads only to eventual destruction.34 They die a futile and frightening death, but Vergil nevertheless provides these rash young men with a grand epitaph.

Fortunati ambo! si quid mea carmina possunt, nulla dies umquam memori vos eximet aevo, dum domus Aeneae Capitoli immobile saxum accolet imperiumque pater Romanus habebit.

Fortunate pair! If my song has any power, no day will ever wrest you from the memory of time, so long as children of Aeneas ring the Capitol's unshaken rock and a Roman father maintains sovereignty. (9.446-49)

3 3 The critical issues are nicely drawn by Johnson, Darkness Visible, 8-16. Cf. Syme, The Roman Revolution, 459-68 and Williams, "What is Hap-pening to Interpretation of Virgil's Aeneid?"

3 4 Not everyone reads the episode this way. See Pavlock, TAPA 115 (1985) 207-24, now restated in her Eros, Imitation, and the Epic Tradition, 87-112, which includes references to the full range of scholarly opinion. Vergil's de-partures from Homer are discussed by Heinze, Vergils epische Technik, 216-19.


This coda, however ironic, is also traditional. Critics rightly hear in it echoes of Ennius, who added a sixteenth book to his Annals to immortalize the exploits of two valiant brothers.35 Epic conventions as well as epic values are being invoked and perhaps called into question. Yet the shadow of Ennius here brings us back to the use of fragmentary evidence and raises an important point of critical method. Reading irony into Vergil's lines depends on a "dark" read-ing of the entire Nisus and Euryalus episode. Had this epitaph sur-vived without the details of the preceding narrative, would we ever suspect Vergil of questioning traditional sensibilities? The words alone seem to confirm, not challenge, what Conte calls the epic code. The desire to read the passage ironically originates in a complete text and a heightened sensibility to the excessive furor critics now find throughout Vergil's battle scenes. If the picture were not so full— if we, like F. W. Bateson, turn the poet's complete thought into a fragment—would we still dare to question the ostensible meaning of his words?

It is difficult to judge the tone of Vergil's Doloneia solely from al-lusions to it in the secondary tradition. Though Servius comments extensively on Nisus and Euryalus, he never remarks on the grotesque slaughter and greed for booty which make us so uneasy. Macrobius shows that Vergil turned Homer's corpse-strewn battlefield into a beach full of sleeping drunks (Sat. 5.9.7-9), but while that alteration may give us pause, its significance would be difficult to determine without context. Only one other hint of trouble survives among the grammarians. The passage occurs early on, when Ascanius promises the pair rewards out of proportion to their stature and beyond his power to bestow.

bina dabo argento perfecta atque aspera signis pocula, devincta genitor quae cepit Arisba, et tripodas geminos, auri duo magna talenta, cratera antiquum quern dat Sidonia Dido.

I will give two silver cups with figures in relief, which my father captured when he took Arisba, a pair of tripods, two great golden talents, and an ancient bowl which Sidonian Dido gave, (9.263-66)

Ascanius exhibits the same boyish vainglory that will soon destroy Euryalus: the list of riches is evocative and telling. Nonius (366L),

35 Mazzoli, Athenaeum 42 (1964) 331-33. Klingner, Virgil, 561-64, is more sentimental than ironic about this epitaph, as was Servius ad loc.: "mire horum mortem non luctu aut misericordia, sed felicitatis testimonio pro-sequitur."

Ruins 23

however, quotes the passage only down to pocula, omitting the relative clause that first suggests Ascanius' excess, while Servius, who preserves that tell-tale clause as a lemma, was too interested in the origin of the cups to query Ascanius' right to dispose of them. Servius himself found nothing amiss with these rewards. Given only this testimony, we too could note the odd meaning of aspera and explain Aeneas' complex dealings with the town of Arisba without ever questioning the appropriateness of Ascanius' promise. Though the ancient testimony provides vague hints of oddity in this episode, it does not encourage speculation over them.

There is thus no proof of irony here, and silence may seem the prudent course. Yet that is not a course free of its own methodological problem. Though a refusal to interpret may save us from risk, it does not necessarily save us from error. A wrong interpretation of the "fragment" may distort the record, but no interpretation certainly distorts it. A bare outline of the narrative without inquiry into its tone offers scant indication of its significance. We miss an important piece of evidence if we fail to notice that Vergil's gratuitous inclusion of this episode creates an opportunity to question the heroic values on which the entire poem is based. Is it therefore irresponsible to inter-pret as well as annotate a fragmentary text? Consider a second Vergilian "reconstruction."

Nonius Marcellus illustrates the verb parcere in the sense "to save" with another quotation from the Aeneid:

PARCERE, servare. Vergilius lib. X: argenti atque auri memoras quae multa talenta, natis parce tuis.

PARCERE, to save. Vergil, book 10: save for your sons those gold and silver talents that you mention. (590L)

This sounds like the refusal of a bribe or ransom, and since Book 10 is a battle book, we may surmise that a ransom is at issue. The circum-stances would remain unclear until we noticed Macrobius' observation that Aeneas refused Magus' offer of ransom just as Achilles, still grieving for Patroclus, refused the pleas of Lycaon in Iliad 21:

et cum ille genua amplectens supplex vitam petisset, respondit, . . . belli commercia Turnus

sustulit ista prior iam turn Pallante perempto. sed et insultatio Achillis in ipsum Lycaonem iam peremptum in Tarquitium a Marone transfertur..

When the suppliant, grasping his knees, begged for life, he replies, Turnus swept away


the bargaining of war at the moment Pallas was destroyed. So Achilles' taunt to Lycaon is borrowed and transferred by Vergil to Tarquitius. (Sat. 5.2.16)

Macrobius' quotation appears both to complete the line provided by Nonius and to supply a context for it. The temptation to join them is irresistible—and correct. They are in fact the lines we know as Aeneid 10.531-33. We might still jump to some wrong conclusions about Pallas—no ransom was involved in his death—but that is a secondary consideration. We have at least restored three continuous lines of Vergil's "lost" poem and supplied a plausible context for them.

This would itself be an advance in knowledge, and editors of Naevius—who rarely see three continuous lines intact—might well rest content with such a discovery. Yet a further question about our Vergilian "fragment" is also worth asking. Can we not explain this pitiless streak in Aeneas? Is it simply an explicit Homeric borrowing in another distinctly Iliadic context, or is it meant to give us pause, creating another of those dark Vergilian moments? An answer to this interpretive question is within our grasp, for something else survives to tell us.

Aeneas' refusal of ransom with a scornful reference to the belli commercia recalls a famous fragment of Ennius in which Pyrrhus spurns a Roman attempt to rescue captives after the battle of Heraclea (183-90 S).36

Nec mi aurum posco nec mi pretium dederitis: non cauponantes bellum sed belligerantes ferro, non auro vitam cernamus utrique. vosne velit an me regnare era quidve ferat Fors virtute experiamur, et hoc simul accipe dictum: quorum virtuti belli fortuna pepercit eorundem me libertati parcere certum est. dono—ducite—doque volentibus cum magnis dis.

Neither do I ask for gold nor offer me a ransom: not touting war but waging it with steel, not gold let us run our risks. Whether you shall reign or I—whatever Fortune wills— we shall test by strength, and take this as a fact:

36 The association of Vergil's passage with this fragment finds further support in Servius ad 10.532, who observes that parce in Vergil's sense is also found "apud Lucilium et Ennium." It occurs in line 188 of this fragment. See Skutsch, Annals, ad 184, who also notes that cauponantes appears to be a caique on καπηλεύω. Such formations are characteristic of the early poets, who used them to build an exotic epic vocabulary. Vergil's much easier word again puts distance between him and them.

Ruins 25

whose strength the chance of war has saved, their iiberty I certainly do save. I grant them—take them—I give them up as is the great gods' will.

These lines are cited by Cicero in De Officiis with appreciative comments on their nobility of sentiment ("regalis sane et digna Aea-cidarum genere sententia," 1.38). Pyrrhus' dismissal not just of the ransom but of the very idea of ransom excites Cicero's admiration. In thrusting Aeneas back to an Achillean viciousness, Vergil has separated his warrior from the grand and regal attitude in Ennius that he has himself taken pains to recall. Aeneas' conduct is thus at odds with its Roman epic exemplar and with the praise that a reader like Cicero heaps upon it. Good grounds may perhaps be found within the context of the Aeneid for Aeneas' fighting frenzy, but his cruelty toward Magus explicitly sets him apart from his predecessor in the Roman tradition. It is indeed a dark, or at least problematic moment in the Aeneid. The "fragmentary" record here does not deceive. It needs only to be read with the requisite imagination.

We shall need to focus comparable powers of imagination on the true fragments of Vergil's predecessors, which means bringing them out from behind his shadow. Our standard histories too often subject these poets to a teleological bias, judging them against the greatness Vergil achieved as if their only significance lay in establishing the standards he would surpass. This tendency seriously—and un-necessarily—distorts the record. The history of Roman epic should not be read as a story of growth and decline, as if the tradition culminated with the Aeneid and might better have ended with it. To admire the subtlety of Vergil's hexameters and his skill in suborning Homer to his new purpose does not require us to disparage the norms from which he departed (and to which his successors would often return). The innovative brilliance we so justly admire in Vergil actually sets his poem apart from the tradition we too often think synonymous with it. The Aeneid is more accurately seen not as the telos of epic achievement at Rome but as only one more influential, yet anomalous stage in its ongoing evolution. "We get the perspective all wrong," wrote L. A. MacKay, "if we call Lucan a rebel against Vergil; on the contrary, in terms of strict Romanity, Vergil was the rebel, Lucan a reactionary, in literature as in politics."37 Getting a clearer and truer perspective requires sympathetic attention to effects Vergil avoided and norms he ignored. To escape the teleology still so

3 7 MacKay, CSCA 8 (1975) 147. Essays in CHLL nevertheless present the history of Silver Age epic as a set of (unsuccessful) responses to the Aeneid and consistently mistakes what is really a critical position for a "fact" of lit-erary history. Contrast Hardie, The Epie Successors of Virgil, who makes a significant advance by recognizing the inadequacy of this view.


palpable in modern histories of Latin literature, we need a fresh understanding of that literature's beginnings. We must read early poetry on its own terms and trace as best we can both the development of and the impetus for what became its norms and its characteristic effects.

Once allowances are made for the peculiar demands of fragmentary texts, the techniques of literary criticism are sufficient for identifying their favorite devices and assembling the evidence we need for tracing the evolution of their style. Discerning the impetus for that evolution, however, is not strictly a literary matter. This part of the problem has social and political dimensions, for the growth of epic poetry in Latin is closely connected to changing tastes and expanding horizons in second-century Rome and a corresponding shift in attitudes toward literary production. Much of the evidence for these developments is to be found among the facts and fictions transmitted about poets, their supporters, and their audiences in the middle and late years of the Republic. This sort of material, gath-ered from a variety of sources and dealing with a variety of cultural activities, will bring us very close to the historians' domain, but I do not mean to linger in their realm or to usurp their prerogatives. This literary history remains at heart an exercise in poetics. Its con-struction nevertheless begins in chapter 2 by considering the cultural climate that fostered production of Rome's first epics, and it will return to issues of politics and patronage in chapter 5. These discussions provide the necessary frame for two more traditionally literary chapters, which outline the developing aesthetics of epic style in Latin. Chapter 6 then joins the cultural and aesthetic strands of the inquiry by using cultural values to explain the consequences of Cicero's poetic choices as he produced what turned out to be the last Republican epics, while chapter 7 seeks to measure by example and analogy what the early epic poets accomplished and to understand the eventual limits of their achievement.

Juxtaposing discussions of text and social context in this way entails a decidedly eclectic approach to the interpretive problem the epic fragments present. It may even seem to distract attention from the texts that give the whole enterprise its raison d'être. This is an inevitable risk in narrative history, which, as David Perkins has observed, cannot escape "a practical conflict between the opposed claims of two aesthetic forms, the literary work one is describing and the narrative one is constructing. The first requires pause for critical responsiveness; the second, coherence and momentum."38 Yet we need both the description and the narrative. It would be wrong to offer a

38 Perkins, Is Literary History Possible?, 46. Cf. 127-29 on the rival claims of text and context.

Ruins 27

narrative account of epic artistry at Rome without demonstrating the art in Rome's early epics, and it would be self-indulgent to present the fruits of close reading without explaining why such fragmentary poems still deserve that reading. Since no account of Roman literary development, however, can afford to ignore Rome's first experiments with epic poetry, the attempt to understand those experiments is at least worth making. Eclecticism will, I hope, then seem a small price to pay.

2 Reconstructions

Latin literature began not with epic but with public spectacle, and it was fostered as a matter of public policy. The year was 240. The Romans, having just defeated Carthage in the First Punic War, were quick to flex cultural as well as political muscle. They expanded the native ludi Romani, a festival that had long offered musical and stage performances, to include formal plays in the Greek style. A man named Livius Andronicus was commissioned to furnish Latin scripts for the occasion and to oversee the productions.1 We know just a little about this Andronicus. The name itself suggests a Greek freedman of the Livian gens, and that idea makes some sense. A certain M. Livius Salinator was one of the decemviri sacris faciundis who presided over the Saecular games of 236. He could easily have held this office by 240 and recommended a family retainer for the new theatrical commission. Later sources support that inference: Andronicus was said to have reached Rome as a slave from Tarentum, to have tutored the sons of Salinator, and to have won his freedom by virtue of his talents.2 Those talents launched the first literary career at Rome and the first literary genres. Andronicus continued writing plays—the titles of nine tragedies and three comedies survive—and his experiments bore lasting fruit. Plays came to be performed at numerous Roman festivals by the late third century, and succeeding

1 Liv. 7.2.3-13; Val. Max. 2.4.4. The developments of 240 and their ration-ale are unclear in the sources. Gruen, Studies, 84-85, associates the expan-ded ludi directly with celebration of the victory over Carthage and suspects deliberate imitation of the great Greek festivals. The new program was in any case sufficiently grand to attract Hiero of Syracuse, who came with a large gift of grain (Eutrop. 3.1).

2 Cic. Brut. 72; Jer. Chron. ad a. 1830; Suet. Gram. 1.1. Discussion can be found in Suerbaum, Untersuchungen, 1-8 and 297-99; Waszink, ANRW 1.2 (1972) 873-74; Gruen, Studies, 80-92. References to Livius the decemvir are in MRR 223.

28

Reconstructions 29

playwrights followed his lead in adapting Greek plots and Greek meters to the new Roman conditions.

Andronicus himself lived on to become the grand old man of Roman letters, and the last chapter in his personal story contains further evidence of the close ties between literary and civic endeavors in the third century. The date was 207, when the second war with Carthage once again plunged Rome into crisis. The two consuls of 208 had died fighting Hannibal. A second army under Hasdrubal was then pre-paring to cross the Alps. The specter of a united Carthaginian force in Italy threatened disaster, and Rome could find no comfort in a sudden rash of reported prodigies. They were uniformly frightening: stones falling from heaven at Veii, a wolf inside the gates at Capua, assorted temples struck by lightning, the birth of a hermaphrodite. Etruscan soothsayers were consulted. Priests at Rome called for ela-borate expiatory rites, including a procession of twenty-seven virgins to march through the city singing a hymn. Andronicus was asked to write it. The chorus was still in rehearsal when lightning struck the temple of Juno Regina on the Aventine. Further rites were performed, and at last the procession itself, uniting magisterial and priestly power in an extraordinary display, wound its way through Rome.3

We know nothing else of Andronicus' hymn—Livy, for one, found it too uncouth to quote—but we do know the result of these extensive measures. The evil omens ceased, and the consuls of 207 destroyed Hasdrubal's army in a battle near Sena on the river Metaurus.

Relief at Rome equaled the earlier panic. The consuls celebrated a triumph and an ovation for their victory, and Andronicus received his own unique tribute. The imperial scholar Festus preserves the rele-vant facts (446L).

Itaque cum Livius Andronicus bello Punico secundo scripsisset carmen, quod a virginibus est cantatum, quia prosperius respublica populi Romani geri coepta est, publice adtributa est ei in Aventino aedis Minervae, in qua liceret scribis histrionibusque consistere ac dona ponere; in honorem Livi, quia is et scribebat fabulas et agebat.

At the time of the Second Punic War, Livius Andronicus wrote the hymn which virgins sang, and because Roman affairs afterwards began to im-prove, the temple of Minerva on the Aventine was publicly granted to him, in which writers and actors could gather and make offerings. This was done in honor of Livius, since he both wrote plays and acted in them.

The right to meet and make offerings in a temple are the marks of an official Roman collegium, and both the creation of this guild and its

3 Liv. 27.36-37. See MacBain, Prodigy and Expiation, 65-71, and for the central role of the hymn, Fraenkel, Horace, 379-80.


designation as an honor provide important clues to the literary climate of the period.

By 207, Roman theater had become a lively commercial enterprise. It required state sanction for its productions and looked at least in part to state funding, but its most striking characteristic was an extensive professional network of playwrights, actors, and producers. The rapid expansion of ludi scaenici to include a growing number of state festivals and privately sponsored shows indicates not just an aristocratic interest in enhancing their appeal, but the availability of a theatrical community to meet the growing demand for produc-tions and to handle their increasing technical sophistication. Greek acting companies had long been organized as professional guilds, the so-called σύνοδοι των περί, τον Διόνυσον τεχνιτών, which brought theater to all parts of the Mediterranean world. The troupes per-forming in Latin became equally polished.4 Such late plays of Plautus as Casina and Pseudolus require extraordinary levels of instrumental and vocal skill and presuppose elaborate conventions of staging and stage business. This was clearly not work for amateurs or dilettantes. Roman theater people were highly trained, and the acceptance of all scribae into the new collegium provides interesting and important evidence for their developing social position.

In the third century, the term scriba embraced a range of occu-pations that included clerks and bookkeepers as well as authors. Their common characteristics were literacy and a certain upward mobility. The first descendant of a freedman to win a curule chair, an aedile of 304 named Cn. Flavius, had been scriba to Ap. Claudius the Censor; a scriba of the elder Scipio, C. Cicereius, reached the praetor-ship in 173 and was a man of real distinction.5 Theirs was not a disreputable calling. The fact that Andronicus himself was known as both scriba and actor should suggest not the low status of the former but the comparative respectability of the latter. To think, as some remain inclined to do, that "poets were lowly creatures, even the best of them of modest means and quite dependent on the magistrate-class" reflects too narrow and elitist a view of Roman society.6 No

4 Pickard-Cambridge, The Dramatic Festivals of Athens, 279-87; Jory, Hermes 98 (1970) 224-53. For expansion of the ludi scaenici in this period, see Taylor, TAPA 68 (1937) 284-304. Theatrical activity was flourishing throughout Italy: the most proficient (and ambitious) may well have gravi-tated to Rome. See Rawson, PBSR 53 (1985) 97-113.

5 MRR 168 and iii.92 (Flavius); 406 η. 2 and 408 (Cicereius), Martina, La-be ο 26 (1980) 166-67, perhaps infers too much from likely exceptions to the norm, but Horsfall, BICS 23 (1976) 79-81, is unnecessarily contemptuous of scribae.

6 MacMullen, Historia 40 (1991) 423, "Authorities agree," begins the sup-porting note, "on the lowly status of poets in the century beginning with

Reconstructions 31

itinerant hack or compliant retainer wrote that hymn to Juno and reaped that official reward. The later Roman prejudice against the world of theater should not be read into this more neutral testimony from an earlier time.7 What Andronicus won for his peers was not just official acknowledgment of the Roman theatrical profession, but also confirmation of their status as independent professionals.

How independent? Did the aristocrats who paid the bills influ-ence the content and style of their productions? Andronicus himself was almost certainly a freedman of the powerful Livii: not only was Livius the decemvir probably responsible for that first dramatic commission, but a Livius—the same man or his son—was the domi-nant consul of 207. The Livian connection may continue to explain Andronicus' high public profile, Yet the commissions of 240 and 207 were not within the gift of a decemvir or consul. As matters of state policy, they reflected collective decisions and involved no partisan interests. Andronicus served the Senate in these matters, not the Livii, and his reward came directly from the state.8 This is one hint of a certain independence, and the subsequent history of drama suggests an even greater degree of financial and political freedom within the theatrical community. While official approval was always necessary for staging plays at public festivals, and aristo-cratic connections were doubtless needed to win the private contracts for funeral and votive games, this type of support has left no mark on the surviving scripts. Neither Plautus nor Caecilius has been as-sociated with any political figures, and what contemporary allusions have been suspected in their texts are, as befits the comic genre, more critical than laudatory.9 The biographical fantasies eventually spun around Plautus suggest, if anything, a struggling and independent agent. Gellius' story that he earned his living through some kind of

Livius Andronicus (240-140 B.C.)," but the citations that follow are modern. One might claim on analogous authority that UCLA professors are lowly creatures, even the best of them of modest means and quite dependent on the legislative-class. Yet we do not necessarily feel or act lowly. Neither, I suspect, did Livius Andronicus.

7 Cf. Livy's comment that he is setting out the origin of the ludi scaenici "ut appareret, quam ab sano initio res in hanc vix opulentis regnis tole-rabilem insaniam venerit" (7.2,13). Dramatists had a collegium separate from authors by the early first century, and we begin hearing more negative comments about actors under the Principate. See Tamm, Opuscula Romana 3 (1961) 157-67, and the prosopographical data in Garton, Personal Aspects of the Roman Theater, 231-65 and ANRW 2 30.1 (1982) 580-609.

8 A similar conclusion is reached by Gruen, Studies, 91-92. 9 On the thorny issue of topicality in Plautus, see Harvey, CW 79 (1986)

297-304, and Earl, Historia 9 (1960) 234-43. The issue is now redefined by Gruen, Studies, 124-57.


stage work ("in operis artificum scenicorum," 3.3.14) may preserve an allusion to the Greek-style technitai of the later third century, and recurring references to the selling of plays remind us that theater people had to be self-supporting. Access to the aristocrats who commissioned performances normally involved the assistance of es-tablished peers in the trade like the actor-managers Publilius Pellio and Ambivius Turpio; Caecilius himself was eventually cast as such a sponsor for Terence.10 Our evidence for patronage in the early Roman theater thus centers on the relationship between producers and playwrights, not between the theatrical community as a group and the official sponsors of ludi scaenici. The magistrates' power to grant and withhold contracts does not seem to have affected the conception or execution of these early literary enterprises.

The notorious exception to this fairly coherent picture is Gnaeus Naevius. He was a native of Campania, not an ethnic Greek like Andronicus, and he probably enjoyed limited citizen rights, the civitas sine suffragio. He was also a man of prodigious originality. The traditional date for his first stage production is 235, and his career was long and successful.11 Over thirty titles survive. His great gift was for comedy, but he also invented a style of play dramatizing Roman subjects, the so-called fabula praetexta. We know just enough of this new genre to complicate the picture of Naevius. One play was called Clastidium and presumably dealt with M. Claudius Marcellus' victory over the Gauls before that town in 222. It may well have celebrated Marcellus' winning of the spolia opima—he was consul and killed the Gallic chief Viridomarus—and his subsequent triumph: its sole surviving line suggests a victorious general who has safely returned home ("vita insepulta laetus in patriam redux"), as does its one other attested word, vitulantes ("celebrating"). The date and circumstances of its production are unfortunately unknown. An association with the triumph itself is usually dismissed on the grounds that living Romans could not be represented on the stage. A production in either 208, at Marcellus' funeral games, or 205, when his son dedicated the temple to Honor and Courage that Marcellus had vowed at Clastidium, is usually preferred. That guess is then used to

10 This romantic story in the ancient vita Terenti is either supported or inspired by the pairing of Terence and Caecilius in the second prologue to Hecyra. On the organization and financing of plays, see Beare, Roman Stage, 164-70. Cf. Jer. Chron. ad a. 1817: "scribere fabulas solitus [Plautus] ac vendere," and Horace's sneer about Plautus' commercialism at Ep. 2.1.175-76.

11Gruen, Studies, 92-93 and 104-5, offers critical review of the Naevian biography.

Reconstructions 33

claim Naevius as a client of the Marcelli, writing a play to serve their partisan interest.12

So portentous a conclusion lacks firm support. It makes major in-ferences from the hypothetical dating of Clastidium, equates the commissioning of a play with formal obligations of clientele, and, most seriously, makes an unwarranted assumption about the nature of the fabula praetexta. These plays were not all and not necessarily so explicitly topical. In addition to his Clastidium, Naevius wrote a Romulus. (Lupus is likely to be an alternate title.) In the next century, Ennius wrote a Sabinae as well as an Ambracia, and Accius wrote an Aeneadae. There was more to the genre than political puffery and familial prestige. Its celebration of mythical as well as contem-porary deeds suggests a literary form that used the stage, the first medium for serious literary expression in Latin, to establish the traditions as well as the glory of Roman accomplishments.13 The closest kin to Naevius' Clastidium is, as we shall see, his Bellum Punicum, an epic that deliberately set contemporary events in a mythological context. It too singled out individual Romans for praise (and blame), but its impetus came from Naevius' own experience and pride in the achievements of the First Punic War. And the Bellum Punicum had no patron. There is no basis in either work for casting Naevius as the champion of any single noble house; nor should other, still more famous evidence be used to claim the opposite.

The most famous story about Naevius involves a quarrel with the powerful family of the Caecilii Metelli. He is said to have been im-prisoned for slandering them and then, exiled from Rome, to have died a lonely death at Utica. These troubles are traced back to a line of verse, "Fato Metelli Romae fiunt consules." Taken at face value, the story provides evidence of a poet's engagement in contemporary politics and the dire consequences to him of speaking freely. It is widely cited to support the claim that Roman authorities kept a tight rein on theatrical expression and decorum: "one could get in serious trouble, meaning jail, for ignoring those rules of presen-

12 Jocelyn, Antichthon 3 (1969) 34. The ingenious reconstruction of Mat-tingly, Historia 9 (1960) 432-38, is unconvincing. Historical references to the events can be found in MRR 233. The presumption that drama avoided the living is founded on the fragments of Cic. Rep. 4.10, though Cicero says only "veteribus displicuisse Romanis" and the context seems to be comedy. Dis-cussion in Marmorale, Naevius poeta, 50-51, 130; see 202-3 for the fragments themselves.

1 3 Zorzetti, La pretesta e il teatro latino araico, 61-62, thus rightly pairs Clastidium and Romulus in his general discussion of the praetexta as a celebration of Roman achievement. Beare, Roman Stage, 39 and 41-42, makes an artificial distinction between historical and legendary plays.


tation."14 Yet the story does not withstand scrutiny. Its two central elements, the verses and the imprisonment, are worth considering in detail.

We owe knowledge of the verse exchange to the impudence of Gaius Verres, which enabled Cicero to raise a laugh at the expense of Verres' ally Q. Metellus Creticus, consul designate at the time of Verres' trial in 70: "The story went that Verres used to say that you were elected consul not by fate, as the rest of your family was, but through his influence." ("Nam hoc Verrem dicere aiebant, te non fato, ut ceteros ex vestra familia, sed opera sua consulem factum," Ver. 1.29.) A commentator of the fifth century A.D., probably drawing on the work of Q. Asconius Pedianus four centuries earlier (p. 215 Stangl), explains the allusion.

dictum facete et contumeliose in Metellos antiquum Naevii est "fato Metelli Romae fiunt consules," cui tunc Metellus consul iratus versu responderat senario hypercatalecto [sic], qui et Saturnius dicitur: "dabunt malum Metelli Naevio poetae."

An old saying of Naevius was directed wittily and insultingly at the Metelli, "By fate are the Metelli elected consuls at Rome." The consul Metellus became angry at this and replied in a hypercatalectic iambic verse [sic], which is called Saturnian: "The Metelli will make trouble for the poet Naevius."

The famous exchange between Naevius and the Metelli thus comes to us through the filter of late Republican politics, which often gave new political twists to old lines of verse. A politically charged audience at the Floralia of 57, for example, used lines of Accius to voice their support for Cicero. A similar display at Pompey's expense had occurred in 59 at the ludi Apollinares. A line from Pacuvius' tragedy Armorum iudicium, "Did I rescue them to become my des-troyers?" ("men servasse, ut essent qui me perderent!"), was recited to great effect at Caesar's funeral.15 The only thing attested for Naevius' line is that Verres probably turned it against a Metellus, and Cicero certainly did. Nothing here confirms that Naevius him-self intended a barb: the insulting contrast between fato and opera is

14 MacMullen, Historia 40 (1991) 422. The most important sources are Gell. 3.3.15 (imprisonment) and Jer. Chron. ad a. 1816 (exile, pulsus Roma factione nobilium ac praecipue Metelli). That Plautus refers to this incident at M G 211 is but a scholarly assumption. No explicit contemporary evidence confirms the historicity of the quarrel. Discussion can be found in Suerbaum, Untersuchungen, 31-42; Gruen, Studies, 96-106.

15 Cic. Sest. 117-26; Att. 2.19.3; Suet. Caes. 84, cf. App. BC 2.146. Discussion and further examples are in Nicolet, World of the Citizen, 367-73.

Reconstructions 35

in Cicero's parody, not the original. Fato alone is not so easy to read pejoratively.

Nor do we really know that the line enraged the Metelli. The scholiast of course says so, as had Caesius Bassus four centuries ear-lier ("Metelli . . . ab eo versu lacessiti," GLK 6.266), but they, like most scholars since, probably failed to hear in the Metellan reply the less than horrific ira of comedy, where phrases like "malum vobis dabo" abound.16 There is more wit (of a kind) than anger in this ex-change. Nor, finally, do we know that a stage performance, whether palliata or praetexta, was directly involved. The scholiast simply quotes "a saying" (dictum). It could have come from a play, but this is dangerously like attributing the Metellan reply to an epic because it is a Saturnian.17 Iambic verse was never restricted to the stage. The couplet aimed at Caninius Rebilus, consul suffectus for a single day in 45 B.C., offers a nice example of such a circulating iambic barb,

vigilantem habemus consulem Caninium qui in consulatu somnum non vidit suo.

We have a wide awake consul in Caninius, who never saw sleep in his consulship. (Macrob. 2.3.6)

While the historicity of an exchange between Naevius and the Metelli seems beyond question—what could have motivated a forgery?—both its tone and the circumstances surrounding it remain unclear.

Not that explanations have been lacking. Gellius' tale, which may derive from Varro, is best known: Naevius was imprisoned for slandering leading men of the state (principes civitatis) and then made amends by delighting authorities with two plays he wrote there, Hariolus and Leon. The story has a superficial appeal. Varro is a reputable source. The titles sound authoritative. A Roman concern with slander may be traced back as far as the Twelve Tables, and offenses against public morality could indeed land their perpetrators in jail. During the second Punic War, for example, the banker L. Fulvius was imprisoned on senatorial order for wearing a garland in

16 PL Per. 847, cf. 817, 827: a host of examples are in Lodge, Lexicon Plan-tinum, 2:19 s.v. malus. Gruen, Studies, 98-100, detects an allusion in fato to the election of Q. Caecilius Metellus to the consulship of 206 by somewhat dubious means.

1 7 Flintoff, Latomns 47 (1988) 598-99, wrongly reads Naevius' verse as a Saturnian. Though it can be broken into uneven cola on the Saturnian pattern, the first colon lacks the caesura korschiana typical of seven-syllable cola, and the second ends with a distinctly iambic cadence unparalleled among authentic Saturnians. See Cole, YCS 21 (1969) 19-20 (caesura) and 39-41 (cadence).


public.18 Yet Gellius' report is hardly more than the doublet of a story he records just above in the same chapter and specifically attributes to Varro, the story that Plautus wrote three plays (Satarionem et Addictum et tertiam quandam) while working in a mill after he had lost his money on a trading venture. No one today believes that tale. Mills, like pillories, prisons, and trading ventures are standard elements of Roman comedy. (A glance through the Plautine lexicon s.v. pistrinum is again instructive.) The story is fanciful, a reminder that what Varro assembled was not necessarily independent evidence for the careers of comic poets. Such anecdotes simply reflect the long-standing ancient tendency to supply missing biographical data from the literary works at hand. There is thus no more reason to believe that Naevius wrote Hariolus and Leon in a prison cell than to believe that Plautus wrote three plays—none of which Varro included in his canon of authentic works—while chained to a mill.19

Naevius' exchange with the Metelli has engendered a fantasy. Some may still wish to credit it, but it is not the strength of ancient evidence that compels such belief. Handbooks would do better to popularize Cicero's wonder at the license of Greek comedy and his explicit statement that Roman dramatists kept politics off the stage.

sed Periclen, cum iam suae civitati maxima auctoritate plurimos annos domi et belli praefuisset, violari versibus et eos agi in scaena non plus decuit, quam si Plautus noster voluisset aut Naevius Publio et Gnaeo Scipioni aut Caecilius Marco Catoni maledicere.

It was no more fitting for Pericles, when he had presided with supreme authority over his city in war and peace for many years, to be insulted in verses performed from the stage than if our own Plautus or Naevius had wanted to slander Publius and Gnaeus Scipio or Caecilius to slander Marcus Cato. (Rep. 4.10.11)

Nothing said of the poet Naevius offers evidence to the contrary. Efforts to find more veiled political significance in the numerous

fragments of his comedies have been no more convincing, and in recent years they have rightly fallen from favor. Topical allusions claimed for his palliatae lack conviction when set against the genre's con-ventions. The best example comes from Aulus Gellius. Many (later) readers, he says, saw a reference to Scipio Africanus in the lines,

18 Plin. Nat. 21.8. Further examples, none very like the case of Naevius, are discussed by Eisenhut, ANRW 1.2 (1972) 268-82. The law forbidding malum carmen in Hor. Ep. 2.1.152-55 (cf. AP 283-84) may recall the Twelve Tables, but whether the original reference was to slander or magic remains uncertain. Its association with theater history is in any case tenuous. See Brink, Horace on Poetry II, 196-99.

19 Leo, Plautinische Forschungen, 70-78, recognizing the scholarly topoi at work, saw these tales for what they were. Cf. Fraenkel, RE Suppl. 6 (1935) 625.

Reconstructions 37

etiam qui res magnas manu saepe gessit gloriose, cuius facta viva nunc vigent, qui apud gentes solus praestat, eum suus pater cum palliod unod ab arnica abduxit.

Even he who often performed great deeds of valor, whose exploits still live on, who alone stands out in all the world, him his father carried from his girlfriend with a single cloak.

(108-110 R = Gell. 7.8.5)

Scipio, ever a controversial figure, did in fact raise eyebrows for sporting the pallium in Sicily, but the stock miles gloriosus might wear similar garb, and the fragment's structure, with its ironically grandiose description followed by a sudden deflation, even finds a parallel in the plight of Menander's soldier Polemon.

ο σοβαρός ήμΐν άρτίως και πολεμικός, 6 τάς γυναίκας ουκ έών εχειν τρίχας κλάει κατακλινείς.

Our haughty hero, so belligerent just now, who won't let women keep their hair, has collapsed in tears at lunch. (Perike iromene 172-74)

A seemingly bold line like "libera lingua loquemur ludis Liberalibus" ("We'll talk with free tongue at Freedom's games," 113R) is also too easily paralleled by the boasting slaves of Roman comedy to be read as a serious statement of artistic freedom. Naevius did no more than reflect the ambience of his time.20 He was, like the other poets of third-century Rome, an independent entrepreneur working freely and creatively with the dramatic forms of the day.

So commercial and entrepreneurial an environment, however, inevitably affected th,e character of the plays it fostered. The need to attract and hold the kind of audience lured to the varied attrac-tions of a public festival bred a fundamental conservatism in the Roman theatrical tradition. Once a successful formula emerged, there was little incentive to change it. Style and technique certainly advanced as professional competence grew, but the basic tone and quality of the dramatic genres remained constant. Tragedy kept a consistent emphasis on grandeur and spectacle, while comedy perfected and then formalized the brilliantly exaggerated style we know best from Plautus. The Roman audience's expectations of what a performance entailed, both technically and intellectually, became

20 Wright, Dancing in Chains, 56-57. Wright ignores the historical fact of Scipio's Greek affectations (cf. Liv. 29.19.12), but if accepted, Gellius' state-ment would at best only bring us to the inconclusive kind of argument that surrounds discussions of Plautine topicality.


too well established to encourage experimentation. The old formulae worked too well. Two incidents of the 160s illustrate the nature of that inherent conservatism and its consequences for the development of Roman literature.

The first episode occurred in 167 at a musical variety show spon-sored by the praetor L. Anicius Gallus. The historian Polybius, recent-ly arrived at Rome with a group of Achaean hostages, was probably an eyewitness to the proceedings. His account survives in abbreviated form, incorporated into a long section of Athenaeus' Deipnosophistae which is dedicated to the jests and buffoonery of powerful men.21

Anicius, says Polybius, brought leading Greek artists, including four of the world's most distinguished pipers, to Rome for the occasion, and he erected a vast stage for them in the Circus Maximus. He set the featured musicians at the front with a chorus. When pipers and dancers began to perform in their customary way, however, Anicius found the entertainment insufficient. He sent lictors out to liven things up, and they ordered the performers to form lines and advance in mock battle against each other while the pipers blew in unison as hard and discordantly as they could. The Greek performers were at first perplexed by these instructions. Once they had caught Anicius' drift, however, they entered in with a will, and as the audience cheered them on, pipers and dancers resorted to fisticuffs, first in their rows and then singly. Boxers and solo dancers were then sent out to join the display, and all struggled wildly to a cacophony of trumpets and horns. A performance by tragic actors eventually fol-lowed this extraordinary spectacle, but a disgusted Polybius wryly withheld further description lest readers think he was joking.

The episode reads like a paradigm of Roman boorishness. Athenaeus' dinner guests laugh heartily at what, punning on the name Άνίκιος, he called "these Invincible spectacles" (ταΐς Άνικίοις ταύταις θέαις), while Polybius doubtless found the show offensive. Yet Polybius, as a new and involuntary resident of Rome, was no dispassionate observer, and his own inexperience with Roman ways fed a natural but perhaps hasty indignation. His testimony should be set within the context of Roman preferences in musical entertainment and Roman styles of performance. That may be difficult to do, but it is not impossible. Andronicus' hymn to Juno Regina provides evidence of a kind. Its performance was carefully managed: there were rehears-als under the poet's supervision and a planned march. The twenty-

21 Athen. 14.613d-615e = Polyb. 30.22.1-2. Anicius is put in the company of Philip II, Demetrius Poliorcetes, Lysimachus, and Sulla, a raucous group. Cf. Athen. 6.260-61c on the humor of Philip and Demetrius. For a somewhat different treatment of this episode, see Gruen, Culture and National Identity ,215-18.

Reconstructions 39

seven virgins were probably arranged in three groups of nine (ter novenae, says Livy), and they held a rope that would have estab-lished and maintained their formation as they marched through the city. The procession itself must have been slow and stately—the girls wore long robes—and the hymn was, as we might expect, as note-worthy for its pronounced beat as for its music. A group performance arranged in three lines was also a feature of the more elaborate procession that traditionally opened the ludi Romani. Among its many components was a display of armed dancers arranged in three groups by age (men, youths, and boys). They performed warlike maneuvers to the accompaniment of pipes and lyres and were then followed by other dancers dressed as satyrs, who mimicked the intricate steps of the earlier formations.22 These native displays were parades rather than stage shows, but Anicius' insistence on setting lines of dancers in motion is precisely what disrupted their Greek style of performance and aroused Polybius' indignation. Commanding the four pipers to play in unison—the aulos was not normally an ensemble instrument—probably also reflects the Roman predilection for the sound of massed instruments. The loud music, mock combat, and moving lines conformed to Roman tastes and expectations. Anicius' show was no random display, and it was not entirely devoid of rationale.

Polybius certainly implies a near riot, but a near riot is not what he describes. Numerous signs of control are implicit in his account: lictors on the scene to keep order and relay the praetor's instructions, supplementary acts waiting in the wings and advancing on cue, facilities constructed to suit the occasion. Anicius erected so large a stage not to be extravagant but to suit the performance. He built it in the Circus, not the forum, because only the Circus afforded sufficient space for the many acts and could accommodate the audience in tiered seats. An impromptu stage in the forum or Campus Martius would have elevated only the performers, making it harder for spectators to appreciate the choreography and the intricate sequence of acts. Anicius wanted them to look down to the lines of performers. The site was selected for an optimal view of a planned spectacle. It was certainly not a Greek show, which is why it offended Polybius, who came to it with inappropriate expectations. It was a Roman show fashioned from Greek elements, and that is the key to its significance.

2 2 Liv. 27.37.7-15 (the hymn); D.H. 7.72.5-11 (the ludi Romani). Hellenic artifices appeared in Rome as early as 186 for games held in that year by Fulvius Nobilior and Scipio Asiaticus, but we cannot be sure that they were actually stage performers (Liv. 39.22.2 and 22.10). Their relevance to Anicius' show is therefore uncertain.


Anicius' manipulation of these Greek artists was no gratuitous, unmotivated entertainment. It occurred in the aftermath of a grand but problematic military triumph. In 168, Anicius had conducted a swift and decisive campaign in Illyria. Within a single month, he made King Gentius his prisoner and broke forever the power of Rome's pesky and intermittently piratical neighbor across the Adriatic. His triumph the following year came with all the trimmings: a royal family in chains, vast piles of booty on display, enthusiastic soldiers saluting their general. Everything was perfect except the timing. There had already been one great triumph at Rome that year. The city had witnessed Aemilius Paullus' celebration of his famous victory at Pydna, and the inevitable comparison did not favor Anicius. Gentius was no Perseus, Illyria no Macedon, and Anicius no Paullus. "People," says Livy, developing the contrast, "found everything more similar than equal" (45.43). But Anicius did not abandon the limelight. Close upon the heels of his distinguished yet subordinated triumph came this second public entertainment. He was still laboring in the shadow of Aemilius Paullus. After Pydna, Paullus had stayed on in Greece to preside over the political restructuring of Macedon, and he enhanced his prestige throughout the east by sponsoring an elaborate festival in the Greek style at Amphipolis. A military conqueror, he said, also had to be an impresario. The comment became famous, and events at Amphipolis confirmed its truth.23 Anicius doubtless agreed, but the respect he had to win was at Rome. He therefore forced his imported talent to accommodate native tastes. Though the show was presumably successful in that aim, Polybius' displeasure was not its only legacy. Such a success did not bode well for any attempt to bring Greek standards to a Roman performance, which is what Terence attempted in the year 160.

The occasion was the public funeral of Aemilius Paullus, a lavish spectacle sponsored by his sons, Fabius and Scipio. The actual event was an aborted production of a play called Hecyra. This play has a curious, even notorious production history. Program notes preserved with the script report that the play was offered for the first time at a state festival, either the ludi Romani or Megalenses, in 165. An error in programming, or simply bad luck in drawing a crowd, prevented the show from getting off the ground. An inattentive audience had its mind on other things. Terence himself refers to this failure in the prologues to two subsequent productions.

23 Liv. 45.32.11: "vulgo dictum ipsius ferebant et convivium instruere et ludos parare eiusdem esse, qui vincere bello sciret." Cf. Polyb. 30.14.

Reconstructions 41

ita populus studio stupidus in funambulo animum occuparat.

The crowd dumb with desire wanted a tightrope walker. (Hec. 4-5)

quom primum earn agere coepi, pugilum gloria (funambuli eodem accessit exspectatio) comitum conventus, strepitus, clamor mulierum fecere ut ante tempus exirem foras.

As soon as I began the show, a love of boxers (hope of tightrope walkers also intervened) attracted such a crowd, clamor, women's shouts, that I had to exit prematurely. (Hec. 33-36)

Since the play was still practically new, Terence had no compunction about offering it for sale again in conjunction with his Adelphoe for the funeral games of 160. Yet this attempt also failed. The second prologue, written for a third and successful production, offers a reason. The speaker is Terence's producer, Ambivius Turpio.

primo actu placeo; quom interea rumor venit datum iri gladiatores, populus convolat, tumulantur clamant, pugnant de loco: ego interea meum non potui tutari locum.

I was satisfied at first. Then a rumor got about of gladiators to appear. A crowd gathers: they mill about and shout and fight for places. I meanwhile could not hold my place. (Hec. 39-42)

The prospect of gladiators evidently drew a new crowd into the theater, and their struggle to find places disrupted the performance beyond recovery. Only the third attempt later that year, this time again under state sponsorship, drew an audience able to see the play through to the end.

These two failures have long puzzled critics. Recent studies give Hecyra itself good notices and accept Terence's explanation at face value.24 The logistics of public festivals at Rome, which subjected acting troupes and their audiences to the inconveniences of im-

2 4 Gilula, Athenaeum 59 (1981) 29-37, and Sandbach, CQ 32 (1982) 134-35, interpret the evidence of the prologues correctly but accept it uncritically. Contrast Gruen, Culture and National Identity, 210-15. Recent favorable discussions of Hecyra include Gilula, SCI 5 (1979/80) 137-57, Konstan, Roman Comedy, 130-41, and Slater, CW 81 (1988) 249-60.


provised theaters and the vagaries of rival entertainments, is easy to blame. No production could hope to continue before a jostling crowd eager for a tightrope act or armed combat. That at least is Terence's own testimony, but it should give us pause. Its deliberate bias demands skepticism. Terentian prologues are all highly rhetorical set pieces that adapt the style and manner of forensic argument to advance the dramatist's interests.25 They never record strict truth, though there is surely truth in them. This play must indeed have failed twice, but not necessarily for the reasons given.

Hecyra's first failure is known only from Terence's tendentious report of boxers and tightrope walkers, but there is independent evidence concerning conditions at the funeral games of Aemilius Paullus. These games should, following Roman usage, more properly be styled gladiatorial offerings (munera gladiator urn), which means, among other things, that the professional combatants hired for the occasion were not marginal to the entertainment. They were the main attraction. We know that Paullus' sons had the fortunes of their adoptive families behind them, and they lavished those resources on an extraordinary show. The gladiators performed over a three-day period and cost at least thirty talents. That was a sum equal to half the entire estate of Aemilius Paullus, and it dwarfs the 8,000 denarii, itself a record of sorts, awarded for two performances of Terence's Eunuchus in the previous year.26 Our own cultural preferences should not distort the original Roman priorities. The games for Paullus were not ludi scaenici with gladiators attached, but a gladiatorial dis-play with plays attached. How could so complex, professional, and expensive an enterprise confuse the arrangements for gladiators and players? While Paullus' sons, as private citizens, did not have lictors available to keep order as the praetor Anicius had done, they would hardly have allowed their arrangements to lapse into pandemonium, even on a small scale. Their father, after all, was the man who said that a good general should also know how to stage shows. The logical procedure was to reserve a separate day for the stage plays, as the younger Flamininus apparently did when offering similar munera for his father in 174 (Liv. 41.28.11). Fabius and Scipio may well have done the same. Certainly no trouble dogged the production of Adelphoe, the Hecyra's companion piece on the program. Terence's claim in the prologue is therefore difficult to accept at face value. His reference to gladiators more likely hides the failure of Hecyra

2 5 Goldberg, Understanding Terence, 31-60. 26 Polyb. 31.28.1-5. There were 6,000 denarii to the talent. Suet. Vita Ter. 3

reports "octo milia nummorum" as the price paid for Eunuchus ; the Repub-lican nummus was the silver denarius. See Gilula SCI 8/9 (1989) 74-78.

Reconstructions 43

behind the best-known feature of a famous occasion. The real cause of that failure presumably lies elsewhere.

In commissioning Terence to provide plays for them, Fabius and Scipio were hiring the most popular and expensive poet of the day, and also the most hellenized. Terence consistently restrains the extravagance of the traditional comic style: both his diction and his lack of song set him closer to the Greek style of New Comedy than the Latin one. Yet he also knew how to appeal to Roman values and Roman prejudices: Adelphoe rewards the crusty old farmer Demea and mocks his urbane, indulgent brother, Micio. Its hellenized style does not obscure its Roman moral. Hecyra is far more daring in con-ception, and the more conservative our expectations, the more radical it appears, Judged by the norms of Roman comedy, the play dramatizes the wrong problem, mocks the wrong characters, and de-velops the wrong plot. Modern critics have ample grounds for their admiration: it is bold, inventive, and sensitive. A Roman audience, however, could easily think otherwise. They preferred to see their values and expectations confirmed. Hecyra's surprises tread the boundaries of their tolerance, and that is why it took three attempts to get it accepted. It was too inventive an exercise for the funda-mentally conservative genre of Roman comedy, and the ultimate proof of its failure is the ultimate failure of its genre. The dramatists who came after Terence did not continue his experiment. They re-verted to the older comic style, and because they could never equal Plautus' genius, their genre never regained its earlier vigor.27 There was no great comic poet at Rome after Terence. His hellenization of Roman comedy bore no lasting fruit, and by the end of the second century, public spectacle had ceased to be the vehicle for serious poetic expression in Latin. This was hardly the end of serious poetry, but only because serious poetry had already begun to move in a different direction.

That direction was toward epic. The pioneers were still Andronicus and Naevius, but their experiments in narrative verse introduced quite different literary possibilities from their innovations on the stage. As a genre for private circulation rather than public per-formance, epic developed a different appeal, a different relationship to Greek models, and a very different internal dynamic. Whatever oral precedents may have encouraged its growth have left no clear traces. The elder Cato apparently believed that Romans of much earlier times had praised famous men at banquets by singing verses in turn to a pipe accompaniment, and Cicero, always keen to develop an evolutionary argument, associated this practice with the subsequent

2 7 Wright, Dancing in Chains, 153-81.


growth of epic.28 That, however, was only his guess. Neither Cicero nor Cato nor any other Roman of record had ever heard such songs himself or had any clear idea of what they were like. Their testimony is necessarily vague and indirect, but it nevertheless gave rise to theories of a preliterary bardic tradition at Rome. That idea, bandied about since the Renaissance, was embraced in the early nineteenth century by the historian B. G. Niebuhr because it explained the legendary and romantic qualities of so much early Roman history: Niebuhr imagined heroic lays on such themes as Coriolanus and Verginia that were passed down through oral tradition among the great families of the city. His argument was subsequently taken up with much enthusiasm by Lord Macaulay, who had his own, quite vivid notion of how such poetry should sound.

Then out spake brave Horatius, The Captain of the Gate:

"To every man upon this earth Death cometh soon or late.

And how can man die better Than facing fearful odds,

For the ashes of his fathers, And the temples of his Gods . . . ("Horatius," Stanza 27)

The subjects of Macaulay's famous Lays of Ancient Rome (1842) are certainly as Roman as Livy. Their form, however, owes much more to Sir Walter Scott. Macaulay's interest in folk poetry simply led him to create "border ballads" on Roman themes, and their powerful evo-cation of ancient traditions is therefore less a monument to antiquity than to the Victorian imagination. Though they continue to mes-merize scholars of a certain temperament, the spell must be broken.29

28 Cic. Brut. 75, cf. Leg. 2.24.62, TD 1.2.3. His fullest reference to Cato's testimony, tied to a different but equally tendentious argument, comes at TD 4.2.3: "Gravissimus auctor in Originibus dixit Cato morem apud maiores hune epularum fuisse, ut deinceps qui accubarent canerent ad tibiam clarorum virorum laudes atque virtutes." Varro ap. Nonius (107L) says that pueri modesti did the singing either a capella (assa voce) or cum tibicine. Less direct testimony appears in Val. Max. 2.1.10 and Quint. 1.10.20. The Brutus passage makes clear that Cato was speaking of a time "multis saeculis [i.e., generations] ante suam aetatem." Plutarch's report that around Cato's own table, πολλή μεν ευφημία των καλών και αγαθών έπεισήγετο (Cat. 25.3), probably misinterprets this older testimony.

29Thus Williams, CHLL, 55: "the widespread evidence for the existence of such lays in human societies at all times and in all places (excellently gathered by Macaulay) makes it attractive to posit their existence in Rome too" (cf. p. 59). Yet Macaulay neither gathers nor evaluates evidence in a modern scholarly way. For perspective on this issue, see Momigliano, JRS 47 (1957) 104-14, and Zorzetti, "The Carmina Convivalia," 289-95.

Reconstructions 45

Nothing sung around Roman dinner tables ever sounded like Macaulay's "Horatius."

The ancient testimony does not support these bardic fancies, which make two very dubious assumptions. The first, implicit in any claim of influence on later literature, is that the ancient carmina developed a recognizable style and a stable content. No trace of such style or content survives. Cato himself describes only improvised and probably amateur efforts: he tells us elsewhere that itinerant professionals who offered their services at banquets were mocked as flatterers.30 If the Romans had a bardic tradition, it must have been a tradition without bards. The second assumption is that the banquet songs were narrative. Cato again says no such thing; he speaks of "laudes atque virtutes/' not of "res gestae." His report of singing in turn actually sounds much more like the Greek skolia, songs that could praise such clarissimi viri as Harmodius and Aristogeiton without turning themselves into ballads.31 Cato's description may also recall the lyric paean, a verse form which could be excru-ciatingly laudatory, especially when addressed to the living. The preferred meter for paeans was the ithyphallic. That rhythm has strong affinities with the versus quadratus of Roman triumphal song, and Greek sources could confuse the two.32 That is probably why Plutarch, for example, mentioned both triumphal songs (μελή) and paeans sung at Marcellus' triumph in 222 (Marc. 8.2). Yet neither

3 0 Cato ap. Gell. 11.2: "poeticae artis honos non erat: si quis in ea re stu-debat aut sese ad convivia adplicabat, crassator vocabatur." Cf. Festus 86L: "grassari antiqui ponebant pro adulari. grassari autem dicuntur latrones vias obsidentes; gradi siquidem ambulare est, unde tractum grassari, videlicet ab impetu gradiendi." Préaux, Latomus 25 (1966) 710-25, thus equates gras-satores with πλάνοι. Martina, Labeo 26 (1980) 155-60, nevertheless makes too much of their likeness to comic parasites.

3 1 PMG 895-97. The styles of singing skolia are described by Athen. 15. 694b, Plut. Mor. 615b, and Σ Pl. Grg, 451 e. Narrative poetry was not normally sung to the pipe. Zorzetti, CJ 86 (1991) 312-15, rightly stresses the lyric character of the so-called carmina convivalia; Cole's response at CJ 86 (1991) 378-82 provides independent confirmation of the present argument.

3 2 One example is the paean performed at Athens in 291 in honor of Demetrius Poliorcetes, which embarrassed Duris of Samos, who recorded the text, and Athenaeus, who quotes it (7.253d = CA 173-75). Flamininus was recipient of a similar (dactylic) compliment at Chalchis ca. 195; Plut. Flam. 16 quotes a few lines. Testimony for such paeans—fifteen in all—is gathered in Cameron's forthcoming Callimachus and His Critics, to which I owe much. The personal address in Ennius, "vel tu dictator . . . esto vel consul" (4-5V), perhaps echoes this style, and Vahlen attributed the fragment to his Scipio. Courtney, Fragmentary Latin Poets, 26, finds its hexameters more appro-priate to the Annals.


paeans nor triumphal songs were narratives. Evidence for narrative poetry at Rome begins only in the third century, and it is all literary. By the time Plautus' Sosia tells his mock-heroic tale of Amphitruo's exploits against the Teleboians, he is able to parody the conventions of an epic style already in place.33 Cicero's observation that the Romans were slow to appreciate poetry probably reflects the em-barrassing fact that they were slow to produce any poetry worth respecting (TD 1.2.3). Narrative poetry came to Rome late, and it was a conscious invention. Its inventor, and the inventor of so much else, was Livius Andronicus.

At some point in the third century—the precise when and why are beyond recall—Andronicus produced a Latin version of Homer's Odyssey. The poem did not win lasting acclaim. Cicero called it a work of Daedalus (Brut. 71), by which he meant "primitive," and Horace could muster only faint damns in its praise.

non equidem insector delendaque carmina Livi esse reor, memini quae plagosum mihi parvo Orbilium dictare; sed emendata videri pulchraque et exactis minimum distantia miror

Not that I attack Livius' poems and think they should be destroyed, which I remember Orbilius beating into me as a boy; but for them to be thought polished and pretty and near perfection astonishes me. (Ep. 2.1.69-72)

A generation that expected poets to lick their verses into shape like a she-bear grooming cubs could hardly approve a work of only inter-mittent felicity. Nevertheless, though Andronicus was no Vergil and his Odussia no Aeneid, the former was still a poet and the latter still a poem. It was not written to be the school text that Horace knew, and it could never have been anyone's crib. For one thing, Andronicus did not tell Odysseus' story at Homer's length. Though surviving frag-ments have been identified with twelve different books of the original, this Latin version apparently fit on a single roll: no source indicates book divisions.34 Its verses were short, too. Andronicus adapted Greek quantitative meters for his plays, but he chose the

33 Awipfo, 203-47, a recitative performed to the tibia. The play can hardly be dated before ca. 200. See Lelièvre, Phoenix 12 (1958) 117-24, and Harvey, Athenaeum 59 (1981) 480-89.

34 The one partial exception is Priscian, GLK 321.6, introducing a quotation with the words "Livius Andronicus in I Odissiae ponit . . . " What follows is a fragmentary hexameter. Priscian seems to have known a (later) hexameter version of Andronicus' poem, which he quotes four times (fr. 37-40) without distinguishing it from the Saturnian original. See Mariotti, Livio Andronico, 55-59.

Reconstructions 47

solemn and formal Saturnian verse, the old Latin meter of hymns, epitaphs, and incantations for his epic. Though the choice may have limited his stylistic options, the Roman epic tradition nevertheless begins with this poem.

Andronicus brought undeniable skill to the enterprise. He found adroit Latin equivalents for Greek vocabulary and syntax, and he used his schoolmaster's knowledge to clarify the text before him.35

His is the learning of a bilingual teacher experienced in the rudi-ments of Homeric exegesis. It should not be mistaken for true Alexandrian erudition, but it has been. The mistake began innocently enough with Eduard Fraenkel's demonstration that the epic's archaic diction differs from the colloquialism of Andronicus' dramatic fragments. Fraenkel suggested that since archaism is a Hellenistic mannerism, its presence in Andronicus' poem reflects the influence of his Tarentine origin. The artistry of early Roman literature would then owe much to contemporary Greek practice.36 The syllogistic quality of this argument has not prevented its extension. Scevola Mariotti, taking Suetonius' report that Andronicus both "interpreted" and "presented" poetry to mean that he was a professional gram-maticus, likened him to the fifth-century poet Antimachus of Colophon, who produced an "edition" of Homer as well as an epic Thebais.37 Antimachus' own poetry certainly prefigured the artful and allusive mannerisms of Alexandria, and solemnity is indeed to be found in both the Thebais and Odussia. Mariotti therefore ascribed Andronicus' weighty diction to Hellenistic influences, and he found in Andronicus' occasional expansion of his Greek model—as in the statement that ocean billows soon ruin even a powerful man—a reflection of those subjective and pathetic qualities thought typical of Hellenistic but not Homeric epic.38 Antimachus also seems to have

3 5 Thus fr. 17, "simul ac lacrimas de ore noegeo detersit" renders the aorist participle of Od. 8.88, δάκρυ* όμορξάμενος κεφαλής απο. At Od. 8.379, ταρφέ' άμειβομενω becomes "nexebant multa inter se" (fr. 20) after Σ V ad. loc. See Fränkel, Hermes 67 (1932) 306-07.

3 6 Fraenkel, RE Supp. 5 (1931) 605: "Die junge römische Dichtung ging, so stark sie sich auch älteren Mustern zuwenden mochte, doch als Kind ihres Zeitalters aus hellenisticher Kunstübung hervor/'

3 7 Mariotti, Livio Andronico, 16-17, building on Suet. Gram. 1.2: "nihil am-plius quam Graecos interpretabantur [sc. Livius et Ennius], aut si quid ipsi Latine conposuissent praelegebant." Büchner, SO 54 (1978) 61, goes so far as to call Andronicus poeta doctus. We do not know what Antimachus' edition entailed. References to it among the Homeric scholia are gathered by Fraser, Ptolemaic Alexandria, vol. 2b, 648 n. 12.

38 " v i r e s c u i s u n t magnae / topper confringent inportunae undae," fr. 18.2-3. Mariotti, Livio Andronico, 35; cf. Büchner, SO 54 (1978) 47-48.


originated the so-called dialect gloss, the use of contemporary dialect words to color poetic diction, and this trait too has been found in Andronicus. Where Homer, for example, used the Aeolic εννεπε to open the Odyssey, Andronicus' corresponding insece has been iden-tified as an Umbrian formation. "In short," goes the argument, "Livius might be described by the same phrase which Strabo (14.657) used of Philetas of Cos: ποιητής αμα καΐ κριτικός—i.e. the typically Hellenistic combination of poet and scholar."39

This is an appealing but ultimately unconvincing notion. The first objection to it is chronological. According to Suetonius, who is not so much reporting facts as constructing an argument from them, erudition on the level associated with Antimachus and Philetas did not reach Rome until a generation or more after Andronicus, and it then took some time to feel at home. He dated the beginning of this process to about 168, when the noted philologist Crates of Mallos arrived on a diplomatic mission from Pergamum. Crates promptly stepped in a drain, broke his leg, and spent his convalescence introducing Romans to the professional Studium grammaticae. Serious literary exegesis at Rome thus began with Crates and those who followed his example.40

The first of these imitators, Suetonius continues, were C Octavius Lampadio, whom he credits with the book divisions of Naevius7

Bellum Punicum, and Q. Vargunteius, who expounded (pronuntiabat) on Ennius' Annals. Neither of these, however, was a fully professional grammarian ("hactenus tamen imitati [sc. Cratem]"), and probably neither was active until the later second century. Not until half a century or more after Crates' visit would Aelius Stilo and his son-in-law Servius Claudius set grammatica on a firm and orderly footing at Rome ("instruxerunt auxeruntque ab omni parte gram-maticam," 3.1). This account by Suetonius, which remains our main source of information on these points, may not be entirely satisfactory, but it clearly provides no basis for attributing Hellenistic standards of learning to Livius Andronicus.

The use of Antimachus as a paradigm for Andronicus' stylistic experiments should also give us pause. Precious little of his hex-ameter Thebais or his elegiac Lyde survives, and while that is no

Quintilian, cited by Mariotti (42), mentions Antimachus' vis et gravitas et minime vulgare eloquendi genas (10.1.53) but also his singular lack of grace.

3 9 Sheets, A]Ρ 102 (1981) 62. Sheets finds four dialect glosses in the epic fragments of Andronicus. Whether κριτικός actually means "scholar" in anything like the modern sense is at best uncertain. See Brink, Horace on Poetry iii, 414-19, and, for the negative side of Philetas' learning, Cameron, CQ 41 (1991) 534-38.

40 Suet. Gram. 2.1 with Kaster, Commentary, ad loc. Cf. Garbarino, Roma e la filosofia greca, vol. 2, 356-62, and Quinn, ANRW II.30.1 (1982) 97-99.

Reconstructions 49

great loss to letters, it has made his name too easily taken in vain. Yet hints remain to keep us honest. A fragment of the Thebais points the way. It may come from the proemium.41

δή τότ' Έρεινύς ήλθεν άσήτορος [Οίδιπόδαο "Αϊδος έκπρολιποΰσα θοον δόμον [

Then came the Fury of [Oedipus?] sick at heart, having foresaken the looming house of Hades...

The description of a divinity leaving one place for another is a familiar topos in ancient poetry; Antimachus puts his own stamp on the commonplace by employing a singularly recondite vocabulary. In the first line, the word άσήτωρ, "sick at heart" is unique. It is probably his own coinage and is generally, though not universally, explained as an agent noun from άσάω. The participle έκπρολιποΰσα in the second line is much less obscure, but only slightly less rare. The verb appears only once in all of Homer, part of the highly charged and memorable description of the Greeks emerging from the Trojan horse: ίππόθεν έκχύμενοι, κοίλον λόχον έκπρολιπόντες ("streaming from the horse, forsaking their hollow sanctuary/'Od. 8.515). Its only other appearance in literature before Antimachus, also a participle and with perhaps a Hesiodic echo, is at Theognis 1136, which de-scribes how all blessings save Hope have forsaken earth for Olympus. Antimachus has thus chosen a word at once rare, clear, and apposite. The act of forsaking is harmful in every case, and each appearance is participial. He borrows both connotation and con-struction to heighten his own epic moment.

All this is rather adroit, though Antimachus' fondness for neo-logisms and oddities can also get out of hand. An admonitory example comes from the Lyde: ενθα Καβάρνους θήκεν άβακλέας όργειώνας (fr. 67 Wyss). It must mean something like, "there he positioned the Ka-barnoi, wagon-driving holy men." The Καβόφνοι were priests of De-meter in Parium. Άβακλεύς, another unique agent noun, is apparently formed from the equally unique word άβακλή (a wagon?), and όργειών is a rare poetic form of an Attic legal term for religious celebrants that later became a metaphor for priests. All this is a heavy load of learning for a single line. Small wonder that Callimachus thought the Lyde flabby and turgid (παχύ γράμμα και ού τορόν, fr. 398Ρ) and that Catullus would call Antimachus tumidus (95.10) 42

41 Fr. 187 Wyss, identified now in P. Oxy. 2516 fr. 4 = SH no. 65.1 quote just the first two lines with Maas' supplement offered exempli gratia.

4 2 Krevans, Hellenistica Groningana 1 (1993) 156-59, explains Callima-chus' complaint in terms of the Lyde's "florid language and metrical rough-

50 Epic in Repiiblican Rome

This is obviously not the language of Andronicus. It cannot be. Antimachus' style is, for better and worse, rooted in and made possible by a literary language with a long history. Naturally so, for when stories are well known and literature abundant, the poet can put greater emphasis on the telling than on the tale. Allusions assume prior contexts. Coinages and obscurities challenge the reader's ex-perience as well as the writer's cleverness. It is a poetics of retrospect, built upon and incorporating the history behind it. There was no such history and no such need in third-century Rome. Latin literature was Andronicus' to begin. His appeal can still lie in the story. The sense of style we detect in his poem does not require us to assume a specifically Hellenistic aesthetic. His epic indeed sounds different from his comedies, but the result is hardly Antimachean. He employs a scholion to avoid an obscurity, not to create one. The pathos of his inportunae undae is, as we shall see, more a technical function of Saturnian composition than a reflection of Hellenistic sensibility. Close attention to the linguistic facts of early Latin has denied the presence of dialect glosses in the Odussia.43 Andronicus' epic found other sources of appeal.

The story he retold was of immediate interest to Romans of the later third century. The Odyssey was not just a more exciting and varied tale than the Iliad; it was also one with strong Italian associations. Many of its episodes were traditionally located in Italy and its environs. Travelers could visit memorials to Odysseus' com-panions Elpenor on the coast of Latium and Drakon in Leucania. Post-Homeric tradition identified the promontories at Misenum and Baiae with Odysseus' men Misenos and Baios, and sons of Odysseus and Circe named Ardeias, Antias, and Romos were made the eponymous founders of Ardea, Antium, and Rome. Their better-known son Tele-gonus, hero of the epic Telegonia by Eugammon of Cyrene, had strong associations with Tusculum and Praeneste. Nor was Odysseus himself entirely outside the Roman story: in the fifth century, Hellanicus apparently joined Odysseus with Aeneas as a founder of Rome.44 This fusion—and eventual confusion—in the stories of Greek and Trojan

ness." Catullus contrasts tamidus Antimachus with the parva monimenta of his friend Cinna. Quint. Inst. 10.1.53 reflects a similar verdict.

4 3 Kearns, AJP 111 (1990) 40-52. 4 4 For the western settings of the Odyssey, Str. 1.2.18. Tomb of Elpenor,

Thphr., HP 5.8.3, Plin. 15.119; heroön of Drakon, Str. 6.1. Eponymous sons recorded by Xenagoras of Heraclea ap. D.H. 12.72.5 (FGH 240 F 29); traditions of Telegonus in D.H. 4.45.1, Liv. 1.49.9, Festus 116L, Hor. Epod. 1.29 and C. 3.29.8, Prop. 2.32.4. Odysseus is identified as founder of Rome in Hellanicus ap. D.H. 1.72.2. See in general Phillips, JHS 73 (1953) 53-67, Solmsen, HSCP 90 (1986) 93-110, and Gruen, Culture and National Identity, 17-21.

Reconstructions 51

origins were both appealing and helpful to the Romans. Links to the great age of heroes established Rome's place in the Greek world, while the legend of Trojan origin through Aeneas also marked its difference. Romans could thus assimilate Greek cultural influences without surrendering their own identity. Small wonder that Andronicus' convenient version of Odysseus' adventures found a readership and, very quickly, a successor.

Andronicus opened the world of epic to Latin literature. Naevius made it Roman by elevating current events to epic proportions. He wrote of a war he knew at first hand (fr. 44), but its telling did not lack ties to Andronicus' Odussia. Naevius also chose the Saturnian meter, and he also looked back to the Greek heroic age. The poem included the story of Aeneas' flight from Troy (fr. 5-7) and echoes of Odyssean adventures (fr. 15). There were divine debates and inter-ventions of the epic type (fr. 9, 14). Naevius also mentioned Aeneas' visit to Carthage (fr. 21-23) and the founding of Rome (fr. 25-29). Yet his combination of mythic and historical narrative was not simply a history of Rome from the fall of Troy to the peace of 241. It was too compact and too intricately structured to ramble in annalistic fashion. Naevius shared Andronicus' sense of scale: not until at least a generation after its composition did C Octavius Lampadio, Rome's first attested editor, divide the Bellum Punicum into the seven books known to later generations (Suet. Gram. 2, cf. Santra ap. Non. 250L).

Many of its fragments are identified by these book numbers, and a curious pattern emerges from the original sequence. The first book refers explicitly to the consul Valerius' expedition to Sicily in 263 (fr. 3) but also describes Aeneas fleeing from Troy (fr. 6). Aeneas' visit to Carthage came in Book 2 (fr. 23), and he had landed in Italy by Book 3 (fr. 26). The historical narrative had certainly resumed by Book 4, which mentions a Roman invasion of Malta in the early 250s (fr. 32). As unassigned fragments are fitted within this framework, it becomes clear that the legendary material was embedded in the historical narrative.45 How Naevius moved back into legendary time remains uncertain, but attention often centers on fragment 4:

inerant signa expressa, quo modo Titani, bicorpores Gigantes magnique Atlantes Runcus ac Purpureus, filii Terras

4 5 Strzelecki's demonstration in De Naeviano Belli Punici Carmine quaestiones selectae commands belief. See also Rowell, AJP 68 (1947) 21-46, and the preface to Strzelecki's first edition of the fragments, Cn. Naevii Belli Punici Carminis quae supersunt (Wroclaw 1959) 54-79. His Teubner preface (1964) is much abridged. If, as Strzelecki suspects, Lampadio's book divisions were determined more by length than by content, the exercise was perhaps more mechanical than truly "critical."


Figures were represented on it, how the Titans, double-bodied Giants and great Atlases, Runcus and Purpureus, sons of Earth

The curious plural Atlantes put Hermann Fränkel in mind of the atlantes that were a famous architectural feature of the Temple of Zeus at Agrigentum.46 These figures, which Romans called telamones (Vitr. 6.7.6), were a set of colossal male statues over twenty feet tall that, like the Caryatids of the Athenian Erectheum, supported the architrave of the building. Fränkel suspected that familiarity with the architectural term had led Naevius to use it in describing a Gigantomachy. There was in fact a Gigantomachy depicted on the east pediment of the temple, and Fränkel therefore identified the fragment as coming from a description of that sculpture. Readers, having followed the consul Valerius to Sicily, then stand before the temple at Agrigentum. Since the west pediment of this temple illustrated the Homeric heroes at the fall of Troy (D.S. 13.82), one of Naevius' Romans could have recognized Aeneas among them. One association would bring on another, leading to the story of his flight and his voyage to Italy. The effect would be similar to Aeneid 1, where the pictures of Troy on Dido's Temple of Juno prefigure Aeneas' narrative in Book 2.

Fränkel's engaging suggestion has not won, and really cannot win, universal acceptance. The association of Naevius' Atlantes with the figures at Agrigentum is not explicit in the text, while the signa, which are, are not necessarily statues. Embossed, painted, or em-broidered figures are also possible. The Gigantomachy has therefore been explained with rival claims to probability as appearing on a shield, a ship's beak, or a garment, and once the fragment is divorced from the Temple of Zeus, its structural significance vanishes.47 Yet the tentative quality of Fränkel's suggestion cannot obscure a basic fact: some such device must have joined the historical and legendary narratives. The link was not purely chronological. The Bellum Punicum had an intricate structure, and thus starts to look like a very interesting poem.

That inherent interest has again encouraged claims of Alexandrian influence, and this time the case is a little stronger. Naevius used historical and mythic material in his work, and there was genuine Hellenistic precedent for each usage. These precedents—first history,

46 Fränkel, Hermes 70 (1935) 59-61. The structural implication of this iden-tification is developed by Klotz, RhM 87 (1938) 190-92, and by Rowell, AJP 68 (1947) 32-39 and AJP 78 (1957) 424-25.

4 7 Fraenkel, JRS 44 (1954) 14-17 (a shield); Warmington, Remains of Old Latin, vol. 2, 67 (a ship); Wimmel, WS 83 (1970) 84-100 (an embroidered robe).

Reconstructions 53

then mythology—are worth considering briefly. Recounting contemporary events in epic verse was certainly a practice, though not an invention, of the Hellenistic world. It began as early as Choerilus of Samos, a contemporary of Antimachus, who wrote an epic Persica on events of the early fifth century and was recruited by that century's end to sing the praises of the Spartan general Lysander (Plut., Lys. 18.7). Alexander, following Lysander's lead, indulged his Achillean aspirations by keeping a brace of versifiers on hand to immortalize his deeds in epic. None was much good as a poet, but the seed was sown. By the time Ptolemy II defeated a contingent of rebellious mercenaries in 274, there was a poet on hand to record his success.48

This kind of occasional poetry, however, was probably neither epic in length nor necessarily epic in form. The poem for Ptolemy was written in elegiac couplets; poems written for public recitation, as were many of these, were unlikely to be very long. The widespread modern belief that the Hellenistic age produced many long hexameter poems on historical or contemporary deeds finds little support in the actual record of Greek texts and testimony.49 Though poetry on historical themes would eventually become common at the courts of Antioch and Pergamum and inspire—if that is the word— the topical epics of first-century poets like Archias and Varro of Atax, there is no immediate connection to be made between this kind of occasional verse in the third century and Naevius' national epic.

Mythological material in poetry often centered, as it apparently did in the Bellum Punicum, on the founding of cities. The so-called ktisis poems had a long history and enjoyed a considerable vogue in the Hellenistic world. The use of foundation myths as a theme for poetry can be traced back to Homer's brief excursus on the history of Rhodes at Iliad 2.653; in Hellenistic times such poetry generated considerable appeal.50 Much of it was written by itinerant (and

4 8 PHamb. inv. 381 = SH 958, a papyrus of the iii B.C. I follow the identification in SH, but other explanations are possible; see ad locv p. 460. The Alexander poets were of near-legendary badness. Curtius Rufus 8.5.8 calls the Argive Agis "pessimorum carminum post Choerilum [sc. of Iasus] conditor." The tradition is hardly kinder to Anaximines, Cleo of Sicily, and Pyrrhon of Elis.

4 9 Ziegler, Das hellenistische Epos is the main popularizer of this view, but see the criticisms in Otis, Virgil, 396-98, and now Cameron, Callimachus and His Critics, chap. 7. The only significant historical epics we know of are Choerilus' Persica and Rhianus' Messeniaca. Rhianus was a near contem-porary of Livius Andronicus. For these two, see Hainsworth, The Idea of Epic, 60-67.

5 0 Cairns, Tibidlns, 68-86, gathers the evidence for the theme of ktisis in poetry and defines the principle features of the emerging genre.


otherwise unknown) versifiers who traveled through the eastern Mediterranean praising—and being rewarded by—the cities they visited: records of their successes survive through inscriptions erected by the grateful cities whose origins they exalted. A more literary side to such activity might be inferred from the career of Apollonius, who wrote ktiseis celebrating Alexandria, Caunus, Cnidus, Nau-cratis, Rhodes, and Lesbos. The related phenomenon of embedded aitia became a famous feature of his Argonautica.51

Whether Naevius' combination of mythical and historical nar-ratives actually owes anything to these Greek precedents is entirely unknown. Choerilus may have been the first to join myth to history: his Persica at least alluded to the story of Boreas and Orithyia.52

Others presumably followed his lead, and it has become common-place to interpret the blending of mythic and historical elements in Catullus and Vergil as an Alexandrian mannerism. This says nothing about Naevius. We can no more leap from the Bellum Punicum to the Persica, explaining one spotty set of fragments by recourse to an even spottier set, than we can leap from Naevius to Vergil, who certainly found things to borrow (and to better) in the Bellum Punicum.

Vergil's own style of Alexandrianism has nevertheless been read back into Naevius. Encouragement of a kind comes from Macrobius' testimony that the famous storm of Aeneid 1 and the idea of Aeneas' visit to Carthage were lifted from Naevius. Whether Naevius actually used a Dido episode as a mythic motivation for the Punic War is unknown and therefore much debated, but the germ of the idea was doubtless present.53 And so was, according to some critics, a certain delicacy of sentiment. It is most often claimed for fragment 23, an unassigned fragment that Strzelecki places in Book 2.

blande et docte percontat, Aenea quo pacto Troiam urbem reliquerit

kindly and carefully asked how Aeneas left the city of Troy

5 1 Hutchinson, Hellenistic Poetry, 93-96. Testimony and fragments of Apollonius' ktiseis are found in CA 5-8. On all this see chap. 7.3 of Cameron, Callimachus and His Critics.

52 Colace, Choerili Samii Reliquiae, 49-58; Suerbaum, Untersuchungen, 18-20.

53 Varro's report ap. D.S. ad 6.682 that Anna, not Dido perished for love of Aeneas could have come from Naevius. For discussion of the Dido episode in Naevius, see Wigodsky, Vergil and Early Latin Poetry, 29-34. To the further references there, add Luck, ICS 8 (1983) 267-75. Macr., Sat. 6.2.31 (= fr. 14 St.) is quite explicit about the storm: "hie locus totus sumptus a Naevio est ex primo libro Belli Punici."

Reconstructions 55

The subject of percontat is unknown. Many would like it to be Dido: blandus is a word thought suitable to women and doctus particularly suited to Dido. We might even recall Aeneid 1.670-71, "nunc Phoe-nissa tenet Dido blandisque moratus / vocibus," and hear a Naevian echo.54 A Vergilian sensibility is then claimed for Naevius. This pleasant notion, however, is hardly secure. The arguments for it are too easily inverted.

Vergil's borrowings from the Bellum Punicum are far less striking than his many departures from Naevian precedent. Naevius' Aeneas left Troy with only one ship, which was built by Mercury (fr. 7). Anchises was himself a seer, his powers the gift of Venus, and he lived to reach Italy with Aeneas (frs. 8, 9, 25). Romulus was the offspring of Aeneas' daughter (fr. 27). Abandoning these traditions enabled Vergil to expand his narrative and heighten its Homeric associations. Aeneas' fleet of ships, which he is so conspicuously unable to lose, stands in pointed contrast to Odysseus' voyage of increasing isolation. Vergil's Anchises can prophesy only from the Underworld, which necessitates the famous echo of the Homeric nekyia, and his death in Sicily provides the opportunity for funeral games on the Homeric model. Separating Aeneas' arrival in Latium from the eventual founding of Rome by descendants from Alba keeps the Aeneid in the older, Homeric world of heroes. Naevius not only told a simpler, more direct story of Rome's foundation; he also lacked those explicit associations with the Homeric poems that are so characteristically Vergilian. Mariotti's claim that Naevius deliber-ately united characteristics of the Iliad and Odyssey thus finds little support in the fragments.55 That synthesis was actually Vergil's.

The case for an "Alexandrian" Naevius comes down to the basic fact of a foundation story set into the historical narrative. This is not much grist for the mill. The Bellum Punicum is itself no ktisis, only cousin to one, and its historical narrative lacks the focus on an individual—a Lysander, an Alexander, or a Ptolemy—that seems to characterize Hellenistic historical poems. Naevius' epic had no single patron and no single hero. It could praise the deeds of individual Romans (frs. 32, 34), but it could also reveal their excesses (fr. 37). Its echoes of Greek materials and Greek devices should not distract attention from its essentially Roman qualities. This was a national epic just as Naevius' fabula praetexta was a national

5 4 Summarizing discussion can be found in Barchiesi, Nevio epico, 477-82; Mariotti, Bellum Punicum, 29-31, is especially astute.

5 5 Mariotti, Bellum Punicum, 13: "Nevio ha avuto l'intenzione, al pari di altri epici antichi, di riunire in unico poema le carratteristiche dominanti delV Iliade e delV Odissea." (His emphasis: the point is central to his argu-ment.) See his summarizing discussion in Belfagor 20 (1965) 39-42.


drama, and the sense of communal rather than individual achievement is not a Hellenistic trait. It is distinctly Roman. Naevius simply presses Greek traditions into Roman service, making legendary Trojans the ancestors of contemporary Romans and recasting the gods of Greek epic as the guarantors of Roman greatness. The closest literary parallels to the Bellum Punicum are therefore to be found not at Alexandria but among his own stage creations, like Lupus, on the city's founding, and Clastidium, dramatizing a con-temporary military achievement. Roman literature from the beginning showed a will of its own.

Efforts to turn Andronicus and Naevius into poets with Alexandrian sensibilities misconstrue the intellectual climate of their time. Horace's famous claim that conquered Greece subdued its savage conqueror, bringing the arts to rustic Latium ("Graecia capta ferum victorem cepit et artes / intulit agresti Latio," Ep. 2.1.156-57) is not a fact of Roman literary history. It is an interpretation. As heir to the intellectual and artistic ferment of the late Republic, Horace might well cast the development of Latin literature as a process of increasing Hellenization, but we do not need to see the developments of our period with Horace's eyes. Romans of the second century certainly did not. Their relations with Hellenism, as Arnaldo Momi-gliano reminds us, were never simply those of master and disciple: "They [the Romans] acted from a position of power and effortlessly preserved a strong feeling of their own identity and superiority. They paid the Greeks to teach them their wisdom and often did not even have to pay because they were their slaves." The result was what Momigliano rightly calls "Latin Hellenism," never quite the same as the Greek variety and not to be measured by its fidelity to the original.56 There is no need to assimilate the Saturnian poets to the patterns of a later age or to foist the rudiments of an Augustan aesthetic upon them. Yet the claim of a strong Hellenistic influence on the early Latin poets has a worthy motive. It arises from a belief in, or at least a search for, quality in those poets. Its advocates are willing to take them seriously, which is something other Latinists have been singularly reluctant to do.

The usual alternatives to a Hellenistic reading of early Roman epic have been generally unappealing or unflattering. Here, for example, is Brooks Otis, hardly an insensitive reader of Latin poetry:57

5 6 Momigliano, Alien Wisdom, 10-11. 5 7 Otis, Virgil, 20. The other common alternative is silence. Thus Newman,

The Classical Epic Tradition, 106, devotes barely a paragraph of a very big book to epic between Apollonius and Vergil.

Reconstructions 57

The first Roman epic (apart from Livius Andronicus' crude version of Homer) was the Bellum Punicum of Naevius (written toward the end of the third century B.C.). This was, as Ennius somewhat rudely declared, the poetry of fauns and fortune-tellers (Fauni vatesque), a highly unsophis-ticated composition in the native Saturnian measure.

Calling Andronicus7 poem "crude" and Naevius' poem "highly unsophisticated" casts assumptions as facts, and to dismiss the Bellum Punicum as "the poetry of fauns and fortunetellers" mistakes Ennius' description of its meter for a judgment on its quality. Yet it is easy to join Otis in assuming that short, isolated fragments in an old and occasionally peculiar style conceal little of merit. Their artistry is not easily grasped. Otto Skutsch once put the critic's dilemma well: "Literary criticism in the narrow sense can . . . find little scope in these scanty remains. The commentator is restricted to elucidating as far as possible the biographical and historical background of the poet's work and explaining the individual fragments."58 That opinion still prevails, but Skutsch's alternatives have revealed limitations of their own. Biographical and historical criticism, with their fixation on such dubious facts as Naevius' exile, divert attention from the poetry that is, after all, his true claim to attention. "Explaining the individual fragments" has come to mean reconstructing their lost contexts or, as is now more often the case, debating the current inventory of conjectures. It is a scholarship of suppositions and hypotheses. Can no more be said? Must literary criticism "in the narrow sense" be silent in the face of this fragmentary evidence?

No. The literary history we have been reconstructing has an internal dynamic as well as a factual base. It is a complex mix of poetic features and poetic ends, of combinations that faded from the scene and others that endured. The literature that began as public spectacle was losing its creative impetus at Rome just as literature for private recitation—epic—was achieving its first great success. Terence's Hecyra failed barely a decade after Ennius' Annals succeeded, and the Annals, not the Hecyra, set the future course for Latin poetry. Yet the Annals did not itself arise from nothing. That earlier poetry of fauns and fortunetellers requires more than apology or condescension. The cumulative artistry of Andronicus and Naevius, as well as Ennius, made Latin epic possible. Understanding the nature of their artistry and its effect on the growth of epic requires both the practical criticism of the fragments and attention to the dynamics of literary development in second-century Rome. The texts must be set in aesthetic and cultural contexts. The literary critic still has work to do.

5 8 Skutsch, review of Marmorale's Naevius poeta, CR 65 (1951) 174.

3 Saturnian Aesthetics

Horace knew what he thought of Saturnian verse, and the thought was not kind. When Greece imposed its arts on rustic Latium, an early and welcome casualty among the natives was the Saturnian, whose demise he likened to the draining of a marsh.

sic horridus ille defluxit numerus Saturnius, et grave virus munditiae pepulere.

Thus that harsh Saturnian rhythm drained away, and elegance drove off the brackish scent. (Ep. 2.1.157-59)

The greater appeal of the Greek import proved undeniable and un-stoppable. In the course of the second century, a Latinized form of dactylic hexameter became the recognized epic meter at Rome, and elegiac inscriptions began replacing Saturnians on the tombs of the ever-progressive Scipios. Yet the stylistic changes that brought Latin verse from Saturnian thud to hexameter flow did not immediately or entirely eclipse the work of the earliest epic poets. Even while Horace was heaping his faint damns on Andronicus' Odussia, Romans were still reading Naevius' Bellum Punicum, and the horridus nume-rus was still heard as poetry.1 What kind of poetry was it? How did it contribute to the development of an epic aesthetic at Rome?

Appreciating the artistry of Saturnian poets is no easy task. Not only are their poems in ruins, but the meter itself remains something of a mystery. Two thousand years have passed since anyone has really understood it. The name had apparently become traditional by the time of Cicero's friend Varro (L 7.36), and controversy over its nature and origin is nearly as old. Is its governing principle

1 As Horace himself admitted, Ep. 2.1.53-54: ''Naevius in manibus non est et mentibus haeret / paene recens?" For the reception of Naevius' poem in antiquity, see the preface to Strzelecki's Polish edition (Wroclaw 1959) 9-18 and Barchiesi, Nevio epicof 12-29 and 38-52.

58

Saturnian Aesthetics 59

quantitative or accentual? Was the meter itself imported from the Greek world or native to Italy? Because ancient grammarians as-sumed that Greek metrical practices reflected universal principles of language, they worked assiduously, though unconvincingly, to ex-plain the Saturnian in quantitative terms. It was a fruitless struggle. Caesius Bassus, writing in the time of Nero, argued strenuously for the meter's Greek origin but was nevertheless forced to acknowledge that Greek authors must have used it quite differently from their ostensible Latin successors ("a Graecis enim varie et multis modis tractatus est"). The scansions he proposed to explain its quantitative basis necessarily defy consistency. Three centuries later, Marius Victorinus concluded his own, largely derivative discussion of the same problem with the admission that the question was still contro-versial ("apud grammaticos super hoc adhuc non parva lis est").2

Modern scholarship has enjoyed little more success. One small but characteristic example may demonstrate the extent and difficulty of the problem. Eduard Fraenkel, similarly convinced that the Saturnian was a Greek import, adduced as evidence for that claim the opening line of the Hymn to Dictean Zeus, which is now usually dated to the fourth or third century B.C.3

ίώ μέγιστε Κούρε, χαΐρέ μοι Κρόνειε. Ο greatest Youth, Hail son of Kronos

Fraenkel pointed not only to the line's iambic scansion, which re-sembles a verse like "malum dabunt Metelli Naevio poetae," but also to its striking coincidence of metrical cola with syntactic units, a recurring phenomenon among the Saturnian fragments. Naevius' "noctu Troiad exibant capitibus opertis" is an example. The Greek line, concluded Fraenkel, "is versus Saturnius purus putus. Caesius Bassus could have wished for no better model of his ideal form,"

At first glance, the case is compelling, but some further considerations weaken it considerably. The Greek line Fraenkel

2 GLK 6.265 (Caes. Bass.); GLK 6.138 (Mar. Vict.). The ancient testimonia are conveniently gathered by Luiseiii, II verso Saturnio, 105-14, with discussion on pp. 13-103. Palmer, Roman Religion, 173-85, speculates on the origin of the verse's name. For a helpful survey of modern work, see Waszink, ANRW 1.2 (1972) 875-85. More in Barchiesi, Nevio epico, 310-27; for perspective on earlier modern theories, see Timpanaro's introductory essay to the reissue of Pasquali, Preistoria della poesia romana, 7-80. What will follow owes most to Cole, YCS 21 (1969) 3-73, which is seminal. Blänsdorf, "Metrum und Stil/' 41-69, is also good on metrical features, but rather less interested in artistic effects and the relationship between meter and sense.

3 Fraenkel, Eranos 49 (1951) 170-71. The Greek text is most easily found in CA, 160-61, but I follow the colometry of West, Greek Metre, 148.


quotes opens the hymn's refrain, which in its entirety has more accurately been printed like this:

ίώ μέγιστε Κοΰρε χαΐρέ μοι Κρόνειε

παγκρατές, γαν ος βέβακες δαιμόνων άγώμενος· Δίκταν ές ένιαυτον ερπε και γέγαθι μολπαΐ.

Ο greatest Youth, Hail son of Kronos

the almighty, who walks the earth as leader of gods: Come to Dicte for a year and rejoice in song.

Things suddenly look quite different. Fraenkel's claim of a Greek antecedent here for the Saturnian requires not just the metrical and syntactic correspondence of two consecutive cola, but a common metrical context for the Greek and Latin patterns as well. The full quotation shows that this larger context is lacking. The enjambment of Κρόνειε / παγκρατές moves us immediately to lines of quite differ-ent metrical shape which do not preserve the strict coincidence of syntactic and metrical units that is crucial to Fraenkel's case. The parallel he has adduced ultimately fails. Though Saturnian rhythms are certainly reminiscent of iambo-aeolic patterns, their recurrent metrical configurations are not in fact to be found here or anywhere else in the extensive corpus of extant Greek verse.4

A narrower approach based on the close analysis of the Latin evidence alone has proven no more helpful. The scansion of Saturnian verses remains as intractable for us as it was for Caesius Bassus, and the number of such verses to consider is now even smaller. Only some 125 authentic Saturnians survive. They reveal no normative pattern, either quantitative or accentual, on which to base a coherent metrical analysis, and vagaries of transmission often make essential details of their prosody uncomfortably problematic. It has also proven exceed-ingly difficult to test any analytic hypothesis against the norms of the language, a step necessary to show that recurring patterns of heavy and light syllables are indeed deliberately sought features of the meter rather than mere consequences of Latin phonology.

Given these uncertainties, the Saturnian's basic organizing princi-ples must remain a mystery, but a few formal features of the meter are clear enough to provide an adequate operational definition. Thus Thomas Cole describes the Saturnian as "any archaic Latin verse that is used stichically and divided by a caesura into two parts, the first of which contains five to nine syllables, the second (usually one to three syllables shorter than the first) five to eight syllables" (p.

4 Detailed discussion in Cole, YCS 21 (1969) 49-59.


10). This need not mean that the verse pattern was—like the French alexandrine—actually based on syllable count. We can only be sure that an arrangement in asymmetrical cola determined by a prominent caesura is its most apparent and dependable feature.5

A discussion of poetic technique in Saturnian composition is therefore necessarily incomplete: effects based on extended rhythmic patterns or incremental repetitions will have vanished through lack of context, and the relationship of style to content may now often elude us. Other mannerisms, however, are plain enough to encourage other kinds of aesthetic analysis. The role of caesura and line end in defining units of sense, the way ideas are expressed in the resulting cola, and the devices of sound and syntax that unite cola into complete verses will repay attention. Here the alexandrine line, with its formal division into two six-syllable cola, provides a helpful analogy. Metrical and grammatical boundaries may coincide, as in the opening of Racine's Phèdre:

Le dessein en est pris: je pars, cher Théramène, et quitte le séjour de l'aimable Trézène.

The poet, however, may also create a much more elaborate tension among the metrical, syntactic, and dramatic demands of his verses:

Thésée La fortune à mes voeux cesse d'être opposée, Madame; et dans vos bras met...

Phèdre Arrêtez, Thésée,

Et ne profanez point des transports si charmants. (913-15)

What may initially seem a rigid and highly artificial metrical form thus proves to be quite flexible and can generate considerable dramatic power.6

No early Roman poet, of course, wrote like Racine. Extant Satur-nians cannot claim this level of technical dexterity in manipulating the demands of meter and sense, nor were Saturnians used for drama, but the shape of the verse did create some stylistic opportunities of this sort. Even lines as unpromising as this dedicatory inscription of

5 "Caesura" is, I think, the appropriate term for the place where two metrical cola meet. The distinction made in metrical handbooks between caesura within a metron and diaeresis between metra is modern and seems unhelpful here, where cola are discernible but feet are not. See West, CQ 32 (1982) 292-97.

6 On the drama see Maskell, Racine, 125-32; for the technical consid-erations in alexandrine composition, Howarth, "L'Alexandrin classique/' 341-54.


L. Mummius reveal features of note. Here, as throughout this chapter, I have added space at the caesurae to make the colon-structure apparent at a glance.

duct(u) auspicio imperioque eius Achaia capt(a) Corinto deleto Romam redieit triumphans.

under his leadership, auspices, and command, Achaia was captured: with Corinth destroyed, he returned to Rome in triumph. (ILLRP, no. 122)

Ablative phrases are set in the first colon of each line and finite verbs in the second; the two initial cola are linked to their shorter companions by eius looking back in the first line and Romam looking ahead in the second. Such stylistic features arise from established characteristics of the verse form. And there are others. Under-standing the Saturnian aesthetic may be scarcely less problematic than the evidence, but it is certainly no more so.

Yet the earliest Saturnians offer little hint of literary potential. The elogium for L. Cornelius Scipio (cos. 259) may be nearly con-temporary with his death, which would date it to ca. 240-30 B.C. The original is carved in stone—a larger titulus above it is painted— and sets out one verse per line. The right edge has suffered damage, though the restoration is fairly certain.

Hone oino ploirume cosentiont R[omane duonoro optumo fuise viro Luciom Scipione filios Barbati consol censor aidilis hic fuet a[pud vos hec cept Corsica Aleriaque urbe dedet Tempestatebus aide mereto[d

This man most Romans agree was the very best of good men, Lucius Scipio, the son of Barbatus. He was consul, censor, aedile among you, He captured Corsica and the city of Aleria, He fittingly dedicated a temple to the Weather. (ILLRP, no. 310)

The caesurae here define short cola that are themselves often significant phrases: the set of titles in line 4, the pair of actions in line 5. Only the fifth line puts a complete clause in a single colon. Verb and complement are usually separated by the caesura. Each line, however, represents a complete thought, and the lines them-selves are arranged into larger units of meaning. If, as seems likely, filios (i.e., filius) in line 3 is the carver's erroneous expansion of what was intended to be an accusative filio in his exemplum, the elogium divides neatly into sections of three lines each, the first (with Scipio


as object) extolling his ancestry and the second (with Scipio as sub-ject) recounting his achievements.7

Such a division into three-line units is certain for the slightly later elogium to Scipio's father, Barbatus (cos. 298), which probably took this one as its model. An earlier, painted inscription in prose was erased from Barbatus' impressive tufa sarcophagus and replaced with the following incised verses:

Cornelius Lucius Scipio Barbatus Gnaivod patre prognatus fortis vir sapiensque quoius forma virtutei parisuma fuit consol censor aidilis quei fuit apud vos Taurasia Cisauna Samnio cepit subigit omne Loucanam opsidesque abdoucit

Cornelius Lucius Scipio Barbatus father Gnaeus' offspring, a brave and educated man, whose beauty matched his character. Consul, censor, aedile was he among you. Taurasia, Cisauna, Samnium he conquered. He subjugated all Lucania and brought back captives. (ILLRP, no. 309)

Beside features of phrasing and arrangement similar to those em-ployed for the younger Lucius, this elogium provides physical evi-dence for a strong sense of verse structure. Erasing the earlier inscription had left a long but relatively narrow space for this new one, forcing the carver to incise these six verses continuously on just four lines. He nevertheless marked line ends with a long dash, as if anxious to declare that these were indeed verses.8

The common stylistic features of these elogia, however, probably owe less to art than to the exigencies of verse composition itself. Since a verse is commonly the metrical equivalent of a sentence, each line quite naturally conveys a coherent thought, and since cola are best defined as metrical features determined by grammatical boundaries, we can hardly be surprised when Saturnian cola coincide with

7 For the carver's error, see Ernout, Recueil de textes latins archaïques, 16; Van Sickle, "First Hellenistic Epigrams at Rome," 146 n. 25. It is hard to accept the otherwise violent division of the Roman name formula between sentences and with change of case. Note the following elogium, where the formula was so strongly felt that the carver omitted to mark the verse division after the first line, which divides pieces of the proper name. In any case, I fail to see the arrangement in couplets that is claimed for these poems by Van Sickle, AJP 108 (1987) 41-55, and "First Hellenistic Epigrams," 144^8.

8 Discussion and plate in Gordon, An Illustrated Introduction to Latin Epigraphy, 80-82; for the punctuation, see especially Vine, Archaic Latin In-scriptions, 352-53. The Scipionic inscriptions are reproduced as a group in Degrassi, Imagines, no. 132-38.


grammatical phrases or even short clauses.9 The Saturnian cadence, however, is not so common. Indo-European spoken meters generally avoid divisions of equal length, probably to avoid a false sense of line end at the caesura, and bipartite lines normally keep the first colon shorter than the second.10 Yet Saturnian cola may be iso-morphic, and where one colon is indeed longer, it is invariably the first. The cadences produced therefore tend to be highly regular line after line, and the meter encourages decrescendo over crescendo. Thus, while the artistry of these early inscriptions may appear rudi-mentary, the artists' medium does not itself seem very tractable. Though adequate for brief statements and catalogues, how can a sequence of such verses develop a more complex idea? Can the Saturnian be shaped into something more appealing?

The opening line of Andronicus' Odussia reveals a different order of craft.

Virum mihi, Camena, insece versutum Tell me, Camena, of the clever man

Comparison with his famous model enables us to pry into the Saturnian poet's aims as well as his techniques.

"Ανδρα μοι εννεπε, Μοΰσα, πολύτροπον, δς μάλα πολλά Tell me, Muse, of the clever man who many things . . .

Andronicus shows at once his capacity for close but clever translation. Thought and word order here are, within the shorter compass of the Saturnian, almost identical to the original; the translator is adept at matching Greek syntax with effective Latin equivalents.11 Small changes, however, also recast the original thought in distinctly Roman terms. One of the Camenae, nymphs associated with a spring outside the Porta Capena, substitutes for Homer's Muse. Andronicus thus conjures up a suitably native divine collective with its own sacred spring and a name that may sound a bit like carmen.12 With insece he offers a rare Latin word of similar meaning, sound, and accent to Homer's own uncommon εννεπε; versutum, built on verto, renders πολύτροπον with a Latin version of the same metaphor. And

9 Allen, Accent and Rhythm, 114-18. 10 Even the ostensibly symmetrical elegiac pentameter distinguishes the

effects of caesura and verse end by allowing metrical substitutions in its first colon but not in the second. See Nagy, Comparative Studies, 99-101.

11 Thus fr. 17, "simul ac lacrimas de ore noegeo detersit," renders the ao~ rist participle of Od. 8.88, δάκρυ' όμορξάμενος κεφαλής άπο, and Ο ci. 10.395, άνδρες δ' αιψ' έγένοντο becomes "topper facti homines" (fr. 25, where editors wrongly print Festus' facit despite the clue provided by the Greek original).

12 Waszink, Opuscula Selecta, 89-98.


Andronicus, unlike the Scipios' elogists, does not allow the verse form to govern the sense. He controls both. Reversing the order of Homer's vocative and verb enables him to accommodate the Saturnian caesura while setting the naturally enclitic vocative in its customary place. He also links his cola with the alliteration of virum / versutum to prevent the isolated rigidity of colon-structure all too typical of the elogia.

Other fragments show how he combines alliteration with word order to create a calculated relationship between the two cola of a single line, whether that relationship be parallel,

argenteo polubro aureo eglutro with a silver basin, a golden pitcher (fr. 6)

or chiastic,

ibidemque vir summus adprimus Patroclus and then the best and very first of men, Patroclus (fr. 10).

This technique for linking cola by sound and sense promptly appears in nonliterary Saturnians as well, as in the opening lines of another Scipionic elogium, this one probably written for a son of Scipio Hispallus (cos. 176) and thereby dated to ca. 180-70.

magna sapientia multasque virtutes aetate quom parva posidet hoc saxsum. quoiei vita defecit, non honos honore, is hie situs, quei nunquam victus est virtutei.

Great wisdom and many virtues few years had he whom this stone holds. His life was short, but not his honor, he lies here who never lacked in virtue. (ILLRP, no. 312)

Its word play may suggest the verbal dexterity (if not the humor) of Plautus, but the versification echoes Andronicus.

Another device for bridging the caesura developed from the avoid-ance of Greek patronymics, which never found a comfortable Latin equivalent. Athena, for example, addresses Zeus in Odyssey 1 as "our father, the son of Kronos" (πάτερ υμέτερε Κρονίδη, 45, 81), which Andronicus rendered with characteristically flexible fidelity as "pater noster, Saturni filie . . . " (fr. 2). This semantic bridge became a Saturnian mannerism, as in,

sancta puer Saturni filia regina blessed child, Saturn's daughter, queen (fr. 12)

apud nympham Atlantis filiam Calypsonem with the nymph, Atlas' daughter, Calypso (fr. 13)


nam diva Monetas filia docuit for the goddess, Mnemosyne's daughter, taught (fr. 21)

or, varying the pattern, Mercurius cumque eo filius Latonas Mercury and with him the son of Leto (fr. 19)

None of these iines has a single, clearly identifiable Greek proto-type. Suggestions have of course been made, but πότνια "Ηρη (Od. 4.513), the phrase usually adduced as the model for fragment 12, does not carry conviction. Other attempted matches are still weaker and can lead critics astray. Fragment 13, for example, is an oft-cited example of Andronicus' use of patronymics for solemn effect because it is usually matched with the recurring Homeric phrase νύμφης έν μεγάροισι Καλυψους ("in the halls of the nymph Calypso," 4.557, 5.14, 17.143). His addition of Atlantis filiam seems gratuitous, and thus stylistically significant, until we question the match. It has no ancient authority, and there are other possibilities. An equally compelling model for the line could be Odyssey 1.51-52,

νήσος δενδρήεσσα, θεά δ' έν δώματα ναίει, "Ατλαντος θυγάτηρ όλοόφρονος . . .

a wooded island, and a goddess dwells in a house there, destruction-minded Atlas' daughter . . .

or Odysseus' repetition of the thought at 7.245-46,

ενθα μεν "Ατλαντος θυγάτηρ, δολόεσσα Καλυψώ, ναίει έϋπλόκαμος . . .

there Atlas' daughter, artful Calypso with beautiful tresses dwells . . .

This could be an example of conflation, what Italian scholars like to call "contaminazione a distanza," but Andronicus may just as easily have compressed as expanded his original.13 The Saturnian line has in any case transformed the original beyond recognition.

The patronymic was just one kind of Greek expression that lacked a ready Latin equivalent. To render characteristic Homeric metaphors in succinct but comprehensible Latin, Andronicus moved still further from his model. Zeus, for example, answers Athena's appeal for help in Odyssey 1 with a common Homeric expression.

13 Cf. Fraenkel, RE Supp. 5 (1931) 604; Mariotti, Livio Andronico, 33 and 35 n. 51. Such contaminazione is sometimes taken as an Alexandrian trait in Andronicus, but for no good reason. It is also found in Plautus.


τέκνον έμόν, ποιόν σε επος φύγεν ερκος οδόντων; my child, what word escapes the barrier of your teeth? (64)

Andronicus is again as faithful as possible in matters of vocabulary and word order.

mea puera, quid verbi ex tuo ore supra fugit? my child, what word escapes up out beyond your mouth? (fr. 3)

Yet while fugio easily assumes the metaphoric sense of its Greek cognate, Latin cannot so readily describe a "barrier of teeth." Supra perhaps echoes the idea of an obstacle surmounted, but Andronicus must translate the meaning rather than the words. Os offers him a far easier metaphor for "lips." So too with a typical Homeric expression of fear:

κσΛ τότ' Όδυσσηος λύτο γούνατα και φίλον ήτορ and then Odysseus' knees loosened and his own heart

Andronicus again preserves what he can.

igitur demum Ulixi cor frixit prae pavore only then did Odysseus' heart freeze for fear (fr. 30)

He, too, begins with narrative particles and keeps the name in the genitive at the caesura. He maintains the extended notion of "heart" and, perhaps most important of all, he avoids prosaic circumlocutions or explanations. Andronicus renders a metaphor with a metaphor. The metaphor itself, however, must change to something Latin can more easily bear. Frigesco may be bold—it has only the metaphoric sense of waning enthusiasm in Roman comedy—and the effect is heightened by the alliteration of prae pavore.14 The resulting figure is comfortably Latin, and it bore significant fruit. The sense of os as "lips" was taken up in Ennius' phrase suaviloquens os (304) and entered the language. Vergil would better handle Homer's notion of limb-loosening fear, but only with Andronicus' help: "extemplo Aeneae solvuntur frigore membra" ("Aeneas' limbs at once weakened with cold," Aen. 1.92).

One final example of the altered metaphor is preserved, and it tells us something else about Andronicus' technique. The original is in Odyssey 8. Alkinoos has commanded the young men Halios and Lao-damas to entertain their guest with a solo performance.

όρχείσθην δή επειτα ποτι χθονι πουλυβοτείρη ταρφέ' άμειβομένω* κούροι δ' έπελήκεον άλλοι έσταότες κατ' άγώνα, πολύς δ' ΰπο κόμπος όρώρει.

14 Fantham, Comparative Studies in Republican Latin Imagery, 13-14.


And then the pair danced upon the much-nourishing earth, exchanging places often. The other young men standing by beat time, and a great sound arose. (378-80)

The sole surviving Latin line from this passage indicates a striking change.

nexebant multa inter se flexu nodum dubio They wove much between themselves with tangled bend of knots (fr. 20)

All that remains of the Greek original is the adverbial accusative multa and perhaps a hint of the dual in inter se. The complex dance itself has become a matter not of exchanges made back and forth, but of a weaving in and out. Perhaps Andronicus had difficulty with άμειβομένω referring to dance steps—it more commonly introduces an exchange of speeches—or feared that his readers might have such difficulty. Certainly he rendered not Homer, but an annotator's gloss on the phrase that is preserved among the minor scholia: πυκνώς πλέκοντες εις αλλήλους εναλλασσόμενοι.15 Andronicus, teacher as well as poet, evidently read Homer with a schoolmaster's notes and employed a metaphor taken not from the text but from its margin.

Another kind of accommodation to a Roman readership involves the matter of epic decorum. The easy familiarity with which gods and men in the Odyssey communicate across the barrier of mortality undergoes a curious change in this Latin version. At Odyssey 3.102ff., for example, Nestor recounts for Telemachus the high cost of the Trojan campaign.

. . . ενθα δ' επειτα κατέκταθεν δσσοι άριστοι, ενθα μεν Αϊας κείται Άρήϊος, ενθα δ' Άχιλλεύς, ενθα δέ Πάτροκλος, θεόφιν μήστωρ ατάλαντος, ενθα δ' έμός φίλος υιός, άμα κρατερός και άμύμων, Άντίλοχος . . .

. . . then perished there so many of the finest men. There Ajax lies, the man of Ares, and there Achilles, and there Patroclus, equal to the gods as counselor, and there my own dear son, both strong and flawless, Antilochus . . . (108-112)

Andronicus' tenth fragment seems to come from this passage.

ibidemque vir summus adprimus Patroclus and then the best and very first of men, Patroclus

15 Σ V ad 379, recognized by Fränkel, Hermes 67 (1932) 306-7. Fränkel's similar claim for fr. 23, based on an M scholion, is less compelling: it lacks a close verbal similarity, and the identification of Andronicus' original for this line is in any case uncertain.


îbidemcjue strongly suggests that Andronicus has repeated Nestor's sequence, but the poet evidently shied away from calling Patroclus "equal to the gods as counselor." That half-line epithet is applied in the Iliad to Priam (7.366), Peirithous (14.318), and Patroclus (17.477), and in the Odyssey to Neleus (3.409) as well as Patroclus, and in different ways it suits them all. It is not without meaning or point, but Andronicus discards it. He chooses instead an epithet still em-phatically distinguished (longe primum, says Gellius, our source for the line), but also decidedly human in its measure. Karl Büchner aptly compares it to the elogium for Atilius Caiatinus (cos. 258), which calls him "populi primarium virum."16 Andronicus appears reluctant to equate his mortal hero with the gods. A similar scruple seems to operate in fragment 7, which also has a fairly certain model.

tuque mihi narrato omnia disertim and you tell me everything clearly

αλλ' άγε μοι τόδε είπε και άτρεκέως κατάλεξον but come, tell me this, explaining it clearly straight through (Od. 1.169)

Scholarly attention here has centered on Andronicus' compression of the two verbs of telling, but a more significant question involves his one omission. What has happened to Homer's αλλ' αγε? Latin has the same expression. Characters in Plautus frequently say things like "come now, tell me your name" (e.g., "agedum nomen tuom primum memora mihi," Tri. 883). Why has Andronicus left it out? An answer lies in the context and in his new sense of epic decorum. Telemachus here is being gracious to the disguised Athena. The polite but collo-quial moment in Homer is evidently too colloquial for the new epic audience. The Latin Telemachus is much more formal in his address to a god. A colloquialism disappears, and the more solemn future imperative enters.

Could the colloquial exhortation perhaps have appeared in the lost preceding line? Certainty is impossible when lines survive without context, but another fragment tells against the possibility. At Odyssey 8.138-39, Laodamas remarks on the destructive power of the sea.

ού γαρ εγώ γέ τί φημι κακώτερον άλλο θαλάσσης άνδρα γε συγχεΰσαι, ει και μάλα καρτερός ειη

I say for sure that nothing's more destructive than the sea: it overwhelms a man, however powerful he may be

16 Büchner, SO 54 (1978) 45. The elogium is preserved ap. Cic. Sen. 61, where the cognomen is rendered Caiatinus. See MRR 206-7.


Andronicus renders these lines as follows (fr. 18):

namque nullum peius macerat humanum quamde mare saevom: vires cui sunt magnae, topper confringent inportunae undae

for nothing wounds a mortal worse than a savage sea: he whose strength is great, the remorseless billows shatter at once.

This comparative structure parallels Homer fairly closely, and Andronicus has even preserved an appropriate nuance of vocabulary: macerat, like συγχεΰσαι, takes its metaphoric sense from the effect of liquids. Yet tone and emphasis have shifted. The colloquial earnestness woven into Laodamas' ού γαρ έγώ γε τί φημι is, like Telemachus' casual grace, entirely gone. Once again, Andronicus' notion of epic decorum is not Homer's.

Neither is his organization. This is our longest fragment of the Odussia, and it offers a rare opportunity to see how a complex thought is expressed within the comparatively rigid Saturnian structure. Andronicus expands the thought in peius with saevom and inportunae, and he doubles the verb (macerat, confringent). Such expansion is not simply, as some have thought, a product of Roman antipathy to seafaring or a reflection of Hellenistic pathos.17 There is a technical reason. Homer's concessive ει και μάλα καρτερός εϊη, coming almost as an afterthought, poses a challenge to the ordered sequence of Saturnian cola. Andronicus meets the challenge by re-ordering the thought. He expands on his model to give himself space, beginning with a more general statement than Homer's (humanum replacing άνδρα) and then narrows the thought with a specific example (cui). Word order heightens the effect. Postponement of the relative, a familiar mannerism in hexameter poetry, juxtaposes vires and mare saevom in their unequal rivalry. The movement from gen-eral to specific restates and, in the process, emphasizes the primary thought. It becomes the functional equivalent of Homer's concessive clause. A complex idea is thus expressed by short cola in asyndeton.

The asyndeton itself is of interest because the strong stop occurs at a caesura, not a line end. Andronicus strives to link thoughts between lines. Actual enjambment is rare, but lines frequently open with conjunctions. The connectives namque and quamde here are typical. Enclitic -que, which is only rarely attached to the first word of a hexameter verse, occurs six times in the thirty apparent line openings of the Odussia. Other lines begin with neque and atque. Ten more begin with sequential adverbs like nam, igitur, and topper. A small

17 Büchner, SO 54 (1978) 47-48; Mariotti, Livio Andronico, 35.


group of single lines that contain clause breaks at the caesura prob-ably also reflect this tendency to prefer strong stops there rather than at line end. These include:

in Pylum deveniens, aut ibi<dem> ommentans

coming to Pylos or remaining there (fr. 9)

quando dies adveniet, quern profata Morta est

when the day comes which Morta has foretold (fr. 23)

topper facti homines ut prius fierunt the men at once made as formerly they were (fr. 25)

There is nothing quite like this among subliterary Saturnians. Epitaphs often use relative pronouns to connect thoughts over a line break, but we do not find such widespread use of conjunctions. A closer parallel comes from historical texts, as in Cato's sequence "eum sustulere, isque convaluit saepeque post illam operam rei publicae fortem atque strenuam perhibuit" ("They took him up, and he re-covered, and often after did strong and rugged service for the state," Orig., fr. 79P). The demands of sequential narrative seem to cross the boundary between poetry and prose.

Saturnian composition, however, does not necessitate parataxis. Fragment 15 preserves a different type of complex sequence.

ibi manens sedeto donicum videbis me carpento vehentem domum venisse

Sit waiting there until you see I have come home, carried in my cart

The context seems to be Nausikaa's instructions to Odysseus toward the end of Odyssey 6, and editors usually cite lines 295-96 as the model.

ενθα καθεζόμενος μεΐναι χρόνον, εις ο κεν ημείς άστυδε ελθωμεν και ικώμεθα δώματα πατρός.

Sit waiting there a time until we have come to town and entered my father's house

This may then lead to puzzlement over the logic, or perhaps the topography, of the situation. How can Odysseus remain where he is and yet see Nausikaa return home? It helps to read on a little further.

αύταρ έπήν ήμέας ελπη ποτ! δώματ* άφΐχθαι, κα! τότε Φαιήκων ϊμεν ές πόλιν ήδ' έρέεσθαι δώματα πατρός έμοΰ μεγαλήτορος Άλκινόοιο.


but when you think that we have reached the house, then enter the Phaeacians' city and ask for the house of my father, greathearted Alkinoos.

Nausikaa, in rather leisurely fashion, says two things to Odysseus: Wait here until I get home. When you think sufficient time has passed, follow after me. Andronicus compresses these thoughts. He translates the first part of 295 closely, but videbis has the figurative sense "think" and actually renders ελπη of line 297.18 The corre-spondences are really quite close:

ibi manens sedeto - ενθα καθεζόμενος μειναι χρόνον donicum videbis me ...domum venisse ~ έπήν ήμέας ελπη. . . δώματ'

άφΐχθαι

Where Homer builds his thought gradually, however, Andronicus' verses are much more condensed. Each colon contains a discrete verbal idea (sitting, seeing, riding, arriving). Structural devices join them to develop a single complex sequence: ibi. . . donicum unites the first line around the caesura and me . . . venisse bridges the second. The central idea (donicum videbis) is surrounded by the injunction and the in-finitive clause.

The cart, Andronicus' apparent addition to the passage, is itself of interest. It may contain a hint for the literal of mind that Nausikaa is not alone (Homer's verbs are true plurals) or that the trip home will take some time. It certainly reflects another aspect of Andron-icus' epic decorum. Nausikaa's entourage in Homer carried a good deal of laundry piled into a four-wheeled wagon (απήνη, 69, 252). We might expect Latin plaustrum, the familiar wagon of Roman every-day life. Instead we get carpentum, the two-wheeled cart of the rich. It is the more suitable conveyance for a king's daughter, if not necessarily for a load of wash. In the choice of detail here, as in the manner of dialogue, Andronicus seeks a uniformly elevated tone, altering or omitting those graphic and homely touches that anchor Homeric epic in the familiar realities of daily life. He creates a kind of epic dignity that looks not back to the Odyssey but forward to the Aeneid.

The fact of translation has offered this extraordinary glimpse into the workings of Andronicus' craft. Fragments responding to the chal-lenges of a known model reveal the tricks of word order and vocabulary used to reproduce their content. Andronicus translates closely when it suits him, develops new mannerisms to present com-

18 Büchner, SO 54 (1978) 50-51; Traîna, Vortit barbare, 24-26. The ex-tended sense of video is fairly common in Plautus (e.g., Most. 1051: "ego me video venire in meo foro") and may even be present in a sententia of Ap. Claudius, "amicum cum vides, obliviscere miserias" (2).


plex thoughts in Saturnian cola, and creates the sense of decorum in diction and behavior that he thought appropriate to epic situations. An examination of Naevius' poetic technique must necessarily take a different tack, though Greek influence of a kind has also been sus-pected in his work.

The Bellum Punicum apparently began with an invocation containing the line "novem Iovis concordes filiae sorores" ("nine like-minded daughters of Jupiter, sisters," fr. 1). Scevola Mariotti thought he heard an echo here of Hesiod's Theogony; εννέα θυγατέρες μεγάλου Διος έκγεγαυιαι ("nine daughters born of great Zeus," 76). For Naevius7 concordes, Mariotti looked to Theogony 60, which describes the children of Mnemosyne as "nine like-minded maidens" (έννέα κούρας όμόφρονας.)19 This suggests the kind of careful, clever translation characteristic of the Odussia, where word position and sense can, with the help of caiques, follow the original closely, while alliteration and homoioteleuton shape the units of this quite differ-ent verse form. The sequence θυγατέρες . . . έκγεγαυιαι may even help explain the near redundancy of filiae sorores. The identification, however, is problematic. Naevius' Greek antecedent is not so readily pinpointed, for the epic mannerisms of Homer and Hesiod had long since joined the common stock of Greek poetic language. Echoes and imitations abound. The learned Antimachus, for example, began his hexameter Thebais with his own invocation to the Muses that contained the line έννέπετε, Κρονίδαο Διος μεγάλοιο θύγατρες ("Sing, daughters of great Zeus, the son of Kronos," fr. 1 Wyss). The Hesiodic echo is too strong and too sly in Antimachus to permit the confident presumption of any one Naevian model over another, nor has any other fragment of the Bellum Punicum brought a Greek echo to mind. More illuminating signs of literary pedigree lie in a different direction.

The shape of Naevius' first fragment recalls lines like Andronicus' fragment 12 ("sancta puer Saturni filia regina") and their tendency to bridge the caesura with a patronymic phrase. A simple description, "ferunt pulcras creterras, aureas lepistas" ("they bear beautiful bowls, golden goblets," 54) echoes the content, structure, and stylistic ornament of Andronicus' "argenteo polubro aureo eglutro" (6). Naevius may also arrange his thoughts chiastically around the caesura, as in fragment 8: "res divas edicit, praedicit castus" ("he sacred rites decrees, declares a state of chastity," cf. Od., fr. 10). These are, if not signs of explicit modeling, at least indications of a common approach to the challenge of verse composition that marks the emergence of a distinct epic style in Saturnians. Because many fragments of Naevius'

19 Mariotti, Livio Andronico, 53-56.


poem are slightly longer than the remains of Andronicus' Odussia, we can see something of the effects that such features could create.

Epithets so easily fill the short cola that in Naevius descriptive phrases may well extend over several lines, as in fragment 10:

senex fretus pietatei deum adlocutus summi deum regis fratrem Neptunum regnatorem marum

The old man buoyed by piety called upon the god, brother of the gods' great king, Neptune, ruler of the seas

This layering of epithets provides a way to avoid the predictable diminuendo so common in sub literary Saturnians. The thought here, however, may seem in danger of rambling to no purpose: we might have expected greater emphasis on the old man and the prayer to come than on this expanded object. Yet the technique has merit.

Other fragments show how Naevius used this layered style to build power and a certain tension into his lines.

dein pollens sagittis inclutus arquitenens sanctus love prognatus Pythius Apollo

Then mighty with arrows, the famous bowholder, blessed son of Jupiter, Pythian Apollo (fr. 20)

Readers might well suspect at once that the god to be associated with bow and arrow will be Apollo, but that expectation only heightens the effect as Naevius sets one epithet in each colon and brings the sequence to a climax with the proper name. The famous Giant-fragment (4) represents a more elaborate form of the same technique:

inerant signa expressa, quomodo Titani, bicorpores Gigantes, magnique Atlantes Runcus ac Purpureus, filii Terras

Figures were represented on it, how the Titans, double-bodied Giants and great Atlases, Runcus and Purpureus, sons of Earth

The sequence of three generic titles, Titani . . . Gigantes . . . Atlantes, is each set in a two-word phrase and again culminates with proper names. Naevius then rounds off the expression, which must have been the expanded subject of the quomodo-clause, with the dignified archaism of filii Terras.20 The crescendo of the lines can thus suggest

2 0 Fraenkel, J RS 44 (1954) 14-17, offers a masterful discussion of this syntax.


movement even though the thought is standing still. The technique involved is not so far removed from the true progression of Vergilian descriptions, as in the scene confronting Aeneas and the Sibyl on the threshold of Dis.

vestibulum ante ipsum primisque in faucibus Orci Luctus et ultrices posuere cubilia Curae, pallentesque habitant Morbi tristisque Senectus, et Metus et malesuada Fames ac turpis Egestas, terribiles visu formae, Letumque Labosque

Before the entrance, at the very jaws of Orcus, Grief and the avenging Troubles chose their rooms, and pale Diseases live there and grim Old Age, and Fear and seductive Greed and ugly Want, terrible shapes to see, and Death and Toil (Aen. 6.273-77)

Vergil of course has certain advantages. The greater compass of the hexameter enables him to alter the length of individual phrases by choosing which nouns to modify, and it is easier for him to group them in a deliberately varied sequence. Yet he, like Naevius, creates his effects by deciding where in the line to position each proper noun, and he too rounds off his set of figures with the archaic sound of Labos in the final foot.

The static quality of such elaborately ornamented fragments of the Bellum Punicum is to some extent an accident of their preservation. Each gives us only the subject of a subordinate clause. The vagaries of citation and survival keep us from seeing how Naevius developed over a span of verses the complex sentences of which they formed a part. Layered expressions of the kind we have seen are in any case a matter of choice. The syntax of fragment 23, for example, is quite similar to the structure of the Giant-fragment.

blande et docte percontat, Aenea quo pacto Troiam urbem liquerit

kindly and carefully asked how Aeneas left the city of Troy

It too opens a thought in the first colon and puts the central idea in a subordinate clause (quo pacto ~ quo modo). Only the elaboration of the subordinate subject is lacking.

Other stylistic devices were available to heighten the effect of actions. Fragment 5 describes how the wives of Anchises and Aeneas fled the destruction of Troy.

amborum uxores noctu Troiad exibant capitibus opertis fientes ambae, abeuntes, lacrimis cum multis


Their two wives fled Troy by night, heads covered, weeping both, leaving with many tears

Each colon, shaped as so often by alliteration and homoioteleuton, contains a discrete idea, while the nearly parallel ablative phrases link successive verses. A combination of participles and finite verbs creates the action. The picture reveals itself piece by piece to dramatic and poignant effect. It is perhaps one place where the greater sweep of an Ennian hexameter—"fientes plorantes lacru-mantes obtestantes" ("weeping, wailing, crying, supplicating," An. 498) contains a similar idea—may not represent an improvement.

Not all of Naevius7 effects, however, depend on the juxtaposition of such small pieces. In fragment 25, entire verses, not their con-stituent cola, are the primary units of meaning.

postquam avem aspexit in templo Anchisa, sacra in mensa penatium ordine ponuntur; immolabat auream victimam pulcram.

After Anchises saw the bird in flight, offerings to the household gods were duly set out. He began sacrificing a fair, golden victim.

Here there are no significant breaks in phrasing at the caesurae, but strong ones at line ends. The integrity of each verse is further emphasized by separate alliterative patterns in the first and third lines and by the grammatical shift in the second, where the plural verb stands apart from its surroundings. Each verse is devoted to a discrete action. The effect may remind us of the Catullan hendecasyllable, another meter that seems to fall into cola and lines rather than feet:

soles occidere et redire possunt: nobis cum semel occidit brevis lux, nox est perpetua una dormienda.

Suns can set and rise again: for us, when once our brief light sets, the night is one continuous sleep. (5.4-6)

Given an analogous opportunity, Catullus creates a similar cor-respondence of sense and line.

The Anchises fragment seems to come from a narrative rather than essentially descriptive passage, and we might well expect a certain difference of style in such a context. The degree of difference, however, has itself become a significant crux of Naevian criticism. Friedrich Leo saw in the dignified directness of such narrative passages what he called "an elevated Chronicle style" which he


distinguished from the more artful effects manifest in the women's flight from Troy (fr. 5) and in Naevius' numerous references to a divine apparatus. The stylistic distinction he perceived has since been treated as a defect or, if condescension is equated with kindness, as an inevitable consequence of archaic crudity. In the words of one noted Vergilian: "We cannot expect at that era any special sensi-tivity to style, any strong sense of the incongruity between annalistic history and Homeric narrative."21 Yet we have already found a greater sensitivity to style in Saturnian epic than is often credited to Andronicus and Naevius. We have also seen that the structure of the Bellum Punicum integrated a legendary (i.e., Homeric) narrative into the historic account of the First Punic War. How great a stylistic difference would Naevius tolerate between the two elements of his complex story? Was his narrative insensitive to the artful manip-ulation of language? The very idea of a Chronikstil and the stylistic incongruity it introduces merit further consideration.

A Latin "official style" was indeed in place by the later third century, discernible in both public pronouncements and the parodies they engendered. It often featured bold declarations of conquests with the details strung together in ablative phrases, accounts of spoils taken, and the name and titles of the conqueror. Livy preserves a long dedicatory inscription typical of the form. It begins,

Ti. Semproni Gracchi consulis imperio auspicioque legio exercitusque populi Romani Sardiniam subegit. in ea provincia hostium caesa aut capta supra octoginta milia . . .

Under the command and auspices of Tiberius Sempronius Gracchus, the legion and army of the Roman people subdued Sardinia. In that province over 80,000 enemy were either killed or captured . . . (Liv. 41.28.8)

This particular example dates only from 174, but the style was suf-ficiently well established to be parodied over a generation earlier by Plautus (Amph. 188-96). Caesar would eventually bring such a style to literary elegance, but its older public form could be no more plain than it was modest.

In 179, for example, the censor M, Aemilius Lepidus dedicated to the Lares Permarini a temple that he claimed had been vowed by L. Aemilius Regillus at the battle of Myonnessus eleven years before.

21 Otis, Virgil, 21. Cf. Schanz-Hosius, Geschichte der römischen Literatur, 53: "Poetische Schönheit zeichnet keines der wenigen Fragmente aus; es klingt wie eine versifizierte Chronik, wo nur die Taten reden/' The distinction of styles formulated by Leo, Geschichte der römischen Literatur 1, 80-81, and developed by Fraenkel, RE Suppl. 6 (1935) 639 is actually more subtle and sympathetic to Naevian art than this.


Livy records the inscription set over its doors and copied for the temple of Capitoline Jupiter.22

duello magno dirimendo, regibus subigendis, caput patrandae pacis haec pugna exeunti L. Aemilio M. Aemilii filio [ . . . ] auspicio imperio felicitate ductuque eius inter Ephesum Samum Chiumque inspectante eos ipso Antiocho, exercitu omni, equitatu elephantisque, classis regis Antiochi antehac invicta fusa contusa fugataque est, ibique eo die naves longae cum omnibus sociis captae quadraginta duae. ea pugna pugnata rex Antiochus regnumque [ . . . ] eius rei ergo aedem Laribus permarinis vovit.

For ending a great war, for subduing kings, as a way to establish peace, this battle [was granted to] L. Aemilius, the son of M. Aemilius, when he took the field. Under his auspices, command, good fortune, and leadership near Ephesus, Samos, and Chios, while Antiochus himself watched them, the entire army, cavalry, and elephants, the fleet of King Antiochus though previously undefeated was routed, shattered, and put to flight, and on that same day forty-two warships with all their crews were captured. With this fight fought, King Antiochus and his kingdom [were conquered]. For this victory he vowed a temple to the Lares of the Sea.

This formidable, "annalistic" record of achievement does not lack stylistic embellishment. A phrase like "auspicio imperio felicitate ductuque eius" was already formulaic (cf. Amph. 196), but Lepidus' inscription also contains stylistic features of its own. It is deliberately and conspicuously alliterative. The archaic form duello is doubtless chosen for its effect, as is the so-called figura etymo-logica, itself a favorite device of early Latin, in pugna pugnata. The grouping "Ephesum Samum Chiumque" may merely serve to place the battle geographically, but "fusa contusa fugataque est" is certainly an artfully powerful tricolon that plays on the repetition of syllables and the increasing length of its components.

This is writing for maximum effect. In fact, the "chronicle style" here has blended so easily into literary composition that the grammarian Caesius Bassus, plucking "duello magno dirimendo, regibus subigendis" from its clearly unmetrical context, claimed it was a Saturnian (GLK 6.265). He thought he saw another in a triumphal inscription of Acilius Glabrio, "fundit fugat prosternit maximas legiones," where the tricolon is equally striking. Its context is un-known, but Mummius' dedication of 144 on the temple of Hercules Victor apparently was in Saturnians, though the stone does not

22 Liv. 40.52.5-7. I offer an eclectic version of a lacunose and problematic text whose difficulties, more of syntax than sense, are masked in the translation. The stylistic argument here is not compromised by the condition of the text, though more could doubtless be said along these lines if the text were sound.


punctuate for meter.23 This "official style" created such flamboyant effects by employing precisely those ornamental devices charac-teristic of literary compositions. The stylistic gap between official and literary narrative is not so very large. Neither was necessarily plain, and Naevius could thus more easily integrate historical narrative into his epic than Leo's formal distinction would suggest.

A trace of that integration remains in the alliteration across the caesura of even the "prosaic" fragment 3, referring to the Roman expedition to Sicily in 263.

Manius Valerius consul partem exerciti in expeditionem ducit

Manius Valerius the consul led part of his army on a foray

The unembarrassed pride of the Roman honorific inscriptions finds its best parallel, however, in one notoriously difficult fragment of the Bellum Punicum. The source is an entry in Nonius' lexicon (129L):

CONCINNARE, conficere vel colligere. Naevius Belli Poenici lib. IV: transit Melitam Romanus exercitus, insulam integram urit, populatur, vastat, rem hostium concinnat.

CONCINNARE, to complete or collect. Naevius, Bellum Punicum Book IV: The Roman army crossed to Malta, the untouched island it burned, devastated, destroyed, finished the enemy's affairs.

The lines as quoted are too long and ungainly to be satisfactory Saturnians. The most pleasing correction was proposed early in this century by Carl Thulin, who restored cola of the requisite proportions:

transit Melitam exercitus Romanus. insulam integram urit vastat populatur, rem hostium concinnat.

The fragment as restored shows not just the kind of tricolon and syllable play we often find in military inscriptions but also a thought sequence characteristic of Saturnian composition.24

We have already seen how a poet may present an incomplete idea in the first colon of a line and then either resolve or explain that idea

2 3 Degrassi, Imagines, no. 61. I doubt very much that the two lines quoted by Bassus were actually written as Saturnians, though they are accepted as such by Cole, YCS 21 (1969) 16.

2 4 Thulin, Italische sakrale Poesie und Prosa, 32-33 η. 7. Strzelecki prints unsuitable cola for this fragment, but his apparatus is a model of clarity.


in the second. The epitaph for Scipio Barbatus (no. 309) does this with the line "Taurasia Cisauna Samnio cepit," for Taurasia and Cisauna are actually cities in Samnium.25 Naevius resolves the modi-fiers "novem lovis concordes" with the nouns "filiae sorores" (fr. 1) and identifies the "sanctus love prognatus" as "Pythius Apollo" (fr. 20.2). This mannerism is also found in historical contexts such as fragment 2, which apparently refers to the declaration of war with Carthage: "scopas atque verbenas sagmina sumpserunt." The sagmina, tufts of grass used by the Fetial priests in declaring war or making treaties, stand apart here in the second colon to explain the "scopas atque verbenas."26 In fragment 33, the first colon ("simul atrocia pro-icerent") makes little sense without the object and subject nouns provided in the second ("exta ministratores"). The same is true of fragment 32. The last line describes the violence done to Malta with three verbs in the first colon and then summarizes the result of that violence with the declaration that the enemy has been discom-moded.2 7 The marked parataxis of the fragment may recall the flatness of official prose, but Naevius' distribution of ideas in these lines actually reflects a familiar epic mannerism. Strong stops occur only at the caesurae, a pattern that we have observed in Andronicus (18) and that can be found in both mythological (8, 23) and historical (32, 34) fragments of the Bellum Punicum. The poet creates a tension between thought sequence and verse structure by forcing the reader over the line ends through enjambment while slowing the pace at the caesurae.

Stylistically, fragment 32 is thus of a piece with Saturnian epic as a whole. Naevius may vary his art to fit his subject—good poets

25 Samnio is thus accusative; we need not understand "(in) Samnio" as printed by Van Sickle, "First Hellenistic Epigrams at Rome/' 147 and n. 30.

2 6 Barchiesi, Nevio epico, 388-89 thus rightly translates "pressera ranos-celli e verbene come sagmina." Leo, Mariotti, Marmorale, and Strzelecki (a formidable group!) all read verbenas as genitive singular with sagmina, but "tufts of grass of a shoot" makes little sense. Livy's references to the fetiale ritual at 1.24.3-6 and 30.43.9 treat sagmina and verbenae as synonyms; he does not modify one with the other. See Ogilvie, Λ Commentary on Livy, 110— 12 and 127-29. Schwarte, Historia 21 (1972) 206-23, associates the fragment with the signing of a treaty, not a declaration of war.

27 Concinnat has been thought problematic and tempted emendation, but Nonius' synonyms make the meaning sufficiently clear. The verb de-notes an action taken to frustrate another person, either by upsetting some-one's plans (conficio) or by appropriating something for oneself (colligere). Thus Hegio complains, "confecisti omnis res ac rationes meas" (PL Capt. 673). A character in Titinius says, "qui fugere polsi hinc spolia colligunt" (182R). Mariotti, "Concinnat in Naevius," 1-5, finding Naevius' synonyms in-compatible, nevertheless opts for an "ironic" sense of concinnat.


generally do—but he is never entirely without art. Taken as a group, the fragments of his poem show stylistic variety, not incongruity. The genuine oddity of fragment 32 lies in its subject. The Roman attack on Malta that it describes cannot have been a very distinguished achievement of Roman arms. Polybius ignores it completely. Only Augustine's student Orosius records that the consul Atilius sacked Lipara and Malta, and it remains unclear whether this Atilius was A. Atilius Caiatinus (cos. 258) or C Atilius Regulus (cos. 257). The legitimate military objective must in any case have been Lipara, which Caiatinus failed to secure in 258 (Polyb. 1.24.8-13), and Marino Barchiesi may be right to associate this foray to Malta with some-thing closer to piracy than legitimate warfare.28 Naevius perhaps set a noble face on a less than noble body of fact, a possibility with ramifications for the study of Ennius, though the Bellum Punicum certainly did not shrink from reporting Roman failures and excesses. The rhetorical flourish of "superbiter contemtim conterit legiones" ("haughtily and contemptuously he wore down the legions," 37), for example, probably had as its subject P. Claudius Pulcher (cos. 249), whose arrogance toward his own troops at Lilybaeum is rivaled only by his infamous contempt for the sacred chickens off Drepanum.

The key fact in all this is that Naevius' historical narrative wore a stylistic garb not so very different from his mythological passages. One final example may summarize the point. Fragment 34 is gen-erally associated with events of 260, the only time we know the praetor urbanus to have held a military command.

virum praetor adveneit, auspicat auspicium prosperum

The men's praetor arrives, he takes propitious auspices

We find some familiar features here: strong break at caesura rather than at verse end, ubiquitous alliteration, auspices graced by à figura etymologica. The crowning flourish is the odd phrase virum praetor. Mariotti took praetor back to its root in prae-itor and explained the phrase as a caique on the Homeric epithet "lord of men" (άναξ ανδρών). This clever suggestion has been dismissed as inappropriate to "a terse, factual statement about a Roman magistrate," but such an objection puts the matter exactly backward.29 Factual statements

2 8 Orosius 4.8.5: "Atilius consul Liparam Melitamque insulas pervagatus evertit." Thus Cichorius, Römische Studien, 39 and Barchiesi, Nevio epico, 337-38. Orosius' source was probably Livy.

29 Mariotti, Bellum Punicum, 72, doubted by Skutsch, CR 72 (1958) 47. The usual alternative is to emend virum in the manuscripts of Nonius to verum,


among the surviving fragments are rarely without ornament of some kind. Naevius' decision to write a poem about the war with Carthage made this struggle quite literally an epic conflict, and the possibility of a Homeric echo here should not be so quickly dismissed. We cannot press a distinction between "chronicle" and "poetic" styles. There is a stylistic unity to the Bellum Punicum that reflects Naevius' desire to extend the epic manner pioneered by Andronicus to include the narration of historical events. The Saturnian aesthetic expands its scope in this poem, though the impetus for that expansion would soon take Roman epic in a different direction.

Later generations knew what purports to be Naevius' own epitaph, a curious little poem that Gellius, who preserves it, thought "full of Campanian arrogance."30

immortales mortales si foret fas flere, flerent divae Camenae Naevium poetam. itaque postquam est Orchi traditus thesauro obliti sunt Romae loquier lingua Latina.

If immortals could mourn for mortals, the divine Camenae would mourn the poet Naevius. After he was passed to Orcus' treasury, they forgot at Rome how to speak Latin.

The lines employ the kind of word play, alliteration, and phrasing that so often give Saturnian composition its splendor. It is a fine and effective epitaph, but it proved deficient as prophecy. Though the Camenae doubtless mourned Naevius, Roman poets did not stop speaking Latin. They did, however, start speaking it differently, and the epic style necessarily evolved in the process.

vixdum, or visum. For the historical context, see Cichorius, Römische Studien, 32-33.

30 Gell. 1.24.2: "epigramma Naevi plenum superbiae Campanae." For its authenticity, see Suerbaum, Untersuchungen 31-42 (possibly genuine), Courtney, Fragmentary Poets, 49-50 (not genuine), and more generally Gruen, Studies, 92-93.

4 Hexameter Aesthetics

When Aeneas and his men land at Cumae, the Sibyl instructs them to provide proper burial for their shipmate Misenus. They promptly set off to gather timber for a funeral pyre.

itur in antiquam silvam, stabula alta ferarum. procumbunt piceae, sonat icta securibus ilex, fraxineaeque trabes cuneis et fissile robur scinditur, advoluunt ingentes montibus ornos.

They go into an ancient wood, tall home of beasts. Spruce crash, the holm-oak rings with ax-blows, ashen timbers and the oak tree split with wedges fall. They roll down giant rowans from the hills. (Aen. 6. 179-82)

This description owes a double debt: a remote one to Homer, who de-scribed the felling of oaks for the funeral pyre of Patroclus (II 23.114-20), and a nearer debt to Ennius, whose Roman soldiers brought down a more varied forest at Heraclea to fuel a pyre for casualties of the battle there.

incedunt arbusta per alta, securibus caedunt, percellunt magnas quercus, exciditur ilex, fraxinus frangitur atque abies consternitur alta, pinus proceras pervortunt: omne sonabat arbustum fremitu silvai frondosai.

They enter a tall grove. They strike with axes, they overturn great oaks. The holm is chopped, the ash is shattered and tall fir laid low. They uproot lofty pines: the whole grove echoed with the clamor of the leafy wood. (175-79)

Vergil recalls Ennius through the borrowing of significant details, the strategic placement of key words, and more generally through his greater interest in the trees than the axmen. He deftly modernizes the prosody (thus fraxineae replaces the archaic scansion fraxinüs), and he lightens the metrical effects, or at least brings his own pas-

83


sage to closure with a less extraordinary set of spondees. One great poet thus pays homage to another and in doing so declares both the ancestry and the progress of Latin epic.

These passages have attracted considerable attention, usually with a nod to Vergil's "improvements" on his predecessor: Ennius, to borrow J. K. Newman's phrase, displays "too much art," which Vergil had the good taste to polish and restrain.1 Yet the differences we perceive between Vergil and Ennius, significant as they are for tracing the development of their genre, are nevertheless based on other, still more significant features that they share. Their common faith in the seriousness of the epic enterprise and willingness to use Homeric material for their own ends are facilitated by their common meter, which was Ennius' innovation. His introduction of the hex-ameter—and the consequent eclipse of the Saturnian—set a new and enduring direction for both the intellectual and technical develop-ment of Latin epic. It was not, however, an innovation entirely with-out precedent. While the dactylic hexameter was indeed new to Latin epic, the adaptation of Greek quantitative patterns was not entirely new to Latin poetry. Iambic and trochaic verses had been the mainstay of Roman drama from its beginning, and by the early second century Plautus' mastery of quantitative rhythms had enabled him to compose long and complex lyrics. Ennius' own tragedies, which them-selves became Latin classics, were equally skillful. Dramatic prosody, however, was based on the norms of spoken Latin: its flexible treatment of final -e and -s, its frequent elisions and con-tractions, and its tendency to shorten long but unemphatic syllables reproduced a casual style of pronunciation that Roman dramatists, unlike their Greek predecessors, found acceptable for both tragic and comic diction, whether spoken or sung. Plautus could write, "quid hoc hie clamoris audio . . . ?" (Trin. 1093), where iambic shortening in the first metron gets the verse moving, and so could Ennius in his tragedy The Ransom of Hector (Hectoris Lytra): "quid hoc hie clamoris, quid tumulti est?" (163J). A similar informality carried over into early hexameters: the chatty Hedyphagetica, Ennius' translation of a culi-nary guide by the fourth-century poet Archestratus, scanned "glau-cumque aput Cumas" at verse end. Lines employing such scansions were preeminently speakable, which was why dramatists found this prosody attractive, but these echoes of colloquial Latin were unsuitable for a work like the Annals.

1 Newman, Augustus and the New Poetry, 93-95. The teleological bias is difficult for critics reared on Vergil to escape. For detailed technical com-parison of the two passages, see Austin, Aeneid VI, 93-95; for their sig-nificance in the history of epic, Williams, Tradition and Originality, 263-67, and Aicher, C/85 (1989/90) 218-22.

Hexameter Aesthetics 85

Heroic epic required a more formal style. Andronicus and Naevius had therefore abandoned Greek quantitative models and sought the requisite formality in the Saturnian. They were, as we have seen, successful in that endeavor, but at the cost of a certain rigidity in ex-pression. Replacing their Saturnians with the more flexible dactylic hexameter required a different quantitative prosody that could move the language some distance from the rhythms and connotations of ordinary speech. For epic, Ennius therefore developed stricter scansions that greatly restricted the use of elision. He came to put only short-voweled syllables in short positions, and he avoided iambic shortening. He did not entirely abandon the norms of speech: a line scanning "clamor ad caelum volvendus per aethera vagit" (545), a rhythm Vergil would find unacceptable, accurately reflects the vowel quantities and pronunciation of the second century. When he did offer highly artificial sounds, as in that extraordinary ending sîlvâï fröndösäl (179), the context suggests conscious and calculated experimentation with metrical effects. Not only do the archaic genitives of the line slow the passage to closure, but the metrical units of the last two lines are arranged chiastically: 4 sp + da + sp / sp + da + 4 sp. By manipulating strict rules of scansion and natural tendencies in pronunciation, Ennius created a new formal style that avoided grotesque artificiality while still expanding epic's range.2

The resulting technical advantage for expressing ideas and telling stories did not lie simply in the hexameter's greater length. The Annals averages between fourteen and fifteen syllables per line, hardly more than the thirteen syllables of a normal Saturnian; some heavily spondaic lines in Ennius actually have as few as twelve. The real improvement lay in freedom from the Saturnian's rigid colon structure, which gave the hexameter poet greater flexibility in shaping sentences and greater variety in their rhythm. Dramatic verse had long brought a certain syntactic plasticity to the inter-weaving of clauses. Thus, while Naevius' epic Saturnians may seem rather stiff and formal, his iambic senarii could produce more supple Latin, like this:

vos, qui regalis corporis custodias agitatis, ite actutum in frondiferos locos, ingenio arbusta ubi nata sunt, non obsita.

2 The details of Ennian prosody may be found in Skutsch, Annals 46-61, and Gratwick, CHLL, 66-70. There is general but astute discussion of the Latin poets' task in adapting Greek metrical patterns in West, Creek Metre, 186-90, and Allen, Accent and Rhythm, 335-47. Andronicus and Naevius are sometimes thought to have been simply unequal to the task of writing quantitative hexameters, but the only certain fact is that they did not do so. See Mariotti, Bellum Punicum, 83-86,


You who stand as guards about the royal person, go at once into the leafy groves where trees have grown by nature, not design.

(Naevius, Lycurgus, 21-23R)

There could be nothing comparable in epic until the new hexameter form enabled Ennius to weave similar complexities into his verse.

Hannibal audaci cum pectore de me hortatur ne bellum faciam, quem credidit esse meum cor suasorem summum et studiosum robore belli

Hannibal with presumptuous spirit urged me not to fight, he whom my own heart knew to be its greatest advocate and devoted to the might of war. (371-73)

The gain for narrative, especially in a long poem, is obvious.3

The aesthetic gain was even greater, for Ennius could n o w assim-ilate a wider range of Homeric influences than his predecessors could manage. To translate a Homeric poem as Andronicus had done meant creating a new formal style for the old story. Ennius could be much more direct in turning the Homeric style itself into Latin poetry. This process might involve something as simple as borrowing a formula (e.g., πατήρ ανδρών τε θεών τε - patrem divomque hominumque) or as complex as adapting an extended simile. Twice in the Iliad, for example, a warrior enthusiastic for battle is l ikened to a spirited horse at liberty.

ώς δ' οτε τις στατος ϊππος άκοστήσας έπι φάτη δεσμδν άπορρήξας θείη πεδίοιο κροαίνων είωθώς λούεσθαι έυρρεΐος ποταμοΐο κυδιόων • υψου δέ κάρη εχει, άμφι δέ χαΐτοα ώμοις άίσσονται· ό δ' άγλαίηφι πεποιθώς, ρίμφα έ γοΰνα φέρει μετά τ' ήθεα και νομον ίππων

as when a stalled horse, well-fed at the manger, has broken his bond and gallops down the plain to bathe in his accustomed place at a fine-flowing river, proudly: he holds his head high, his mane floats over his shoulders. Trusting in his strength, his knees soon take him to familiar ground, the realm of horses.

(IL 6.506-11 = 15.263-68)

Ennius' version of this simile is preserved by Macrobius.

3 The greater syntactic sophistication of the Annals over its Latin pre-decessors in epic is thus the product of metrical rather than linguistic developments. A comparable phenomenon gave drama the stylistic edge over contemporary prose. See Goldberg, Understanding Terence, 192-95.


et tum, sicut equos qui de praesepibus fartus vincla suis magnis animis abrumpit et inde fert sese campi per caerula laetaque prata celso pectore; saepe iubam quassat simul altam, spiritus ex anima calida spumas agit albas

and then, as when a horse, glutted at the manger, breaks his bonds with all his strength and takes himself out to the green pasture and lush meadows with proud heart: he often tosses his mane high, his spirit, hot and eager, works up a foamy lather. (535-39)

The debt to Homer is certainly striking.4 Ennius preserves not only details of the pampered horse running proudly on the loose but also its exuberant energy through the bumps and dips of his rhythm and enjambments which press the thought on from line to line. Yet there are also hints of an interesting change. The point of Homer's simile lies not just in the horse's energy but in its desire for an accustomed river and favorite meadow. Its natural element is not the manger, where it must be tethered, but the open space, where it runs free. So the hero described by this simile, whether Paris or Hector, is revealed to be rushing back to his natural element, the battlefield. Ennius' horse shows a similar energy and pride, but the details of the Latin simile develop differently. The Roman meadow may be lush and pleasant, but it is not the specifically favored or accustomed place. It is no more welcome for the freedom it affords than the manger was desired for the food it provided. Pride, not nature, has broken this horse's bonds. The simile is preserved without context, but Ennius has probably altered its connotation to fit a new situation. He certainly did so in another simile, also drawn from a battle scene.

A fragment from Book 15 records the courage of a Roman tribune at the siege of Ambracia in 189.

undique conveniunt velut imber tela tribuno: configunt parmam, tinnit hastilibus umbo, aerato sonitu galeae, sed nec pote quisquam undique nitendo corpus discerpere ferro, semper abundantes hastas frangitque quatitque. totum sudor habet corpus, multumque laborat, nec respirandi fit copia: praepete ferro Histri tela manu iacientes sollicitabant.

From every side the missiles pour like rain upon the tribune: they strike his shield, its boss resounds with javelins, his helmet makes a brazen sound, but no man on

4 Von Albrecht, Hermes 97 (1969) 333-45, examines the correspondences at length.


any side can tear his body by straining with the sword. Steadily he strikes and slashes at the teeming spears, sweat soaks his body and he labors hard, there is no time for rest: the Istrians press on and hurl their shafts with flying point. (391-98)

Macrobius preserves this passage because it also has a Homeric model, that moment in the Iliad when Ajax retreats from the Achaean ships. Homer's scene was striking for its vivid portrayal of the hero laboring under a shower of Trojan arrows (II. 16.102-11). In suiting this description to his own needs, Ennius does more than avoid Homer's characteristic verbal repetitions and some of his more graphic details; he also alters Homer's basic organization. The ori-ginal passage had focused squarely on the hero crouched behind his towering shield.

ο δ' άριστερον ώμον εκαμνεν εμπεδον αίέν εχων σάκος αίόλον, ούδ' έδύναντο άμφ' αύτφ πελεμίξαι έρείδοντες βελέεσσιν

and his left shoulder grew weary always holding firm the glittering shield, but they could not move him from his place, though they hurled their shafts (106-8)

Ennius moves from the shower of missiles in the first four lines to his hero's efforts to ward them off, and then back again to the enemy. The change of focus must be deliberate, for the entire passage is care-fully structured. Ennius varies the speed of his lines effectively and uses caesura and diairesis to articulate his phrases. He also deals artfully with larger units of sense. These eight lines break into two perfectly symmetrical sections. Each consists of a single end-stopped line, then three short cola spread over the next one and a half lines, and then another single idea, this time run over the next one and a half lines.5 Such technical dexterity makes for very effective poetry, but the effect is again caused by a departure from the Homeric model. Emphasis in Ennius lies on the attackers' onslaught. We read more about their missiles and their efforts than of the immovable hero who fends them off. The governing image has had to change because Ajax's towering σάκος has of necessity become the small, round parma of a tribune as he faces Istrian weaponry in a contemporary battle. We are in the modern world facing modern conditions, though incorporation of the Homeric echo raises this Roman soldier to epic stature.

5 For more on the artistry of the passage see von Albrecht, Römische Poesie, 26-31, and Williams, Tradition and Originality, 687-89.


Evoking these Homeric connotations and associations enables Ennius to enrich both the tone and scale of Roman epic. His technical innovation thus creates a new alliance with the Greek tradition, a change that he foreshadowed at the very beginning of his poem. The proemium to Book 1 had replaced the Italic Camenae of Andronicus and Naevius with the Olympian Muses (1) and deliberately put us in the world of Greek poemata, not the traditional Latin carmina (12). Nor was Ennius content simply to sound like Homer. He also claimed a special relationship with his illustrious predecessor. Like Hesiod and Callimachus before him, Ennius began his poem by describing a dream that provided both inspiration and guidance. An extra-ordinary teacher emerged from the Acherusia templa to bring him an extraordinary message. This teacher was Homer himself: "visus Homerus adesse poeta" (3). The vision he went on to describe remained famous in Latin literature down through Augustan times. We know it best from Lucretius, whose reference provides our most explicit testimony.6 After hailing Ennius as the first Roman to have fetched the Muses' crown from Helicon, Lucretius recalls the famous moment in the Annals,

unde sibi exortam semper florentis Homeri commémorât speciem lacrimas effundere salsas coepisse et rerum naturam expandere dictis.

He records that the shade of ever-famous Homer, gushing salty tears, rose up from Acheron and began instructing him in the nature of the universe. (1.124-26)

Homer weeps for joy at the sight of his successor and explains Ennius' gift by setting forth a doctrine of transmigration in what Horace would later call a Pythagorean vision ("somnia Pytha-gorea").7 Souls pass into other suitable bodies after death. Homer's own soul was once incarnated as a peacock ("memini me fiere pavom/' 11); it now lives on in Ennius. The Roman poet thus claims not simply to imitate, but actually to be Homer. Greek and Roman epic traditions merge through a fusion of the Greek and Roman poets themselves. It is a bold and wonderful notion, particularly wonderful for its claim of

6 Lucr. 1.112-26 contains several apparent echoes of Ennius. Other references to the passage include Lucil. 1189; Cic. Rep. 6.1; H. Ep. 2.1.50.52; and Prop. 3.3.1-16, after which echoes became fainter until the grammarians started annotating the later classics. The reconstructions of Ennius' proemia have been hotly debated. I follow Skutsch, Annals, 147-67 and 370-78, whose discussions are an admirable mix of detail, cohesion, and good sense. Detailed analysis and bibliography in Suerbaum, Untersuchungen, 46-113.

7 Hor. Ep. 2.1.52. For Homer's tears, see Timpanaro, RFÏC 119 (1991) 30-43, with references to earlier discussion.


fusion rather than displacement as the force for change. Ennius' style certainly involved a new start, but it was not all new. He did not entirely abandon the achievement of his Roman predecessors.

This fact is easily forgotten, for Ennius himself stressed the distance between himself and the older epic poets. We have already noted his adoption of the Greek terms poemata and Musae for their traditional Latin equivalents. At the beginning of Annals 7, when his history of Rome had reached the First Punic War, Ennius took the opportunity inherent in that milestone to declare quite explicitly his independence from the older epic style. He apparently felt no need to repudiate Naevius completely by retelling his story, but he did abandon the Naevian manner of telling it. The proemium to this book called attention to the resulting technical distance between himself and his predecessor (206-7).

scripsere alii rem vorsibus quos olim Faunei vatesque canebant

Others have given an account in rhythms which the Fauns and seers once sang.

In keeping to the old meter, Naevius had bound his poem to an archaic style of limited capability. Ennius looks out on a much broad-er literary horizon. He has scaled the heights of the Muses and been inspired to fashion a new diction befitting that grander perspective. The exact words of his claim are lost, but the thought, as recon-structed by Otto Skutsch, went something like this (208-9):

nam neque Musarum scopulos escendit ad altos nec dicti studiosus fuit Romanus homo ante hunc.

for no Roman scaled the Muses' lofty crags or was careful with his speech before this man.

This announcement reflects a careful balance between study and inspiration as the sources of his achievement.8 The Annals was indeed a work of genius, which is what the Muses here represent, but it was also a work of calculation and skill.

The nature and source of that skill have aroused great interest and considerable speculation. Though not actually contemptuous of his predecessors—Naevius, after all, was also a poeta—Ennius is clearly conscious of his special gifts and of the learning that nourished them.9 His claim to be dicti studiosus seems to recall the φιλολογία of

8 This reconstruction, proposed by Skutsch, Annals, p. 374, is based on disjointed quotation of the lines by Cicero at Brut. 71-76 and Or. 171.

9 Naevius wears poeta (not vates) like a cognomen in both contemporary verse (the Metellan barb and the epitaph) and the later prose tradition (Var.


Alexandria, and this echo has been heard as a commitment to the aesthetic standards and literary experiments we associate with the age of Callimachus. It is an attractive hypothesis. Ennius' own exper-iences growing up in Magna Graecia and his work as a teacher at Rome could well have interested him in contemporary Greek literature. He knew Greek drama well. He translated Euhemerus, whose ideas subsequently gained some currency at Rome, and he shows familiarity with the teachings of Pythagoras and Em-pedocles. He experimented with Latin versions of works by the Sicilian poets Archestratus and Epicharmus. His own Epicharmus may even have included a dream vision: Cicero not only quotes a line appropriate to such a context, "nam videbar somniare med ego esse mortuum" ("for I seemed to dream that I had died"), but also associates it explicitly with the dream of Annals 1 (Luc. 51). What such earlier experiences with Greek literature meant for the com-position of his epic, however, remains unclear.

It is by no means certain that φιλόλογος was in fact an Alexandrian catch-word in Ennius' day—the first man known to have claimed the title was Eratosthenes, his older contemporary—or that Ennius means to associate himself here with Alexandrian literary canons.10 Signs of Alexandrian influence have in fact proven difficult to find in the fragments. The Greek models that inspired many of the Annals' stylistic experiments are Homeric, not Hellenistic. Thus when Ennius uses the figure Greeks called apocope, the dropping of a final syllable, he follows Homer's practice at line end: endo suam do (587) - ήμέτερον δώ. When stretching this figure to create a more original formulation, he takes care to signal his meaning with the help of a significant adjective: altisonum cael (586).11 He seems to be building directly on Homer. In Ennius' hands, the Greek device passes as a

ap. Gell. 17.21.45, Cic. Sen. 20). Ennius' break with the Naevian tradition as described by Newman, Augustus and the New Poetry, 64-66 and 100-101, is greatly overdrawn.

10 Suet. Gram, 10. Cf. Pfeiffer, History of Classical Scholarship, 156-59, and Skutsch, Studia Enniana, 6-7 (a somewhat tendentious discussion). For a more measured account of Ennius' erudition, see Feeney, The Gods in Epic, 120-23.

11 Ancients certainly read δώ as an abbreviated form of δώμα, though modern explanations of the form actually vary. See Heubeck, A Com-mentary on Homer's Odyssey, vol. 1, p. 99. A third example of apocope in Ennius, laetificum gau (585) is the least likely to be genuine. See Skutsch, Annals, 726-28. An example of extraordinary tmesis, "saxo cere comminuit brum" (609 Vahlen), which Zetzel, AJP 95 (1974) 138-40, claimed as a clear sign of Alexandrian influence, is spurious. Thus Skutsch, 787-88. There are no authentic examples of such tmesis in Ennius. Fruyt, Glotta 69 (1991) 243-46, does not address the problem of authenticity.


homegrown product, nurtured by the thoughtful study of his predecessors and accommodated to the requirements of his own time. It is not in either case an expressly Callimachean innovation or, to put things less specifically, not an innovation that declares any Hellenistic pedigree, and our inability to apply such a label is itself of some importance. While no great harm has come from seeking specifically Alexandrian influences on the Annals, the search for them has distracted readers' attention from something at least as important and much more palpable. Despite his hints to the contrary, the main stylistic influence discernible in Ennian verse is a Latin one. The Saturnian tradition also lived on in him.

This should hardly be surprising. As Wallace Lindsay observed over half a century ago, "a poetical diction for Rome had been formed during successive generations by Saturnian writers: it was this diction, these phrases, these turns of speech . . . that Plautus and Ennius transferred to Greek moulds. It is natural to seek in Saturnian Verse the reason for this or that departure of Plautus from Menander's type or of Ennius from Homer's type."12 Yet, as Lindsay also knew, the debt is easier to suspect than to demonstrate. Metrical features in Ennius that originate in Naevius' practice have proven impossible to isolate from the more general linguistic consequences of verse composition in Latin. Correspondences that fuel only dubious arguments for cause, however, remain sure evidence of effect. Some Ennian techniques certainly recall the earlier style. Coincidences stemming from a common Latin base no doubt lurk among them, while others may well be deliberate echoes. Whatever their causes, the cumulative effect on readers is the same. There are places where Ennius, though writing in the new style, clearly recalls the sound of his predecessors. His innovations then wear a familiar guise, and that familiarity was itself a factor in the success of his revolution.

One of the most widely recognized correspondences between Ennian and Saturnian verse involves placement of the caesura.13 What metricians call penthemimeral caesura, a word end falling after the first position of the third foot, occurs in nearly 88 percent of Ennius' verses but in something less than 45 percent of Homer's. A line like "Livius inde redit magno mactatus triumpho" ("Livius returned from there honored with a great triumph," 299) may therefore sound un-Homeric, but its opening rhythm will recall a line like Naevius' "silvicolae homines bellique inertes" ("wood-dwelling men and unskilled in war," 11). It also sets a single thought in each of two cola much like Naevius' "noctu Troiad exibant capitibus opertis" ("they

12 Lindsay, Early Latin Verse, 8-9. The discussion of nondramatic prosody on pp. 1-10 remains valuable.

13 Witte, RhM 69 (1914) 217-18.


left Troy by night with covered heads/7 5.2). The prevalence of penthemimeral caesura in the Ennian hexameter not only leads to formulae of distinct metrical shape, like olli respondit (31, 113) and succincti gladiis (425, 527), but contributes to the patterning of whole lines.

contendunt Graecos, Graios memorare soient sos They compare the Greeks; Graii they habitually call them. (357)

accipe daque fidem foedusque feri bene firmum Take and give assurance and have a treaty firmly made (32)

A discrete thought stands in each colon, while sound and sense bridge the caesura. Such patterns are familiar from Andronicus and Naevius.

quando dies adveniet, quem profata Morta est when the day comes which Morta has foretold (And. 23)

res divas edicit, praedicit castus. he sacred rites decrees, declares a state of chastity (Naev. 8)

It would be rash to infer from this correspondence that Ennius favored penthemimeral caesura under Saturnian influence, or that Naevius and Andronicus taught him to order his thoughts in this way. The hexameter is itself a bipartite verse. A caesura normally divides it into two dactylic cola, and metrical cola normally correspond to units of sense. The continued prevalence of penthemimeral caesura in hexameter poetry after Ennius suggests more the influence of Latin word shape and stress accent than a conscious imitation of earlier epic verse. Yet the effect of this similarity is significant. Its continuation of traditional patterns enables Ennius' imported meter to echo sequences of sound and thought that his audience already knows.

The use of alliteration to link Ennian cola has also been claimed as a debt to Saturnian composition.14 Alliteration is indeed a feature of all early epic, and Ennius sometimes carries it to extraordinary lengths. The line "O Tite, tute, Tati, tibi tanta, tyranne, tulisti" (104) is notorious. The sound play of the line is, in origin, natural enough: an alliterated name (Titus Tatius), a common phrase (tute tibi), and an emphatic word (tyranne). The word order necessary to produce this effect, however, disrupts the natural phrasing of the idea. The line is carried by sound more than sense, and by the early first century B.C. alliteration on this scale was understandably thought excessive.15

14 Bartalucci, SCO 17 (1968) 99-122. 15 Rhet. Her. 4.18, faulting its "litterae nimia adsiduitas." Note the con-

trast with Skutsch's Greek parallel, OT 371: τυφλός τά τ' ώτα τον τε νουν τά τ' ομματ' ει. Sophocles' alliterative pattern is set by the key word τυφλός but


A more restrained though also striking relationship between sound and sense occurs in the invocation of Book 10.

insece Musa manu Romanorum induperator quod quisque in bello gessit cum rege Philippo

Tell, Muse, what each commander of the Romans accomplished in the war with King Philip (322-23)

Appearance of the rare verb insece invites comparison with Andron-icus' "virum mihi Camena insece versutum." While Andronicus could accommodate only a single verbal idea in his single Saturnian, Ennius integrates a more complex thought over two lines in something closer to Homeric fashion: he need not restrict his thoughts to the length of individual cola. Yet Ennius, like Andronicus, also uses alliteration to enhance the integrity of each verse. The sound pattern runs across the line in the first case; it employs homoioteleuton at colon ends in the second. A connection between the two lines is made by setting manu extraordinarily far from gessit. The surprise is mitigated by a resonance that is undeniably pleasing, if not entirely natural.

Such striking dislocations of normal word order occur most often in the versification of proper names. The most interesting of these is the praise of Cornelius Cethegus in Book 9.

additur orator Cornelius suaviloquenti ore Cethegus Marcus Tuditano collega Marci filius. is dictus popularibus ollis qui tum vivebant homines atque aevom agitabant flos delibatus populi Suadaique medulla.

The sweet-speaking orator Cornelius followed, Cethegus Marcus, to Tuditanus the colleague, of Marcus the son. He was called by countrymen, the folk then living and drawing out their years, the prized flower of the people and soul of Persuasion. (304-8)

Ennius may be taking pains here to distinguish this Cethegus (cos. 204) from the more distinguished C. Cethegus (cos. 197), a man better known for military prowess than for oratory. He has certainly gone to extraordinary lengths to incorporate the complete Roman name, M. Cornelius M. f. Cethegus, into the hexameter. The fragment itself appears self-contained: additur is clearly a beginning and the bold archaism of Suadai medulla suggests closure. The whole passage,

carried through the line by articles and connective particles. The nouns make a different kind of contribution, especially with the expected όμματα delayed to the end. The line therefore has both force and direction, while the Ennian line has only force.


with its symmetrical division between identification and praise, recalls the equally balanced, six-line elogia of the Scipios, and the closest parallel to this extended hyperbaton of a full Roman name is in fact from the Saturnian epitaph for the son of Africanus, the man who adopted Scipio Aemilianus.

quare lubens te in gremiu Scipio recip[i]t terra, Publi, prognatum Publio, Cornell

Therefore the grieving earth, Scipio, takes you to her breast, Publius, offspring of Publius, Cornelius (ILLRP, no. 311.6-7)

Ennius has succeeded in fitting Roman formalities and Roman man-nerisms into the hexameter, and there is more to this effort than technical bravado. The new verse form has once again been made to sound traditional. The traditional sound accommodates the new form to uniquely Roman requirements.

Such observations, however, provide little more than confirmation of a truth put most succinctly by Horace:

. . . si solvas "postquam Discordia taetra Belli ferratos postes portasque refregit/' invenias etiam disiecti membra poetae.

. . . if you dissolved "After foul Discord flung back the iron posts and gates of War/' you still would find the pieces of a poet torn apart. (Sat. 1.4.60-62)

The techniques of meter, diction, and syntax that create Ennius' distinctive epic sound can be easily recognized and, with only minor adjustments of taste, duly admired. Yet a practical criticism rooted in the minutiae of these fragments has clear and frustrating limits. Ancient literary criticism may have thrived on the analysis of isolated lines, but modern readers prefer thinking in larger units. Once we have tuned our ears to the weighty rumble of Ennius' epic machine, we naturally begin wondering how and what that machine was designed to accomplish. It somehow sustained an extremely long narrative poem: these Annals ran to fifteen books in the original design. Some contained perhaps a thousand or so lines, and Ennius eventually added three more books devoted to more recent achieve-ments. How did they tell their story? What can we say of Ennius as a teller as well as a sounder of epic deeds? Two passages can help us, each offering its own glimpse into Ennius' narrative style. Both come from his first book.

The story of Rome's founding, which Naevius had embedded in his account of the First Punic War, begins the narrative of Ennius' more inclusive Annals. The poets' version is not quite the story we know


from Livy and Plutarch. Later writers followed Fabius Pictor, Rome's first native historian, in br idging the chronological gap between Aeneas and Romulus with a line of Alban kings. Naevius and Ennius ignored this complication. They had a daughter of Aeneas be the t w i n s ' mother . 1 6 They call her Ilia, though her role is much like Livy's Rhea Silvia: a Vestal virgin made pregnant by Mars. Con-signed to the Tiber after giving birth, she is rescued b y divine intervention and eventually married off to the river Anio.

In our first extended fragment of the Annals, Ilia tells her half-sister of a terrifying dream that seems to prefigure these events. She has apparently called out in her sleep, rousing the sister and an old servant woman, who rush into her bedroom with a light.

et cita cum tremulis anus attulit artubus lumen. talia turn memorat lacrimans, exterrita somno: 35 "Eurydica prognata, pater quam noster amavit, vires vitaque corpus meum nunc deserit omne. nam me visus homo pulcer per amoena salicta et ripas raptare locosque novos. ita sola postilla, germana soror, errare videbar 40 tardaque vestigare et quaerere te neque posse corde capessere: semita nulla pedem stabilibat. exim compellare pater me voce videtur his verbis: 'o gnata, tibi sunt ante gerendae aerumnae, post ex fluvio fortuna resistet/ 45 haec ecfatus pater, germana, repente recessit nec sese dedit in conspectum corde cupitus, quamquam multa manus ad caeli caerula templa tendebam lacrumans et blanda voce vocabam. vix aegro cum corde meo me somnus reliquit." 50

Quickly the old woman, limbs atremble, carried in a light. Then, roused from sleep, she recalls with tears what had occurred: "Eurydica, my sister whom our father loved, strength and life have now completely left my body. A handsome man appeared to take me to a pleasant reedy river bank, a strange locale. And afterwards alone, my sister, I seemed to wander and 40 to track you slowly and to seek you, though unable to fulfill this wish: no path would guide my foot. Then my father's voice appeared to call me with these words: 'What troubles, daughter, you must bear before good fortune saves you from the river/ 45 Our father, sister, saying this, at once withdrew

16 The annalistic and poetic traditions are contradictory but came to con-taminate each other in the second century. See Strzelecki, RFÏC 91 (1963) 440-58.


and could not bring himself to come into my sight, although I stretched my hands to heaven's azure realm in tears while saying much with gentle voice. Sleep has now reluctant left me with a sorry heart." 50

The passage employs a wide variety of stylistic devices. The meter itself creates a breathless, uncertain quality through the use of long dactylic runs, weak caesurae, and one line without any caesura at all (42). Such effects are well suited to reflect Ilia's agitation.17 Ver-bal effects are equally rich. Alliteration is extensive, and especially effective because it is so carefully controlled. Nearly every line has its own pattern. The first, with a different set of sounds in each colon, is especially striking, and its repetitions are enhanced by the correspondence of metrical ictus and word accent. Dislocations for alliterative effect contribute to sense as well as sound: raptare (39) not only alliterates with ripas but also isolates locosque novos, which becomes the key phrase for the next part of Ilia's narrative. Word placement is consistently precise. The dream itself falls into three parts: the abduction (38-39), the wandering (39-42), and the appari-tion of Aeneas (43-49). Each is introduced by a form of videor. The repetitions of prognata, pater, germana, lacrimans, and corde are similarly deliberate. The difficult phrase corde capessere at line 42, which Skutsch thinks must mean something like "though desiring to reach you," is particularly bold.18 Ennius perhaps softens the shock with corde cupitus in line 47, an echo whose easier construction yields a similar sense. Other notable phrases will reappear in fragments placed later in the story. The formula hate ecfatas (46) also opens line 57. Words eventually spoken to comfort Ilia echo Aeneas' words to her in the dream ("Ilia, dia nepos, quas aerumnas tetulisti," 60), and the prediction of a son elevated to heaven ("in caerula caeli templa," 54-55) recalls her last gesture of despair (48).

The overall effect is highly dramatic, as Ennius, the consummate tragedian, doubtless intended. Many of the devices used so effec-tively here recur in his tragedies, which contained notable scenes of

1 7 Ennius may himself pun on the metrical effect in line 42 if pes -"metrical foot" or perhaps "verse." As Skutsch says ad loc., "ut semita nulla pedem, ita caesura nulla vers um stabilit."

18 Cupitam capessere ("to reach the one desired") is Skutsch's gloss on the phrase; capessere with the instrumental ablative is difficult to construe as a verb of perception. Vine, Glotta 67 (1989) 123-26, would emend to colla capessere, which has almost everything to commend it except, I think, the sense. Ennius' verbal sequence looks to a word for reaching, not for an act taken once the sister is reached. If the text itself is wrong, Skutsch suggests that corde may conceal the sister's name (or perhaps a vocative such as cara?). Ilia is prone to such interjections.


divine possession and prophecy. Cassandra, for example, had an elaborate scene of prophetic trance in Ennius' Alexander that in-cluded these lines:

mater, optumatum multo mulier melior mulierum, missa sum superstitiosis hariolationibus; tnequei me Apollo fatis fandis dementem invitam ciet.

Mother, by far the best mother of mothers, I am carried aloft on soothsayer's breezes; Apollo drives me unwilling, mad to fateful utterances. (34-36J)

The passage features a similar use of single-line alliteration—the consonant sounds of missa look both forward and back—and line 36, which explains the extremely difficult phrase superstitiosis hario-lationibus, is an internal gloss analogous to corde capessere / corde cu-pitus. There is also marked repetition throughout the scene: forms of optumum, mater, and pater reappear in the lines following these. Ennius uses the combination of sound patterns and allusive detail to create an aura of tension and mystery.

Ilia's dream also enhances an extraordinary narrative moment. Her apparent ineffectiveness before these kaleidoscopic shifts of scene and character reflects a well-known feature of dream psychology. She suffers the dreamer's helpless exhaustion that Vergil would later depict, with a nod to this passage, in the last and famous simile of the Aeneid.19 Her transitions of time and place (nam, postilla, exim) are deliberately vague, and this very in-coherence establishes her credibility. Yet while dreams are common in epic, this dream does not function in the usual epic way. Epic dreams, like their counterparts in drama, are usually prophetic or admonitory, or they reveal the dreamer's mental state. The two functions sometimes combine, as in Penelope's dream about her slaughtered geese, which portends the death of the suitors even as it reveals her ambivalent feelings for them (Od. 19.535-58). Hellenistic epic came to exploit this double function. In the third book of Apollonius' Argonautica, for example, Medea's confused loyalties torment her in sleep: she fears for Jason's safety when he is commanded to harness Aeetes' bulls. Her dream, like Ilia's, presents a complex sequence of events that are vitally but only vaguely connected.

19 "non corpore notae / sufficiunt vires nec vox aut verba sequuntur" (Aen. 12.911-12, a passage beginning at 908). For Vergil's debt to Ennius, see Steiner, Der Traum in der Aeneis, 20-22, Skutsch, Annals, 193-95, and Krevans, HSCP 95 (1993) 257-71.


She dreamt that the stranger had accepted the challenge, not in the hope of winning the ram's fleece—it was not that that had brought him to Aea—but in order that he might carry her off to his own home as his bride. Then it seemed that it was she who was standing up to the bulls; she found it easy to handle them. But when all was done, her parents backed out of the bargain, pointing out that it was Jason, not their daughter, whom they had dared to yoke the bulls. This led to an interminable dispute between her father and the Argonauts, which resulted in their leaving the decision to her—she could do as she pleased. And she, without a moment's thought, turned her back on her parents and chose the stranger. Her parents were cut to the quick; they screamed in their anger; and with their cries she woke. (Arg. 3.616-32, tr. Ε. V. Rieu)

The dream both represents Medea's conflicting passions and pre-figures her eventual action. Her own thoughts and those she projects on others dominate the action in the dream and reveal the state of mind that will motivate analogous events in the actual narrative.

Ilia's dream lacks this motivational aspect. The anonymous figure who whisks her away is not the projection of her own longings. If Otto Skutsch is right to find in the adjective pulcer a hint of his divinity, we may even surmise from the outset that Ilia is the victim of external rather than internal forces. Aeneas' last words to her cer-tainly suggest as much. This is not an essentially psychological moment. When Vergil alludes to this passage in describing Dido's nocturnal anxieties, he restores the psychological function of the epic dream, and the resulting contrast is significant (Aen. 4.465-73). Dido's helpless wanderings are, like Medea's harnessing of the bulls, the symbol of her mental torment, not the record of steps taken in the world of fact. The dream Ennius creates stands apart from such analogues. It provides not a motivation but a description, however oblique, of actions that will indeed take place, and this presents a problem. The dream is elaborately prophetic, and Skutsch rightly wonders how Ennius could fulfill this prophecy in the following narrative without redundancy or anticlimax.

What exactly was Ilia's story? Our fullest version appears in Dionysius of Halicarnassus (1.77-79): the Vestal goes to fetch water at a grove sacred to Mars and is raped by the god. The details of that encounter are suitably melodramatic. There is sudden darkness, a portentous rumble from heaven, and then a superhuman apparition reveals itself to Ilia and proclaims great things in store for the twin products of their union. This, says Dionysius, is the story most authorities tell, and that report is probably correct. The late antique Origo gentis Romanae (20.1) expressly attributes an abbreviated form of this tale to Fabius Pictor and Vennonius, an annalist of the later second century. The Greek mythographer Conon took up the tale a century later, and Ovid used it at the beginning of Fasti 3. The melo-


drama probably originated with Diocles of Peparethus, the first writer on Roman antiquities and a man prone to the dramatic and fictitious.20 It seems to recall the myth of Tyro's rape by Poseidon, which produced the twins Neleus and Pelias. That was a story well known in antiquity from Homer (Od. 11.235-59), Hesiod (Catalogue, fr. 30), and Sophocles (Tyro), and it had its own vogue at Rome: at least one Latin tragedy was based on the story of Neleus. Did Ennius follow Ilia's surrealistic dream with a similarly melodramatic narrative?

So elaborate an account of Ilia's pregnancy could be worked into an epic, and three small fragments might support this hypothesis.

unus erit quem tu tolles in caerula caeli/Templa There will be one whom you will raise to heaven's azure realm (54-55)

These could come from a speech of consolation spoken after the event by Mars: Ilia in 56 would then be vocative, and the troubles referred to in 60 would be the oppression by her uncle that has forced her to Vestal status. By this account, Mars would be her benefactor, not her oppressor. The perfect tense of the related fragment xxxviii ("cetera quos peperisti / ne cures") would then place it in a different scene, perhaps spoken after Ilia's marriage to the river Anio. Yet a recon-struction along these lines suffers from its redundancy: two encounters for poor Ilia and two exculpatory apparitions. Why should Ennius invent a dream for Ilia if he was going to retell a story his audience already knew from Diocles or Fabius Pictor? Why resurrect Aeneas if he was going to give her Mars?

There is a better alternative. The delicacy of the dream narrative suggests that Ennius may simply have suppressed details of Ilia's rape, just as Ovid, who rarely skirts the sensational, would even-tually do. His Ilia, like Homer's Tyro, sleeps through the crucial moment (Fasti 3.21-22). Ennius' discretion led him to combine ele-ments of the old tale in a different way. Ilia's merciful sleep and the

20 Diocles, no. 820 in FGH 3, is known only from passing references in Athenaeus and in Plutarch, who identifies him as a source for Fabius Pictor. Vennonius is hardly more than a name (HRRf CCVilli), and attributions in the Origo are often suspect. There is rather more of Conon (no. 26, FGH 1, p. 209). For Fabius and Diocles, see Timpe, ANRW 1.2 (1972) 942-45. Skutsch, Annals, 194 traces this tangled web with more certainty than the evidence warrants. These relationships, however likely, are almost entirely conjectural.

. . . at Ilia reddita nuptum

. , . but Ilia given to marriage (56)

Ilia, dia nepos, quas aerumnas tetulisti Ilia, descendant of gods, what troubles you have borne (60)


divine apparition that consoles her come from the tradition, but he has altered their order and function to create a different version of the story. There was no trip to a spring or any of the usual mythological clichés. Ilia never actually encounters Mars, which is why she calls only upon Venus in her time of need (xxxvi). Ilia's dream as we have it is, I think, the narrative itself: she is describing what has already happened as if it were a dream. The encounter is then all the more terrifying for her inability to describe it clearly. The verbal echoes so characteristic of Ennius' style link her unusual account with the following prophecies, which then confirm the truth of her experience. Its significance dawns gradually on the reader, creating a true revelation through a bold narrative stroke. That cryptic boldness may even explain why Ennius' version of the Ilia story has left no trace in the later tradition. It was too difficult a sequence of ideas to lift from its context, but it is also testimony to the dramatic power of a poet who was also, after all, the leading tragedian of his day.

Our second long narrative passage, the augury of Romulus and Remus, introduces comparable elements of drama and surprise by different means. Its governing device is antithesis, which is another common feature of Ennius7 style, A simple, normative example of the pattern is found in the famous speech of Pyrrhus scorning Roman efforts to ransom the prisoners of Heraclea (183-90).

Nec mi aurum posco nec mi pretium dederitis: non cauponantes bellum sed belligerantes ferro, non auro vitam cernamus utrique. vosne velit an me regnare era quidve ferat Fors virtute experiamur, et hoc simul accipe dictum: quorum virtuti belli fortuna pepercit eorundem me libertati parcere certum est. dono—ducite—doque—volentibus cum magnis dis.

Neither do 1 ask for gold nor offer me a ransom: not touting war but waging it with steel, not gold let us run our risks. Whether you shall reign or I—whatever Fortune wills— we shall test by strength, and take this as a fact: whose strength the chance of war has spared, their liberty I certainly do spare. I grant them—take them—I give them up just as the great gods will.

The fragment employs some familiar stylistic ornaments: homoio-teleuton, alliterations created by a boldly disjunctive word order (era . . . Fortuna) or an equally bold collocation (dono ducite doque), and repetition (virtus, parco). Its parallelism and antithesis are even


more striking, their effect heightened by the metrical structure. The first four lines are end-stopped, and each contains its own antithesis (nec . . . nec; non . . . sed; ferro, non auro; vosne an me). Pyrrhus7 high-principled dictum then spans two lines with similar beginnings and identical rhythm. The change of construction from one to the other implies the subordination of me to belli fortuna. The last line features some extraordinary metrical effects. Spondaic dono, which makes the previous statement explicit, sounds almost final, while the diaeresis of ducite sets this interjected imperative apart from its context. Skutsch marks this isolation with the first set of dashes: the two verbs of giving (dono . . . doque) certainly go together. His insistence on taking the cura-phrase with ducite, which is what his final dash is meant to indicate, is far less certain. The parallels he claims for cum volentibus dis with the imperative are not particularly close, and the result is awkward. Pyrrhus' doque more appropriately has the modifier, which is bound to it by lack of metrical pause as Ennius accelerates the line before bringing it to a sudden, spondaic end.

An equally striking combination of rhythm and antithesis occurs in a fragment from Book 8 (247-53).

proelia promulgantur, pellitur e medio sapientia, vi geritur res; spernitur orator bonus, horridus miles amatur; haud doctis dictis certantes, nec maledictis miscent inter sese inimicitias agitantes; non ex iure manu consertum, sed magis ferro— rem repetunt regnumque petunt—vadunt solida vi

battles are decreed, good sense is driven from the field, by force is business done; the honest orator is spurned, the uncouth soldier loved; not arguing with thoughtful speech, nor just by hurling insults back and forth do they stir up their quarrels; not asserting rights by law, but rather with the sword— they press their claim and seek success—they hurry on with force

unchecked

The passage reflects a basic dichotomy of Roman public life.21 As Cicero, thinking of this passage, would put it, "two skills can bring men to the highest point of honor. One is military command, the

21 The passage is usually thought to describe conditions at Rome during the Hannibalic War (Skutsch, Annals, 432-33), but Pascucci, "Lo scoppio delle ostilità/' 111-14, attributes it to Carthaginian attitudes at Saguntum. The present stylistic argument is unaffected by this question.


other oratorical success."22 Ennius' antitheses seem equally straight-forward, but the pairings in this fragment are never actually quite equivalent. The semantic, though not syntactic, match for sapientia is vis, which makes the line less symmetrical in thought than in structure. The chiasm of the next line is exact, but the rhythm is not. Spondaic lingering over orator (like doctis dictis in the next line) yields to the greater fluency of horridus miles; putting the negative idea in the second half of each line allows the correspondence of metrical and natural stress to emphasize the key words vi and horridus miles. The negative statement in 250-51 never has a positive correlate; ex iure is superseded by ferro. This subversion is, of course, Ennius' point; positive values are both explicitly dismissed and implicitly compromised by their manner of expression. The language itself is weighted toward violence. Proelia promulgantur—Skutsch's restoration of these words from Cicero's context is almost certain—is an oxymoron: laws, not battles, are rightly promulgated. Pellitur is a military term looking ahead to vis in the following clause, and in ex iure manu consertum, a legal formula for asserting claims of ownership, we find the echo of manum conserere ("to begin fighting") that leads to ferro and solida vi. The ars oratoris boni never receives fair expression, let alone equal status.

The kind of weighted antithesis we find in these small fragments is central to the success of our second major example of Ennius' nar-rative style. The famous augury that confirmed Romulus as the sole founder of Rome occupies our longest and probably finest surviving passage of the Annals. The text looks like this in Skutsch's edition; I abbreviate his apparatus.

Curantes magna cum cura tum cupientes regni dant operam simul auspicio augurioque. in fmonte Remus auspicio sedet atque secundam solus avem servat. at Romulus pulcer in alto 75 quaerit Aventino, servat genus aitivolantum. certabant urbem Romam Remoramne vocarent. omnibus cura viris uter esset induperator. expectant veluti consul quom mittere signum volt, orrmes avidi spectant ad carceris oras 80 quam mox emittat pictos e faucibus currus: sic expectabat populus atque ore timebat rebus utri magni victoria sit data regni. interea sol albus recessit in infera noctis. exin Candida se radiis dédit icta foras lux 85

2 2 Cic. Mur. 30; "duae sint artes quae possint locare homines in am-plissimo gradu dignitatis, una imperatoris, altera oratoris boni." Cicero then goes on to quote Ennius.


et simul ex alto longe pulcerrima praepes laeva volavit avis, simul aureus exoritur sol cedunt de caelo ter quattuor corpora sancta avium, praepetibus sese pulcrisque locis dant. conspicit inde sibi data Romulus esse propritim 90 auspicio regni stabilita scamna solumque.

74 in monte] in Murco Sk. 81 pictos F. Brown, A. Cameron: pictis AVE 82 ore timebat] or&i mebat (littera post i erasa) B: ora tenebat ς

Carefully then taking great care, desiring a kingdom, they together take the auspices and augury. On the Murcus Remus sits in wait and watches for a favorable flight alone. Noble Romulus on high 75 Aventine seeks and watches for the high-soaring race. They were contending whether to call the city Rome or Remora. All men care which one becomes their ruler. They wait as when the consul is about to give the signal, all watching anxiously the starting gates 80 to see how soon he sends the painted chariots from the barrier: so the people wait, worry on each face for their affairs, for which should be given the victory of highest rule. Meanwhile the sun had set into the depth of night. Then struck by rays the shining light showed itself openly 85 and at once on high from far away a beautifully winged left-moving flight advanced. Just as the golden sun arises, there came from heaven a dozen blessed bodies of birds, settling on the place with noble wings. Thus Romulus saw that given to himself alone, 90 approved by auspices, are the base and bulwark of a kingdom.

The f ragment contains problems of t ransmiss ion as wel l as interpretation. Skutsch has introduced some major improvements, but not entirely in the text as printed. His own emendation in Murco is almost certainly right, and his endorsement of pictos currus almost as surely wrong. Murcus, the early name for a separate height southeast of the Aventine, provides the perfect counterpart to Romulus ' position on the Avent ine proper. The emendat ion pictos, in contrast , is insensitive to the context. The crowd's attention is fixed on the barriers themselves just before they open for the start of the race, not on the chariots hidden behind them. The jaws of these barriers are the object of attention, and they require the epithet.2 3 Finally, the

2 3 For the Murcus, see Skutsch, Stadia Enniana, 64-70. No reviewer of Skutsch's edition has liked pictos in the text. I paraphrase the refutation of Gratwick, CR 37 (1987) 164. Skutsch wrongly assumes that the car ceres were necessarily iron and unpainted. See Humphrey, Roman Circuses, 133 and 649 n. 9, for doubts on that detail. We can contrast Sidonius Apollinaris 23.325-41, which describes horses snorting and stamping behind the barrier


Renaissance emendation ora tenebat offers good—some say better— sense than the manuscript reading, but Skutsch rightly points to echoes of this passage in Vergil and Ovid that confirm the idea of fearful anticipation here {Aen. 12.716-19; Met. 9.46-49). We should, however, probably join Vahlen in putting a comma after rebus (83): the enjambment makes rebus slightly less feeble than modern editors are inclined to think.24

Despite such difficulties, the passage has a power that commands respect and admiration. A cursory reading might suggest the kind of structural balance we noticed in Pyrrhus' speech. Participles of similar sound anchor the first line, and the whole passage of course contrasts Romulus on one hill with Remus on the other. Yet a strict antithesis never develops; it instead yields to something more subtle and dramatic. We should be on our guard at once, for those initial participles are not truly parallel, as Pyrrhus' cauponantes . . . belli-gerantes were. Cupientes has regni as a complement. The sense thus spills over into the next line, and all through the passage such potential parallels are deliberately and significantly broken. The initial description is typical.

in Murco Remus auspicio sedet atque secundam solus avem servat. at Romulus pulcer in alto quaerit Aventino, servat genus altivolantum. certabant urbem Romam Remoramne vocarent.

The first two sentences are structurally and semantically equivalent: each has a single subject and a double predicate. But in a passage so full of repetition and echo, variations gain particular importance. Here only Romulus and his hill—the high bird hill, altus Aven-tinus—have adjectives, and avem of the first sentence is replaced by genus altivolantum in the second.25 This ornamentation weights the thought toward Romulus, so the last, explanatory line quite natu-rally puts Roma before Remora. From this point on, details center on

at the start of a race. Because that poem is addressed to the charioteer Con-sentius, it naturally describes the scene from the participant's point of view, not the start as spectators would perceive it. The chariot simile recurs in Ennius, An. 463-4.

2 4 Skutsch omits punctuation after rebus because he finds no parallel for timebat constructed with both dative and indirect question, an objection perhaps answered by Ter. An. 419: "nostrae timeo parti quid hie respondeat/' Gratwick, CR 37 (1987) 168, would relocate the line after 73, an extreme remedy creating problems of its own.

25 The etymological play may go back to Naevius. So Var. L. 5.43 ( - BP, fr. 29): "Aventinum aliquot de causis dicunt. Naevius ab avibus, quod eo se ab Tiberi ferrent aves." For the recurring sounds involving avis and altus, see Dominik, "Ennius' Annales/' 54-55.


Romulus' augury. We completely lose sight of Remus. As in the orator/miles passage, the expectation of equality is thwarted, and that thwarted expectation is itself central to the narrative.

So far, however, Ennius has only established a scene and situation, as his imperfective tenses indicate. The setting sun and subsequent first light bring us from the exposition to the event itself. The poet heightens the tension inherent in that moment by withholding crucial information until the last possible moment. As in an actual augury, we first perceive only something in the sky (ex alto) and far away (longe), then that it is winged (praepes), and only afterward a confirmation that it is indeed a bird (avis).26 The augury is im-mediately explained at greater length (87-89), but only after the sign itself comes clear are we given the final essential fact: that sign has been given to Romulus. What happened to Remus now hardly mat-ters. He slips from the narrative just as surely as he slips from history. The last line offers a strong echo of the second, but where there was a double subject in line 73, there is now only the one.

But what of the extraordinary simile at lines 79-81? The original focus of the passage, and of the contest it describes, is the rivalry be-tween Romulus and Remus. Not until line 78 do we learn of a worried audience for this contest, and their rapt attention is immediately explained in what seems to be a markedly anachronistic way. Ennius has drawn his simile from one of the most popular public events of Republican Rome, the chariot race at the ludi Romani. He has con-jured up a memory of his own time to describe an anxiety felt before the city has even been founded. The chronological anomaly is particularly striking since he has just employed the pseudo-archaism induperator, a metrically convenient coinage with a contrived whiff of ancient days. Now he suddenly brings us forward to an inde-terminate present. There are epic parallels for this fluid sense of time, but, as in the case of Ilia's dream, the parallels are inexact. Homeric similes can certainly seem anachronistic: the river Xanthus, assaulted by Achilles, seethes like a kettle full of meat, although, as the scholiast duly notes, Homeric heroes never cook meat in pots.27

Homer has borrowed a detail from contemporary life to illustrate an important heroic moment. This sort of juxtaposition is an important device in Homer that extends beyond the formal simile. Fantastic deeds can be made credible by emphasizing the qualitative change between past and present.

26 Compare the famous snakes that come from Tenedos to seize Laocoön in Aen. 2: twin things (203), serpents (204), their motion described (205-8), then heard (209-11), with their fiery eyes visible as they reach the shore.

2 7 Σ A ad II. 21.362. I borrow the example from J. B. Solodow, who has kindly shared his thoughts on epic anachronism.


Aeneas took a rock in his hand which two men could not lift, such as men are now, but he easily hefted it alone.

(II 20.285-87, cf. 12.383, 447)

Homer reminds us that his heroes lived in a different, greater time and could achieve what is now the impossible.

Ennius also writes of heroes, but not to foster such distance between them and his readers. Just the opposite. Describing the tribune at Ambracia like Ajax at the ships equates the Roman soldier and the epic hero. In the present passage, Ennius has deliberately done what he can to obscure the distance between legendary past and con-temporary reality. The technique is familiar from Vergil: the simile of Aeneid 1.148-54, in which Neptune calms the stormy sea like a Roman magistrate restraining a mob, is a famous example. Ennius, however, is much bolder in his merging of past and present. His verbal repetitions tie the simile so closely to the narrative that it is actually difficult to know where one stops and the other begins. Skutsch naturally—and probably rightly—thinks the simile begins with veluti and ends with resumption of the narrative after sic ex~ pectabat, but other possibilities are defensible. The populus of line 82 could still be the crowd at the chariot race, and a strong stop is some-times placed after timebat. Whatever interpretation we prefer, there remains an ambiguity that blurs the boundary between past and present. Ennius merges the populus of legendary and contemporary time, and with them the ideas of past and present, of old Romans and modern ones, are similarly joined.

This is not all, for the blending of past and present extends beyond the formal simile. Populus is never simply a Latin word for crowd, and it is only secondarily a generic word for people. Its inevitable hint of that legal entity, the populus Romanus, brings with it connotations of community and civic purpose.28 This too is deliberate and appropriate. Save for the voting assemblies, there were few opportunities and fewer places for the people of Rome to gather as a group. One such place was the Circus Maximus. Unlike theatrical entertainments, which were comparatively impromptu affairs that attracted mixed and random audiences to temporary structures, events at the Circus involved a large and permanent building with formal seating organized by predetermined groups. By the late first century, three tiers of seats accommodated 150,000 spectators in sections re-

2 8 Cic. Rep. 1.25: "populus autem non omnis hominum coetus quoquo modo congregatus, sed coetus multitudinis iuris consensu et utilitatis communione sociatus." Cf. Liv. 39.15.11 for the traditional limits set on public gatherings.


served for each of the thirty Roman curiae, the oldest civilian grouping at Rome. The structure itself was certainly smaller in Ennius' day, but the seating arrangement by curia was a tradition that the Romans themselves traced back to the days of Tarquinius Priscus.29 A place at the Circus therefore brought with it a reminder of civic position as well as civic pride. In choosing the chariot race for his simile, Ennius recalls not just a moment of familiar expectation but a sense of community, an exciting time when Romans gathered as Romans in celebration of that community. And they gathered, of course, in just that place between the Palatine and Aventine where their ancestors would have had their best view of the twins' rivalry. The augury as Ennius describes it thus determines and recalls the very idea of Rome. The legendary, the historic, and the contemporary are treated as a unit.

Combining heroic trappings and modern details in this way allowed Ennius to raise the story of Rome to epic proportions, and his success had an important sequel. Ennius' fusion of past and present prefigures the famous pattern of anachronisms and historical al-lusions in the Aeneid. Vergil's manipulation of time has indeed become a familiar feature of his art, but it is worth recalling that modern critics were slow to understand and accept this technique of reaching out from legendary to contemporary occasions. In a famous passage of his Laocoön, for example, Lessing faulted the description of Aeneas' shield in Aeneid 8 as a contrived, static, and even obse-quious intrusion into the narrative. 30

His objection came in two forms. The first was a charge of vulgar jingoism.

The clever courtier who decks out his material with every sort of flattering allusion is everywhere in evidence, but not the great genius who relies entirely on the intrinsic strength of his work and scorns all external means of awakening interest. Consequently, the shield of Aeneas is an insertion, intended solely to flatter the national pride of the Romans; an alien stream turned by the poet into his own river in order to stir it up.

A second criticism lay in the richness of Vergil's detail, which Lessing read as a vain attempt to better his Homeric model.

For why was it necessary to introduce the whole of Roman history on a shield? Homer was able to make his shield the very essence of all that

2 9 D.H. 3.68, cf. Liv. 1.35.8. For seating arrangements, see Humphrey, Roman Circuses, 69-70 and 76-77, and for the significance of the curiae, which cut across distinctions of wealth and birth, Momigliano, JRS 53 (1963) 108-12.

30 Lessing, Laocoön, 96-97 and 216. For modern responses to Lessing, see Johnson, Darkness Visible, 111-14, and Hardie, Virgil's Aeneid, 336-39.


happened in the world by means of but a few pictures. Does it not almost seem that Vergil, since he could not surpass him in the execution of the shield or in the choice of subjects on it, wanted at least to exceed him in the number of his subjects? And what could have been more childish than that?

Lessing took Aeneas' uncomprehending acceptance of the shield as proof of its irrelevance, and he therefore missed the larger symbolism of Vergil's famous coda,

miratur rerumque ignarus imagine gaudet attollens umero famamque et fata nepotum.

He marvels and delights in images he cannot understand, raising to his shoulder the fame and fortune of his heirs. (Aen. 8.730-31)

Because he finds no rationale for Vergil's digression, Lessing cannot see beyond the literal significance of the burden Aeneas now assumes.31

Lessing's first complaint hardly troubles us today. Sympathetic readers no longer attribute the passage to Roman vanity. Vergil's extended ekphrasis is hardly an isolated phenomenon: the images on the shield and Aeneas' inability to understand them are but part of a series of complex and evocative allusions to the Roman future in Italy. The description is therefore not an inorganic intrusion. Lessing's preoccupation with artistic objects in the Laocoön has forced him to consider the shield apart from its larger context, and that separation has led him astray. His second complaint, however, is not so easily dismissed. Interruptions of this type come at a certain cost. To give sufficient prominence to these contemporary allusions, Vergil must halt his narrative. Setting his story back in heroic time did not re-strict him to heroic time, but stepping out as he does from the world of Aeneas calls attention to the gulf in knowledge that separates his characters and his readers. Even as we admire Aeneas' dedication, we may find ourselves regretting his ignorance. What encouraged Vergil to expose his hero to his readers' condescension?

An answer, or at least a precedent, lies here in Ennius. Turning Roman history into epic was more than an artistic achievement. It also had a social dimension. Ennius' poem not only immortalized the Roman past but gave new status to the Roman present. Elevating contemporary events to Homeric proportions attracted readers to the epic genre and encouraged its particularly close identification with those aristocratic readers who were themselves responsible for the deeds it glorified. The development of such a relationship doubtless began with the Bellum Punicum, which found a way to combine

31 For the symbolism of this act, see Hardie, Virgils Aeneid, 369-75.


legendary and contemporary events in a continuous narrative, but Naevius had told a story of only limited scope. The broader sweep and sophistication of the Annals completed the legitimization of Roman achievements as a theme for poetry. The incorporation of Homeric echoes gave them special significance, while Ennius' ability to carry his poem from mythical origins to modern times brought a new coherence to the Roman record and a heightened awareness of its greatness. By the time Vergil put the identification of past and present to his own use, the link between literary and contemporary events was already an established feature of epic narrative. Readers of Roman epic had already come to identify themselves with the world it portrayed. The creation of that identification is itself of interest, and it must now take our discussion back once more from the technical devices of epic poetry to the social attitudes that en-couraged these literary developments.

5 Poetry and Patronage

Midway through Plautus' Truculentus, the soldier Stratophanes makes a grand and delightfully disingenuous entrance.

Ne expectetis, spectatores, meas pugnas dum praedicem: manibus duella praedicare soleo, haud in sermonibus. scio ego multos memoravisse milites mendacium: et Homeronida et postilla mille memorari potest, qui et convicti et condemnati falsis de pugnis sient.

Don't expect me, spectators, to boast about my battles: with fists do I proclaim my deed, not words. I know that many soldiers have immortalized mendacity, sons of Homer capable of claiming countless things, who later stand convicted and condemned for faking fights. (482-86)

We treat his comment today as a literary, perhaps even meta theatrical joke. This confident soldier, glinting before us in full kit, plays to our expectations of the braggart warrior. His speech may bring Pyrgopolynices to mind, or Therapontigonus Platagidorus, or any of those other comic gloriosi whose words prove so much bolder than their deeds. For Plautus' own audience, however, the joke was as much topical as literary.

By the late third century, real generals were publicizing their victories at Rome with nearly the vigor that had won them abroad. Though state interests must have come first, private interests apparently came a close second: the race for individual glory was fiercely competitive. Military claims and political counterclaims were therefore often in the air. Among the resulting conflicts was a famous case in 190 in which two proconsuls, M.' Acilius Glabrio and Q. Minucius Thermus, petitioned for triumphs. Glabrio's petition was granted, but Thermus was not so fortunate. Cato, aroused to scorn by Thermus' arrogant behavior in Liguria, attacked him in a strident speech that came down to later generations under the title De falsis pugnis. The petition was denied after heated debate. Though Stratophanes' joke about words and deeds is probably not an allusion

111


to this particular case, neither is it simply a literary conceit. The comic effect grows from an issue of genuine public interest and public policy.1

Thermus' political ordeal was hardly unique. An even longer and louder controversy—and one with more definite literary impli-cations—dogged M. Fulvius Nobilior, the consul of 189, when he claimed a triumph for campaigns in Aetolia and Cephallania. He had laid siege to Ambracia, captured it, and then forced the Aetolians to accept peace on his terms. As proconsul in 188, he then moved successfully against the Cephallanians. There were com-plications, though. Ambracia was not actually taken by storm. The Ambraciotes had put up stiff resistance, and they surrendered to the besieging army only when the support of their Aetolian allies finally wavered. The city came to terms, and in 187 the Ambraciotes actually sent a delegation to the Senate complaining of Fulvius7 subsequent pillage of their city. Those emissaries found a powerful ally in the consul M. Aemilius Lepidus, no friend of Nobilior. The Senate proved sympathetic to their complaint, voted them significant compen-sation, and—at Lepidus' further instigation—declared that Ambra-cia had not been captured by force. That decree, combined with the consul's continued opposition, delayed and beclouded Fulvius' claim. The triumph was finally voted only after he returned to Rome himself and pressed the case while Lepidus was abroad. Fulvius promised lavish games in honor of Jupiter Optimus Maximus, then hastily celebrated the triumph itself before Lepidus could return to Rome and interfere.

The whole affair was at the very least undignified, and even the votive games Fulvius held in 186 could not erase the memory. Cato raked it all up a few years later in a speech probably aimed at Fulvius' censorship of 179. One of his targets was the victory crown so prominently displayed in Fulvius' triumph, though no city had been captured nor enemy camp destroyed: "iam principio quis vidit corona donari quemquam cum oppidum captum non esset aut castra hostium non incensa essent?" (ORF3 fr. 148). A second target was Fulvius' literary aspirations: Cato berated him for having taken the poet Ennius to Ambracia.2

1 ORF3 fr. 58. See also Astin, Cato, 59 n. 27 and Liv. 37.46.1-2. Enk, Trucu-lentus, vol. 1, 28-30, and vol. 2, 117-19, rightly dismisses efforts to find a specific reference here to Thermus. Gruen, Studies, 129-40, offers insightful review of the politics surrounding triumphs and ovations in this period and its relevance to Plautine comedy.

2 Cic. TD 1.3: "obiecit [oratio Catonis] ut prob rum M. Nobiliori, quod is in provinciam poetas duxisset: duxerat autem consul ille in Aetoliam, ut scimus, Ennium/' Cf. Arch. 27 and Astin, Cato, 110 n. 22. The major historical

Poetry and Patronage 113

The implication—at least to modern ears—is that Fulvius, like one of Plautus' Homeronides, was grooming a publicist for his achievements. There is circumstantial support for such an interpre-tation. Ennius was by then a major poet, and Fulvius' achievements were eventually celebrated in some serious poetry. A play called Ambracia may have figured in the games of 186, which featured some kind of artistic entertainments as well as wild animal shows. Its plot and genre, however, remain unknown, for its few surviving lines are uninformative. More important was the Annals. The campaign that Cato ridiculed was the very one in which Ennius' heroic tribune fought like Ajax at the ships (391-98) and where, as the poet claimed,

occumbunt multi letum ferroque lapique aut intra muros aut extra praecipe casu

many meet their death by sword and stone within the walls or out, pitched headlong (389-90)

This is not pure fiction. Ferocious fighting raged among the siege works at Ambracia and is described in horrific detail by Polybius and Livy. Ennius was certainly not inventing battles, but the heroic dignity of his verses does stand in striking contrast to the political wrangling that was to follow.

That contrast presents a major problem of literary and social inter-pretation. How we reconcile the lasting achievement of Ennius' poetry with the contemporary realities that nurtured it affects both our reading of the Annals and our understanding of its appeal to second-century Romans. Yet Ennius' priorities are difficult to recon-struct. He would in any case have taken military success seriously, for such seriousness came with the genre. Comedy might joke about fake battles and inflated honors, but epic necessarily saw the issue differently. Conferring Homeric grandeur on the Aetolian campaign was the way to make poetry out of contemporary events, and poetry, as Cicero would observe in the famous opening of his Laws (1.4), claimed standards of truth quite independent of history's obligations. Even so, Ennius' new enterprise could not entirely escape certain contemporary ramifications: his account was elevating Fulvius' achievements to epic stature just as Fulvius' political opponents were advancing a much dimmer view of them. The harried consul must doubtless have welcomed the resulting good press. Was good press therefore part of Ennius' design? Should we go even further and think his artistic achievement only coincidental to his politics? Our own

sources are Polyb. 21.25-31 and Liv. 38.4-10 (the campaign); Liv. 38.43-44 and 39.4-5 (the triumph).


rather cynical age seems to think we should. Though Cato's com-plaint about the poet at Ambracia was directed at Fulvius and not Ennius, the poet's reputation has, in long retrospect, been the ultimate victim of Cato's attack. Preoccupation with the historical and proso-pographical evidence required to set the fragments of his epic in plausible sequence has encouraged scholars to assume and then to reconstruct a political agenda for the poet as well as a structure for his poem. They call Ennius poeta cliens and interpret his epic as an instrument of Fulvian propaganda.3 This dark or at least demeaning approach to Rome's first major poet warrants review. The very point of his Annals is at issue, and so is the nature of its influence on the subsequent course of Latin literature.

Biographical testimony certainly makes clear that Ennius enjoyed prestigious social connections.4 A Calabrian from Rudiae, in the heel of the Italian boot, he claimed descent from the legendary king Messapus. He was never a slave like Andronicus or a provincial hustler like Plautus. He was at least as proud of his Italian origins as Naevius was, and he liked to say that he had three hearts because he spoke Greek and Oscan as well as Latin.5 During the later days of the Hannibalic War, he apparently served with Calabrian auxili-aries in Sardinia, where he met the young quaestor Cato. In 204, Cato brought Ennius to Rome, and, as plebeian aedile in 199, he may well have offered a first dramatic contract when he supervised the ludi plebeii of that year.6 In any case, Ennius soon became a major force in the Roman theater and wrote plays continuously until his death in 169. Though his comedies failed to impress posterity, his tragedies were masterful. There were at least twenty-one. Several became

3 Both the term and its implications are modern. Romans, out of polite-ness if nothing else, would probably have said amicus, even if cliens was meant. See Sailer, Personal Patronage, 8-15; Quinn, ANRW 2.30.1 (1982) 116-22. The political interpretation of Ennius' work is developed most fully by Martina, QFC 2 (1979) 13-74, whose conclusions are accepted passim by Skutsch. A précis of the case for Ennius' putative ira et Studium is presented by Skutsch, Β ICS 23 (1976) 75-78, with fuller discussion in the commentary ad loc.

4 For a general survey of the biographical evidence and its conventional interpretation, see Jocelyn, ANRW 1.2 (1972) 991-96; a more critical dis-cussion paralleling my own appears in Gruen, Studies, 106-22.

5 Sil. 12.393; Gell. 17.17.1. The pride in Calabrian roots may also surface in the Annals, where someone says, "nos sumus Romani, qui fuimus ante Rudini" (525).

6 Liv. 32.7.13. Badian, "Ennius and His Friends," 155-59, doubts the explicit testimony of Nepos Cat. 1.4 that Cato brought Ennius to Rome, but he has convinced few. See the comments by Jocelyn and Suerbaum that follow Badian's presentation, 200-202, and Skutsch, Annals, 1.


classics of the Roman theater, read and produced into Augustan times: fondness for them inspired Cicero in particular to extensive quotation that preserves hints of their genius. Ennius was thus a highly suc-cessful member of the same theatrical community that had fostered the talents of Andronicus and Naevius, but his background enabled him to move into quite different social circles. He eventually had his own household on the Aventine, then a favored quarter for the burgeoning and socially mobile middle class. Several anecdotes link his name with the aristocracy. He could afford to keep a maid, who once kept Scipio Nasica at bay when he came calling, and he owned gardens where he walked with his neighbor, Sulpicius Galba. Fulvius Nobilior, of course, chose to take him to Aetolia in 189, and he probably owed his eventual Roman citizenship to Fulvian con-tacts. There was also a story that his portrait bust adorned the tomb of the Scipios.7

What are we to make of this evidence? Its most striking features are surely the range of Ennius7 acquaintances and the ease with which he moved among them. There is no hint here of political partisanship. The network of contacts crosses and even contradicts the political divisions among Scipionic and anti-Scipionic factions often thought to dominate the politics of this period. Ennius7 literary works, like this biographical testimony, also refer to a full gamut of contemporary public figures. Only the tragic fragments completely lack contemporary allusions: the political coloring given tragic per-formances in Cicero's day was not a feature of Roman tragedy in this, its most creative period. Other works more naturally touch on con-temporary affairs. The meager fragments of a poem called Scipio sound something like a triumphal song, and Ambracia presumably had relevance to the campaign of 189.8 The Annals was by its very nature full of references to distinguished Romans. These references, however, do not justify claims that he tailored the poem to partisan interests. No pattern of reference, much less bias, is discernible among its fragments. Nearly half the poem told of events before Ennius' lifetime: Books 1-3 dealt with Aeneas and the kings, Books 4-6 with the conquest of Italy and war with Pyrrhus, and Book 7, in deference

7 Cic de Or. 2.276 (Nasica); Acad. Pr. 2.51 (Galba); TD 1.3 (Nobilior); Brut. 79 (citizenship); Arch. 22 and Liv. 38.564 (bust). Badian, "Ennius and His Friends," 183-85, casts legitimate doubt on Cicero's identification of a (too) young Q. Fulvius as Ennius' sponsor, though not on the likelihood of a Fulvian connection. For the Aventine locale, see Badian, 166-68; for the bust, Suerbaum, Untersuchunge, 210-13.

8 Scholz, Hermes 112 (1984) 183-99; Suerbaum, Untersuchungen, 239-^8. The political reading of Ennius' career leads Martina, QFC 2 (1979) 17, to date the Scipio after 180.


to Naevius, dealt but briefly with the First Punic War. Only the later books are likely to reflect partisan feeling, but in them we find praise for such disparate figures as Cornelius Cethegus (304-6), Aelius Paetus (329), and Fabius Maximus (363). Cicero explicitly absolved Ennius of partisanship in the praise of Cethegus ("nulla suspicio est amicitae causa esse mentitum," Brut. 58), and attempts to find Fulvian bias in other fragments are unconvincing. Two of these are worth considering in detail.

Perhaps the most ingenious attempt involves Ennius' treatment of the famous reconciliation between Fulvius and his political foe, M. Aemilius Lepidus. The two men had a long history of personal enmity. Fulvius twice kept Aemilius from the consulship, and Aemilius had, at the very least, tarnished Fulvius' Aetolian victory. Their joint election to the censorship of 179 was therefore a potential embarrassment to all concerned, but the rivals rose to the occasion. Immediately upon assuming office, they staged a grand and highly public reconciliation in the Campus Martius that ended with a joyous procession to the Capitol, an episode that later generations made an exemplum of unselfish civic devotion. We might have expected Ennius the poeta cliens to offer a version of this event stressing Fulvius' unselfishness on the occasion. The story, which he set into Book 16 of the Annals, has unfortunately not survived, but Cicero makes it quite clear that Ennius actually put Lepidus at the center of things: "and so that famous M. Lepidus, who was twice consul and pontifex maximus, was praised not only by public memory, but also by history and by the voice of that most illustrious poet because, on the day he was elected censor, he at once restored amity with his colleague M. Fulvius, his most bitter rival." Turning such testimony into evidence of a pro-Fulvian bias has necessitated the invention of an elaborate scenario in which a politically weak Fulvius must have sought the alliance with Aemilius, a circumstance which the poet then obligingly disguised by attributing the initiative to the other side.9 Critics have had to make black into white and turn Ennius' generosity against him.

A rather more compelling case for Ennian bias involves a line from Book 10 that refers to the consuls of 200, P. Sulpicius Galba and C. Aurelius Cotta:

9 Martina, QFC 2 (1979)21-37, endorsed by Skutsch, Annals, 572-74. The account in Livy 40.45.6-46.15 offers no support for such conjectured motives. In addition to Cic. prov. cons. 20, the reconciliation story is told by Val. Max. 4.2.1 and Gell. 12.8.5. Literary critics should be aware that the reconstructed factional politics underlying interpretations of this sort is now losing credi-bility among historians. See Brunt, Fall of the Roman Republic, 443-58.


Graecia Sulpicio sorti data, Gallia Cottae Greece falls by lot to Sulpicius, Gaul to Cotta (324)

Strictly speaking, Cotta's province was Italia, not Gallia, which went that year to the praetor L. Furius Purpurio. Consul and praetor came into conflict when Furius won a major victory over the Gauls at Cremona with the consul's army before Cotta could arrive on the scene. The enraged consul then packed him off to Etruria, but Furius continued on to Rome and claimed a triumph for his victory (Liv. 31.10-11, 47-48). In his commentary on the line, Otto Skutsch suspects Ennius here of taking Cotta's side in the resulting controversy. Furius, he claims, was an inimicus of Fulvius Nobilior, and so Ennius supports Cotta's claim to preeminence by associating him with the Gallic com-mand. The interpretation is subtle, but it falters on both historical and literary grounds. There is no evidence that Furius really was persona ingratissima (Skutsch's phrase) to Ennius or even to Fulvius, only a presumption based on the fact that Livy, writing of events in 187, calls him an inimicus of Fulvius' consular colleague Cn. Manlius Vulso. The relationship of Furius' triumph in 200—the first ever awarded a praetor—to a factional dispute of the early 180s is entirely unknown, and a "Fulvian" position on Furius is impossible to determine. On the literary side, we should note not just the elegant balance but also the emphasis inherent in Ennius' substitution of Graecia / Gallia for the technically correct Macedonia / Italia. Book 10 dealt with Sulpicius and the war against Philip of Macedon, not events in Gaul or the wrangle over Furius' triumph.10 The very order of clauses here indicates as much: a different emphasis would have led Ennius to a different formulation. The long shadow of the poeta cliens has obscured the salient features of the line, which are poetic rather than political.

Other curious aspects of the Annals also merit greater attention to their artistic rather than exclusively political rationale. Emphasis on Fulvius' Aetolian campaign, for example, furnished an elegant climax to Book 15 and can justify its Homeric qualities on what are essentially poetic grounds. Ambracia had been Pyrrhus' capital. Ennius thus ended the account of foreign wars with this final victory over Rome's first great foreign enemy. His attention to the Muses also makes artistic and structural sense, though a Fulvian initiative again provided the opportunity. Fulvius himself took an interest in their cult that went beyond political posturing. He housed monuments to both Muses and Camenae in a temple on the Campus Martius that

10 The book had its own proemium emphasizing this content: "Insece Musa manu Romanorum induperator / quod quisque in bello gessit cum rege Philippo" (322-23).


became known as the templum Herculis Musarum11 The Annals, too, recognized the association of the Muses, poetic patrons of Ennius' Hellenized epic world, with the Camenae of older, Saturnian epic.

Musas quas memorant nosce nos esse Camenas Know that we Camenae are those they call the Muses (487)

If Skutsch is right in assigning this fragment to Book 15—Vahlen had placed it in the exordium to Book 1—Ennius' original plan would have ended the poem with the dedication of Fulvius' new shrine. The work would thus return in its finale to that praise of the Muses characteristic of the major proemia to Books 1 and 7.

The subsequent extension of the Annals with three more books of bella recentia might better suggest the workings of a propaganda mill, but its putative beneficiaries have been impossible to identify. The valiant Caecilii featured in Book 16 are otherwise unknown; the reconciliation of Fulvius and Aemilius Lepidus put the emphasis, as we have seen, on Aemilius. Battle fragments have a distinctly Homeric, which is to say literary, ring. Historically minded critics have had to claim that the Caecilii were but surrogates for A. Man-lius Vulso (cos. 178), a younger brother of Fulvius' consular colleague, who badly needed good publicity for his Istrian campaign, and that the emphasis on Aemilius disguised Fulvius' political weakness, but such arguments reflect rather than support the hypothesis of a dutiful poeta cliens.12 We do better to remember the priorities implied by the unbalanced cola in Ennius' own famous claim that Rome's success depended on its traditional values first and then on individuals: "moribus antiquis res stat Romana virisque" ("On its ancient character Rome stands, and on its men," 156). If Skutsch has correctly attributed this line to the famous story of the consul T. Manlius, who executed his brave but impetuous son for disobeying his magisterial command, we have explicit support in the poem for an important principle: national achievement—and the broad sweep of the Annals made it Rome's preeminent national poem—must take precedence over individual gloria. This is the same broadly based sense of patriotism found in the famous epigram preserved by Cicero:

Aspicite, ο cives, senis Enni imaginis formam: Hie vestrum panxit maxima facta patrum.

11 Nobilior's dedication of what was probably a portico to an existing temple is best dated to his censorship in 179, but the issue is complex and problematic. See Skutsch, Annals, 144-46 and 313-14, and Stndia Enniana, 18-29.

12 Badian, "Ennius and His Friends," 185-87, and Martina, QFC 2 (1979) 37-44, for Vulso; for Aemilius, see Martina and Skutsch above, n. 9.


Gaze, fellow citizens, on aged Ennius' form and image: He set to verse your fathers' greatest deeds. (TD 1.34)

Some have thought that the statue in question graced the tomb of the Scipios—perhaps put there on Africanus' instructions—but the call is in any case to Romans as a community. It would have been the Scipios who were appropriating the poet's glory, and not the other way around. Patrum is deliberately and proudly general.13

There is thus no sound internal evidence to support the claim that Ennius' epic advanced partisan causes. The reverse is far more likely: the incorrect and unhelpful label poeta cliens has encouraged tendentious readings that are based less on the evidence than on certain presuppositions about Roman social conventions and insti-tutions in the middle years of the Republic. The historical record, however, is as fragmentary as the literary one and yields no clear idea of how patronage functioned in Ennius' day. The evidence is too late and too diffuse. Letters Cicero wrote to advance the careers of ambitious juniors are quite a different kind of testimony from Horace's poems of thanks to Maecenas; forensic narratives in which Cicero furnishes details of domestic and political relationships reflect a different purpose and a different perspective from the observations of an inquiring outsider like Dionysius of Halicarnassus, Such spotty and diverse images cannot produce an intelligible picture of a social system so delicate that perceptions and realities must always have been difficult to distinguish. This should not surprise us. Even students of modern patronage, with vastly more evidence to evaluate and active informants to interview, continue to wrestle with the conceptual difficulties inherent in the subject. Classicists with only the limited, skewed testimony of a vanished culture before them can hardly hope to do better.14 Yet enough information is available to question some of the assumptions that too readily color discussion of Ennius' career.

1 3 The vocative cives implies Ennius' own citizenship: he is a proud part of the community he glorifies. As discussed in Chap. 1, 16-18, the story of the portrait bust may be based on a fabrication or misunderstanding. The epi-taph is more likely to be genuine. See Courtney, Fragmentary Poets, 42-43, and Jocelyn, ANRW 1.2 (1972) 1023-24.

14 On this problem see the programmatic essays by Silverman, "Patron-age as Myth," 7-19, and Weingrod, "Patronage and Power," 41-51. Both sociologists illustrate their essays with modern Italian examples that will not surprise Latinists. Wallace-Hadrill, "Patronage in Roman Society/' 63-87, attempts to bring their lessons into his discussion of Roman patronage. The references to court poetry and propaganda often found in discussions of Roman literary patronage are in any case anachronistic. See White, Promised Verse, 95-109.


We know that patronage in Republican Rome never involved a rigid set of social roles or the expectation of specific favors bestowed and received, and because no formally established system of obli-gations linked patrons and clients, the Romans never developed a technical vocabulary to describe relations between them. The term cliens itself did not denote either a permanent relationship or a legal status in the sense that, say, liber tus did. Necessarius, amicus, and hospes could all refer to—or disguise—the kind of dependency we conveniently but rather loosely think of as clientela.15 All these social relationships were fluid: clients at any level of society could have multiple patrons at the same time and could transfer loyalties from one to another as conditions, or their own needs, dictated. Nor was the exercise of patronage restricted to relations between aris-tocrats and their social inferiors. Mechanisms for distributing power or controlling the access to power through an exchange of favors operated across, but also within the various levels of Roman society. Ennius' documented associations with Cato and Fulvius and with multiple Cornelii and Sulpicii reflect this fluidity; they are not necessarily signs of shifting loyalties, purchased talent, or the sub-ordination of aesthetic to political considerations. They do not even prove economic dependence, and Ennius' success in the entrepreneurial world of Roman theater may have made such dependence unneces-sary. Patronage, as the sociologists remind us, is in any case a social system of great and subtle complexity, not an aggregate of discrete personal relationships.16

What this means in practical terms is that we cannot use the bare fact that Ennius had many aristocratic acquaintances to support the assumption of gratuities received and specific causes championed in return. To envision Ennius passing from one patron to another and tailoring his verse to the political needs of the moment misconstrues his social position and certainly oversimplifies the mechanisms of Roman patronage. The record of broad social connections that survives is simply the mark of a man making his own way in society. Clientela as commonly understood is therefore not the right term for describing relations between a poet of Ennius' stature and his many admirers. How, then, should we envision the social position of this gifted poet,

15 Sailer, "Patronage and Friendship/' 50-55; Brunt, Fall of the Roman Republic, 392-400; White, Promised Verse, 27-34. Our more rigid sense of clientela as a formal arrangement derives largely from the description at D.H. 2.9-11, itself an artificial and anachronistic construct. See the extended analysis of Brunt, 400-424.

16 Johnson and Dandeker, "Patronage: Relation and System," 219-42, stress the importance of system over individual relationships for under-standing patronage. Such a system does not require quid pro quo exchanges.


whose stage writing provided an independent livelihood impossible for a Vergil or Horace and who was yet on easy terms with a wide range of distinguished public figures? We need a different model for the kind of relationship possible between men like Ennius and Ful-vius, and a famous fragment of the Annals may offer a glimpse of the required alternative.

The scene is probably Cannae. Cn. Servilius Geminus, consul in 217 until the defeat at Lake Trasimene, has just made a speech, perhaps to the Celtiberian deserters from Hannibal.

Haece locutus vocat quocum bene saepe libenter mensam sermonesque suos rerumque suarum consilium partit, magnam quom lassus diei partem fuisset de summis rebus regundis consilio indu foro lato sanctoque senatu; quoi res audacter magnas parvasque iocumque eloqueretur +et cunctat malaque et bona dictu evomeret si qui vellet tutoque locaret; quocum multa volup

gaudia clamque palamque; ingenium quoi nulla malum sententia suadet ut faceret facinus levis aut mala: doctus, fidelis, suavis homo, iucundus, suo contentus, beatus, scitus, secunda loquens in tempore, commodus, verbum paucum, multa tenens antiqua, sepulta vetustas quae facit, et mores veteresque novosque ttenentem multorum veterum leges divomque hominumque prudentem qui dicta loquive tacereve posset: 285 hunc inter pugnas conpellat Servilius sic:

280 iucundus Sk. : facundus Gell

270

275

280

Saying this, he calls upon a man with whom he often at his pleasure shares his table and his conversation and his thoughts on private matters, tired having spent a large part of the day 270 engaged in counsel on the great affairs of state in the forum broad and hallowed senate-house: to whom he spoke quite freely of subjects great and small, and jokes, letting go on matters good and bad to say if he were so inclined and knowing they were held in safety; 275 with whom much pleasure

joys private and in public; whose character no passing evil thought can turn to do an evil deed: a learned, loyal, charming man, delightful, modest, happy, 280 discerning, timely with the fitting word to say, pleasant, of few words, knowing many old-time things which time has buried, and customs old and new, and laws of many ancient gods and men, knowing


when to speak and when to hold his tongue on matters said; 285 this man between the battles Servilius addressed this way:

This passage is among our longest fragments of the Annals and would in any case, despite its textual difficulties and apparent lacuna, be a significant example of Ennius7 descriptive style. It has long claimed special attention, however, because Gellius reports that the antiquary Aelius Stilo had seen Ennius' self-portrait in the pas-sage. Stilo, the teacher of Varro and Cicero, is an esteemed source and presumably had good reason for this identification, though modern scholars sensitive to the vagaries of biographical criticism should be allowed their skepticism.17 A self-portrait need not have been inten-tional to be helpful. It was certainly not literally true. Whether this Servilius was indeed, as Skutsch argues, the consul of 217 or, as Badian would prefer, the consul of 252 and 248, the passage must deal with a time before Ennius7 own appearance on the Roman scene. The poet is at best, if we accept Stilo's testimony, giving literary form to a social relationship congruent with his own circumstances, but that is still enough to help us.

The modest congeniality of this so-called Good Companion is not unparalleled in epic, though the parallels are not poets. Otto Skutsch, with the poeta cliens ever in mind, thought he saw in the passage a Hellenistic epic topos of the dependable social inferior attending a royal patron, but his evidence for such a topos consists only of a verbal parallel in Terence and a snippet of Ptolemaic encomium.1 8 There are other, clearer, and less subservient good companions to be found. We might recall Patroclus comforting a sulk-ing Achilles in his tent (iL 9.189-91), or even the faithful Achates offering solace to a troubled Aeneas ("multa inter sese vario sermone serebant/7 Aen. 6.160). These are hardly servile examples, nor does Servilius' companion lack dignity. Though the mention of table talk in the passage may remind us of Cato's complaint about grassatores,

17 Gell. 12.4: "L. Aelium Stilonem dicere solitum ferunt Q. Ennium de semet ipso haec scripsisse picturamque istam morum et ingenii ipsius Q. Enni factam esse." Stilo's information presumably derives from reports current after Ennius' death. His testimony is defended at length by Badian, "Ennius and His Friends," 180-83, and accepted by Skutsch, Annals, 450, but note the doubts expressed by Jocelyn and Suerbaum in the discussion fol-lowing Badian's article, 206-7. Identification of speaker and context here follows Skutsch, Annals, 447-50. It is likely, but conjectural; contra Badian, 174-77 and 180.

18 Skutsch, Studio, Enniana, 92-94, repeated in the commentary, 450-51. Skutsch's figure of "the king's trusted companion" would equally well fit the confidants discussed by Lattimore, CP 34 (1939) 24-35, who themselves have claims to nobility.


this is not the portrait of a flatterer. Scitus and commodus, which in other circumstances befit the parasite,19 here mean only "discerning" and "pleasant/' while the other adjectives of this favorable context become the catchwords of later generations. Doctus, suavis, and iucundus will come to denote qualities central to the neoteric world of, say, Catullus 50; suo contentus and beatus provide the theme of Horace's first Satire and the essence of his poetic persona.

Nor is Ennius' companion just an after-hours friend. While the companion seems initially to have taken no part in state affairs ("de summis rebus regundis"), his good qualities are evidently valued at all times. We are far from casual conversation. This portrait comes at a dangerous time, and praise is not what Servilius seeks. It is, if any-thing, his lack of servitude that makes the companion so valuable. Knowledge, discretion, and perspective are the relevant qualities, and he is a good listener. This description is flanked by Servilius' own speeches: the companion himself may never have spoken a word. These are the qualities that young Cato perhaps recognized in Sardinia, that Scipio Nasica and Sulpicius Galba found worth culti-vating at home, and that Fulvius Nobilior valued on campaign. The "good companion" could hold his own in such company.

The passage thus has significant resonances within Ennius' own time. It also carries important implications for the development of Roman literature. The relationship it describes between public figure and urbane companion need not imply social equality, but it does suggest a kind of countervailing power. Roman aristocrats provided the subject and much of the audience for early poetry. The poet's claim to strength lay not just in recounting their achievements but in creating a context that declared and confirmed their significance. Epic's narrative scope provided the perfect medium for this venture. Casting a wide range of deeds in the epic mold raised them all to an antique and Homeric grandeur, thus fulfilling aristocratic desires for personal honor. Yet the very sweep of the Annals subordinated indi-vidual achievements to the greater cause of Roman gloria. The inevitable result was a national epic, not a Fulvian or Fabian or Scipionic one, that subsumed the potentially rival claims of poetry and politics. The balance Ennius struck between them insured both the appeal and the integrity of his poem, but however simple such a merger may seem in retrospect, it was new in the second century. Earlier Latin poetry had, like drama, been public and general in its appeal or, like the Bellum Punicum, more impersonally nationalistic. Ennius introduced the first open partnership between aristocratic and

19 "Scitum hercle hominem!" says Parmeno of the parasite Gnatho (Ter. Ειιη. 254); Cicero describes Apronius, a minion of Verres, as "aliis inhumanus ac barbarus, isti uni commodus" (Ver. 3.23).


poetic interests. What features of the cultural climate encouraged him in that endeavor? The evidence is all indirect and anecdotal, but one such anecdote provides a clue to the relationship between aristo-cratic institutions and the circulation of ideas in Ennius7 time.

In the year 181, an odd discovery at the city's edge won a place in Roman history. The fullest account is in Livy, who tells a story along these lines (40.29):

Workmen plowing land on the Janiculum that belonged to a scriba named L. Petillius uncovered two stone chests of inordinate size. Each was sealed with lead and bore an inscription in Latin and Greek, one pur-porting to contain the body of Numa Pompilius, second king of Rome, and the other claiming to contain his books. The first casket proved to be empty, but in the second were found two bundles of books in pristine condition. The first was a set of seven Latin books on pontifical law; the other bundle of seven, in Greek, were Pythagorean treatises. The books began circulating among Lucius' friends and soon came to the attention of the urban praetor, Q. Petillius. A quick reading convinced him that they were harmful to the state religion. He thought they should be burned, but he offered Lucius the right to appeal his decision without prejudice and to petition for their recovery. The scriba took the case to the tribunes of the people, who in turn referred the matter to the Senate. The Senate held a hearing that confirmed the praetor's decision, but it offered financial compensation to the owner. Lucius refused this compensation, and the books were publicly burned in the Comitium.

The episode, though surely not the library, is authentic, but it is also strange: neither the discovery nor its aftermath can be quite what it seems. Scriba and praetor sound like client and patron, and there is good reason to suspect that they were working in collusion. Livy hints at this relationship not just by the shared gentile name—the scriba is called Cn. Terentius by other, older sources—but by the fact of Lucius' debt to the praetor for his position.20 He was presumably Petillius' agent, and not just for the act of discovery. The two men jointly ar-ranged to bring the matter before the Senate, and this suggests that they were acting not in hope of reversing the praetor's decision but of confirming and even publicizing it. The action seems planned to insure public attention for Lucius' discovery and a public burning of these books. It must have been a staged demonstration . . . but of what?

20 Liv. 40.29.10: "et erat familiaris usus, quod scribam eum quaestor Q. Petillius in decuriam legerat." Cassius Hemina ap. Plin. NH 13.27.84 and Varro ap. Aug. CD 7.34.5 call the scriba Terentius. Livy's version could have arisen from an ambiguity like "in agro scribae Petilii" in a source. The an-cient evidence is conveniently gathered by Garbarino, Roma e la filosofia greca, vol 1, 64-69; discussion in vol. 2, 244-56. There are now major treat-ments by Rosen, Chiron 15 (1985) 65-90; Pailler, Bacchanalia, 623-67; Gruen, Studies, 163-70.


The Romans' notorious antipathy to Greek influences is often re-called, and the affair has been associated with other, more famous senatorial actions such as the banning of the Bacchic cult in 186 and the expulsion of philosophers and rhetoricians from the city in 161.21

The target, according to this interpretation, is arcane Greek philosophy. Further support for this argument comes from the fact that the praetor Q. Petillius was apparently one of the two Fetillii who, as tribunes in 187, queried the financial conduct of L. Scipio and initiated the Scipionic trials of the 180s. The Petillii were widely viewed at the time as agents of Cato, and Cato's personal hostility to things Greek is seen at work behind the scene here as well. This explanation is neat and simple, but the motivation does not hold up very well under scrutiny.

For one thing, Cato himself can no longer be so easily typed: our perception of his anti-Hellenism has been radically revised.22 Evi-dence of his hostility to contemporary Greeks remains. He certainly called them "a thoroughly worthless and unbridled race" ("nequis-simum et indocile genus"), and his bitter injunction against Greek doctors is notorious. He hastened the philosopher Carneades' depar-ture from Rome in 155, and he ridiculed the Hellenic posturing of such contemporaries as Postumius Albinus. Yet Cato also had another side. His public comments included witty and informed allusions to Greek literature, and his formal writings show the influence of Xenophon and Demosthenes. He kept a Greek teacher in his household. His son had a Greek education and eventually married a daughter of Aemilius Paullus, whose own sympathies for Greek culture were open and sincere. Such complexities and contradictions in attitude and behavior reflect the intellect of a lively, educated, and astute public figure who lived out a long life in exciting and challenging times. Cato was too complex a man to be readily categorized. We can no longer associate him with a simple and reactionary anti-Hellenism. While his involvement in the affair of Numa's books certainly remains a possibility, anti-Hellenism was not likely to be its cause.

Nor was Pythagoreanism a particularly suitable target for anti-Greek sentiment. Ennius' own use of Pythagorean motifs in his pro-emium and Fulvius' attention to the calendar, a distinctly Numan preoccupation, testify to the acceptability of such interests and their

2 1 Garbarino, Roma e la filosofia greca, vol. 2, 255-56; Scullard, Roman Politics, 171-72; Pailler, Bacchanalia, 669-77. Prowse, G&R 11 (1964) 36-42, fails to see that the entire episode, not just the initial discovery of the books, was arranged. Pailler pays scant attention to the politics of the incident.

2 2 Astin, Cato, 157-81, offers detailed discussion of the evidence. On the complex intellectual climate of the period, see Gruen, Hellenistic World, vol. 1, 250-72; MacMullen, Historia 40 (1991) 419-38.


currency in Roman intellectual life. The figures of Numa and Pythag-oras fascinated the Romans, and association of the two was perhaps inevitable. Numa was a semi-mystical, learned figure whom Romans revered as a second founder. He gave Rome its laws and its religion, its calendar and its sense of community. As Livy would eventually put it, Numa brought Rome from the status of armed camp to true city (1.21.2). The persistent though chronologically impossible claim that he was a student of Pythagoras identified the cultural development that Numa symbolized with the intellectual traditions of southern Italy, where Pythagoreanism flourished. This association gave the Sabine Numa a kind of intellectual pedigree. He could thus more easily be called the wisest of Romans, just as Romans acknowledged Pythagoras as the wisest of Greeks. As early as the third century, a statue of Pythagoras with that very label was erected in the Comitium, presumably near the scene of the later book burning, and Pythagorean ideas had long circulated freely. Cicero thought he saw their influence as early as a carmen of Appius Claudius Caecus, an unlikely but, for him, credible possibility.23 Distinguished Roman families who proudly traced their lineage back to Numa did not shun the Pythagorean connection. Aemilii and Pinarii bore the cognomen Mamercinus and claimed descent from Numa's son Mamercus, himself allegedly named for a son of Pythagoras. By the time of the Hannibalic War, the Pomponii and Calpurnii were making similar claims for themselves.24 There was luster in the notion, however unstable its foundation.

It is thus particularly significant that all of Numa's books, not just the Greek ones, were burned. The inevitable and convenient variant, that the pontifical books were saved and just the philosophical ones destroyed, appears only in the imperial moralist Valerius Maximus (1.1.12). Republican sources agree on the complete destruction of Numa's purported legacy, and none regrets the loss. Why not? The assumption of an anti-Greek demonstration explains neither the association with Numa nor the fate of the Latin books. Neither Greek philosophy in general nor Pythagoreanism in particular was the likely target of this demonstration. The key details lie else-

2 3 Plin. NH 34.12.26 and Plut. Numa 8.20 (the statue); Cic. TD 4.2.4 (Αρ. Claudius). On Roman Pythagoreanism, see Garbarino, Roma e la filosofia greca, vol. 2, 221-30, and Burkert, Philologue 105 (1961) 240-46. The asso-ciation of Numa and Pythagoras is treated in detail by Panitschek, GB 17 (1990) 49-65.

24 Gabba, "Considerazioni sulla tradizione letteraria," 159-61. The associ-ation of Numa with Pythagoras persisted into the first century despite its inherent and oft-exposed impossibility. See Cic. Rep. 2.28; de Or. 2.154; D.H. 2.59.1-2; Liv. 1.18.1-3.


where, but recognizing their significance requires a broader cultural context.

The affair of Numa's books took place toward the end of a decade in which the Senate was continually asserting its authority to adjudicate rival claims of public obligation and personal prerogative. Two other notorious events of the time may serve to illustrate the pervasiveness of the problem. The controversy of 187 over L. Scipio's account books represents the political dimension of the issue. As best we can tell, the facts were these: as consul in 190, Scipio won a. decisive battle at Magnesia over the Syrian king Antiochus. He acquired not only the cognomen Asiaticus but also 500 talents from Antiochus as the first installment on a proposed war indemnity of 15,000. Scipio paid his troops with the money and left responsibility for the rest to his successor, Cn. Manlius Vulso. Back in Rome, the Petillii demanded an audit of his expenditure. Scipio refused to make his accounts public, for generals traditionally dispensed booty at their discretion. Lucius' more charismatic and irrepressible brother then intervened. In what became famous examples of histrionic self-importance, Scipio Africanus publicly questioned the tribunes' right to query the disposition of 500 talents from a man who had brought the state 15,000. He answered a similar question before the Senate by sending for his brother's account books and destroying them on the spot.25 Such conduct was hardly a lasting defense. Further action against the two Scipiones was the almost inevitable consequence, and not simply because such arrogance inflamed their political foes. The Petillii had actually raised a legitimate question.

The distinction between military booty under a general's personal control and state funds, for which he was responsible to the Senate, demanded debate and clarification. It was a recurring issue bigger even than Lucius and Publius Scipio or the events of a single consulship. The suspicion that Cato lay behind the Petillii's action is true to his position in a long series of debates over this question. In 190, Cato's attack on M.' Acilius Glabrio had claimed, among other things, that gold and silver objects seen among the spoils of Thermo-pylae were not displayed in Glabrio's triumph for that victory. The subsequent furor compelled Glabrio to withdraw his candidacy for the censorship. Other speeches from the 180s dealt with the distribution of military booty and the display of enemy spoils for

2 5 A narrative account is in Li v. 38.50-60, with anecdotal versions in Polyb. 23.14.7-11, Gell. 4.18.7-12, Diod. 29.21, Val. Max. 3.7.1e. The actual sequence of events remains obscure. See Scullard, Roman Politics, 290-303, and Balsdon, Historia 21 (1972) 224-34.


private ends.26 These controversies reflect a growing tension between personal prerogatives and civic responsibilities, a tension inevitable in a social system that saw the exercise of public office—especially its military exercise—as the primary measure of personal achievement. As Cato himself would say, "the demands of right, law, liberty, and the state should be exercised in common: glory and honor as each man has acquired them for himself."27 The balance between public and personal gain was always precarious, and the Senate alone had the power to maintain it. Yet political wrangles over triumphs and account books were not the only signs of collision between public and private interests. There was also a social dimen-sion to the problem, which is reflected in a second famous incident of the time.

The Bacchanalian affair of 186 was not a purely aristocratic matter, nor was it primarily a challenge to religious freedom or the acceptability of Greek cultural influences.28 The consul who assumed primary responsibility for the investigation, Sp. Postumius Albinus, came from a family of notorious philhellenes. The Senate treated the cult as a criminal conspiracy, and the famous decree drafted to curb its influence was careful to avoid speaking ill of the god or—at least in theory—forbidding his worship. The problem in 186 was largely social, for the cult of Dionysus was not a recent import. It had been in Italy long enough to take on the characteristic organization of a Roman collegium and to recruit a membership cutting across the normal boundaries of wealth, occupation, civil status, and back-ground.29 It was therefore especially difficult to suppress, which is why the official investigation took the better part of a year and claimed the attention of both consuls, a rare interruption in the long sequence of consular commands abroad. Even after promulgation of the Senatus consultum de Bacchanalibus, sporadic outbreaks of Bacchist

26 Cato, "In M/ Acilium Glabrionem" (ORF3 66; cf. Liv. 37.57.13-18); "Uti praeda in publicum referatur" (ORF3 98); "De praeda militibus dividenda" (ORF3 224-26); "Ne spolia figerentur nisi de hoste capta" (ORF3 97), and cf. references to his own conduct in fr. 173 and 203. See in general, Shatzman, Historia 21 (1972) 188-94.

27 ORF3 fr. 58: "iure, lege, libertate, re publica communiter uti oportet: gloria atque honore, quomodo sibi quisque struxit." The context is unknown. On the aristocratic attitude to conquest and praise, see Harris, War and Imperialism, 10-41; Gruen, Studies, 69-72.

28 Pailler, Bacchanalia, offers an exhaustive but diffuse treatment of the affair. Rather closer to the present point is Gruen, Studies, 34-78.

29 North, PCPS 25 (1979) 85-103; Gruen, Studies, 54-56. The wonderfully lurid story in Liv. 39.8-19 must be balanced by the facts implicit in the SC and other documentary evidence. See Cova, Athenaeum 52 (1974) 82-109, and Pailler, Bacchanalia, 333-98.


activity recurred to plague authorities. Repressive measures con-tinued into the late 180s, in Apulia to the south and Etruria to the north, as well as at Rome and in Campania.

What finally roused the Senate to action against the cult was its combination of complex organization and secrecy. This explains the charge of coniuratio. Romans tolerated foreign ideas best by assimilating them. When the Asian cult of the Magna Mater reached Rome in 205, for example, it came in response to an official invitation and enjoyed an elaborately orchestrated reception. Because the ecstatic worship of Cybele was as non-Roman as could be, Rome civi-lized her cult, building a temple and instituting formal games in her honor even as it curtailed her rituals and limited access to her priesthood. In the process, Cybele's cult was made open and official and thus safe. Bacchist worship remained private and secret, and that made it potentially dangerous. The Senate was acting to curtail a widespread and excessively invisible activity. As in the affair of L. Scipio, the state would not tolerate a source of power beyond its supervision.

This concern with the growth of private power in Roman society provides the context for that curious discovery on the Janiculum. The affair was engineered to put a set of books into private circulation and to bring the fact of that circulation to the Senate's notice. The praetor Petillius could presumably have dealt with the matter himself: all the Senate really did was take his deposition and endorse his judgment. By appealing his original decision, scriba and praetor were insuring not just a more public end for the books but the highest confirmation of the state's right to take them out of private hands. The full sequence of events from private discovery to public destruction proclaimed the Senate's ultimate authority over the circulation of potent ideas. That is itself significant for the development of literature, and there is an equally significant corol-lary. For ideas to circulate freely, the circulation needed to be open. Official encouragement, or at least the encouragement of those who were also officials, was therefore to be welcomed, not avoided. This brings us back to the poem of Ennius and to the cultural interests of Fulvius Nobilior. Nothing implicates them directly in the affair of Numa's books, nor need we think that either of them consciously read our own lessons into it.30 Yet the episode reflects conditions and

3 0 Herrmann, Latomus 5 (1946) 87-90, actually claims Ennius as the author of the (forged) books. He has convinced few, though Pailler, Bacchanalia, 695-98, would like to implicate Fulvius and Ennius in the affair. Ennius' Pythagoreanism cannot be taken very far: it was probably more a poetic than philosophic or religious stance. See Timpanaro, RFIC 119 (1991) 28-29.


attitudes in Roman society which certainly touched them and which can help explain their activities.

To be successful, literature at Rome required a public dimension, and for books to circulate widely, they had to reflect as well as guide the goals and sensibilities of their readers. Literature needed the social sanction that came from the expression of shared interests and communal feelings. Ennius7 work gained this acceptance by appealing to the Romans7 own developing sense of community. This is not the same as saying that Ennius pandered to aristocratic tastes and preoccupations. As a Roman of independent means and great civic pride, he shared those interests, and in the process of sharing them he became their popularizer. Yet the Annals was no isolated monu-ment. It gave literary form to a cultural development that also found expression in the public works of at least one public figure.

Fulvius Nobilior was himself no boorish gloriosus: he apparently took a genuine interest in antiquarian research. As late as the sixth century A.D., the Greek writer Ioannes Lydus knew the work of a "Fulvius77 that claimed Numa's authority for a discussion of astral phenomena. No other Fulvius has been advanced as the likely author, and such an interest accords well with Macrobius7 testimony that Nobilior compiled a set of fasti and deposited them in the Temple of Hercules of the Muses. Whether Fulvius was the actual author or simply the official sponsor of that work, he was evidently keen to associate his name with the project.31 Still more significant is the public posture revealed by the temple housing this document. It too owed something to Fulvius, for it was the tangible symbol of his effort to integrate and institutionalize a Roman cult of the Muses with older Italic traditions. There was already a temple of Hercules the Great Guardian on the south end of the Campus Martius near the Circus Flaminius. During his censorship of 179, Fulvius apparently added the portico to this temple that changed forever its character in the popular mind. He introduced two cult items. One was a statue of Hercules as protector of the Muses that had come to Rome with the spoils from Ambracia. The association of Hercules and the Muses is more apt than may at first appear. Hercules7 role as Musagetes had a considerable history, perhaps fostered under Pythagorean influence. The characterization is found in Greece proper and in Greek cities of southern Italy, especially Pythagoras' city of Croton. The particular statue brought back by Fulvius may in origin have been a choregic

31 Lydus, De ostentis 16a (no. 300 in Garbarino, Roma e le filosofia greca, vol. 1, 143-44), concluding ταΰτα μεν ούν Φούλβιός φ[ησιν, έκ τ]ών του Νουμά ί[στορήσας]; discussion in Garbarino, vol. 2, 256-58. Macr. 1.12.16: "nam Fulvius Nobilior in fastis, quos in aede Herculis Musarum posuit . . . dicit," but dicit should not automatically be equated with authorship.


monument.32 Hercules was also, like Odysseus, one of those Greek figures whose legend incorporated strong Italian associations, and this made him an especially appropriate conduit for the importation of Greek ideas. With Hercules as their protector, the Muses were assured a place at Rome

The second object displayed had native connotations. Beyond the Porta Capena and just south of the Caelian hill was a grove asso-ciated with Numa and identified in Roman legend as the place where the king met with his mysterious adviser, the nymph Egeria, The grove featured a perennial spring and, in early times, a small monument to the Camenae, those water spirits whom Andronicus and Naevius had enlisted as their own divine patrons. When this shrine was eventually struck by lightning, it was relocated in the Temple of Honor and Courage. Fulvius took the opportunity that his building campaign provided to move this aediculum Camenarum to his own expanded temple of Hercules Musarum.33 Camenae and Muses were brought under one roof. The newly styled temple thus gave public expression and public sanction to the same cultural synthesis that Ennius was developing through his epic. That was surely no accident, but neither is it a matter of cause and effect, of mutual flattery, or of self-congratulation. Ennius did not end Annals 15 with the dedication of this temple simply to flatter Fulvius any more than Fulvius dedi-cated the temple to honor Ennius. Each action had its own rationale and its own integrity. These actions shared roots in a broader cultural development at Rome, one that simultaneously nourished Ennius' artistry and awakened Fulvius' interest in artistic endeavors. Epic could thus treat both current events and larger issues of national identity without being, in a narrow or partisan sense, political. Poetry and public affairs had simply found a common ground, though that discovery marks a major turning point in Roman literary history.

Earlier Roman literature had been nurtured by the openness and wealth of Roman society, but it also stood apart from the sources of power in the state. Drama, as an entertainment incorporated into the public festivals, had begun with official encouragement and had always required a kind of license to perform, but the partnership be-tween the theatrical community and its aristocratic sponsors lay only on the surface.34 Theater scripts were not conceived as literature—

32 Marabini Moevs, BdA 12 (1981) 1-58; Skutsch, Annals, 144-46. 33 Serv. ad Aen. 1.8 (the aediculum); for the grove, Liv. 1.21.3, and Ogilvie,

Commentary on Livy, 102-03. For the parallel interests of Fulvius and Ennius, see Gruen, Studies, 117-18, with further references there.

34 Gruen, Culture and National Identity, 188-97, rightly challenges the oft-repeated assertion that theatrical performances were crucial to the careers of aspiring politicians.


that status came only later and in retrospect—while the com-mercialism of Roman drama in the third century kept the dramatists themselves on the margins of Roman society. Their collegium was indeed conceived as a mark of honor, but it was still essentially a tradesmen's guild. Their art was similarly marginalized. The mythological postures of tragedy offered bold and impressive lan-guage but remained rooted in the popular taste for spectacle and melodrama; comedy laughed loudly though obliquely at the excesses of its day, but did so best by playing on the Greek conventions of its genre. Because neither form developed a strong intellectual base, drama failed to bring poetry into the mainstream of Roman culture. Epic did. Ennius showed not just how Latin could be made into high poetry but also how epic could become the literary vehicle for a growing Roman sense of cultural identity. Naevius' Bellum Punicum was the pioneer, but it had been a work of limited scope and, if only for technical reasons, of limited success. The Annals completed the process of establishing a truly Roman epic tradition. Its Homeric echoes gave special significance to Roman achievements, while Ennius' ability to carry his poem from mythical origins to con-temporary events brought a new coherence to the Roman record and a new sense of its greatness. With Ennius, epic found both its form and its public. The immediate acceptance of his experiment had two major consequences.

The first of these involved the people who were or became Rome's poets. Roman literary figures began coming in from the margins of Roman society. Respectable birth, independent means, and immense talent had enabled Ennius to move easily in high circles. He was the first poet at Rome to do so, but his example did not remain unique for long. He died in 169. By 160, Terence would publicly claim the friend-ship and support of Roman aristocrats. His effort to adapt comedy to the changing tastes of that aristocracy was, as we have already seen, a failure. The palliata genre proved unequal to the task, and Terence's hellenizing style engendered no successors. Yet his own progress up the social ladder furthered the process of poetry's grow-ing social acceptance. He, like Ennius, was able to amass property, and his daughter was thought to have married an equestrian.35

Within a generation, native Romans were writing poetry of note, and poetry itself ceased to be the occupation of social outsiders. The greatest poet of the mid-second century, the satirist Lucilius, was an eques, and an ever-increasing proportion of Roman poets enjoyed the status of knights and even senators.36 By the end of the century, a collegium poetarum apparently separate from the guild honoring the

35 Ad. 15-21; Vita Ter. 99. See Gilula, SCI 8/9 (1989) 74-78. 36 White, Promised Verse, 5-12.


poet-actor Andronicus was meeting in Fulvius' Temple of the Muses with the aristocrat Julius Caesar Strabo, himself a tragedian of some note, in attendance/7 while the hellenized epigrams of Lutatius Catulus (cos. 102) were beginning to suggest the new cultural climate that would in time nourish the literary interests of, among so many others, Varro and Cicero. This growing interest in poetry, an interest that parallels the aristocracy's own developing achievements in history and oratory, first becomes manifest with Ennius' social acceptance and artistic renown.

A second consequence of this enhanced status was less uniformly wholesome. The open circulation of epic poetry and recognition of its appeal led to an increasing appreciation of its utility. That could be edifying. Cicero would eventually put the positive side of things well when, in defending the poet Archias, he would call poetry not only pleasant but also useful to the man of affairs.

Quam multas nobis imagines non solum ad intuendum verum etiam ad imitandum fortissimorum virorum expressas scriptores Graeci et Latini reliquerunt, Quas ego mihi in administranda re publica proponens, animum et mentem meam ipsa cogitatione hominum excellentium conformabam.

How many examples of the very finest men have Greek and Latin writers provided and left not only for us to study but also to imitate. So I myself in conducting public business consider these examples and mold my character and thought by the very contemplation of excellent men.

(Arch. 14)

He will claim Scipio, Laelius, Furius, and even Cato as adherents to this view (16) and Ennius as the poet who first convinced them of it (22). It is a noble sentiment, and perhaps even a true one, but such utility could also have a darker side.

Praise is not so very far from flattery. Cicero is equally unabashed about this fact because he does not see a problem in it. Everyone is quite naturally out for glory, he claims. The more distinguished the man, the keener the desire: "We are all motivated by a desire for praise, and the better each man is, the more he is attracted by glory" ("trahimur omnes studio laudis, et optimus quisque maxime gloria ducitur," 26). Cato had voiced a similar sentiment (fr. 58), and it even occurs in Ennius ("omnes mortales sese laudarier optant," 574). That love of glory, however, could pose a serious challenge to the integrity of epic. Ennius managed to avoid embracing partisan causes, and even

37 Val. Max. 3.7.11; Ascori. 1.1. The change of venue from Minerva to the Muses and a membership that seems not to have included actors suggest a separate organization, but nothing about this collegium is certain. See Hors-fall, BICS 23 (1976) 81-86, and Gruen, Studies, 89-90.


Archias found ways to avoid awkward commissions, as Cicero would learn to his dismay. Yet Cicero and his audience clearly expected epic's poets to further the interests of epic's patrons, and modern critics are only following that lead when they project the career of an Archias back into the time of Ennius and Fulvius. We have seen how misleading and unhelpful that projection can be. Archias certainly gained fame and fortune by traveling through the Greco-Roman world and praising—in Greek—the deeds of successive public figures.38

Ennius, however, was not an itinerant poet of this kind. As a teacher and dramatist, he did not have to live off the proceeds of praise. He was not a poet for hire in the sense that Archias was, but something besides a disparity in talent and ethnic background accounts for the difference between them.

Though the fragments of Republican epic after Ennius are especially meager, we can nevertheless see how the success of his Annals fostered an idea of epic that eventually, and perhaps inevitably, led to the failure of the last famous epic of this period, Cicero's own ill-starred poem on his consulship. The genre begun by a freedman from Tarentum reached, as we shall now see, its next important milestone with the novus homo from Arpinum.

3 8 Cic. Arch. 4-6. Cf. Quinn, ANRW 2.30.1 (1982) 122-24, and the testi-monia gathered in SEI, no. 194-200. No Latin poet has left the record of a comparable career.

6 Ciceronian Sirens

What little we know of Ennius' immediate successors provides little help for tracing the details of epic's development, but there is no doubt of its growing popularity. By the later second century, a time when revivers of Plautus' Casina would lament the demise of drama's flos poetarum (19), epic verse began appearing with some frequency at Rome. Accius, that man of many talents and interests, found time to write Annales, though the meager remains of his poem suggest myth and festival rather than history. A poet named Hostius, exact dates and connections unknown, wrote a Bellum Histricum that probably de-scribed a campaign and triumph in 129 by the consul C. Sempronius Tuditanus. Cn. Matius and the shadowy Ninnius Crassus both trans -lated the Iliad into Latin hexameters; a hexameter Odyssey, which Priscian would eventually confuse with Andronicus' poem, probably also dates from this period. The consul Lutatius Catulus had a friend named Furius of Antium, who wrote an epic that later grammarians hated: its single, six-line fragment is full of neologisms that may pre-figure neoteric innovations. And all this came well before Varro of Atax and Furius (Bibaculus?) celebrated Caesar's campaigns in epic verse and Volusius penned the notorious Annales whose cacata carta would arouse Catullus' enthusiastic scorn (36, 95).1

Though the aims and expectations of this motley group must remain unclear, their fragments nevertheless preserve sure signs of the continuing tradition within which they worked. The subject, vocabulary, and very sound of a line like Hostius' "percutit atque

1 Hostius' poem is cited largely by Servius and Macrobius for its echoes in Vergil; Accius comes to us via the grammarians Priscian and Festus. For Hostius and Furius, see Bardon, La littérature latine inconnue, 178-81, and Wigodsky, Vergil and Early Latin Poetry, 98-101; for Matius and Ninnius, Bardon, 161; for the second Odyssey, Mariotti, Livio Andronico, 56-58; for Accius, Leo, Geschichte der römischen Literatur, 391; for the second Furius, Bardon, 349-50, and Courtney, Fragmentary Poets, 199-200. The panegyric turn first noticeable in later Republican epic is discussed by White, Prom-ised Verse, 78-82.

135


hastam pilaris prae pondéré frangit" ("he struck and, pressing down the spear, broke it with his weight," fr. 1) produce a distinctly Ennian ring. The diction and layered structure of his fourth fragment may, with a little stretching, recall Naevius.

dia Minerva <simul>, simul autem invictus Apollo arquitenens Latonius . . .

Divine Minerva now, and now invincible Apollo, bow-bearing son of Leto . . .

We saw a similar effect in the Bellum Punicum.

dein pollens sagittis inclutus arquitenens sanctus love prognatus Pythius Apollo

then commander of arrows, famous, bow-bearing, blessed son of Jupiter, Pythian Apollo (fr. 20)

It is hard to read Accius' line, "fraxinus fixa ferox infensa infinditur ossis" ("the firm fierce ash aimed is splintered to the heart," fr. 4), without recalling the trees of Ennius' funeral pyre (177, cf. 620). Even that seemingly innovative fragment of Furius is full of traditional effects: "increscunt animi, virescit volnere virtus" ("courage grows, wounds revive their valor") revels in the kind of sound play, etymol-ogical figure, and colon arrangement popularized by his Saturnian predecessors. Epic diction had clearly achieved a stability that gives these scattered remnants an air of consistency, if not cliche, and the limited range of their effects may also in part explain their failure to maintain a readership. Certainly none of these poets came close to rivaling Ennius' achievement, and nothing more substantial can be said of the genre they represent until we come to Cicero, when things get interesting once more.

Cicero's habit of self-promotion does not often win him friends, but it has preserved his place in our history. He became faute de mieux both a translator and a composer of epics, and thereby hangs a tale of some significance. The translations date from the mid-forties B.C., when Cicero's public career had left the rails. He consoled himself for his imposed political inactivity during Caesar's dictatorship by writing philosophical tracts that often illustrate their arguments with citations from Homer and the Attic tragedians. The decorum for such formal discourse required translation rather than quotation in Greek, and Cicero—ever the deft, clever, and even fidus interpres when he chose—made the most of these opportunities. The result is a set of Latin excerpts true to the form and dedicated to the content of

Ciceronian Sirens 137

their originals.2 Yet they are also highly conventional. While Cicero's original compositions may strain against the traditions of Latin epic, the translations—more fluent and more faithful than Andronicus'—pay close attention to Roman norms of epic style and content. They therefore offer insight into how Romans read and understood Homer in the waning days of the Republic and how they expected Latin epic verse to sound.

In the last book of his treatise De finibus bonorum et malorum, for example, Cicero argues that the human passion for knowledge is in-satiable and extraordinary. Children and even adults will brave inconvenience, discomfort, and danger in their eagerness to learn, and they find compensation for those risks in the pleasure that comes from knowing. Proof of a kind may be drawn from Homer: the Sirens lure sailors to destruction not simply by the sweetness of their song but with the promise of superior knowledge. To dramatize the point, Cicero offers his own version of the Sirens' appeal to Odysseus in Odyssey 12.

ο decus Argolicum, quin puppim flectis, Ulixes, auribus ut nostros possis agnoscere cantus? nam nemo haec unquam est transvectus caerula cursu, quin prius adstiterit vocum dulcedine captus, post, variis avido satiatus pectore musis, doctior ad patrias lapsus pervenerit oras. nos grave certamen belli clademque tenemus, Graecia quam Troiae divino numine vexit, omniaque e latis rerum vestigia terris.

Glory of the Argives, why not turn your helm, Ulysses, so you may appreciate our song with your ears? No one has ever traveled this sea-green course without first stopping, caught by the sweetness of our voices. Afterwards, his eager heart filled with our varied songs, he continues on, a wiser man, to his ancestral shores. We know the grim struggle of war and disaster which Greece with divine aid visited on Troy, and the signs of all things on the wide earth. {Fin. 5.49 = fr. 30)

This passage is a nice example of his skill. It is a close and careful bit of work, as a glance at the original will confirm.

2 For the origin and use of quotations in these tracts, see Jocelyn, YCS 23 (1973) 79-98. The Roman aversion to literal translation is sometimes over-stated. See Brock, GRBS 20 (1979) 69-87, and Traîna, Vortit barbare, 55-61. For convenience, Cicero's verse is cited from Büchner's revision of Morel, Fragmenta poetarum latinorum, but Courtney, Fragmentary Poets, 152-78, which does not include the translations, is now indispensable.


δεύρ' αγ' ιών, πολι>αιν' Όδυσεύ, μέγα κΰδος 'Αχαιών, νήα κατάστησον, 'ίνα νωϊτέρην οπ' ακούσης. ού γάρ πώ τις τηδε παρήλασε νηϊ μελαίνη, πρίν γ' ήμέων μελίγηρυν από στομάτων οπ' άκούσαι, άλλ' ο γε τερψάμενος νειται και πλείονα εΐδώς. ϊδμεν γάρ τοι πάνθ\ οσ' ένι Τροίη εύρενρ Άργεΐοι Τρώες τε θεών ίότητι μόγησαν, ιδμεν δ' οσσα γένηται έπί χθονι πουλυβοτείρη.

Come this way, illustrious Odysseus, great glory of Achaeans, halt your ship to hear our pair of voices. Never has anyone sailed by here in his black ship until he has heard the honey-toned sound of our voices, and then continues on delighted and better informed. We know everything that before wide Troy Argives and Trojans endured by divine will; we know all that happens upon all-nourishing earth. (12.184-91)

Cicero's prosody is up to date. He avoids the spondaic openings of En-nius. He prefers end-stopped lines to enjambments, and the rhythms are those of his contemporaries Lucretius and Catullus.3 Much also remains, however, to recall Andronicus. The epic style and decorum established by the early Latin Odussia have evidently survived the intervening centuries and the shift from Saturnian to hexameter verse.

Like Andronicus, Cicero has the confidence and the skill to render Homeric epithets gracefully (e.g., ο decus Argolicum - μέγα κΰδος Αχαιών) or to omit them entirely (e.g., πολύαινος). He too can find comfortable equivalents for Greek idiom (e.g., doctior - πλείονα είδώς) and can maintain a close line-by-line correspondence. Only the construction of satiatus forces an extra line into the sequence. Cicero is even sensitive, though perhaps subconsciously, to a Greek nuance. The claim of Homer's Sirens to universal knowledge is couched in the language of the Muses, and while Cicero cannot reconstruct the stylistic features that create this association for a Greek audience, he substitutes the word itself in his fifth line.4 He also follows Andronicus in substituting a more formal Latin diction for a characteristic Homeric colloquialism. Just as Andronicus omitted expressions like αλλ' αγε (fr. 7) and ού γαρ εγώ γέ τί φημι (fr. 18) from his translation, so Cicero here ignores δεύρ' αγ' ιών. He also

3 For the details of Ciceronian prosody, see Ewbank, The Poems of Cicero, 40-71, and the cogent summary of Courtney, Fragmentary Poets, 149-52.

4 Musis should not, however, be capitalized; cf. Lucr. 4.589: "fistula silvestrem ne cesset fundere musam." For the association of Sirens and Muses in Homer, see Pucci, Arethusa 12 (1979) 127, noting IL 2.484 and the repeated ϊδμεν of Hes. Th. 27-28.


manipulates Homeric ornament to create a favorite Latin effect: the color word has moved from Homer's ship to Cicero's sea, making possible the kind of alliterative verse end that Latin favors.

This tendency toward a greater formality of tone and structure is inescapable in these Latin versions of Homer. The change is even more pronounced in his rendering of a passage from Iliad 19 where Odysseus urges Achilles to eat before charging into battle (226-29).

λίην γαρ πολλοί και έπήτριμοι ήματα πάντα πίπτουσιν • πότε κέν τις άναπνεύσειε πόνοιο; άλλα χρή τον μεν καταθάπτειν ος κε θάνησι νηλέα θυμδν έχοντας, έπ' ήματι δακρύσαντας.

Too many men fall every day and thickly: when could anyone find respite from grief? We must bury the man who has died, staying hard of heart when we have wept for a day.

Cicero embedded a translation of these lines into the discussion at Tusculan Disputations 3.65 (fr. 29).

namque nimis multos atque omni luce cadentis cernimus, ut nemo possit maerore vacare. quo magis est aequum tumulis mandare peremptos firmo animo et luctum lacrimis finire diurnis.

We see too many men falling every day, so that no one can leave off grieving. It is therefore right to commit the departed to their tombs with a firm spirit and to end grief with a day's tears.

The line correspondences are again close and the translation deft, but the feel of the Latin verses is very different from their original. The loose and easy style of Odysseus' speech becomes much stiffer in Latin. A formal result clause replaces the syntactically vague link between Homer's first two sentences, and "quo magis est aequum" makes explicit the logical connection of his άλλα χρή. Diction too is of a different order. Cicero can simply be more precise: maeror is as much an interpretation as a translation of πόνος. Other changes are more complex. Cicero's omni luce and lacrimis diurnis are not only more elaborate locutions than their originals, but their very difference from each other avoids the kind of verbal repetition that is so char-acteristic of Homeric diction. Cicero has gone to some lengths to intro-duce a distinctly Roman copia verborum.

The plasticity of Homeric syntax is not always so firmly handled. From a lost work called De gloria, for example, comes a fragment in


Gellius that quotes Hector's words to Ajax when proposing the duel of Iliad 7.

"άνδρός μεν τόδε σήμα πάλαι κατατεθνηώτος ον ποτ' άριστεύοντα κατέκτανε φαίδιμος "Εκτωρ." ώς ποτέ τις έρέει· τό δ'έμον κλέος οΰ ποτ' όλεΐται.

"This is the tomb of a man long dead whom once fighting bravely shining Hector killed/'

So one will say, and my fame will never perish. (89-91)

The best Cicero could do was apparently this: "hie situs est vitae iam pridem lumina linquens qui quondam Hectoreo perculsus concidit ense." fabitur haec aliquis, mea semper gloria vivet.

"Here lies one leaving the light of life long since, who once fell, struck by Hector's sword/' Someone will say this. My glory will live always. (fr. 25 = Gell. 15.6.3)

We may well admire the structural fidelity of the first two lines, which capture Homer's sequence of main and relative clauses. The connotation, however, is a little different. Hector was imagining the words of a passing traveler upon seeing the tomb of his victim. Cicero, with no context to restrain him, adopts the epitaph formula "hie situs est" and the typically Roman solemnities of "vitae . . . lumina linquens" and "Hectoreo ense." In doing so, he absolves Hector of the boastfulness that so annoyed scholiasts, and he mutes the emphasis on aristeia that is so typical of his original.5 Yet other changes are not so happy. He is forced into asyndeton in the third line, where the Greek sequence is smoother, and while fabitur certainly captures both the sense and the rarity of the epic future έρέει, it is a painfully odd word for so emphatic a position. The Latinity of "iam pridem lin-quens" ("leaving some time ago"?) may also raise eyebrows. An-dronicus had preserved the coincident force of an aorist participle by turning it into "simul ac lacrimas detersit" (fr. 17, translating δάκρυ' όμορξάμενος), but Cicero is working his adverb even harder to compensate for the lack of a perfect active participle. Grammar here seems to strain under the burden of meaning.

And not only grammar. Diction too struggles with only limited success against the austerity of the Homeric model. It cannot be

5 Σ ad 91b: φιλότιμος δέ και άλαζών και βαρβαροήθης άει ό "Εκτωρ χαρακτηρίζεται. Cicero apparently translated from memory, which was faulty. He thinks the speech addressed to Achilles, leading Gellius to com-ment on his carelessness. Cicero makes a similar mistake at Div. 2.63 (fr. 23), ascribing to Agamemnon a speech actually made by Odysseus.


sufficiently simple. Not even Cicero, for all his linguistic facility, can escape the fact that Latin had already developed distinctly un-Homeric techniques for creating an epic sound. His "Hectorean sword" and "lights of life" reflect a characteristically Roman aesthetic. Such changes in tone mark the transition from Homeric to Roman epic style. This change is even more striking when Cicero is free from the requirements of close translation and adapts a Homeric model to an original context. The result is, if anything, still more traditional in its sound.

In Iliad 12, Hector and Poulydamas pause in their attack on the Achaean ships to witness a powerful and unsettling omen (200-5).

όρνις γάρ σφιν επήλθε περησέμεναι μεμαώσιν, αίετός υψιπέτης έπ' αριστερά λαόν έέργων, φοινήεντα δράκοντα φέρων όνύχεσσι πέλωρον ζωόν ετ' άσπαίροντα- και ου πω λήθετο χάρμης-κόψε γαρ αυτόν έχοντα κατά στήθος παρά δειρήν ίδνωθεις οπίσω* 6 δ' άπό εθεν ηκε χαμαζε άλγήσας όδύνησι, μέσω δ' ένι κάββαλ' όμίλφ, αυτός δέ κλάγξας πέτετο πνοιής άνέμοιο.

A bird appeared to them as they were eager to press on, an eagle flying high, passing to the people's left, bearing a blood-red snake in its talons, monstrous, alive, and breathing still. Nor had it lost its fighting spirit. Rearing back, it struck on chest and throat the eagle holding it. The bird, smarting at the bite, sent it to the ground, dropping it among the fighters, and itself flew shrieking off on the wind's blast.

Events unfold in a logical sequence line by line: a bird appears, an eagle, an eagle with a snake in its talons. The Ciceronian version undergoes some interesting transformations. Cicero originally set the passage in a hexameter poem honoring his fellow townsman Marius, the first man from Arpinum to reach the consulate.6 He liked the result well enough to quote it in De divinatione (1.105 = fr. 20).

hie lovis altisoni subito pinnata satelles arboris e trunco, serpentis saucia morsu, subrigit ipsa feris transfigens unguibus anguem semianimum et varia graviter cervice micantem. quem se intorquentem lanians rostroque cruentans 5 iam satiata animos, iam duros ulta dolores

6 Dates as early as the eighties and as late as the forties have been pro-posed for this poem, none decisively. Summary discussions can be found in Ewbank, The Poems of Cicero, 13-16; Townend, "The Poems," 120-23; Courtney, Fragmentary Poets, 177-78.


abiecit ecflantem et laceratum adfligit in unda seque obitu a solis nitidos convertit ad ortus. hanc ubi praepetibus pinnis lapsuque volantem conspexit Marius, divini numinis augur, 10 faustaque signa suae laudis reditusque notavit, partibus intonuit caeli pater ipse sinistris. sic aquilae clarum firmavit Iuppiter omen.

Here suddenly the winged servant of high-sounding Jupiter from a tree's trunk rose up, wounded by a serpent's bite, and pierced the half-dead snake with her fierce talons, as it darted its spotted head about. She tears it writhing with her bloody beak until, her anger quenched and painful agony requited, she drops it dead and mangled, hurled into the sea, and turns herself from setting to the splendid rising sun. Marius, reader of divine intentions, observed her flying with her feathered wings aloft and marked the favorable signs of rehabilitation and return On the left, the father of heaven himself thundered. Thus Jupiter declared the eagle's omen true.

The language of the passage is very different from its model, for it looks to other standards from another tradition. Cicero may avoid the fits and jerks of Ennian spondees and the archaisms of the older epic, but these verses nevertheless rest firmly on Ennian foundations. Penthemimeral caesura remains the norm that Ennius had estab-lished, and the recurrent alliterative patterns are familiar. Diction too owes much to the traditional style. The elaborate periphrasis of the opening line recalls Ennius' substitution of genus altivolantum for simple avis in the augury of Romulus (76), and the compound adjec-tives also betray Ennian influence. Semianimum . . . micantem is a di-rect echo of Ennius' semianimesque micant oculi (484), while altisonus applies to Zeus an adjective that Ennius attached to objects—for example, altisonum cael (586). Cicero probably means to suggest the Homeric phrase Ζευς ύψιβρεμέτης, but the Ennian equivalent for this adjective, altitonans, does not suit his metrical context.7 He therefore uses the alternative adjective. The effect in this context is, however, decidedly un-Homeric. Where Homer was simple, Cicero has become

7 Ennius employs the genitive at line end: "contremuit templum magnum Iovis altitonantis" (554); Cicero uses the nominative in Cons., fr. 6.36: "nam pater altitonans stellanti nixus Olympo." Altisonus is routinely (but I think wrongly) identified with ύψηχής, a word Homer uses only for the sound of horses. The Greek describes a shrill sound rising high; the Latin refers—as its substantives make clear—to sounds that arise on high. Latin compounds of this type are at least as old as Naevius. See Barchiesi, Nevio epico, 380-83.

5

10


complex. The purely decorative epithet is designed to create a suitably epic aura, which for Cicero and his audience meant Ennian.

The Roman fascination with augury, which has encouraged Cicero to incorporate this image into his Marius, is itself traditional. The augury of Romulus in the Annals, which Cicero quotes immediately after this passage in De divinatione, is the most famous example, but we should recall that Naevius had gone out of his way to make Anchises himself an augur (fr. 25). Both the form and the rationale for Cicero's Homeric adaptation are thus rooted in Roman epic tradi-tion and draw their effect from the recollection of their Roman precedents.

Given time, Roman epic would not have to sound like this, and the sequel is of interest. Cicero's traditionally elaborate diction is not the end of the story. Vergil, as it happens, also admired this Homeric passage. It figures in a battle scene from Aeneid 11, adapted to describe the Etruscan chief Tarchon resisting the counterattack of Venulus.

utque volans alte raptum cum fulva draconem fert aquila implicuitque pedes atque unguibus haesit, saucius at serpens sinuosa volumina versat arrectisque horret squamis et sibilat ore arduus insurgens, ilia haud minus urget obunco luctantem rostro, simul aethera verberat alis:

As when a tawny eagle flying high carries off as prey a snake impaled and held fast in its talons, yet the serpent, wounded, turns its twisting coils back, erect and tense, with rigid scales, and hissing rises up. All the same, the eagle's hooked beak overcomes these struggles as it beats the ether with its wings. (11.751-56)

Borrowing the image recalls its history, and Vergil strengthens that recollection with subtle echoes of the Latin intermediary.8 How he separated himself stylistically from his Latin predecessor is thus especially striking.

First, Vergil returns to a simplicity more characteristic of Homer than of earlier Roman poets. He has turned the passage into a simile, freeing himself to be direct rather than portentous. No epic peri-phrasis here: the eagle may be tawny, but it is not a winged minion of

8 Thus unguibus at the same point in the line and saucius at serpens ~ serpentis saucia. Vergil reserved praepetibus pennis for a more memorable occasion, Daedalus' flight at Aen. 6.15. Cicero's contribution to Latin epic diction should not be underestimated. See Allen, TAPA 87 (1956) 141-43, and Wigodsky, Vergil and Early Latin Poetry, 109-14.


high-sounding Jove. He also keeps to more modest proportions. The passage is short, a small and vivid detail in a larger picture of some-thing else. Second, Vergil suggests without actually employing the old epic diction. He avoids compound adjectives and decorative epithets. He only hints at the old style with volans alte, echoing without duplicating Ennian altivolans as he creates an accurate but idiomatic version of the Homeric original (αίετός υψιπέτης, 12.201). He is also far more restrained and pointed in his alliteration. Com-mentators regularly hear the snake's hissing in the sibilants of Cicero's first three lines, but these begin with just the eagle. The snake first appears only halfway through the second line. If the effect is deliberate, it is misplaced. Cicero has more likely just let the sound run away from him, which is an easy thing to do. There are as many sibilants, for example, in lines 11 and 12, which have nothing to do with the snake. Contrast Vergil, who restricts the sound of hissing to lines 753-55, where draco becomes serpens to enhance the effect (cf. 2.204 and 214, changing angues to serpens). The power of his description comes not simply from the correspondence of ornament and sense, but from the absence of alliteration in the surrounding lines. The point here is not simply Vergil's superiority as a poet. It is a matter of attitude. He has assessed the repertoire of traditional effects and chosen which to avoid, which to apply, and which to suggest without actually applying.

In one important respect, however, Cicero is indeed simpler than either Homer or Vergil. His more elaborate style does not describe a correspondingly complex situation. His eagle, no longer surprised in flight by its purported victim, simply takes its revenge on the snake still in its power. What for Homer was the central action-—the snake rearing back to strike unexpectedly (204-5)—becomes only part of Cicero's preliminary description of the bird ("serpentis saucia morsu"). The scene becomes more violent than horrific as the eagle, perched now in a tree, vents its anger on the offending snake. Cicero gives the eagle an unambiguous victory by manipulating the Homeric details to create this rather different picture. It becomes a simple omen that Marius interprets and Jupiter's thunder confirms.

This preference for the simple statement over the complex sug-gestion brings us back to those Ciceronian Sirens, who do not sing quite the Homeric song. Homer's Sirens know a story of shared suffering, "what Greeks and Trojans endured before broad Troy." That is a story both personal and common: not Greece and Troy, but Greeks and Trojans; not one group taxing another, but two groups linked by their suffering. The Sirens recall the world of the Iliad, which turns on two deaths, one Greek and one Trojan, and climaxes with a scene of joint lamentation as Priam and Achilles find their many differences subsumed by the common experience of their grief.


Knowledge in this epic world is the knowledge of death, the acceptance of death as the ultimate price of things.9 Such knowledge is part of the code that defines the valor of normative heroes like Sarpedon and Patroclus and challenges the equanimity of the excep-tional heroes Achilles and Hector. We see it at work in the structure of the Iliad's final duel. The sight of Achilles bearing down on him in Iliad 22 is at first too much for Hector, whose immediate response is flight. Athena then tricks him into making a stand by taking the form of his brother Deiphobus. She promptly disappears again, but not without freeing him from the terrible and demeaning fear that had originally led her to intervene. Hector eventually finds the strength to face Achilles when he recognizes and accepts the divine intervention for what it is, and that acceptance gives him the courage to die bravely. It is not, however, an easy or an automatic acceptance. The heroic values so simply endured by winners like Ajax and losers like Sarpedon are not so simple for Hector, or for Achilles after him. They both have lessons to learn about the cost of their choices and the limits of their power. Homer's story is not, for them, a simple one. The Iliad is not a simple poem.

Excuse a Latinist for saying something so obvious. It was not so obvious—or at least not so important—to Cicero. He preserves the outward shape of the Sirens' statement with syntax clearly modeled on the Greek: ϊδμεν πάντα becomes "certamen clademque tenemus" and δσα becomes "quam." Yet he does not—cannot—preserve the emotional and moral complexity of the Sirens' allusion. This Latin version is impersonal and unambivalent. It is now a matter of Greece and Troy, not Greeks and Trojans. The war may have been harsh and even horrible (grave), but Cicero nevertheless casts it in terms of a winner and a loser ("Graecia quam Troiae . . . vexit"). That thought follows naturally enough from the idea of disaster in clades, and the form of its expression is itself something of a cliché for Cicero. In a roughly contemporary translation embedded in De divinatione, he uses a similar construction to describe the ships at Aulis, "quae Priamo cladem et Troiae pestemque ferebant."10 The elaboration clearly appealed to him, but there is more to this substitution than a matter of favored vocabulary and ornament. He is not simply deaf to Homeric nuance. He is substituting a Roman nuance, for the Latin version cannot avoid reflecting Roman values.

9 My reading of the Iliad here owes most to Mueller, Mosaic 3 (1970) 86 -103. Pucci, Arethusa 12 (1979) 121-32, details the Sirens' Iliadic language.

10 Fr. 23.7 (Div. 2.63), translating IL 2.304: [νήες] κακά Πριάμφ και ΤρωσΙ φέρουσαι. The relative chronology of these two translations is uncertain but does not affect the present argument. Cicero's poetry displays from first to last a striking unity of style: see Kubiak, Philologus 138 (1994) 63-65.


The Homeric poems were never the embodiment of a meaning independent of the interests and values of their readers. The Roman versions of Homer are therefore not isolated and artificial philo-logical exercises but documents in a history of reception. Changed expectations in changed times inevitably generate new readings that find, either by accident or design, new values in the old texts. Interpretation and assimilation are the inevitable consequences of this process. A literary work, as Hans Robert Jauss reminds us, "is not a monument that monologically reveals its timeless essence. It is much more like an orchestration that strikes ever new resonances among its readers and that frees the text from the material of the words and brings it to a contemporary existence."11 We saw this principle at work as early as Andronicus' Odussia, which invented an original diction and decorum to suit Roman sensibilities. Andronicus pioneered not only the art of translating literature but also of making it at home in its new surroundings. Cicero's translations also quite naturally reflect a Roman version of the epic world. The Homeric passages he renders may often have come to him embedded in his Greek philosophical sources, but as a translator he cannot help but recast them in his own epic idiom. In bringing Homer to a first-century "contemporary existence," he therefore helps us measure the distance between Greek and Roman epic values.

The gap was already wide by the time of Ennius, whose deliberate evocation of Homer does not obscure a basic fact of literary life: the world that Roman poets created in epic verse operated by different rules, aroused different feelings, and cultivated a different appeal from its Homeric prototype. Their poems, for example, consistently lack that Homeric sense of the tragedy in epic deeds. They are not introspective. Roman epic confirms heroic values without questioning them or making inquiry into their price, because the function of epic had itself changed. When Naevius turned to historical narrative and Ennius followed his lead, they were both establishing the content of epic and shaping the terms of its reception. Though Latinists, fol-lowing the poets' own example, often think of their works as annalistic, the poems were not, in either conception or execution, what modern critics mean by "annals." They do not simply provide a sequence of dates and happenings. They more closely resemble chron-icles, coherent though open-ended narratives that are organized chronologically. They have a story to tell that develops a moral vision, a significance that they read into the events recorded.12 What little we know of Greek historical epic shows no signs of a similar

11 Jauss, Toward an Aesthetic of Reception, 21. 12 For the ramifications of this distinction between annals and chronicles,

see White, The Content of the Form, 6-20.


intellectual engagement between poets and their themes. Naevius and Ennius, unlike Rhianus before them and Archias after, were not traveling poets writing at the behest of city governments or ambitious generals. They had made themselves into Romans and show the immigrants' zeal for their new community. Their poems represent subjects of real significance to them and develop a literary form to satisfy their needs. They created epic, at just the time Roman aristocrats began developing history, not simply to record and extoll but to understand the scope and meaning of the Roman achievement. Their deliberate fusion of epic form and historical narrative shifts the function and appeal of epic from tales of individual prowess to the celebration of communal achievements.

This emphasis on the group over the individual has important consequences. For one thing, it rules out tragic overtones. Greek tragic actions, whether depicted on the stage or described in epic verse, develop from the inevitable conflict of equally compelling but contra-dictory moral visions. The Roman subjects chosen by Naevius and Ennius are by their very nature morally simple and therefore un-tragic. They are records of success that attracted readers to the epic enterprise through their patriotic values and happy endings. The only episode of true pathos (but still not tragedy) among the frag-ments of Ennius' Annals, for example, is the story of Ilia, the girl tormented by an all-too-real nightmare (34-50). Much was appar-ently made of her torment (45, 60) and her persecution by Amulius (xxxix), but the gods are on her side. She becomes the object of divine comfort and divine reward. Formidable children are the eventual product of her mysterious encounter, and they are destined to found the city of Rome. Her suffering is thus made good in the next gener-ation, her anguish as an individual subsumed by her distinguished role in the poem's greater story.13 Individual suffering is countered and eventually outweighed by that greater good.

A second consequence is even more significant. Because the subjects of epic held communal rather than largely personal interest, the genre could honor individuals without losing its own integrity. Epic

1 3 This is typical of Roman legend, which may depict the suffering of Roman heroes but never questions the course of action that produces it. Tragedy is impossible in such circumstances. Livy's stories of Brutus (2.4-5) and Verginia (3.44-49) are characteristic of the genre. Their value as moral exempla derives from the inherent Tightness of the painful decisions made. Even the Aeneid, with its undercurrent of sadness, pain, and loss, makes these only the necessary price of success: it is not Dido's epic, or Mezentius', or Turnus'. The Annals too may have treated its losers—Pyrrhus and Hannibal come to mind—with sympathy, but there is no sign of tragedy in the losers' story.


encouraged the merging of individual and group achievements; indi-vidual glory contributed to and even created communal glory. The broader epic's narrative scope, the firmer that identification and the wider its appeal. This was the secret of Ennius' success, as Cicero himself saw clearly. Among his claims in defense of Archias, for example, is the observation that a poet who honors individual Romans also performs a service to the state. "All those Maximi, Marcelli, and Fulvii/' he reminded the jury, "are not honored with-out conferring praise on us all." Ennius' Roman citizenship came as the reward of a grateful citizenry for public services rendered (Arch. 22). This dedication to group values, however, gave epic an inherently conservative outlook. Its narratives follow the normative patterns of their culture, confirming rather than challenging the Romans' view of themselves. Content became as stylized as diction. This stylization presented Cicero with a serious problem when he came to write original epics. Even as he mastered the traditional diction, his eagerness to tell his own story in epic guise came into conflict with the traditional contraints on epic content. Because his own career did not follow the traditional path, it did not furnish a traditional subject for verse. Small wonder, then, that his forays into epic composition failed to win the acclaim he had hoped for them.

We should feel some sympathy for this failure. Not only was it difficult to fit his personal achievement into the communal story, but professional help for telling it proved impossible to secure. Although Archias had willingly celebrated the achievements of Marius and Lucullus in verse, the confident hopes of the Pro Archia proved un-founded. He never produced the Greek poem that Cicero hoped to see on his consulship. A few years later, Cicero approached the historian Lucius Lucceius about writing a monograph that would cover events from the consulship in 63 through his triumphant return from exile in 57. Lucceius was apparently interested, and some documents changed hands. If he ever wrote the monograph, however, it has vanished without a trace.

Cicero was left to be his own publicist, meeting each disap-pointment with a work of his own. In March of 60 he sent his friend Atticus a Greek memoir on his consulship that had literary pretensions, and he thought of adding a Latin version. "As a third item," he warned, "you may expect a poem, not to leave any form of self-praise untried." It was, he added with a wink, a matter not of encomium but of history. By 54, he was telling Lentulus Spinther about a second poem he had drafted on his subsequent time of crisis ("de temporibus meis").14 At least in letters, Cicero could afford to

14 Att. 1.19.10; Fam. 1.9.23. For Archias in 62, see Arch. 28 and Att. 1.16.15; for Lucceius in 55, Fam. 5.12, cf. Att. 4.6.4, 4.9.2, and 4.11.2. Caesar was more


smile a bit at such efforts, but posterity proved less indulgent. The second poem has vanished; the first did little to enhance his reputation. The hexameter poem De consulatu suo quickly gained more notoriety than fame. Its confidence in the greatness of his achievement stretched the application of epic conventions to con-temporary events by enlisting the gods on the consul's side and declaring their willingness to intervene on his behalf. Later gener-ations would deride what critics perceived as the poem's insolentia, and even Quintilian, that consummate Ciceronian, had doubts about its taste.15 All this is certain, but the truth behind such testimony is harder to judge. Little of the poem remains. One long, rather tedious fragment has come down to us as a self-quotation in De divinaiione. A three-line snippet, not itself overly modest, is preserved in another letter to Atticus. Of the rest, we have only four separate lines, which come without context, though not without controversy.

What accounts for the clamor? Some venerable slanders should be set aside. The debate was not over technical ability. Cicero was at the very least a facile versifier. We may think his metrical range limited and therefore repetitious, but that does not seem to have been the Roman view. Nobody criticized Cicero for writing doggerel. That one long fragment—at seventy-eight lines, the longest quotation in extant Latin literature·—illustrates the virtues and limitations of his style (fr. 3). It is a passage in which the Muse Urania describes astral phenomena that foreshadow the crisis of 63. Its effects are tradi-tional. There is a good deal of sound play in phrases like aetheris aeterni (5) and lumine luna (18). The line "lustrasti et laeto mactasti lacte Latinas" ("you purified and graced the Latin rites with joyous stream of milk," 14) not only reflects the same taste for ornamental sound that produced Ennius' famous "O Tite, tute, Tati, tibi tanta, tyranne, tulisti" (104) but also uses internal rhyme and alliteration to link cola across the caesura in a way reminiscent of Saturnian poets. The line "iam vero variée nocturno tempore visae" is an even clearer example of the device (26). Line ends tend toward a certain unifor-mity of metrical shape and word choice, as in the sequence perculsus

fortunate: the time of Cicero's frustration was the very time Varro Atacinus was composing a Bellum Sequanicum on Caesar's campaign of 58. See Courtney, Fragmentary Poets, 235-38.

1 5 Ps.-Sall. In Cic. 4.7 (cf. 2.3); Quint. 11.1.24. See the survey of ancient opinions in Ewbank, The Poems of Cicero, 10-13 and 123-24, and Allen, TAPA 87 (1956) 130-46. No verses of De temporibns suis survive, and ancient testimony often confuses the two poems. So, I think, does Büchner in his presentation of the evidence for them in his revision of Morel, Fragmenta poetarum latinorum, 86. See now Harrison, Hermes 118 (1990) 455-63, and Courtney, Fragmentary Poets, 173-74.


fulmine civis (23), peremit fulminis ardor (41), flammato fulminis ictu (45). Cicero's youthful translation of Aratus' Phaenomena was full of such metrical cliches, and he evidently saw no need to abandon them in his maturity.16 He also ends three successive lines with iden-tical sounds (monebant, ferebant, iubebant, 50-52) and follows them with two passive infinitives (vereri, teneri, 53-54). Such effects in combination can be quite striking, as in lines 60-61:

haec tardata diu species multumque morata consule te tandem celsa est in sede locata

This statue was long delayed and much hindered until at last, in your consulship, it stood in its lofty place.

None of these features, however tiresome to modern ears, violates contemporary standards. Lucretius' poem has many recurrent, metri-cally convenient phrases, including principio at the beginning of lines and fulminis ictu at the end. Hostius created a deliberate jingle at the end of adjacent lines (fr. 3).

non si mihi linguae centum atque ora sient totidem vocesque liquatae

not if I had a hundred tongues and as many mouths and clear voices

Even Vergil could seek similar effects. Anchises' speech at Aeneid 6.724ff. seems to echo Urania (principio caelum . . .), and the poet was not above the repetitions and word-play of his predecesors:

haud aliter Troianae acies aciesque Latinae concurrunt, haeret pede pes densusque viro vir.

No differently did the Trojan line and line of Latins clash. Foot stood to foot and close to man each man. (Aen. 10. 360-61)

There is a qualitative difference here, to be sure, but not a stylistic one. Such devices complement the meaning in a way beyond Cicero's power, but the tricks are the same for all that.17

16 Traglia, La lingua di Cicerone poeta, 246-47, illustrates this tendency. I do not mean to slight the merits of Cicero's Aratea, merely to ignore them. For a sample of what can be done in defense of this work, see Kubiak, CJ 77 (1981/82) 12-22, and SÎFC 8 (1990) 198-214.

17 Cf. Ennius 584: "premitur pede pes atque armis arma teruntur," with Skutsch's note, pp. 724-26, and Furius (Bibaculus?) fr. 10: "pressatur pede pes, mucro mucrone, viro vir." Both Furius and Vergil go back, as Macrobius observed (Sat 6.3.5), to Homer, IL 13.131. The Rhet. ad Her. 4.30-31 urges


No other fragment of De consulatu suo seems particularly inept. Its most notorious line, "o fortunatam natam me consule Romam" ("o for-tunate Rome, born in my consulship/' fr. 12) raised hov/ls from later critics, but not because of its structure, word play, or alliteration. They too were all traditional. Ennius had resorted to the archaic pronoun sum for eum to achieve an analogous effect in the line "astu non vi sum summam servare decet rem" ("by craft, not force, should he conduct affairs of state," 97). Lucretius wrote, "si quos ante domi domitos satis esse putabant" ("[beasts] they thought trained well enough at home," 5.1334). Even Horace, hardly an indulgent critic of Republican poetry, echoes Cicero explicitly in his letter to Augustus: "et formidatam Parthis te principe Romam." Cicero's play on -natam natam is gone, perhaps a casualty of Augustan sensibilities, but Horace does not hesitate to recall this line with a straight face.18

The other infamous line, "cédant arma togae, concédant laurea laudi" ("let arms yield to the toga, let laurels defer to praise," fr. 11), is also technically respectable. The play on cédant / concédant and laurea / laudi is elegant in itself and suits the spondaic rhythm. Meter, orna-ment, and content complement each other. Nor was the vanity of Cicero's poetic enterprise in itself enough to rile his contemporaries. Other consuls figured prominently in their own memoirs. Some dabbled in poetry. Self-praise was never in short supply: even the elder Cato paid himself compliments when nobody else would.19 A parodist would eventually amuse (or dismay) posterity by turning laudi to linguae, which was the version Quintilian knew, but that jab was in the future. We should set against such sniping the comment in Plutarch that Cicero in his prime enjoyed a reputation, however fleeting, as Rome's best poet as well as its best orator (Cic. 2.4).

His main fault as an epic poet lay elsewhere. The line that came back to haunt him aroused a different kind of opposition. Contem-poraries pounced not simply upon the vanity of "cédant arma togae" but upon the values and the politics that it represented. Their jibes

restraint in employing paronomasia and related devices, testimony to their seductive popularity even in oratory.

18 Epist. 2.1.256, cf. Brink, Horace on Poetry, vol. 3, 256-57. Allen, TAPA 87 (1956) 144-47, thinks Cicero himself actually wrote te; the poem was almost certainly in the third person. The impulse to trace both lines to a common model in Ennius is misguided. See Wigodsky, Vergil and Early Latin Poetry, 122-25, on such ephemeral "Ennian contexts."

19 As in the famous praeteritio of the speech De sumptn suo (fr. 173, cf. fr. 203). This tendency became increasingly pronounced. Cf. Leo, Hermes 49 (1914) 164: "Sulla schrieb die Geschichte nicht seiner Zeit, sondern seiner Person/' Cicero did not intend the claim of Arch. 26, "trahimur omnes studio laudis," to be controversial. For standards of self-praise in the late Republic, see Allen, TAPA 85 (1954) 121-44, and Harris, War and Imperialism, 21-23.


require more a social than an aesthetic context. The actual claim Cicero makes in the line is not original to the poem. He has only reworked a sentiment he voiced to the Senate at the time, when he set his success in thwarting Catiline beside the achievements of Afri-canus, Paullus, Marius, and Pompey. They had won military glory; his equal glory lay in saving the city that honored them (4 Cat. 21-22). There was no objection then. It was a deft claim to fame, but not, as it happened, an enduring one. The laurels so proudly worn in the glory days of 63 did not look equally fine by the 50s' harsher light. They had grown shabby, perhaps even ludicrous, by the time of his exile in 58 and became targets of ridicule in the acrimonious ex-changes that followed his recall. By 55, Piso Caesoninus, who had been consul in 58 and then governor of Macedonia, was back in Rome and began complaining of Cicero's vengeful abuse of him. In the course of the ensuing wrangle, he turned the literary trope of De consulat u into an affront to Pompey: how did this novus homo dare to set his achievement over that of Rome's greatest general? Cicero, driven to splitting hairs in public, defended himself by accusing Piso of mis-reading his intention. "I did not mean my toga," he told the Senate, "but the toga as the symbol of peace and repose."20 That reply soon circulated in pamphlet form and helped quiet Piso. It did not lay the criticism itself to rest, however, because it did not really answer the criticism. Cicero's claim of civilian gloria could still raise hackles. In 44, when the stakes had grown so much higher, he had to make public answer to Antony's taunt that the consular toga had in fact yielded to force of arms. He also devoted space in De officiis to defending the claim of civilian virtus over the military kind. This is not, he concedes, the common view, but it is the correct one. He quotes the line with pride in this context, but promptly adds a defense. Had not Pompey himself said publicly that his third triumph would have been in vain without Cicero's achievement? Civilian virtues, he con-cludes, are therefore not inferior to the military kind.21

This more temperate version of the argument first used against Piso is not entirely satisfactory. There are good Greek examples adduced to support it, but Cicero is hard-pressed to summon any corresponding Roman ones. The best he can do is Scipio Nasica, the assassin of Ti~

20 Pis. 73: "Non dixi hanc togam qua sum amictus, nec arma scutum et gladium unius imperatoris, sed quia pacis est insigne et oti toga . . . poetarum more locutus hoc intellegi volui, bellum ac tumultum paci atque otio con-cessurum."

212 Phil. 20; Off. 1.74-78, developing the argument first made at Pis. 72-76. Cicero's friends must have had a different view: Cassius, for example, cites the line with approval, Fam. 12.13.1. Harrer, SP 25 (1928) 77-84, gathers the evidence for its reception.


berius Gracchus, a case he himself knows to be inexact because Nasica had in fact resorted to force.22 The other preeminent togati of this elaborate proof, M. Scaurus and Q. Catulus, actually pale beside their military counterparts, Marius and Pompey. The painful truth is that Roman equivalents of Lycurgus and Solon are not easily found. The historical record, like the popular fancy he attacks, favors res bellicae.

So does epic, which did not normally center on the civilian virtues which Cicero's own career represents. Naevius had honored An-chises' augury, but his was still a tale of armies and battles. Ennius indeed praised the oratory of Cornelius Cethegus, and he detailed the urbane virtues of Geminus' "good companion." The broader context of the poem, however, always kept such moments in perspective. The talk between Geminus and his companion came at Cannae. The praise of Cethegus is merely an elaborate dating formula in a book dedi-cated to the crucial victories of the Metaurus and Zama. Cicero is himself his only good example of triumphant res urbanae, and that case is not entirely clear, either. The repression of Catiline was not without its military side, as Sallust so graphically records, and the eloquent consul of 63 plays a distinctly minor role in that account. There was, after all, some truth to Antony's jibe about the toga yielding to arms. The conspirators who died on the consul's order were not those in the ranks at Pistoria. Cicero's consulship as seen from Cicero's perspective in De consulatu suo could not be the stuff of tradi-tional epic. Cédant arma claims the virtue of a necessity, but it was not a virtue likely to win adherents in the waning years of the Republic. It reflects values that, like Cicero's career itself, managed to be simultaneously untraditional and outmoded.

The Roman epic formula thrived on tales of unambiguous success, or at least on unambiguous tales of success. Though poets did not have to suppress failures, neither did they end with them. Naevius at least mentioned Cn. Scipio's losses at Lipara (fr. 33) and the ill-timed arrogance of P. Claudius Pulcher (fr. 37), but he could set these inglorious moments within the context of the war's ultimate success. Ennius rescued Fulvius' Ambracian campaign from the carping of political rivals by giving it Homeric trappings and subsuming it under the still grander cultural achievement of the newly installed Ambracian Muses. Such narratives could claim the dignity of epic style without forfeiting the moral certainty with which Romans liked to read the record of their achievements. Contemporary criticism would eventually fade; the epic poems would endure. Or so

2 2 "Quamquam haec quidem res non solum ex domestica est ratione— attingit etiam bellicam, quoniam vi manuque confecta est—sed tarnen id ipsum est gestum consilio urbano sine exercitu," Off. 1.76.


Ennius might well think. History, as it happened, offered Cicero no such chance. The insecure achievement of his consulship was over-whelmed by subsequent events, and his poem, equally insecure in its epic context, suffered the consequences of that political failure. When it ceased to be the record of a Roman success, it ceased to function as a Roman epic should. The result was oblivion relieved only by a ridi-cule worse than oblivion.23 Cicero's failure as an epic poet comes as but another manifestation of the values so curiously embedded in his Homeric translations. His original epic enterprise was led astray and ultimately wrecked by the limited range and narrow moral vision reflected in his own Sirens' song.

What lesson should we take, then, from those Ciceronian Sirens? The basic point is obvious. The epic values of the Iliad are not easily cast in Roman terms. What individuals suffer and what greatness they achieve through the endurance of suffering are not central to Roman epic values. It is worth thinking again about Hector's last moments. The courage that leads him to make that final stand against Achilles proves a firmer support than the divine aid that restored it to him. Though abandoned by the gods, his thoughts nevertheless move inexorably from the fear of death to the oppor-tunity that death presents to perform one last great deed (IL 22.304-5).

μή μάν άσπουδί γε και άκλειώς άπολοίμην, άλλα μέγα ρέξας τι και έσσομένοισι πυθέσθαι.

Let me at least not die without a struggle, inglorious, but do some big thing first, that men to come shall know of it.

Courage has succeeded despair. By prevailing over his own fear, he achieves his moment of greatest valor at the time of his ultimate defeat. The courage, like the defeat, is expressed in personal terms. He has given up his role as Troy's defender. The city fades into the background as the narrative focuses on this last great duel. Hector's death may indeed seal the fate of Troy, but it is now Achilles' death that is on his mind and Achilles' end that is prefigured by his own. Claims to heroism in Homer rest on the physical prowess of the individual.

This may be an admirable pattern of behavior, but it is not the Roman one. The Roman hero at such a moment is expected to act differently. When Aeneas is surrounded by the death throes of Troy,

2 3 Thus Tac. Dial. 21, discussing Caesar and Brutus: "fecerunt enim et carmina et in bibliothecas rettulerunt, non melius quam Cicero, sed felicius, quia illos fecisse pauciores sciunt."


he too has an immediate impulse to return to the fray for one last moment of glory,

arma, viri, ferte arma; vocat lux ultima victos. reddite me Danais; sinite instaurata revisam proelia. numquam omnes hodie moriemur inulti.

Arms, my men, bring arms. The last light calls the vanquished. Let me at the Danaans again; return me to new battles. Never shall we die today without revenge. (Aen. 2.668-70)

Aeneas is not granted this moment. Even as he buckles on his armor, he is restrained by his wife and child, awed by Jupiter's sudden flame around the head of lulus, and deflected from his purpose by Anchises' reading of that omen. While Hector, at a vaguely comparable mo-ment in Iliad 6, abandons his family for the battlefield, Aeneas is forced to abandon the battlefield for his family. His shoulders are to bear not the weight of his armor alone but that of his father and his household gods as well. Aeneas abandons Troy, and in doing so he leaves behind both his city and the entire world of Homeric values. Prevented from acting out the role of the Homeric hero, he assumes the role of a Roman one.

The truly Homeric hero of the Aeneid is its great loser, Turnus. The man set aflame by the Fury's torch is gradually undone by a prowess that, under other epic conditions, would have won him the greatest glory. In the Roman context it is only a prescription for failure.24 His escapade at the end of Aeneid 9, for example, is heavy with allusions to Homeric and Roman aristeia. Turnus is shut inside the Trojan camp, wins a long, bloody series of duels in Iliadic style, and meets a last Trojan onslaught with the firmness of both Ajax at the ships and, quite explicitly, the stalwart courage of Ennius' tribune before the walls of Ambracia (806-14, cf. En. 391-98). He eventually escapes his attackers by plunging fully armed into the river, a last act of bravado that may remind the unwary of Horatius Codes at the Tiber bridge. Turnus emerges from this exploit quite pleased with himself—Vergil calls him laetus as he reaches the farther bank—but the poet does not encourage readers to share that pleasure. Behind Turnus' valor lurks a tainting might-have-been. His entry into the Trojan camp had created a golden opportunity:

et si continuo victorem ea cura subisset, rumpere claustra manu sociosque immittere portis,

24 Klingner, Virgil, 548, puts this well: "Turnus ist der glänzendste Krieger der Aeneis, aber eben nur Krieger, Aeneas kämpft nur, weil es sein muß, seine Gedanken und sein Wesen sind auf den göttlichen Willen Jupiters, auf Frieden hingeordnet/'


ultimus ille dies bello gentique fuisset. sed furor ardentem caedisque insana cupido egit in adversos.

And if the thought at once had crossed the victor's mind to smash the bolts by hand and let his comrades in the gate, that would have been the last day for the war and Trojan race. But excitement and a frantic lust for slaughter drove him eagerly against his foes. (Aen. 9.757-61)

Turnus secures his momentary, personal victory at the cost of a last-ing, communal one. Where Ennius' tribune fought as part of an army and Horatius stood bravely in defense of his retreating comrades, Turnus here fights only for himself. The narrator's intrusion— punctuated, as it happens, by the half-line of 761—emphasizes the futility of an episode no less savage and mindless than the destruc-tion of Nisus and Euryalus some three hundred lines before. What value can fighting of this kind really claim?

The doubt Vergil raises over Turnus' exploit is signaled not just by explicit mention of his furor caedisque cupido but also indirectly by the allusion to Ennius' tribune. As in that later reference in Book 10, which contrasts Aeneas' cruelty to Magus with Pyrrhus' magnan-imity, Vergil sets his warrior of the moment apart from the behavior Roman epic had traditionally endorsed. The problem this separation creates for the Aeneid—the famous problem of how Aeneas is and is not like Turnus—becomes, as everyone knows, a major source of the poem's greatness. While the narrative is propelled by the differences between Turnus and Aeneas, its moral tension develops from our grow-ing awareness of their many similarities. Vergil escapes the confines of both Homeric virtue and the fundamentally untragic Roman morality by renegotiating the terms of the epic enterprise. Not content with either the old Roman verities or the even older Homeric heroism, he turns epic in a new direction by creating a story congruent with, but never the same as, either Homer's story or Ennius' values. The Aeneid can thus maintain a keen historical sense without be-coming a historical epic. Nods toward Rome's future greatness—the heroic procession of Aeneid 6, the geography and the shield of Aeneid 8—acknowledge its roots and its function in confirming the communal identity, but the events of its legendary present continually test and challenge the assumptions of traditional epic. Homeric and Roman values are constantly at odds.

The result is something different from anything that had come before, and the change was timely. The Roman values Vergil questions were the very ones that Cicero tried so unsuccessfully to embrace. The final point to be made about Cicero's epic is that it was intellectually empty, and this was so not because he was a bad poet


but because he was a good Roman. He was never alive to poetic and moral subtlety. Thus, when he juxtaposed Ennius' description of Romulus7 augury with his own passage on Marius and the eagle, he was too interested in the similar content of each—as Roman a subject as we can find—to notice the moral ambiguity implicit in auguries that bring about internecine conflict and even civil war.25 So too with De consulatu suo. Had he recognized the moral issues raised by the events of his consulship, had he used his epic to explore the price of his actions, he might at least have flirted with greatness, but Cicero's own faith in the rightness of his cause worked against him. His original song, like his translated Sirens' song, was a simple tale of winners and losers, and a posterity uninterested in those winners could hardly find much appeal in the poetry that glorified them. Having come to such a pass, what future could Roman epic have in store?

2 5 So fixed is Cicero on the surface meaning of his self-quotation that he even attributes the omen as described to divine origin ("quid est illo auspicio divinius . . . ?" Div. 1.106) when, as an adaptation of Homer, it is of course essentially of literary origin.

7 Envoi

I said, Omeros,"

and Ο was the conch-shell's invocation, mer was both mother and sea in our Antillean patois, os, a grey bone, and the white surf as it crashes

and spreads its sibilant collar on a lace shore. Omeros was the crunch of dry leaves, and the washes that enclosed from a cave-mouth when the tide has ebbed.

Derek Walcott, Omeros

Derek Walcott's Omeros is well worth a Latinist's attention. It is not just a poem of modern Caribbean life or some extraordinary classical revenant. Nor is it of interest only for its stock of classical allusions, though they are plentiful: its Achille and Hector vie for their Helen. Philoctete suffers a festering wound on his shin. Trees are felled, men struggle in the surf, voyages are undertaken, and passions flare with conspicuously Homeric coloring. Yet the poem's classical roots are, for all those conscious echoes, not essentially Greek. It has too striking and complex a historical sense. Achille the fisherman also dreams of his African ancestors and their journey in chains across the ocean. The expatriate Major Plunkett studies the Battle of the Saints in com-pulsive detail while the native Hector salvages a bottle from the wreck of de Grasse's flagship, the Ville de Paris. There is the blind sage called Seven Seas, who haunts Ma Kilman's No Pain Café, and there is Omeros himself, Walcott's companion and narrator of chang-ing voices and changing roles, who conducts his own odyssey through modern Europe, Africa, and North America in search of history and his place in it. By constantly juxtaposing the demands of past and present, history and myth, the poem becomes too deliberately and

158

Envoi 159

productively self-conscious tobe simply Homeric.1 It, like its modern Omeros, takes the syllables of tradition to pieces, revalues and reassembles them, and by doing so creates a new poem that is also a significant addition to the literary tradition it appropriates.

As an exercise in literary reception, mythical and historical study, and the creation of a modern Antillean identity, Walcott's poem provides a close analogue to the achievement of Roman epic from the time of Naevius. Like the Bellum Punicum and Annals, it uses the language, images, and conventions of epic to represent and understand the peoples and places that come within its range. Thus its values, as well as its devices, can be strikingly familiar. When Walcott's trip to the volcanic crater of Soufrière takes on the trappings of an under-world journey (pp. 285-94), Omeros quite naturally plays Vergil to Walcott's Dante, and there are clear and deliberate echoes of the Aeneid and Inferno. Yet the closest ancient parallel to its thematic significance—and, for that matter, to the merging of the "I" and the Omeros narrator throughout the poem—is Ennius' somnia Pythagorea in Annals 1. I do not mean to suggest that Walcott intends this echo of Ennius (though I have not asked him), nor does his Nekyia require Ennius' stock of Pythagorean notions, but both the ancient and modern poets use the dream vision to assert the validity and integrity of their endeavor. Each declares his unity with Homer. The earnestness and anxiety with which Walcott faces the shades of dead poets and escapes, with Omeros' help, the selfish pride that doomed them dramatizes the self-effacing honesty that is meant to give the poem its moral authority. Whether the episode succeeds in doing so is another question (its identification with Dante and Vergil certainly flirts with pretension), but the poet's initial unease in this august company is itself part of his message. Ennius' own such encounter doubtless had an analogous inner tension, though the fragmentation of his work has left only the bookish shell of its conceit. Walcott's example nevertheless reminds us of what such allusive and synthetic poetry can achieve. Reading Omeros is as close as we can probably come to experiencing for ourselves the surprises and felicities of the Roman poets' own negotiations with Greek tradition, with history, and with their own present.

Some of the problems this modern analogy raises are also of interest, especially for the discrepancies they reveal between ancient and modern experiences of literature. One of these discrepancies

1 Though Taplin, Arion n.s. 3.1 (1991) 213-26, calls it "profoundly Homeric/' he goes on to trace "a deep ambivalence, a simultaneous em-bracing and fending off" of the Homeric tradition that is far more Roman than Greek. For Walcott's ongoing struggle with the legacy of history, see Martin, Kenyan Review 14 (1992) 197-204.


involves the place of each work in its tradition. Anglophone readers of Walcott's poem—even those without his Caribbean eye and ear— will know at once that they are opening to the most recent chapter in their own literary history. Twelve-syllable lines in three-line stan-zas may suggest Dante, but Walcott also writes with the syntactic plasticity of Eliot at his most Vergilian, and his classicism reflects the free-spirited recreations of contemporary poets like Robert Fitzgerald and Allen Mandelbaum. His style is therefore familiar to twentieth-century readers. Though not necessarily easy reading, the poem at least constructs its new effects from recognizable elements of style and form.

The challenge of reading early Roman epics is quite different, and not just because they are fragmentary poems in a dead language. They represent the first chapter in a history we know best from its glorious middle. We have therefore had to put our expectations and suppositions to one side and reconstruct the genre's history from the beginning. We struggle to find among ruins what whole texts openly display: nuances of diction and poetic mannerisms, the definition of a subject, and the appeal to an audience.

Escape from the teleological bias has made the task a bit easier than it might have been. By taking each poet on his own terms, for example, we have dispelled the illusion of artless beginnings and found ample evidence for continuities in poetic style even as Roman epic moved from Saturnians to hexameters. Ennius' proud eclipse of the verse that fauns and prophets sang (207) is rightly taken to be a clear mark of his progress, but it is not a complete repudiation of his predecessor. He still builds on Saturnian effects. We have already seen them at work in the construction of hexameter cola and his echo of Saturnian elogia. This debt to the older, traditional style is at its most striking when the subjects themselves are most traditional. Some particularly good examples of the phenomenon may be found in the speech-making of gods.

Divine councils were of course an important part of the Homeric baggage that Andronicus first unpacked at Rome. A vocative like "pater noster, Saturnie filie" (fr. 2), generally thought to render Athena's address to Zeus in Odyssey 1 (45: ω πάτερ υμέτερε Κρονίδη), is but a first example of the structural mannerism of balancing modifiers and nouns around the caesura that we have come to recognize in a line like his fr. 12, "saneta puer Saturni filia regina" ("blessed child of Saturn daughter, queen."). The effect lives on, transferred to the requirements of hexameter cola, in Ennius:

optima caelicolum, Saturnia, magna dearum Best of heaven-dwellers, Saturn's daughter, great among goddesses (445)

Envoi 161

o genitor noster, Saturne, maxime divom Oh, our creator, Saturn, greatest of the gods (444)

respondit Iuno Saturnia, sancta dearum replied Juno, Saturn's daughter, blessed among goddesses (53)

The traditional occasion is given a traditional sound.2

Even more important than these recurring forms is the way tradi-tional Roman values are so readily projected on to the divine figures. Thus the wide-ruling Zeus of Homer (ΰπατε κρειόντων) comes to sound much like the Jupiter Optimus Maximus of Roman state religion in a line like Naevius' "patrem suum supremum optumum appellat" ("She calls upon her father, the highest and the best," fr. 17). Nor is this just a matter of poetic ornament. According to Macrobius, the famous scene in Aeneid 1 where Venus pleads for the storm-wracked Trojans and Jupiter consoles her with assurance of their future prosperity in Italy derives from the first book of the Bellum Punicum.3 There may well be some exaggeration in this claim—Vergil's debts in the passage are not only to Naevius—but Naevius must have not only appropriated the epic apparatus for a scene with obvious echoes of Odyssey 1 but also given it a new purpose by making the gods explicit agents for and sanctioners of Rome's coming greatness. This characteristically Roman identification of divine will with national success was then developed in the broader terms of Ennius' Annals, which promised a welcome for Romulus in heaven (54, 110) and prosperity to Ilia's other progeny (61) and endorsed the formal institutions of state religion ascribed to Numa (114-18).4 What we have been seeing throughout this study, then, are stylistic traces of

2 The context for line 53 is the consolation of Ilia in Ennius' first book. The other lines are unplaced, though Skutsch, Annals, 603, makes the attractive suggestion that 445 is from the speech of Jupiter to which Juno in 53 replies, in which case Ennius is also reproducing a characteristically Homeric style of repetition in speech-making.

3 BP, fr. 14 = Macr. Sat. 6.2.31: "hie locus totus sumptus a Naevio est ex primo libro Belli Punici. illic enim aeque Venus Troianis tempestate laboran-tibus cum love queritur et secuntur verba Iovis filiam consolantis spe futurorum." Cf. fr. 15, quoting Serv. Dan. ad Aen. 1.198 to the same effect. For the larger significance of this development in Naevius, see Feeney, The Gods in Epic, 113-15,

4 I choose just a few noteworthy examples. Some of this material was doubtless traditional by Ennius' day, but the deification of Romulus may have been his own contribution. See Jocelyn, "Romulus and the di genitales," 39-65, and for the Annals' depiction of social and religious institutions, his remarks in ANRW 1.2 (1972) 1007-11. For the continued role of the gods in the Annals, see Feeney, The Gods in Epic, 125-28.


an important ideological shift within the epic tradition as Latin poets reconfigured Greek values and literary forms to serve Roman needs.

There were always practical reasons for doing so. The ability of Naevius' poem, for example, to link events of the mythic and the recent past not only showed how epic could both record and explain the vagaries of history but, in the process, made Roman success appear inevitable. This result would have been especially welcome if, as seems likely, Naevius was recalling this first war with Carthage while Romans were fighting a second and their victory was not yet self-evident.5 Ennius too portrayed Roman triumphs as the result of divine favor: Hannibal's fate was sealed in the Annals when Juno withdrew her favor from Carthage (8.xv, xvi). His gods remain major players in the Roman story, and Ennius did not soften the effect of their immanence in historical contexts by recourse to Cato's affec-tation of suppressing the names of military figures.6 Nor was his willingness to praise contemporaries in this way simply a matter of flattering the aristocracy. As we have seen, there are too many names, fulfilling too many functions, to support claims of favoritism and special interests at work. The Annals was a national, not a polit-ical epic. It deliberately put contemporary, historical, and legendary figures on the same canvas, presenting a continuous and seamless portrait of Roman achievement. Whether the Titus addressed is Romulus' Sabine partner (104) or the great Flamininus (335), their words and deeds are recorded in similar style.

These early epics created a deeply satisfying alliance between myth and history not through a process of isolation or subordination but by combining these two strands of the Roman story to produce one coherent national identity. The distinction often claimed in modern handbooks between mythical and historical epic is in any case artificial: the Romans never drew clear lines between fact and fiction. Servius, for example, reflecting grammatical traditions root-

5 Naevius was certainly dead by 201 and perhaps as early as 204. Though the Bellum Punicum was not necessarily a work of his old age (Cic. Sen. 50 should not be so interpreted), the anxieties of the Hannibalic War provide the easiest explanation for this extraordinary undertaking. Binder, "Vom Mythos zur Ideologie," 137-38, nevertheless goes too far in attributing its creation specifically to the trauma of Cannae.

6 Cato's reluctance to name military commanders, both Roman and foreign, attracted the notice of Nep. Ca. 3.3 and then Plin. Ν H 8.11. Traces of the mannerism survive in Or. fr. 83, 86, 87P. There is no basis for attributing this habit to reaction against the vanity of epic. Not all names were sup-pressed, least of all his own. Cato remained, as Livy remarks, "haud sane detrectator laudum suarum" (39.15.9). For his possible motives, see Astin, Cato, 212-13 and 232-33.

Envoi 163

ed in the exegesis of texts like ours, defined the essential difference between historia and fabula as simply the difference between the possible and the impossible (secundum naturam, contra naturam), not between what did and did not actually happen.7 So Livy's famous interest in "what the lives and morals were, through what men and by what skills at home and abroad empire was both engendered and augmented" (Praef. 9) might equally well describe the aims of Ennius' Annals. The fabulous stories of divine origins that Livy calls the special license of poets are matters more of narrative style than of substance. The epicists' use of myth and legend is not a descent into fiction but their way of thinking about reality. This is obviously true of the Aeneid, so famously alive to the contemporary relevance of its antique story, and hardly less so of the Bellum Punicum and Annals.8

Given the opportunities epic affords and the expectations it raises, it would be difficult for epic poets not to mythologize. Here the ancient and modern experiences are again similar.

When would the sails drop

from my eyes, when would I not hear the Trojan War in two fisherman cursing in Ma Kilmari's shop? When would my head shake off its echoes like a horse

shaking off a wreath of flies? (Omeros, p. 271)

The poets' answer is, inevitably, "never," Those "Trojan" fishermen, however, suggest a second significant

contrast between ancient and modern attitudes toward cultural iden-tity and the epic tradition. Walcott's synthesis of a Caribbean self from the disparate traditions of Europe and Africa and past and pres-ent must also negotiate the tensions created by the resulting clash of cultures and the forced inequalities of colonialism and race. Walcott is himself a consummate assimilator and conscious heir to what Leavis called "the Great Tradition," but his English-speaking Omeros finds no welcome in England.9 Nor does cultural assimilation seem possible in the modern world without regret and even resentment

7 Serv. ad Aen. 1.235. Cf. Woodman, Rhetoric in Classical Historiography, 98-101; Feeney, The Gods in Epic, 252-60.

8 Note, for example, the functional correspondence of Ennius' chariot race in the augury fragment (79-81) and the famous magistrate simile of Aen. 1.148-54. Alternatively, a whole narrative may be embedded, as in Naevius' Archaeology and Walcott's Midshipman Plunkett with Rodney at the Saints (Omeros, pp. 77-86).

9 Walcott, Omeros, pp. 193-97. See, in general, Asein, The Literary Criterion 16 (1981) 18-30.


at the sacrifice of language, beliefs, and traditions that it demands. Thus Walcott's ancestors on the slaveship heading west mourn the loss not just of their freedom but of the very words left behind with their African existence.

Their whole world was moving, or a large part of the world, and what began dissolving

was the fading sound of their tribal name for the rain, the bright sound for the sun, a hissing noun for the river, and always the word "never," and never the word "again."

(Omeros, p. 152)

Their dignity, as Walcott declared in a different context, depends on the creation of a new language in what will be their new world.10

Latin poets record no such sentiments, though they too left cultures behind them. One of the most striking facts surrounding the national identity synthesized by Roman epic is that the poets who created it were not themselves native Romans. Andronicus was a Greek slave from Tarentum. Naevius and Ennius came from Oscan-speaking re-gions of southern Italy.11 We nevertheless detect no undercurrents of protest or regret at the suppression of their ethnic identities. A line like Ennius' "nos sumus Romani qui fuimus ante Rudini" ("We are Romans, who were formerly Rudini," 525) would, if anything, suggest the opposite. So does Naevius7 epitaph, with its lovingly declared mastery of the Latin language. Such willing and even proud embrace of a new culture is quite different from the modern response to accul-turation. Whether by accident or design, by their own devices or the vagaries of the slave trade, these talented strangers came to Rome, prospered, and confidently, deftly, and even exuberantly made poetry for the race that took them in. Literature was their gift to the Roman people. That legacy, however, was not without pitfalls for its native-born heirs.

10 Walcott, Dream on Monkey Mountain, 17: "What would deliver him [the New World Negro] from servitude was the forging of a language that went beyond mimicry, a dialect which had the force of revelation as it in-vented names for things, one which finally settled on its own mode of inflection . . . This, not merely the debt of history was his proper claim to the New World."

11 The list is of course not limited to the first epicists: Plautus was from Umbria, Caecilius from Cisalpine Gaul, Terence from North Africa. Rome's ability to attract such talent to its service is eloquently described by Mo-migliano, Alien Wisdom, 7-21. Walcott, incidentally, first came to prom-inence as founder of the Trinidad Theater Workshop. Like Naevius, he uses drama to define a national identity, and—again like Naevius?—he does so in a cultural rather than narrowly political sense.

Envoi 165

The outsider poets found ways to write themselves into their poems: Naevius announced his own participation in the war he glori-fied. Ennius asserted his special authority as an epicist.12 Yet neither, so far as we know, actually told of his own exploits or made himself part of the divine machinery; Ennius certainly kept his en-counter with Homer and revelation of the rernm natura scrupulously distinct from the narrative proper. The poets were also able to strike an acceptable balance between the demands of contemporary subjects and traditional narrative styles. Epic precedent would have accus-tomed readers to the mechanisms for divine intervention that Ennius brings to contemporary battles, but he also managed to accommodate purely political events in the narrative. We know, for example, that he described the public reconciliation of Aemilius Lepidus and Fulvius Nobilior in 179 (16.viii). That event was probably cast as a celebration of domestic harmony, and an equally favorable context was perhaps arranged for the divisive crisis brought about by the Scipionic trials.13 All such descriptions would have entailed par-ticularly circumspect use of the epic apparatus. Ennius proved equal to this task—in part, no doubt, because these events were subsumed into the large and traditional context of the whole poem. As his successors came to write poems of increasingly narrow historicity, however, the dangers of incongruity between the contemporary and the traditional necessarily mounted.

Cicero skirted such dangers in his Marius when he gave to his hero the Homeric omen of the eagle and snake: the association was prob-ably eased by the famous story of the young Marius finding a nest of seven eagles, which were subsequently taken as a portent of his seven consulships.14 Similarly, the Romans" long-standing fascination with omens, not only a commonplace of epic and history but also a feature of civic life, would help justify the elaborate predictions of Urania in

1 2 Ennius was also, of course, an eyewitness to the Ambracian campaign described in Annals 15. For the self-conscious quality of his poem, see Dominik, "Ennius' Annales," 38-48. Naevius' autobiographical statement is attested by Gell. 17.21.45 (BP, fr. 44).

1 3 If indeed he mentioned them. Skutsch tentatively ascribes lines 385-86 to that context, but the attribution requires emendation of the text and a striking anomaly in the poem's chronological structure. (The lines are more frequently attributed to Antiochus in defeat after Magnesia.) Ennius' need to find an appropriate context for so sensitive a subject might explain such a chronological dislocation, but speculation is futile.

14 App. BC 1.75; Plut. Mar. 36. Fr. 18 and 19 may allude to this story. Court-ney's assumption (Fragmentary Poets, 175) that "here the snake represents the Sullans whom Marius massacred on his return" (cf. Pease ad Div. 1.106, p. 292) seems to me overly allegorical, though we can never be sure that no Roman would have made that association.


De consulatu suo. Yet the problem grows keener when the poet writes explicitly of contemporary affairs,15 and keener still when he actually writes about himself. Cicero joked about the possibilities in a letter to his brother Quintus written in September 54, when he was working on a second hexameter poem about his exile and recall. He had some political scores to settle, and in the letter he imagines doing so by calling up a truly extraordinary council of the gods.

Itaque mirificum embolium cogito in secundum librum meorum tern-porum includere, dicentem Apollinem in concilio deorum, qualis reditus duorum imperatorum futurus esset, quorum alter exercitum perdidisset, alter vendidisset.

So I am thinking of including a wondrous episode in the second book of my experiences: Apollo in a divine council saying what a return of two commanders there would be, one of whom had lost an army and the other, who had sold one. (Q-fr· 3.1.24)

The satirist Lucilius had created such a council to lampoon L. Cor-nelius Lentulus Lupus, made princeps senatus in 131, and Cicero may well have had that famous scene in mind. It would have been a nice jab at the imperatores Piso and Gabinius. We cannot tell if he ever took his revenge in this way, for the poem De temporibus suis, either with or without this satiric council, apparently never reached the stage of formal circulation among Cicero's reading public.16 There is evidence of a different kind, however, for the use of divine ma-chinery—and the reaction of readers to it—in the poem on his consulship.

Unfortunately for Cicero's reputation, the reception of that poem has been most profoundly shaped for later generations by the brief but wonderfully abusive Invectiva in Ciceronem that has come down to us under the name of Sallust. That pamphlet delights in turning Cicero's epic against him by deftly employing two techniques. The first, perhaps inspired by Piso's effective play on "cédant arma togae" in the controversy of 55, involves the distortion of lines taken out of context. Both victims of this trick have since become notorious:

15 It proved insurmountable—doubtless for political as well as aesthetic reasons—when Marcus and Quintus Cicero contemplated epics on Caesar's Gallic campaigns. Neither poem was written. See Allen, TAPA 86 (1955) 143-59.

16 Harrison, Hermes 118 (1990) 455-63. Piso and Gabinius, the consuls of 58, had earned Cicero's enmity for standing by as Clodius contrived his exile, and they became frequent targets of his abuse. For the Lucilian concilium deorum, fr. 26-54 Marx, see Gruen, Culture and National Identity, 284-85.

Envoi 167

atque is, cum eius modi sit, tarnen audet dicere: "O fortunatam riatam me consule Romam!" Te consule fortunatam, Cicero? Immo vero infelicem et miseram.

Yet, with a character like that, he nevertheless dares to say, "O fortunate Rome, born in my consulship!" Fortunate in your consulship, Cicero? On the contrary, unlucky and wretched. (Cic. 5)

etiamne aures nostras odio tuo onerabis, etiamne molestissimis verbis insectabere? "cédant arma togae, concédât laurea linguae." quasi vero togatus et non armatus ea quae gloriaris confeceris, atque inter te Sullam-que dictatorem praeter nomen imperii quicquam interfuerit.

Will you even so burden our ears with hatred of you, even so pursue us with those truly awful words, "Let arms yield to the toga, let the laurel yield to the tongue"? As if in fact you achieved those things you boast of in a toga and not in arms, and as if the only difference between you and the dictator Sulla was the power of your office. (Cic. 6)

Not even Cicero was actually quite this vain. The quotations are so wickedly effective because they are deliberate misquotations: Cicero wrote laudi and not linguae, te and not me.17 Since the ammunit ion provided was of insufficient caliber, the author of the Invectiva has remanufactured it to his own specifications.

The second technique—and here we come once more to the matter of gods and counci ls—involves giving a literal twist to Ciceronian figures. There are again two examples:

atque haec cum ita sirit, tamen se Cicero dicit in concilio deorum immortalium fuisse, inde missum huic urbi civibusque custodem absque carnificis nomine, qui civitatis incommodum in gloriam suam ponit.

In spite of this, Cicero nevertheless says that he was present in a council of the immortal gods, from which he was sent to this city and its citizens as a guardian, not an executioner, he who turned the state's distress to his own glory. (Cic. 3)

1 7 The parodies may originate here. See Pasoli, Invectiva, 83-84. Laudi is restored from Cic. Pis. 74 and Off. 1.77 (where inferior MSS. in fact read lin-guae). Alas for Cicero, no self-quotation survived to give the lie to me con-sule, which (like linguae) was the version known to Quintilian and of course Juvenal. Yet the poem was in the third person, and Horace's echo at Ep. 2.1.256 reads te principe. See Allen, TAPA 87 (1956) 144-46. Courtney, Fragmentary Poets, 159, nevertheless prints me, thinking the line part of a speech, perhaps a poetic version of 2 Cat.


sed quid ego plura de tua insolentia commemorem? quem Minerva omnis artis edocuit, Iuppiter Optimus Maximus in concilio deorum ad-misit, Italia exulem umeris suis reportavit

What more of your insolence shall I mention: that Minerva taught you all her arts, Jupiter Optimus Maximus admitted you to a council of the gods, Italy brought you back from exile on its shoulders? (Cic. 7)

The images of Cicero as guardian of the state and returning from exile on Italy's shoulders are drawn from his speeches,18 but the rest of these allusions are doubtless to De consulatu suo. Are we then to imag-ine that the poem depicted the consul as actually summoned to Olym-pus for instructions on how to save the state and being tutored in rhetoric on Minerva's knee?

This is certainly the picture implied by the invectiva, a picture created by taking the poem's devices at face value. The truth was un-doubtedly rather different. The council was most likely a dream vision of the kind Ennius employed in his Annals (and Cicero used to good effect in De re publica), and Minerva's instruction probably refers symbolically to Cicero's student days in Athens, to which Urania also refers (fr. 6.71-76). He was very proud of that experience, which may itself be a fault but is not a poetic fault. Nor did his style for expressing that tutelage probably grate on contemporary ears: Vergil seems to echo it in his own description of the seer Nautes:

tum senior Nautes, unum Tritonia Pallas quem docuit multaque insignem reddidit arte

Then old Nautes, whom above all Tritonian Pallas had taught and given renown in many arts (Aen. 5.704-05)

Yet the deliberate misrepresentations of the Invectiva, though providing poor testimony for the poem, are especially interesting and significant evidence for the educated response to epic conventions in the late first century.

What, after all, is the text called Invectiva in Ciceronem? Quin-tilian thought this document, which is cast as a speech by an aggrieved victim of Cicero's slanders, was an authentic work of Sallust. Modern scholars no longer share that view, nor is the text likely to be the précis of a genuine counterattack by Piso or any of Cicero's other political targets in 54, its dramatic date. It is too bookish to be a political pamphlet, too effete in its line of attack.

1 8 Cic. Dom. 141: "di immortalis suorum templorum custodem ac praesidem . . . cum viderent; Red. Sen. 39: "cum . . . Italia cuncta paene suis umeris reportarit;" Dom. 40 (to Clodius): "quod si fieret, dicebas te tuis umeris me custodem urbis in urbem relaturum." For the interpretation of these passages in the Invectiva, see Harrison, Hermes 118 (1990) 458-60.

Envoi 169

The clever parodies of Cicero's own works and stylistic mannerisms— though not, curiously enough, of the In Pisonem—suggest the decla-mation hall rather than the forum, and Syme rightly identified the work as a rhetorical composition of Augustan date.19 It is an example of impersonation, or prosopopoeia, which Quintilian considered a particularly difficult exercise, since it added to the usual requirements of the suasoria the specific demands of historical char-acterization (3.8.49). Reshaping the given facts of a case to fit the needs of the moment was always among the rhetorician's prerog-atives, and the author of the Invectiva proves himself a master of such invention.

What the work thereby loses in historical value, however, it regains as cultural evidence. The mockery of De consulatu suo is less a critique of the historical Cicero, who is only the notional target of the exercise, than of the futility of depicting contemporary events in epic terms. Epic has become a target for fun, not a medium for cele-brating Roman victories. Its conventions and mannerisms are now only weapons in the rhetoricians' arsenal. As a witness to the reception of epic conventions in the Principate, then, the pamphlet's message is clear: the traditional mannerisms have become stale and even ludi-crous. The world of epic can no longer be taken seriously. Cicero's contemporaries, as we have already seen, rejected the poem's mes-sage. His posterity was also prepared to reject its form.

Our history has thus brought us from Andronicus' experiment in making true poetry from his school text of Homer to the sly and bookish conceits with which rhetoricians made a laughingstock of epic's conventions. A tempting evolutionary scheme suggests itself: the epic diction pioneered by Andronicus leads from the success of Naevius' commemoration of one victory, to Ennius' mature and learned celebration of greatness between the city's founding and what was then the present day, to a century and more of gradual decline and disintegration as poets found themselves trapped between the growing rigidities of epic convention and the disappointments of the Roman present. Epic thrived at Rome when its subject was national success, and the earnestness with which it studied that success pro-vided the intellectual tension and excitement needed to save the poetry from panegyric and the poets from sycophancy. Readers could enjoy the novelty of its poetic form, the coherence it gave to their history, and its confirmation of their own greatness. As the Republic faltered, the genre naturally lost its raison d'être. By Cicero's day, serious poets were therefore abandoning traditional epic for lyric, epigram, and short, attenuated epyllia in a truly Alexandrian style.

19 Syme, Sallust, 314-18, The effort by Pasoli, Invectiva, 13-19, to defend its paternità sallustiana is unconvincing.


Cicero himself might disparage what he called their "newfangled" experiments, but these innovations in fact represented an important new direction for Latin poetry, while Cicero's own old-fashioned efforts only stretched beyond credulity a genre already suffering from lack of purpose. And Cicero, of course, was no poet, only a facile versifier with an ego too big for his talent. No wonder Catullus would mock Volusius' creaking monstrosities and, a generation later, rhetoricians would laugh at Cicero just as Vergil and Horace were shying away from songs of kings and battles. They had all lost faith in the certainties that nourished Naevius and Ennius.

This account is not entirely false.20 A genre so closely associated with the representation of public triumphs faced a dim future in a world that came to lack a suitably unambiguous record of success. Thus when Catullus chose to begin his own hexameter narrative about Peleus and Thetis with a nod to Ennius, he naturally looked not to the Annals, but to a tragedy, Medea exul. In both subject and manner, Catullus is far from the world of traditional Roman epic. This is why his pines, which swim Neptune's flowing billows as the ship Argo, are kept scrupulously distinct from the pinus procerae that fueled the Roman pyre at Heraclea.21 Yet the very possibility of such an experiment in Cicero's own day reminds us that the picture of epic decline, however enticing, remains conceptually flawed: we cannot in fact treat the canon of Republican epics as a closed system. Ennius' Annals was no telos, nor have we witnessed a self-contained process of generation and decay. We have simply recorded the inevitable sequence of changes as poets explored the artistic possibilities of epic form and their audiences read their own aspirations—and eventually their own disillusion—in the poems that they made. The value we put on those experiments, the history we make of them, depends entirely on our point of view. By the end of the Republic, that view was certainly unattractive, but it did not remain so.

The serious and informed embrace of Greek aesthetics by poets like Catullus and Cinna pointed the way. What happened next was precisely the situation Eliot described in "Tradition and the Indi-vidual Talent": Vergil went on to produce "the new (the really new)

20 Otis, Virgil, 20-26 actually constructs a history very like this in order to demonstrate what he calls "the obsolesence of epic" before Vergil. The only significant difference is his conviction that no Latin poetry was artistically successful until Parthenius brought true Hellenism to Rome in the first century.

2 1 Given the verbal similarities of Medea 208-16 and Annals 175-79, Catullus' ability in Poem 64 to echo the former without the latter is especially striking. The Ennian vocabulary noted by Thomas, AJP 103 (1982) 156-57, is more likely to be tragic irr origin than epic.

Envoi 171

work of art" that not only altered the existing order but rejuvenated the genre for succeeding generations. He made from the old wreck of Republican epic a new vehicle capable of exploring the social and moral uncertainties that by the Augustan age increasingly preoccupied the educated classes at Rome. Encouraged by his example, epic would then assume a new sense of purpose, which brought changes not only to the way new epics came to be written, but also to the way the old ones came to be read. All this, however, is more properly the first chapter of another story.22 The story of Republican epic ends, like the Republic itself, with Cicero.

22 The outlines of this story are now elegantly traced by Hardie, The Epic Successors of Virgil

Works Cited Aicher, P. J. "Ennian Artistry: Annals 175-179 and 78-83 (Sk.)." C/85 (1989/90)

218-24. Allen, W., Jr. "O fortunatam natam . . TAPA 87 (1956) 130-46.

. "The British Epics of Quintus and Marcus Cicero." TAPA 86 (1955) 143-59.

. "Cicero's Conceit." TAPA 85 (1954) 121-44. Allen, W. S. Accent and Rhythm. Cambridge: Cambridge University Press,

1973. Asein, S. O. "Derek Walcott and the Great Tradition." The Literary Criterion

16 (1981) 18-30. Astin, A, Cato the Censor. Oxford: Clarendon, 1978. Austin, R. G., ed. P. Vergili Maronis Aeneidos Liber Sextus. Oxford:

Clarendon, 1977. Badian, E. "Ennius and His Friends." In Fondation Hardt Entretiens, Tome

XVII: Ennius, ed. Ο. Skutsch, 149-208. Vandoeuvres-Geneva: Fondation Hardt, 1972.

Baisdon, J. P. V. D. "L. Cornelius Scipio: A Salvage Operation." Historia 21 (1972) 224-34.

Barchiesi, M. Nevio epico. Padua: Antonio Milani, 1962. Bardon, H. La littérature latine inconnue, tome 1: L'époque républicaine.

Paris: Klincksieck, 1952. Bartalucci, A. "La sperimentazione enniana dell'esametro e la tecnica del

saturnio." SCO 17 (1968) 99-122. Bateson, F. W. "Literary History: Non-Subject Par Excellence." NLH 2 (1970)

115-22. Beare, W. The Roman Stage, 3rd ed. London: Methuen, 1968. Berlin, L Vico and Herder, Two Studies in the History of Ideas. New York:

Viking, 1976. Binder, G. "Vom Mythos zur Ideologie: Rom und seine Geschichte vor und

bei Vergil." In Mythos, ed. G. Binder and B. Effe, 137-61. Bochumer Altertumswissenschaftliches Colloquium, no. 2. Trier, 1990.

Blänsdorf, J. "Metrum und Stil als Indizien für vorliterarischen Gebrauch des Saturniers." In Studien zur vorliterarischen Periode im frühen Rom, ed. G. Vogt-Spira, 41-69. ScriptOralia 12, Reihe A: Altertumswissenschaftliche Reihe, Band 2. Tübingen: Gunter Narr, 1989.

Brink, C. O. Horace on Poetry: Epistles Book II. Cambridge: Cambridge University Press, 1982.

Brock, S. "Aspects of Translation Technique in Antiquity." GRBS 20 (1979) 69-87.

Brunt, P. A. The Fall of the Roman Republic and Related Essays. Oxford: Clarendon, 1988.

172

Works Cited 173

Büchner, Κ. "Livius Andronicus und die erste künstlerische Übersetzung der europäischen Kultur/' SO 54 (1978) 37-70.

Burkert, W. "Hellenistische Pseudopythagorica." Philologiis 105 (1961) 226-46.

Cairns, F. Tihullus; A Hellenistic Poet at Rome, Cambridge: Cambridge Uni-versity Press, 1979.

Calder, W. Μ., III. "Karl Lehrs' Ten Commandments for Classical Philol-ogists." CW 74 (1980/81) 227-28.

Cameron, A. Callimachus and His Critics. Princeton: Princeton University Press, 1995.

. "How Thin Was Philitas?" CQ 41 (1991) 534-38. Cichorius, C. Römische Studien. 1922. Reprint, Stuttgart: Teubner, 1961. Cohen, R. "Innovation and Variation: Literary Change and Géorgie Poetry."

In Literature and History, 3 - 4 2 . Los Angeles: Clark Memorial Library, 1974.

Colace, P. R., ed. Choerili Samii Reliquiae. Rome: Bretschneider, 1979. Cole, T. "The Saturnian Verse." YCS 21 (1969) 3-73.

. "In Response to Nevio Zorzetti." CJ 86 (1990/91) 377-82. Collingwood, R. G. The Idea of History. 1946. Reprint, Oxford: Oxford Uni-

versity Press, 1971. Conte, G, Β. The Rhetoric of Imitation. Cornell Studies in Classical Philology

44. Ithaca: Cornell University Press, 1986. . Virgilio. Il genere e i suoi confini. Milan: Garzanti, 1984.

Cova, P. V. "Livio e la repressione dei Baccanali." Athenaeum 52 (1974) 82-109.

D'Anna, G. "Il problema delle origini della poesia latina nel Bruto di Cicerone." Ciceroniana 5 (1984) 81-90.

Degl'Innocenti Pierini, R. Studi su Accio. Quaderni dell'Instituto di filologia classica "Giorgio Pasquali" dell' Université degli Studi di Firenze, 1. Florence: Editrice universitaria, 1980.

Degrassi, A. Inscriptiones latinae liberae rei publicae imagines. Berlin: De Gruyter, 1965.

Dominik, W. J. "From Greece to Rome: Ennius' Annales." in Roman Epic, ed. A. J. Boyle, 37-58. London: Routledge, 1993.

Douglas, A. E. "Intellectual Background of Cicero's Rhetorical ANRW 1.3 (1973) 95-138.

Ducrot, O., and T. Todorov. Encyclopedic Dictionary of the Sciences of Language, tr. C. Porter. Baltimore: Johns Hopkins University Press, 1979.

Earl, D. C. "Political Terminology in Plautus." Historia 9 (1960) 234-43. Eisenhut, W. "Die römische Gefängnisstrafe. ANRW 1.2 (1972) 268-82. Eliot, T. S. Selected Essays. 1950. Reprint, New York: Harcourt, 1964. Empson, W. "George Herbert and Miss Tuve." Kenyon Review 12 (1950)

735-38. Enk, P. J., ed. Plauti Truculentus cum prolegomenis, notis criticis, commen-

tario exegetico. 2 vols, Leiden: Brill, 1953. Ernout, A. Recueil de textes latins archaïques, 3rd ed. Paris: Klincksieck, 1966. Ewbank, W. W. The Poems of Cicero. London: University of London Press,

1933.

174 Works Cited

Fantham, Ε. Comparative Studies in Republican Latin Imagery. Toronto: University of Toronto Press, 1972.

Feeney, D. C. The Gods in Epic. Oxford: Clarendon, 1991. . "History and Revelation in Vergil's Underworld." PCPS 212 (1986) 1-

24. Flintoff, Ε. "Naevius and Roman Satire/' Latomus 47 (1988) 593-603. Fowler, D. "Deviant Focalisation in Virgil's Aeneid." PCPS 36 (1990) 42-63. Fraenkel, E. Horace, 1957. Reprint, Oxford: Oxford University Press, 1970.

. "The Giants in the Poem of Naevius." JRS 44 (1954) 14-17. _ . "The Pedigree of the Saturnian Metre." Eranos 49 (1951) 170-71.

. "Naevius." RE Suppl. 6 (1935) 622-40.

. "Livius Andronicus." RE Suppl. 5 (1931) 598-607. Fränkel, H. "Griechische Bildung in altrömischen Epen II." Hermes 70 (1935)

59-72. . "Griechische Bildung in altrömischen Epen." Hermes 67 (1932) 303-

11. Fraser, P. M, Ptolemaic Alexandria. 3 vols. Oxford: Clarendon, 1972. Fromm, H. Academic Capitalism and Literary Value. Athens: University of

Georgia Press, 1991. Fruyt, M. "Mots fragmentés chez Ennius." Glotta 69 (1991) 243-46. Gabba, E. "Considerazioni sulla tradizione letteraria sulle origini della Re-

pubblica." In Fondation Hardt Entretiens XIII: Les origines de la ré-publique romaine, 133-74. Vandoeuvres-Geneva: Fondation Hardt, 1967.

Galinsky, G. K., ed. The Interpretation of Roman Poetry: Empiricism or Hermeneutics? Frankfurt: Lang, 1992.

Garbarino, G. Roma e la filosofia greca dalle origini alia fine del ii secolo A.C. 2 vols. Turin: Paravia, 1973.

Garton, C. Personal Aspects of the Roman Theatre. Toronto: University of Toronto Press, 1972.

. "Revised Register of Augustan Actors." ANRW 2.30.1 (1982) 580-609. Gilula, D. "How Rich Was Terence?" SCI 8/9 (1989) 74-78.

. "Who's Afraid of Rope-Walkers and Gladiators?" Athenaeum 59 (1981) 29-37.

_ . "Terence's Hecyra." SCI 5 (1979/80) 137-57. Goldberg, S. M. Understanding Terence. Princeton: Princeton University

Press, 1986. . Review of Skutsch, The Annals of Quintus Ennius. CJ 84 (1988) 59-61.

Gordon, A. E. An Illustrated Introduction to Latin Epigraphy. Berkeley: University of California Press, 1983.

Graff, G. Professing Literature, an Institutional History. Chicago: Univerity of Chicago Press, 1987.

Gratwick, A. S. "Ennius' Annales." In The Cambridge History of Classical Literature II: Latin Literature, edd. E. J. Kenney and W. V. Clausen, 60-76. Cambridge: Cambridge University Press, 1982.

Review of Skutsch, The Annals of Quintus Ennius. CR 37 (1987) 163-69.

Greenblatt, S. Shakespearean Negotiations. Berkeley: University of Cali-fornia Press, 1988.

Works Cited 175

Gruen, E. S. Culture and National Identity in Republican Rome. Ithaca: Cornell University Press, 1992.

. Studies in Greek Culture and Roman Policy. Leiden: Brill, 1990. The Hellenistic World and the Coming of Rome. 2 vols. Berkeley:

University of California Press, 1984. Hainsworth, J, B. The Idea of Epic. Berkeley; University of California Press,

1991. Hardie, P. R. The Epic Successors of Virgil Cambridge: Cambridge Uni-

versity Press, 1993. _ Virgil's Aeneid, Cosmos and Imperium. Oxford: Clarendon, 1986. Harrer, G. A. "Some Verses of Cicero." SP 25 (1928) 70-91. Harris, W. V. War and Imperialism in Republican Rome. Oxford: Claren-

don, 1979. Harrison, S. J. "Cicero's De temporibus suis: The Evidence Reconsidered."

Hermes 118 (1990) 455-63. Harvey, P. "Historical Topicality in Plautus." CW 79 (1986) 297-304.

. "Historical Allusions in Plautus and the Date of the Amphitruo." Athenaeum 59(1981)480-89.

Heinze, R. Vergils epische Technik, 3rd ed. 1915. Reprint, Leipzig: Teubner, 1928.

Herrmann, L. "Ennius et les livres de Numa." Latomus 5 (1946) 87-90. Heubeck, Α., S. West, and J. Β. Hainsworth. A Commentary on Homer's

Odyssey. Volume I. Introduction and Books i-viii. Oxford: Clarendon, 1988.

Horsfall, N. "The Collegium Poetarum." BICS 23 (1976) 79-95. Howarth, W. D. "L'Alexandrin classique comme instrument du dialogue

théatral." In Dramaturgies Langages Dramatiques, Mélanges pour Jacques Scher er, 341-54. Paris: Nizet, 1986.

Humphrey, J. H. Roman Cireuses. Berkeley: University of California Press, 1986.

Hutchinson, G. Ο. Hellenistic Poetry. Oxford: Clarendon, 1988. Jauss, H. R. Toward an Aesthetic of Reception. Trans. T. Bahti. Theory and

History of Literature, 2. Minneapolis: University of Minnesota Press, 1982. Jocelyn, H. D. "Romulus and the di genitales (Ennius, Annales 110-111

Skutsch)." In Studies in Latin Literature and its Tradition in Honour of C. O. Brink, ed. J. Diggle et al., 39-65, suppl. vol. 15. Cambridge: The Cambridge Philological Association, 1989.

_. "Greek Poetry in Cicero's Prose Writing." YCS 23 (1973) 61-111. "The Poems of Quintus Ennius." ANRW 1.2 (1972) 987-1026.

. "The Poet Cn. Naevius, P. Cornelius Scipio and Q. Caecilius Metellus." Antichthon 3 (1969) 32-17.

Johnson, T., and C. Dandeker. "Patronage: Relation and System." In Patron-age in Ancient Society, ed. A. Wallace-Hadrill, 219-42. London: Rout-ledge, 1989.

Johnson, W. R. Darkness Visible, A Study of Vergil's Aeneid. Berkeley: Uni-versity of California Press, 1976.

Jory, E. J. "Associations of Actors at Rome." Hermes 98 (1970) 224-53. Kearns, J. Μ. "ΣΕΜΝΟΤΗΣ and Dialect Gloss in the Odussia of Livius

Andronicus." AJP 111 (1990) 40-52.

176 Works Cited

Klingner, F. Virgil Bucolica, Georgica, Aeneis. Interpretationen der Werke Virgils. Zurich: Artemis Verlag, 1967.

Klotz, A. "Zu Naevius' Bellum Punicum." RhM 87 (1938) 190-92. Konstan, D. Roman Comedy. Ithaca: Cornell University Press, 1983. Krevans, N. "Ilia's Dream: Ennius, Virgil, and the Mythology of Seduction."

HSCP 95 (1993) 257-71. . "Fighting Against Antimachus, the Lyde and the Aetia Reconsid-

ered." In Hellenistica Groningana I: Callimachus, ed. M. A. Harder et al., 149-60. Groningen, 1993.

Kroll, W. "Rhetorik." RE Suppl. 7 (1940) 1039-1138. Kubiak, D. P. "Aratean Influence in the De consulatu suo of Cicero."

Philologus 138 (1994) 52-66. . "Cicero and the Poetry of Nature." SIFC 8 (1990) 198-214. . "The Orion Episode in Cicero's Aratea" CJ 77 (1981/82) 12-22.

Lattimore, R. "The Wise Advisor in Herodotus." CP 34 (1939) 24-35. Leavis, F. R., ed. A Selection from Scrutiny. 2 vols. Cambridge: Cambridge

University Press, 1968. Lelièvre, F. J. "Sosia and Roman Epic." Phoenix 12 (1958) 117-24. Lembke, J. Bronze and Iron. Old Latin Poetry from Its Beginnings to 100 B.C.

Berkeley: University of California Press, 1973. Leo, F. Geschichte der römischen Literatur Ι. 1913. Reprint, Darmstadt: Wis-

senschaftliche Buchgesellschaft, 1967. . Plautinische Forschungen, 2nd ed. Berlin: Weidmann, 1912. . "Die römische Poesie in der sullanischen Zeit." Hermes 49 (1914)

161-95. Lessing, G. E. Laocoon, An Essay on the Limits of Painting and Poetry. 1766.

Trans, E. A. McCormick. Reprint, Baltimore: Johns Hopkins University Press, 1984.

Lindsay, W. M. Early Latin Verse. Oxford: Clarendon, 1922. Liu, A. "The Power of Formalism: The New Historicism." ELH 56 (1989) 721-

71. Lodge, G. Lexicon Plautinum. 2 vols. 1924/1933. Hildesheim: Olms, 1962. Luck, G. "Naevius and Virgil." ICS 8 (1983) 267-75. Luiselli, Β. II verso Saturnio. Studi di metrica classica, 3. Rome: Edizioni dell'

Ateneo, 1967. MacBain, B. Prodigy and Expiation: Λ Study of Religion and Politics in

Republican Rome. Collection Latomus, vol. 177. Brussels: Latomus, 1982. MacKay, L. A. "The Roman Tragic Spirit." CSC A 8 (1975) 145-62. MacMullen, R. "Hellenizing the Romans (2nd Century B.C.)." Historia 40

(1991) 419-38. Marabini Moevs, M. T. "Le Muse di Ambracia." BdA 12 (1981) 1-58. Mariotti, S. "Concinnat in Naev. Carm. Frg. 39.3 Morel." In Kontinuität und

Wandel: lateinische Poesia von Naevius bis Baudelaire: Franco Munari zum 65. Geburtstag, ed. U, J. Stäche et al., 1-5. Hildesheim: Weidmann, 1986.

. Livio Andronico e la traduzione artistica, 2nd ed. Urbino: Edizioni QuattroVenti, 1985.

_ . "Letteratura latina arcaica e alessandrinismo." Belfagor 20 (1965) 34-48.

Works Cited 177

. Il Bellum Punicum e Y arte di Nevio. Saggio con edizione dei fram-menti del Bellum Poenicum. Rome: Signorelli, 1955.

Marmorale, Ε. V, Naevius Poeta. Introduzione biobibliografica, testo dei frammenti e commento, 2nd ed. Florence: La nuova Italia, 1953.

Martin, J. "Nightmare History: Derek Walcott's Omeros," Kenyon Review 14 (1992) 197-204.

Martina, M. "Grassatores e CarmentariiLabeo 26 (1980) 155-75. . "Ennio, 'poeta cliens/" QFC 2 (1979) 13-74.

Maskell, D. Racine, A Theatrical Reading. Oxford: Clarendon, 1991. Mattingly, H. B. "Naevius and the Metelli." Historia 9 (1960) 414-39. Mazzoli, G. "II frammento enniano laus alit artis e il proemio al XVI libro

degli Annales." Athenaeum 42 (1964) 307-33. Mette, H. J. "Die römische Tragödie und die Neufunde zur griechischen

Tragödie." Lustrum 9 (1964) 5-211. Miller, J. Hillis. "Presidential Address 1986. The Triumph of Theory, the

Resistance to Reading, and the Question of the Material Base." PMLA 102 (1987) 281-91.

Momigliano, A. Alien Wisdom. Cambridge: Cambridge University Press, 1975.

. "An Interim Report on the Origins of Rome." ]RS 53 (1963) 95-121.

. "Perizonius, Niebuhr and the Character of Early Roman Tradition." JRS 47 (1957) 104-14.

Mueller, M. "Knowledge and Delusion in the Iliad." Mosaic 3 (1970) 86-103. Reprinted in Essays on the Iliad. Selected Modern Criticism, ed. J. Wright, 105-23. Bloomington: Indiana University Press, 1978.

Nagy, G. Comparative Studies in Greek and Indie Meter. Cambridge: Harvard University Press, 1974.

Newman, J. K. The Classical Epic Tradition. Madison: University of Wiscon-sin Press, 1986.

. Augustus and the New Poetry. Collection Latomus, vol. 88. Brussels: Latomus, 1967.

Nicolet, C. The World of the Citizen in Republican Rome. Trans. P. S. Fall a. Berkeley: University of California Press, 1980.

Norden, Ε., ed. P. Vergilius Maro, Aeneis Buch VI, 4th ed. 1927. Reprint, Darmstadt: Wissenschaftliche Buchgesellschaft, 1957.

North, J. A. "Religious Toleration in Republican Rome." PCPS 25 (1979) 85-103.

Novick, P. That Noble Dream: The 'Objectivity Quest' and the American Historical Profession. Cambridge: Cambridge University Press, 1988.

Ogilvie, R. M. A Commentary on Livy, Books 1~5. Oxford: Clarendon, 1965. Otis, B. Virgil, A Study in Civilized Poetry. Oxford: Clarendon, 1964. Pailler, J.-M. Bacchanalia, La répression de 186 av. J.-C. à Rome et en Italie.

Bibliothèque des écoles françaises d'Athènes et de Rome, fasc. 270. Rome: Diffusion de Boccard, 1988.

Palmer, R. E. A. Roman Religion and Roman Empire. Philadelphia: Uni-versity of Pennsylvania Press, 1974.

Panitschek, P. "Numa Pompilius als Schüler des Pythagoras." GB 17 (1990) 49-65.

178 Works Cited

Pascucci, G. "Lo scoppio delle ostilità nelle guerra annibalica secondo il racconto degli 'Annali' di Ennio." In Poesia latina in frammenti, 103-15. Université di Genova Facoltà di Lettere, Miscellanea Filologica. Genoa: Université, 1974.

Pasoli, E., ed. Appendix Saîhtstiana. invectiva in M. Tuïïium Ciceronem. Bologne: Pàtron, 1989.

Pasquali, G. Preistoria délia poesia romana, con un saggio introduttivo di Sebastiano Timpanaro. Florence: Sansoni, 1981.

Patterson, L. "Literary History." In Critical Terms for Literary Study, ed. F. Lentricchia and T. McLaughlin, 250-62, Chicago: University of Chicago Press, 1990.

Pavlock, B. "Epic and Tragedy in Vergil's Nisus and Euryalus Episode." TAPA 115 (1985) 207-24.

. Eros, Imitation, and the Epic Tradition. Ithaca: Cornell University Press, 1990.

Pechter, E. "The New Historicism and Its Discontents: Politicizing Renais-sance Drama." PMLA 102 (1987) 292-303.

Perkins, D. Is Literary History Possible? Baltimore: Johns Hopkins University Press, 1992.

Pfeiffer, R. History of Classical Scholarship, vol. 1. Oxford: Clarendon, 1968. Phillips, C. R., III. "Poetry Before the Ancient City: Zorzetti and the Case of

Rome." C/86 (1990/91) 382-9. Phillips, E. D. "Odysseus in Italy." JHS 73 (1953) 53-67. Pickard-Cambridge, A. W. The Dramatic Festivals of Athens. 2nd ed. revised

by J. Gould and D. M. Lewis. London: Oxford University Press, 1968. Porter, C. "History and Literature: 'After the New Historicism/" NLH 21

(1990) 253-72. Préaux, J. "Caton et l'ars poetica." Latomus 25 (1966) 710-25. Prowse, K. R. "Numa and the Pythagoreans: A Curious Incident." G&R 11

(1964) 36-42. Pucci, P. "The Song of the Sirens." Arethusa 12 (1979) 121-32. Quinn, K. "Poet and Audience in the Augustan Age." ANRW 2.30.1 (1982)

75-180. Rawson, E. "Theatrical Life in Republican Rome and Italy." Papers of the

British School at Rome 53 (1985) 97-113. Romany, F. "Shakespeare and the New Historicism." Essays in Criticism 39

(1989) 271-87. Rosen, K. "Die falschen Numabücher." Chiron 15 (1985) 65-90. Rowell, H. T. "The Original Form of Naevius' Bellum Punicum. " AJP 68

(1947) 21-46. Review of S. Mariotti, II Bellum Poenicum e Varte di Nevio. AJP 78

(1957) 420-27. Sailer, R. P. "Patronage and Friendship in Early Imperial Rome: Drawing the

Distinction." In Patronage in Ancient Society, ed. A. Wallace-Hadrill, 49-62. London: Routledge, 1989.

. Personal Patronage under the Early Empire. Cambridge: Cambridge University Press, 1982,

Sandbach, F. H. "How Terence's Hecyra Failed." CQ 32 (1982) 134-35.

Works Cited 179

Schama, S. Citizens: A Chronicle of the French Revolution. New York: Knopf, 1989.

Dead Certainties (Unwarranted Speculations). New York: Knopf, 1991.

Schanz, M., and C. Hosius. Geschichte der römischen Literatur, vol. 1. Munich: Beck, 1927.

Scholz, U. W. "Der Scipio des Ennius." Hermes 112 (1984) 183-99. Schwarte, Κ.-Η. "Naevius, Ennius und der Beginn des Ersten Punischen

Kriegs." Historia 21 (1972) 206-23. Scullard, Η. H. Roman Politics 220-150 B.C., 2nd ed. Oxford: Clarendon, 1973. Shatzman, I. "The Roman General's Authority over Booty." Historia 21 (1972)

177-205. Sheets, G. A. "The Dialect Gloss, Hellenistic Poetics and Livius Andronicus."

AJP 102 (1981) 58-78. Silverman, S. "Patronage as Myth." In Patrons and Clients in Mediterranean

Societies, ed. E. Gellner and J. Waterbury, 7-19. London: Duckworth, 1977. Skutsch, O. "Notes on Ennius II." BICS 23 (1976) 73-79.

Studia Enniana. London: Athlone, 1968. . Review of Mariotti, II Bellum Poenicum e Γ arte di Nevio. CR 72 (1958)

45-48. . Review of Marmorale, Naevius poeta. CR 65 (1951) 174-77.

Slater, N. "The Fictions of Patriarchy in Terence's Hecyra," CW 81 (1988) 249-60.

Slawinski, J. "Reading and Reader in the Literary Historical Process." NLH 19 (1988) 521-39.

Solmsen, F. "'Aeneas Founded Rome with Odysseus.'" HSCP 90 (1986) 93-110.

Steiner, H. R. Der Traum in der Aeneis. Bern: Haupt, 1952. Striedter, J. "Zur formalistischen Theorie der Prosa und der literarischen

Evolution." Texte der russischen Formalisten, vol. 1, ix-lxxxiii. Munich: Fink, 1969.

Strzelecki, L. "Naevius and Roman Annalists." RFÎC 91 (1963) 440-58. . De Naeviano Belli Punici carmine quaestiones selectae. P o l s k a

Akademja Umiejetnosci, Rozprawy Wydzialu Filologicznego, LXV, 2. Krakow, 1935.

Suerbaum, W. Untersuchungen zur Selbstdarstellung älterer römischer Dichter. Spudasmata 19. Hildesheim: Olms, 1968.

Suetonius. De grammaticis et rhetoribus, Ed. with commentary by R. Α. Kaster. Oxford: Clarendon, 1995.

Syme, R. Sallust. Berkeley: University of California Press, 1964. . The Roman Revolution. 1939. Reprint, Oxford: Oxford University

Press, 1960. Tamm, Β. "Le Temple des Muses à Rome." Opuscula Romana 3 (1961) 157-

67. Taplin, Ο. "Derek Walcott's Omeros and Derek Walcott's Homer." Arion 3

ser. 1 (1991) 213-26. Taylor, L. R. "The Opportunities for Dramatic Performances in the Time of

Plautus and Terence." ΤΑΡΑ 68 (1937) 284-304.

180 Works Cited

Thomas, R. F. "Catullus and the Polemics of Poetic Reference (Poem 64.1-18)." AJP 103 (1982) 144-64.

Thulin, C Italische sakrale Poesie und Prosa. Berlin: Weidmann, 1906. Timpanaro, S. "Ancora su Ennio e le lacrime di Omero." RFIC 119 (1991) 5 -

43. Timpe, D. "Fabius Pictor und die Anfänge der römischen Historiographie."

ANRW 1.2 (1972) 928-69. Townend, G. B. "The Poems." In Cicero, ed. T. A. Dorey, 109-34. London:

Routledge, 1964. Traglia, A. La lingua di Cicerone poeta. Bari: Adriatica, 1950. Traîna, A. Vortit barbare. Le traduzioni poetiche da Livio Andronico a

Cicerone, 2nd ed. Rome: Edizioni dell'Ateneo, 1974. Tuve, R. "On Herbert's Sacrifice." Kenyon Review 12 (1950) 51-75. Tynjanov, J,, and R. Jakobson. "Problems in the Study of Literature and

Language." In Russian Formalist Criticism, ed. L. Matejka and K. Pomor-ska, 79-81. Cambridge: Harvard University Press, 1971.

Uhlig, C. "Tradition in Curtius and Eliot." CL 42 (1990) 193-207. Van Sickle, J. "The Elogia of the Cornelii Scipiones and the Origin of Epigram

at Rome." AJP 108 (1987) 41-55. . "The First Hellenistic Epigrams at Rome." In Vir Bonus Discendi

Peritus, Studies in Celebration of Otto Skutsch's Eightieth Birthday, ed. N. Horsfall, 143-56. BICS Suppl. 51 (1988).

Veeser, Η. Α., ed. The New Historicism. New York: Routledge, 1989. Vine, B. "Corde capessere (Ennius, Ann. 42Sk.)." Glotta 67 (1989) 123-26.

. Studies in Archaic Latin Inscriptions. Innsbrucker Beiträge zur Sprachwisenschaft, Band 75. Innsbruck: Institut für Sprachwissenschaft, 1993.

von Albrecht, M. Römische Poesie. Texte und Interpretationen. Heidelberg: Stiehm, 1977.

. "Ein Pferdegleichnis bei Ennius/' Hermes 97 (1969) 333-45. Walcott, Derek. "What the Twilight Says: An Overture." In Dream on

Monkey Mountain and Other Plays, 3-40. New York: Farrar Straus, 1970. . Omeros. New York: Farrar Straus, 1990.

Wallace-Hadrill, A. "Patronage in Roman Society: From Republic to Empire." In Patronage in Ancient Society, ed. A. Wallace-Hadrill, 63-87. London: Routledge, 1989.

Warmington, E. H., ed. Remains of Old Lathi. Loeb Classical Library. 4 vols. Cambridge: Harvard University Press, 1936-1940.

Waszink, J. H. "Anfangsstadium der römischen Literatur." ANRW 1.2 (1972) 869-927.

. "Camena." In Opuscula Selecta, 89-98. Leiden: Brill, 1979 - C&M 17 (1956) 139-48.

Weimann, R. Structure and Society in Literary History. Expanded ed. Baltimore: Johns Hopkins University Press, 1984.

Weingrod, A. "Patronage and Power." In Patrons and Clients in Mediter-ranean Societies, ed. Ε. Gellner and J. Waterbury, 41-51. London: Duckworth, 1977.

Wellek, R. Concepts of Criticism. New Haven: Yale University Press, 1963.

Works Cited 181

. The Attack on Literature. Chapel Hill: University of North Carolina Press, 1982.

Wellek, R., and A. Warren. Theory of Literature, 3rd ed. New York: Harcourt, Brace & World, 1956.

West, M. L. Greek Metre. Oxford: Clarendon, 1982. "Three Topics in Greek Metre." CQ 32 (1982) 281-97.

White, H. The Content of the Form. Baltimore: Johns Hopkins University Press, 1987.

. "Historical Pluralism." CI 12 (1985/86) 480-93. White, P. "Horace, Epistles 2.1.50-54." TAPA 117 (1987) 227-34.

. Promised Verse. Poets in the Society of Augustan Rome. Cambridge: Harvard University Press, 1993.

Wigodsky, M. Vergil and Early Latin Poetry. Hermes Einzelschriften, Heft 24. Wiesbaden: Steiner, 1972.

Wilkinson, L. P. Golden Latin Artistry. Cambridge: Cambridge UP, 1963. Williams, G. W. Tradition and Originality in Roman Poetry. Oxford:

Clarendon, 1968. . "What Is Happening to Interpretation of Virgil's Aeneid?" Todd

Memorial Lectures 9 (1976). Sydney: University of Sydney, 1982. "The Genesis of Poetry in Rome." In The Cambridge History of

Classical Literature II: Latin Literature, ed. E. J. Kenney and W. V. Clausen, 53-59. Cambridge: Cambridge University Press, 1982.

Wimmel, W. "Vergil und das Atlantenfragment des Naevius." WS 83 (1970) 84-100.

Witte, Κ. "Der Hexameter des Ennius." RhM 69 (1914) 205-32. Woodman, A. J. Rhetoric in Classical Historiography. London: Croom Helm,

1988. Wright, J. Dancing in Chains, The Stylistic Unity of the Comoedia Palliata.

Papers and Monographs of the American Academy in Rome 25. Rome: AAR, 1974.

Wyss, B., ed. Antimachi Colophonii Reliquiae. 1935. Reprint, Hildesheim: Weidmann, 1974.

Zetzel, J. E. G. "Ennian Experiments." AJP 95 (1974) 137-40. Ziegler, K. Das hellenistiche Epos, 2nd ed. 1934. Reprint, Leipzig; Teubner,

1966. Zorzetti, N. "Poetry and Ancient City: The Case of Rome." CJ 86 (1991) 311-

29. , "The Carmina Convivalia." In Sympotica, ed. O. Murray, 289-307.

Oxford: Clarendon, 1990. . La pretesta e il teatro latino araico. Naples: Liguori, 1980.

Concordances

LIVIUS ANDRONICUS

K. Büchner. Fragmenta poetarum Latinorum epicorum et lyricorum . . . post W. Morel. Leipzig: Teubner, 1982.

W. Morel. Fragmenta poetarum Latinorum epicorum et lyricorum. Leipzig: Teubner, 1927.

Büchner Morel Büchner Morel

1 1 21 23 2 2 22 24 3 3 23 11 4 38 24 26 5 34 25 27 6 4 26 28 7 5 27 30 8 7 28 36 9 8 29 37

10 10 30 16 11 13 31 12 12 14 32 29 13 15 33 31 14 17 34 26 15 18 35 33 16 9 36 43 17 19 37 6 18 20 38 25 19 21 39 32 20 22 40 35

182

Concordances 183

NAEVIUS

L. Strzelecki, Cn. Naevii Belli Punici carmen. Leipzig: Teubner, 1964. W. Morel, Fragmenta poetarum Latinorum epicorum et lyricorum. Leipzig:

Teubner, 1927.

Strzelecki

1 2 3 4 5 6 7 8

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

Morel

1

31 32 19 4 5 11 28 13a 12 21 18 17 13 16 20 14 15 29 30 6 10 23 22 3

Strzelecki

26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50

Morel

24 25 27 26 37 38 39 35 36 40 41 45 47 48 44 46 49 50 2 33 42 43 51 54 55

Strzelecki

51 52 53 54 55 56 57 58 59 60 61 62

63 64 65 66 67

Morel

56 58 57 7 52 9 53

Liv. 39 63 8 61 59 60

34

184 Concordances

ENNIUS

Ο. Skutsch. The Annals of Quintus Ennius. Oxford: Clarendon, 1985. J. Vahlen. Ennianae Poesis Reliquiae, 2nd ed. Leipzig: Teubner, 1903.

Skutsch Vahlen Skutsch Vahlen

1 1 63-64 67 2 5 65 68 3 6 66-68 70-72 4 8 69-70 73-74 5 69 71 75 6-7 13-14 72-91 77-96 8-10 10-12 92 97 11 15 93 76 12-13 3-4 94-95 99-100 14 17 96 105 15-16 18-19 97 98 17 20 98 101 18 21 99 104 19 22 100 117 20 23 101 106 21 25 102-3 107-8 22 24 104 109 23-24 26-27 105-9 110-14 25 28 110-11 115-16 26 54 112 118 27 29 113 119 28-29 30-31 114-15 120-21 30 34 116-18 122-24 31 33 119 125 32 32 120 126 33 102 121 127 34-50 35-51 122 136 51 60 123 129 52 61 124 137 53 64 125-26 138-39 54-55 65-66 127 143 56 58 128-29 144-45 57 59 130 134 58-59 52-53 131 130 60 55 132 133 61-62 56-57 133 132

Concordances 185

Skutsch

134 135 136 137 138 139-40 141 142 143-44 145 146 147 148 149 150 151 152 153 154-55 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170-72 173-74 175-79 180-82 183-90 191-94 195-96 197-98 199-200

Vahlen

128 131 135 149 150 147-48 151 152 153-54 159 146 155 156 158 157 161 162 163 501-2 500 169 170 166 167 172 171 173 174 177 178 179 182 186 183-85 416-17 187-91 192-93 194-201 208-10 205-6 180-81 202-3

Skutsch

201 202 203-4 205 206-10 211-12 213 214 215 216 217 218 219 220-21 222 223-24 225-26 227-28 229 230 231 232 233 234-35 236-37 238 239 240-41 242 243-14 245 246 247 248-53 254 255 256-57 258 259-60 261 262 263

Vahlen

204 207 175-76 211 213-17 218-19 222 221 265 223 225 230 231 521-22 260 262-63 266-67 164-65 276 256 524 258 257 280-81 232-33 252 253 62-63 224 254-55 261 259

268-73

294

Var.4-5 287 288-89 278 279 277

186 Concordances

Skutsch

264 265 266 267 268-86 287 288 289 290 291 292 293 294-96 297 298 299 300-301 302 303 304-8 309 310 311 312-13 314 315 316 317-18 319-20 321 322-23 324 325 326-28 329 330-31 332-34 335 336 337-39 340-42 343 344-45

Vahlen

282 283 284 285 234-51 286 292 298 295 296 297 299 227-29 290 300 301 319-20 302 325 303-8 310 311 323 312-13 314 315 316 317-18 321-22 324 326-27 329 330 343-45 331 332-33 340-42 338 334

335-37

348 358-59

Skutsch Vahlen

346 347 348 349 350 351 352 353 354 355 356 357 358 359-60 361 362 363-65 366-68 369 370 371-73 374 375 376 377-78 379-80 381 382-83 384 385-86 387 388 389-90 391-98 399-400 401 402 403 404-5 406 407 408 409

Vahlen

328 346^7 339 349 351 353 354 355 366 363 364 356 365 360-61 352 362 370-72 367-69 378 606 381-83 380 389 386 384-85 387-88 390 391-92 393 394-95 396 397 398-99 401-8 409 425 426 410 411-12 413 423 421 422

Concordances 187

Skutsch

410 411 412 413 414 415-16 417 418 419 420 421 422 423 424 425-26 427 428 429 430 431 432-34 435-36 437-38 439 440 441 442 443 444 445 446-47 448 449 450 451 452 453 454 455 456 457 458

Vahlen

414 415 427 428 433 434-35 418 419 420 424 429 430 431 432 436-37 438 442 440 441 439 443-45 446-47 450-51 452 454 453 7 160 456 491 457-58 264 617 459 140 141 142 526 168 373 488 564

Skutsch

459-60 461 462 463-64 465 466 467 468 469-70 471 472 473 474-75 476 477 478 479 480 481 482 483-84 485-86 487 488-89 490 491 492 493 494-95 496-97 498 499 500-501 502 503 504 505 506 507 508 509 510

Vahlen

356-57 455 226 484-85 486 479 480 495 561-62 503 220 293 274-75 535 496 523 507 587 584 379 472-73 513-20 2 461-62 350 460 463 464 465-66 467-68 103 469 470-71 474 475-76 477 478 481 482 483 487 489

188 Concordances

Skutsch Vahlen Skutsch Vahlen

511 490 621 624 512 492 622 Inc. 28 513 493 623 Inc. 35 514 494 515-16 497-98 Op. Inc. 1 16 517 499 2 Seen. 424 518 309 3-4 Var. 1-2 519 400 5 Var.3 520-21 448-49 6 Var. 6 522-23 374-75 7 Var. 7 524 376 8 Var. 8 525 377 9 Var. 47 589 578 10 Inc. 2 590 579 11 Inc. 3 591 580 12 Inc. 4 592 581 13 Inc. 6 593 582-83 14 Inc. 11 594 585 15-16 Inc. 15 595 586 17 Inc. 16 596 588 18 Inc. 17 597 589 19 Inc. 19 598 590 20 Inc. 20 599 591-92 21 Inc. 22 600 593 22-23 Inc. 23 601 594 24 Inc. 24 602 596 25 Inc. 25 603 597 26 Inc. 26 604 598 27 Inc. 27 605 599 28 Inc. 29 606 600 29 Inc. 31 607 601 30 Inc. 33 608 602 31 Inc. 36 609 603 32 Inc. 37 610 605 33 Inc. 38 611 607 34 Inc. 41 612 608 35 Inc. 46 613-14 611-12 36 Inc. 51 615 613 37 Inc. 52 616 617 618 Spur. 1 An. 212 618 619 2 Inc. 46 619 620 3 An. 573 620 621 4 An. 595

Concordances 202

Skutsch Vahlen

5 An. 609 6 An. 610 7-8 An. 614-15 9 An. 616 10 An. 625 11 An. 626 12 An. 627 13 An. 628

Index of Passages

Andronicus, Livius Odussia fr. 1: 64,94 fr. 2: 160 fr. 3: 67 fr. 6: 65,73 fr. 7: 69 fr. 9: 71 fr. 10: 65, 68 fr. 12: 65, 73,160 fr. 13: 65 fr. 15: 71-72 fr. 17: 47 n. 35, 64 n. 11 fr. 18: 47 n. 38, 70 fr. 19: 66 fr. 20: 68 fr. 21: 66 fr. 23: 71,93 fr. 25 : 64 n. 11, 71 fr. 30: 67

Antimachus Thebais fr. 1: 73 fr. 187: 49 Lyde fr, 67: 49

Apollonius Argonautica 3.616-32: 99

Catullus 5.4-6: 76

Cicero De considatu suo fr. 6: 150

fr. 11: 151,167 fr. 12: 151,167 De re publica 1.25: 107 η. 28 4.10.11: 36 Pro Archia 14: 133 22: 16 n. 22,148 26: 133,151 η. 19 Marius fr. 20: 141 Translations fr. 25 fr. 29 fr. 30

140 139 137

Ennius Alexander 34-36: 98 Annais 3: 89 32: 93 34-50: 96-97 53: 161 54-55: 100 56: 100 60: 100 72-91: 103-4,105 104: 93 175-79: 83, 85 183-90: 25,101 206-7: 90 208-9: 90 247-53: 10,102 268-86: 121 299: 92 304-8: 67,94 322-23: 94,117 η. 10

190

Index of Passages 191

324: 14,117 357: 93 371-73: 86 389-90: 113 391-98: 88 444: 161 445: 160 487: 118 498: 76 525: 114 η. 5 535-39: 87 545: 85 Epigram 17,118

Scipio no. 30: 17

Festus 86L: 45 n. 30 446L: 29

Homer Iliad 6.506-11 (15.263-68): 86 7. 89-91: 140 12.200-5: 141 16.106-8: 88 19.226-29: 139 20.285-87: 107 22.304-5: 154 Odyssey 1.1: 64 1.45: 65,160 1.51-52: 66 1.64: 67 1.169: 69 3.108-12: 68 4.513: 66 6.295-99: 71 7.245-46: 66 8.138-39: 69 8.378-80: 67 12.184-91: 138

Horace Epistles 2.1.53-54: 58 η. 1 2.1.69-72: 46 2.1.156-59:56,58 Satires 1.4.60-62: 95

Hostius fr. 1: 135-36 fr. 3: 150 fr. 4: 136,150

Hymn to Dictean Zeus CA 160-61: 59-60

Inscriptions (ILLRP) no. 122 (L. Mummius): 62, 78 no. 309 (Scipio Barbatus): 63, 80 no. 310 (L. Scipio): 62 no. 311 (P. Scipio): 95 no. 312 (Scipio Hispallus): 65

Livy Praef.9: 163 7.2.13: 31. n. 7 38.56.4: 16 n. 22 40.29: 124 40.52.5-7: 78 41.28.8: 77 45.32.11: 40n.23

Macrobius Saturnalia 2.3.6: 35 5.2.16: 23-24 6.2.31: 53

Naevius, Gnaeus Bellum Punicum fr. 1: 73, 80 fr. 2: 80 fr. 3: 79 fr. 4: 13,51,74,136 fr. 5: 75,92 fr. 8: 73, 93

192 Index of Passages 192

fr. 10: 74 fr. 11: 92 fr. 20: 74, 80,136 fr. 23: 54, 75 fr. 25: 76 fr. 29: 105 n. 25 fr. 32: 80, 81 fr. 33: 80 fr. 34: 81 fr. 37: 81 fr. 54: 73 Epitaph 82 Plays 21-23R (Lycurgus): 85 108-10R: 37

Nonius Marcellus 129L (s.v. concinnare): 79 590L (s.v. parcere): 23

Plautus Truculentus 482-86: 111

Ps.-Sallust Invectiva in Ciceronem 3: 167

5: 167 6: 167 7: 168

Terence Hecyra 4-5: 41 33-36: 41 39^2: 41

Vergil Aeneid 1.148-54: 107 I.670-71: 55 2.668-70: 155 5.704-5: 168 6.179-82: 83 6.273-77: 75 8.730-31: 109 9.263-66: 22 9.446-49: 21 9.757-61: 155-56 10.360-61:150 II.751-56: 143 12.911-12: 98

General Index

Accius, 5,6,33,34,135,136 Acculturation of poets, 163-64 Acting companies, 30, 32 Actors, social status of, 30-31,132.

See also Patronage Aelius Stilo, 48,122 Aemilius Paullus, 40, 42 Alexandrianism

in Andronicus, 47-50 in Ennius, 90-91 in Naevius, 52-56, 73

Alliteration, 65, 73, 79, 93-94, 97,144, 149-50

Ambracia, 87,107,112-13,117,153, 155,165 η. 12

Andronicus, Livius biography, 5,28-29,47 first play at Rome, 5, 28 hymn to Juno, 29, 38-39 Odussia

Alexandrian influence on, 47—18, 50, 66 n. 13, 67-68

appeal of, 50-51, 146 characteristics of, 46-47 hexameter version of, 46 n. 34 style of, 64-72

Anicius Gallus, 18, 38-40, 42 Antimachus, 47-50, 53, 73 Apocope, 91-92 Apollonius Rhodius, 54, 98-99 Archaism, 75, 78, 85, 94,106,151 Archias, 53,133-34,147,148 Aristotle, 6 Art history, 6, 8 n. 12 Atlantes at Agrigentum, 13, 52 Atticus, Pomponius, 5, 8,148,149 Augury, 55,103^8,143,153,157,165 Augustan literary taste, 11, 46, 56, 58

n. 1,171

Bacchanalian affair, 128-29 Bardic lays, 43-46 Bateson, F. W., 12, 14, 22 Biographical conventions, 31-32, 36

Caecilii brothers, 22, 118 Caesar, 34, 77,135,154 n. 23,166 n. 15 Caesura

in hexameters, 92-93, 149 in Saturnians, 60-63, 65, 76, 80

Callimachus, 49, 89, 92. See also Alexandrianism

Caiques, 24 n. 36, 64, 73, 81 Camenae, 64, 82, 89,117,118,131 Carmina convivalia. See Bardic lays Cato, 6,8,43-45,71,151,162

attitude toward booty, 127-28 attitude toward Greeks, 125 and Ennius, 114, 23 and Fulvius, 112-13 and Numa's books, 125-26 and Thermus, 111-12

Catullus, 49, 54, 76,123,135,138,170 Chariot race, 104,106-8, 163 n. 8 Chronicle style, 76-79 Cicero

as literary historian, 3, 5-6, 43-44, 46

as poet De considatu sua, 149-53, 166-68,

169 De temporibus sins, 149 n. 15, 166 Ennian echoes, 142-43 epic values, 144-48, 156-57 Marius, 141^3,157, 165 prosody, 138,149-50 reputation, 148-51, 154 translation, 136-41 Vergil's debt to, 143-44

193

194 General Index

Circus Maximus, 38, 39, 107-8 Clientela, 33, 120. See also Patronage Collegium poetarum, 5 n. 3, 28-29, 31

n. 7,132-33 Collingwood, R. G., 19 Conte, G. Β., 10-11,22 Crates of Mallos, 48

Dante, 159, 160 Dido

in Naevius, 54-55 in Vergil, 52, 99,147 n. 13

Dionysius of Halicarnassus, 6 n. 7, 99,119

Dream vision. See Epic, dreams in

Eliot, T. S., 7, 8,9,11,160,170 Elogia Scipionum. See Saturnian

verse Ennius, Q.

Ambracia, 33, 113, 115 Annals

antithesis in, 10, 101-6 Homeric echoes in, 86-88, 113,

146 political bias in, 115-17,118 proemia, 89-90 Saturnian echoes in, 92-95, 160 somnia Pythagorea, 89-90,159 structure of, 95,115-16,131 Vergil's debt to, 22, 24-25

biography, 114-15,122 as dramatist, 84, 101,114-15, 70 Hedyphagetica, 84 learning of, 90-92 and Naevius, 57, 76, 90 patronage of, 14-15, 113-14 portrait bust of, 16-18,115,119 prosody of, 14-15, 83-85 Scipio, 17, 45 n. 32,115

Epic. See also Historical epic anachronism in, 106-8 aristocratic appeal of, 109-10, 113,

123-24,132,147,162-63 decorum of, 68-70, 138-39 divine councils in, 160-61,166 dreams in, 91, 98-99,159,168 Homeric values in, 144—15, 154 oral precedents for, 43-45

Roman values in, 146-48,153-56, 161-63.

treatment of failure by, 81, 153 Epic code, 10-11, 22 Epithets

in Andronicus, 69 in Cicero, 142-43,144 in Ennius, 144 in Naevius, 74-75, 81-82,161

Fabula praetexta, 32, 33, 35, 55-56 Figura etymologica, 78,81,105 n. 25,

151 Flattery in epic, 116,123,133-34,147-

48,151. See also Grassator Formalism, Russian, 9 Fraenkel, E., 13, 47, 59-60 Fragments

attitudes toward, 4,11,12-16, 57, 160

reconstruction from, 22-25, 26, 160 Fulvius Nobilior, M., 14, 39 n. 22

dispute with Aemilius Lepidus, 112,116,118,165

learning of, 117-18,130-31 patronage of Ennius, 112-15

Gigantomachy, 13, 52 Grammatica at Rome, 48 Grassator, 45 n. 30, 122 Greenblatt, S., 18 n. 24,19, 20. See

also New Historicism

Hannibal. See Punic War Hellenism. See Alexandrianism Heraclea, 24, 83, 101,170. See also

Pyrrhus Hercules Musarum, 118,130-31,133 Hesiod, 49, 73, 89,100 Historical epic, 53, 55-56, 146-47,

162-63,165 Homer

epic values of, 144-45,154 incarnation in Ennius, 89-90 translated by Andronicus, 66-72,

146 translated by Cicero, 136-42 translated by Ennius, 86-88,161 n. 2

General Index 195

Horace, 46, 56, 58, 89, 95,151,167 η. 17. See also Augustan literary-taste

Hostius, 135,136,150 Hyperbaton, 94-95

Iambic lampoon, 35 Ilia, 95-101,147, 161 Invectiva in Ciceronem, 166-69

Jauss, H. R. ,8n. 11,146

Ktisis poems, 53-54, 55

Leavis, F. R., 14,163 Lessing, G. E., 108-9 Literary history

challenge of, 12 n. 17, 20, 26 change in, 8,16,171 evolution in, 7-10,11, 43,169-70 Roman efforts at, 5-6 teleology in, 6, 7, 25, 84 n. 1,160,170

Livius Salinator, M., 5 n. 4, 28, 31 Lucilius, 89,132,166 Lucretius, 89, 150,151 Ludi Romani, 28, 39, 40, 106

Macaulay, T. B., 44-45 Macrobius, 22,23, 54, 86, 88 Mariotti, S., 47, 73, 80 n. 27, 81 Menander, 37, 92 Metelli. See Naevius

Naevius, Gn. Bellum Punicum

book divisions, 48, 51 structure, 51-52 style, 73-82

biography, 5, 32-36, 57,162 n. 5 Clastidium, 32-33, 46, 56 (see also

Fabida praetexta) epitaph of, 82,164 Lycurgus, 85 poeta as title, 89, 90 quarrel with Metelli, 33-35 topicality in, 32, 36-37

New Historicism, 19-21 Nisus and Euryalus, 21-24,156

Nonius Marcellus, 23, 79 Numa Pompilius, 18,124-27,129,

131,161

Otis, B., 56-57, 77,170 n. 20

Paean, 45 Patronage, 119-21, 134. See also

Clientela, Poets of Andronicus, 28, 31 of dramatists, 31-32 of Ennius, 112-13,120-21 of Naevius, 32-33

Patroclus, 23, 83,122 Patronymics, 65-66 Piso Caesoninus, L. Calpurnius, 152,

166,168 Plautus, 31, 36, 46, 77, 84,111-12 Poeta cliens, 114,116,117,118,119,

122. See also Patronage Poets, social status, 30-31,132,164.

See also Actors of Andronicus, 28 of Ennius, 114-15,121-23,132 of Naevius, 32 of Terence, 132

Polybius, 38-39,113 Pompey, 34,152,153 Priscian, 13 n. 18,46 n. 34 Punic War

First, 1, 28, 77, 90, 95 162 Second (Hannibalic), 29, 35, 86,102

n. 21,114,121,162 Pyrrhus, 24-25,101-2,105,115,117,

156 Pythagoreanism, 89,126-27,129 n.

30,159

Quintilian, 8 n. 12,47 n. 38, 49 n. 42, 151,167 n. 17,168,169

Racine, J., 61 Rome, foundation myths, 50-51

in Ennius, 95-101, 103-7 (see also Ilia)

in Naevius, 51-52, 54-55 Romulus, 96,103-6,143,157,

161

196 General Index

Saturnian verse, 47, 51, 58, 85 of Andronicus, 64-73 dedications in, 62, 78, 79 description of, 35 η. 17, 59-62, 64 echoed in Ennius, 92-95 epitaphs in, 62-63, 80, 82, 95 of Naevius, 73-82

Scipio Aemilianus, P., 40, 42-43 Scipio Africanus, P., 16-17,18,30, 36-

37,127-28,165 Scipio Asiaticus, L., 39 n. 22,125,127 Scipionic trials, 17,125,127, 165 Scriba, 30-31, 124, See also Actors Servius, 23, 24 n. 36,161 n. 3,162-63 Shakespeare, W., 12-13 Skolia, 45 Suetonius, 47-48 Syme, R., 21,169

Teleology. See Literary history Terence, 32 n. 10, 40-43, 57,132 Theater, Roman

chronology of, 5-6, 28 commercialism of, 30, 32, 37,120,

31-32 conservatism of, 37-39, 43 politics and, 33-34, 36

Topicality. See also Fabula praetexta

in Ennius, 113-14,165 in Naevius, 32, 36-37 in Plautus, 31, 111-12

Tragedy, Roman politics and, 34, 84,115 sense of, 147 style of, 84,98

Translation. See Homer Triumphal songs, 17, 45,115 Turnus, 155-56

Vergil Aeneid, 15, 21-25, 75, 98,108-9,147

n. 13,154-56 and Cicero, 143-44,150,168 and Ennius, 22, 24-25, 83-84,107,

155 heroism in, 154-56 and Naevius, 54-55, 74-75, 161

Walcott, D., 158-59, 163-64 Wellek, R., 7, 9 n. 13, 20 n. 30 White, H., 19,146 n. 12 Wilkinson, L. P., 9-10

goldberg epic in republican rome

Documents