Sander m goldberg constructing literature in the roman republic poetry and its reception 2005

The Republican literature we traditionally call “early”could be as much a product of the late Republic, when texts were firstsystematically collected and put to new social and artistic u

How the Romans came to have a literature, how that literature reflectednative and foreign impulses, and how it formed a legacy for subsequentgenerations have become central questions in the cultural history of the

Republic Constructing Literature in the Roman Republic examines the problem

of Rome’s literary development by shifting attention from Rome’s writers toits readers The literature we traditionally call “early” is seen to be a productless of the mid-Republic, when poetic texts began to circulate, than of thelate Republic, when they were systematically collected, canonized, and put

to new social and artistic uses Imposing on texts the name and function

of literature was thus often a retrospective activity This book explores thedevelopment of this literary sensibility from the Romans’ early interest inepic and drama, through the invention of satire and the eventual enshrining

of books in the public collections that became so important to Horace andOvid

Sander M Goldberg is Professor of Classics at the University of California,

Los Angeles The author of The Making of Menander’s Comedy, Understanding Terence, and Epic in Republican Rome, he has held fellowships from the

American Council of Learned Societies, the National Endowment forthe Humanities, and the Fulbright-Hays Commission He is a past editor

of the Transactions of the American Philological Association.

L i t e r atu r e i n t h e

Ro m a n R e p u b l i c POETRY AND ITS RECEPTION

SANDER M GOLDBERG

University of California, Los Angeles

cambridge university press Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, S˜ao Paulo

Cambridge University Press

40 West 20th Street, New York, ny 10011-4211, usa

www.cambridge.org Information on this title: www.cambridge.org/9780521854610

C

Cambridge University Press 2005 This publication is in copyright Subject to statutory exception

and to the provisions of relevant collective licensing agreements,

no reproduction of any part may take place without the written permission of Cambridge University Press.

First published 2005 Printed in the United States of America

A catalog record for this publication is available from the British Library.

Library of Congress Cataloging in Publication Data

1 Latin poetry – History and criticism 2 Rome – History – Republic, 510–265 b.c.

3 Nationalism and literature – Rome 4 Poetry – Appreciation – Rome.

5 Authors and readers – Rome 6 Books and reading – Rome.

7 Rome – In literature I Title.

pa6047.g65 2005

871 .0109358 – dc22 2005013006isbn-13 978-0-521-85461-0 hardback isbn-10 0-521-85461-x hardback Cambridge University Press has no responsibility for the persistence or accuracy of urls for external or third-party Internet Web sites referred to in this publication

and does not guarantee that any content on such Web sites is, or will remain, accurate or appropriate.

amico collegae magistroque semper

Preface page ix

Introduction 1

1 The Muse Arrives 20

2 Constructing Literature 52

3 Comedy at Work 87

4 Dido’s Furies 115

5 Enter Satire 144

6 Roman Helicon 178

Retrospective 204

this study developed from the nagging sense, increasinglycommon among students of the Roman world, that the traditional storytold of Roman literature’s origin and early development is deeply unsat-isfactory Challenges to the old verities have become too numerous, tooinsistent, and too convincing to keep the old story in place, but many

of the alternatives now being proposed seem to me to be groundedtoo deeply in modern ideology and not deeply enough in ancient evi-dence Like most New Historicists, I want to speak with the dead, but

I am more eager to hear what they have to say than to tell them what Ithink it means The following pages therefore set the primary evidenceabove the debates being waged over it Scholarly opinions come and go(and sometimes come again), but the evidence endures My presentationreflects that priority, quoting and discussing Roman sources in the textand being as clear as possible about why I read them as I do, but relegatingthe majority of my scholarly debts, disagreements, and suggestions to thenotes Yet this is not a strictly empirical study It owes much to theorists,

in particular to Stanley Fish for its definition of literature and to PierreBourdieu for its understanding of literature’s role in society, and its way

of reading Latin poetry is inevitably influenced by the work of GiorgioPasquali and his successors Though I am obviously not one to unpackand interrogate when I can analyze and ask, this inquiry remains in allsignificant respects, by choice and not just by necessity, a product of itstime

Its approach to literary history is nevertheless a little unconventional,and its findings occasionally run counter to one or another commonlyheld view A new perspective may compel even familiar landmarks toreveal unfamiliar aspects I shall be arguing here that Romans of the lateRepublic had both the concept of and a word for “literature,” but that

imposing this name and function on certain works was often a spective activity The Republican literature we traditionally call “early”could be as much a product of the late Republic, when texts were firstsystematically collected and put to new social and artistic uses, as of themid-Republic, when works were first composed with writing in mind.The literary history that follows therefore pays rather more attention

retro-to readers than is often the case Cicero, the most fully documented ofRoman readers, will loom especially large Horace will acquire his great-est significance as a reader of earlier poetry, and what remains in purposeand in essence a study of Republican literature will nevertheless draw itsfinal argument from the most notorious of Augustan exiles, who foundall too much time to reflect from a distance on the literary life of Rome.One other oddity deserves mention The process of reading and recep-tion in antiquity was of course continuous, but the evidence left of thoseactivities is only intermittent The following chapters focus on what sur-vives, centering on those points in the process that prove most congenial

to investigation One consequence of this decision is a privileging of

poetry over prose Cicero’s sense of litterae no doubt embraced prose as well as verse, and even Cato’s Origines, a pioneering prose work of the

150s, was keenly aware of it own cultural significance Yet the debts oflater Romans to early poetry are, with a few notable exceptions, mucheasier to trace than their debts to early prose, and the reception of poetrythus claims priority here The nature of the evidence also explains why,though I have stressed continuities from one chapter to the next, thereare obvious disjunctions as well I can only build with the material athand

A continuous argument, vaguely chronological despite its avowed trust of chronology, runs through these chapters, but the need for back-reference and recollection allows them to be read separately Since theargument can be complex, a little repetition and an occasional appeal tothe familiar seem a small price to pay for clarity Ancient authors are gen-erally cited from their Oxford editions, the significant exceptions beingHorace and Ovid, who are quoted from the most recent Teubner texts

dis-of Shackleton Bailey (1985) and Hall (1995) respectively, and Cicero’scorrespondence, cited from the Loeb editions of D R ShackletonBailey, though I have maintained the traditional numbering The sourcesfor fragmentary texts are indicated in the notes Translations are my own

As inevitable with a project of this scale, my debts to individuals andinstitutions are considerable, and they are a pleasure to recall The inves-tigation began in 1998 during a term of relative calm as a visitor to the

School of Classics at the University of Leeds, which provided a congenialbase for what became an extensive operation Aspects of its argument haveover the years excited – the verb is deliberately ambiguous – audiencesfrom St Andrews and Exeter to Dunedin and Hobart, Freiburg and Pisa toCharlottesville and Seattle, and I have learned a great deal, though perhapsnot always enough, from the resulting exchanges It is equally pleasant

to acknowledge the fellowship support of the University of California’sOffice of the President, the American Council of Learned Societies, andthe National Endowment for the Humanities for providing leave for writ-ing, and UCLA’s own Council on Research for a timely series of researchgrants Finally, there are the many debts to individuals whose advice andencouragement, suggestions and objections, have not just made this studypossible, but even made it fun The two readers for Cambridge UniversityPress will recognize my debt to them, as will Beatrice Rehl, as demandingand yet supportive an editor as any author could wish I also owe much

to John Barsby, Elaine Fantham, Rolando Ferri, Bob Kaster, J ¨org R ¨upke,and especially Erich Gruen, whose support over the years has meant farmore than a mere dedication can adequately express

Sander M GoldbergLos Angeles

January 2005

An English schoolmaster is shipwrecked on the West African coast Carried inland

by slave traders, he makes himself useful to the most powerful chief of Ife There his old skills as scholar and teacher come to the fore, and, almost by accident, he launches one of the world’s great literatures when he translates Paradise Lost into Yoruba and adapts the plays of Dryden for a local festival.

Who can imagine such a thing? prospero did not recast hisbooks in Caliban’s language or subject them to Caliban’s service.Yet the Romans believed that something nearly this surprising actuallyhappened in Italy in the third century B.C when an educated Greeknamed Andronicus came to Rome as a slave, was taken in by the pow-erful family of the Livii Salinatores, and gave the Romans a literature by

translating the Odyssey into Saturnian verse and staging the first Latin versions of Greek plays at the ludi Romani of 240.1 This account hasbeen so often repeated, and the conscious use of Greek models is socharacteristic a feature of subsequent Latin literature, that even now thefull oddity of the story rarely attracts the attention it deserves Was theRomans’ first literature really poetry of such foreign origin, the gift offreedmen like Andronicus and then Terence and of ambitious provincials

1 Cic Brut 72, Tusc 1.3, Sen 50 Cf Liv 7.2.3–13, V Max 2.4.4 Brut 73 acknowledges

some controversy over these matters, but Cicero’s version of Andronicus’ contribution has prevailed See Gruen 1990: 80–82, Baier 1997: 116–20 (contra Mattingly 1993), and for early Republican attempts at literary history, Fantham 1996: 42–47, Schwindt

2000: 52–121 Andronicus’ Odusia had become a school text by Horace’s time (Ep 2.1.69–72), but there is no evidence for the oft-repeated claim (e.g., von Albrecht

1999: 41–44) that this was his aim in writing it Whether the epic preceded or followed the plays is unknown Mariotti 1986: 16–19 provides excellent discussion of these issues.

like Naevius, Plautus, and Ennius? And, questions of historicity aside,why would Romans be willing to accept and to transmit so peculiar astory of their cultural heritage?

Alternatives should have been possible The story that seized theRomans’ attention emphasizes differences at the expense of equally com-pelling similarities, and if other choices had been made by the tellers,

a somewhat different story might well have developed in its place Inprivileging the world of poetry over the world of prose, for exam-ple, the traditional account sets the mercenary work of Rome’s lowerclasses apart from the personally engaged products of its elite The socialgap between these two worlds of endeavor was considerable ThoughAndronicus may have been a client of the Livii and the beneficiary ofsenatorial largesse, the first Roman to write a history in prose was him-self a Fabius and a senator, and the first to write one in Latin, Cato,was a consul and censor and a public figure for half a century.2 Norwas history the only prose genre to gain prominence among the elite.The oratory of senate and assembly was increasingly preserved in writ-ing and thus available for that range of uses that, as we shall see, began

turning texts into “literature” in the second century Cicero’s Brutus itself

makes a powerful argument for the literary status of oratory and is thusincreasingly appreciated by modern scholars as a serious work of literaryhistory.3

Still more significant is the fact that prose and poetry were not asdiscrete in their practices and in their achievements as an emphasis onsocial distinctions might suggest and not only because poets and aris-tocrats sometimes met as patrons and clients Prose, like poetry, couldalso be inspired and informed by Greek examples, and its developmentwas closely intertwined with the poets’ achievements The prologues ofTerence, to cite one of our less problematic cases, exploit not just thestance but the very language of contemporary oratory, and the com-plexity of Terence’s style in turn prefigures the growing capabilities of

Latin prose Cato’s Origines, to take a more ideologically charged

exam-ple, appears to embrace in the 150s an approach to Roman history that

2 Q Fabius Pictor, the Senate’s emissary to Delphi after the defeat at Cannae in 216, was apparently fluent in Greek and used it for his history (Liv 22.57.4–5, 23.11.1–6;

Plut Fab 18.3; Appian Hann 27), though his motives for doing so are much debated.

See Gruen 1984: 253–55, Momigliano 1990: 88–108, Dillery 2002, with extensive bibliography in Suerbaum 2002: 359–66.

3 Thus in different ways and for somewhat different purposes, Goldberg 1995: 5–9, Hinds 1998: 63–69, Schwindt 2000: 96–121.

can be traced back to Ennius’ Annales.4 The traditional story, howeverconvenient, clearly comes at the expense of significant nuance and detail.Then again, nobody was ever fully at ease with it Even Cicero, whoseexcursions into literary history did most to popularize the traditionalaccount, knew perfectly well that the beginning of the evidence was not

necessarily the beginning of the story Greek poets, as he notes at Brutus

71, existed before Homer The Roman situation was surely no different.There must have been poetry before Andronicus, too, and Cicero’s regretover its loss has become important testimony for the fact of its priorexistence

Atque utinam exstarent illa carmina, quae multis saeculis ante suamaetatem in epulis esse cantitata a singulis convivis de clarorum virorumlaudibus in Originibus scriptum reliquit Cato!

If only those songs survived in which, according to Cato in his Origines,

banqueters many generations before his own time sang in turn the

A reference in the Tusculan Disputations to the same report implies that

Cicero understood these archaic songs to have employed traditionalmelodies rather than to have been improvised anew for each occasion.5Gravissimus auctor in Originibus dixit Cato morem apud maiores huncepularum fuisse, ut deinceps qui accubarent canerent ad tibiam clarorumvirorum laudes atque virtutes: ex quo perspicuum est et cantus tumfuisse discriptos vocum sonis et carmina

That highly esteemed authority Cato said in his Origines that it had

been the custom among our ancestors for those gathered around thetable to sing in turn to the pipe the praises and deeds of famous men

It is thus clear that there were then tunes assigned for the sounds ofvoices as well as lyrics

4 For Terence, Goldberg 1986: 31–60, 170–202, and for Cato’s debt to Ennius, Goldberg

2006 and Sciarrino 2006, important even if we do not accept the argument of Cardinali

1988 that Cato’s work began with a hexameter echo.

5 Cic Tusc 4.3 Discriptos is an emendation for descriptos in the MSS (retained by Peruzzi 1998: 139–40) The general point is unaffected, though descriptos ‘recorded’ would

make it even clearer Cf V Max 2.1.10: “maiores natu in conviviis ad tibias egregia superiorum opera carmine comprehensa pangebant ” There is, however, no inde- pendent support for Cicero’s statement It may simply be an inference from his belief

that the archaic carmina were epic predecessors.

Varro, probably also drawing on Cato’s testimony, imagines a formaltradition of praise poetry that was performed in the context of banquets.6

< sic aderant etiam> in conviviis pueri modesti ut cantarent carmina

antiqua, in quibus laudes erant maiorum, et assa voce et cum tibicine

Respectable boys < were present> at banquets to sing both

unaccom-panied and to the pipe ancient songs containing the praises of ourancestors

These songs, evidently too antique a practice for even Cato’s direct

expe-rience, are the so-called carmina convivalia on which, in the early

nine-teenth century, the historian B G Niebuhr based his famous theory ofheroic lays Niebuhr found in this testimony hints of a lost tradition ofballads, which passed from citizen to citizen, generation to generation as

“the common property of the nation” and could help explain the

sur-vival of archaic legends in the Roman historical tradition The carmina as

he understood them therefore represented a valuable element of populartradition in a record otherwise dominated by patrician annals.7

Niebuhr’s theory, controversial from the outset, today finds few porters Greek parallels suggest a lyric rather than narrative character forthe kind of banquet song Cato recalls, and historians have found moresatisfactory ways to explain the survival of Rome’s earliest traditions.8Yet

sup-the carmina convivalia remain of interest Their mere existence has never

6 Var ap Non 107–8 (De vita pop Rom fr 84 Riposati) Peruzzi 1998: 145–46 claims,

I think unconvincingly, that pueri modesti means specifically “musikalische Knaben.”

The testimony of Cicero and Varro is now generally read as complementary rather than contradictory See Riposati 1939: 187–92 and Zorzetti 1990b: 292–93 The context

of Cato’s remark is unknown It is commonly assigned to book 7, but his preface is

a likely inference from the verbal echo at Cic Planc 66: “Etenim M Catonis illud

quod in principio scripsit Originum suarum semper magnificum et praeclarum putavi, clarorum hominum atque magnorum non minus otii quam negotii rationem exstare oportere.” See Cugusi 1994 for further arguments along this line.

7 Niebuhr 1828: 209–10: “Die G¨aste selbst sangen der Reihe nach; also ward erwartet dass die Lieder, als Gemeingut der Nation, keinem freyen B¨urger unbekannt w¨aren.” A century later, Schanz-Hosius was still fixing Niebuhr’s idea in Roman literary history:

“Ueber den Inhalt der Lieder sind uns keine genaueren Mitteilungen ¨uberliefert Aber die r ¨omische Geschichte bietet uns eine Reihe der sch ¨onsten Sagen dar; diese m ¨ussen doch einmal von Dichtern geschaffen worden sein Wir werden nicht irren, wenn wir annehmen, daß sie mit den Tischliedern zusammenh¨angen” (1927: 23) For the

theory’s appeal to students of German Heldensage, see von See 1971: 61–95.

8 Decisive refutation from the historiographic side came from Momigliano 1957 Cf Cornell 2003 on the origins of the Coriolanus legend, one of Niebuhr’s own examples.

The lyric quality of the carmina is acknowledged by Zorzetti 1990b: 298–301.

been questioned: that poetry preceded history as a record of res gestae and

that dinner parties provide congenial occasions for poetic performancehave been commonplace assumptions since antiquity.9The focus of atten-tion, however, has been shifting An expanding knowledge of early Italy’s

material culture has returned the carmina to prominence by changing the

complexion of what was once largely a philological debate over theirplace in literary history Some of the evidence being used is incontro-vertible A wine trade, for example, is now well attested for Latium in theseventh century, and imported drinking vessels dated to the later eighthcentury have been discovered in domestic contexts in Etruria.10The sig-nificance of this information, however, is not equally clear Whether suchfacts mean that early Romans had a specifically “sympotic” culture and

that the lost carmina were performed at symposia organized in the Greek

style remain problematic inferences Archaeological evidence also seems

to confirm that Italians did not initially recline on couches and did notsegregate the sexes in the Greek manner.11 Nor are the social connota-tions of the Greek symposium entirely clear even in Greek contexts Toclaim both that Italians had that same institution and that it meant thesame thing to them as it did to the Greeks requires a bolder argumentthan everyone is prepared to accept.12

A significant level of literacy is nevertheless traceable to at least the sixthcentury B.C., and linguistic evidence has gradually strengthened the casefor an oral poetics in archaic times that could have shaped important

9 Thus Tac Ger 2: “Celebrant [Germani] carminibus antiquis, quod unum apud illos memoriae et annalium genus est ” Cf Serv ad Aen 1.641, 7.206 Momigliano 1957: 109–11 thought the carmina mentioned by Cato may have survived into the

fourth century.

10 Gras 1985: 367–70, Rathje 1990, and more broadly Cornell 1986: 64–68, Horsfall 1993a: 791–8, and Zorzetti 1991: 312–15 Zaccaria Ruggiu 2003, clearly an important study, appeared too late for consideration here.

11 Rathje 1990: 284–85, confirming the testimony of Ov Fast 6.305–6, V Max 2.1.2, and Var de vit p r 29–30 (Riposati) Cf the skepticism of Holloway 1994: 191–

92 The picture is further complicated by testimony of early Roman actions to curb

drinking by women: V Max 2.1.5b, 6.3.9, Plin Nat 14.89.90, Gell 10.23.3, with

Gras 1985: 386–90.

12 So, in response to Zorzetti 1991, Phillips 1991: 386: “We know comparatively little

about symposia and mousike even in Athens and Sparta, while there is even less evidence for those activities in other cities.” Contrast the caution of Petersmann on the carmina convivalia in Suerbaum 2002: 41–42 with Suerbaum himself on early Rome’s “lyrische

Kultur” (2002: 49–51) Fisher 2000: 356–69 and Wilkins 2000: 202–11 question the exclusively aristocratic connotations of the Greek symposium For the benefits and pitfalls of comparing archaic Greek and Roman cultures, see Raaflaub 1986: 29–37.

elements of what eventually became the Roman literary heritage.13Add

to this the unambiguous ancient testimony for hymns and dances in ritualcontexts, and it becomes clear that verbal art, along with opportunities

to perform it and means to preserve it, was deeply rooted in Romanculture for generations before Livius Andronicus.14 Nevio Zorzetti must

be right in claiming that “the old idea of the typical Roman character,practical and unpoetic, is simply inadequate, besides being unhistorical”(1990b: 295)

In truth, though, that “old idea” was never so widely held Niebuhr,lecturing on Roman literature in the mid-1820s, had already made some-thing much like Zorzetti’s claim:15

Let no one imagine that the Romans were barbarians, before theyadopted the civilisation of the Greeks: their works of art and their build-ings prove the contrary That people must assuredly have attained to

a high degree of intellectual culture, and cannot be conceived to havebeen without some kind of literature, though, of course, different fromthat of the Greeks

What did change profoundly in the generations between Niebuhr and

Zorzetti were the attitude toward Greek culture’s influence on theRomans and the direction of the scholarly gaze For Niebuhr, deeplyinfluenced by J.G Herder, the earliest Roman traditions had of necessity

to be Italic Beneath that confident “of course” in the last sentence ofNiebuhr’s declaration lies Herder’s insistence that a viable literature wasrooted in the experience of the people Anything else was necessarily

insubstantial (Luftblase).16To endure, even an aristocratic literature could

13 On literacy: Cornell 1991: 24–32, Poucet 1989, and more generally Horsfall 1994 For the contributions of historical linguistics to the Romans’ literary prehistory, see Costa 2000: 66–79.

14 So Cic Tusc 4.3, de Or 3.197, Lg 2.22, though Zorzetti 1991: 312–18 goes too far

in adducing “a unified culture of carmina” from such evidence and identifying it with Greek influence The conclusion at de Or 3.197, “maxime autem a Graecia vetere

celebrata” implies a significant difference at least of degree between Greek and Roman practice.

15 Niebuhr 1870: 14 These lectures, delivered from 1826–29, were published mously from students’ notes The English edition of Schmitz quoted here is an inde-

posthu-pendent, fuller witness, not a translation of the Vortr¨age ¨uber r¨omische Geschichte published

by M Isler in 1848.

16 So, e.g., Herder’s essay of 1777, “Von ¨ Ahnlichkeit der mittlern englischen und deutschen Dichtkunst”: “Doch bleibt’s immer und ewig, daß, wenn wir kein Volk haben, wir kein Publikum, keine Nation, keine Sprache und Dichtkunst haben ”

neither precede nor ignore popular tradition This was why Niebuhrwould go on in his lectures to praise Theocritus – the idylls “grew out

of popular song, and hence his poems have a genuineness, truth, and

nationality” – while disparaging the Eclogues for creating “something

which could not prosper in a Roman soil.”17This is now, to say the least,

a very old-fashioned style of argument Roman literary achievementsare no longer thought to stand or fall on their perceived independencefrom Greek models Modern scholarship is so much more appreciative

of Vergil, not to mention of Plautus and Terence, in part because it iswilling to posit a deeper and earlier penetration of Greek culture intoItaly than Niebuhr ever envisioned and to accept, even to admire, theconsequences of its influence

Scholarship is also more ready to focus on the actions of Rome’s eliteand to treat literary activity as an aristocratic phenomenon Thus the con-vivial poetry that Niebuhr saw as a manifestation of popular tradition andthe “Gemeingut der Nation” becomes for Zorzetti “the direct expression

of aristocratic wisdom.”18 The possibility that Roman aristocrats had arich cultural life from quite early times and were so receptive to Greek

influences in the crucial third century because they had long been

recep-tive to them is today neither an improbable nor an undesirable idea tocontemplate Whatever Andronicus actually did for the Senate and theRoman people in 240 B.C., it was surely not to create a literature out ofnothing

What really happened in the third century is not, however, the focus

of this book, nor will it add to the stock of conjecture about Rome’spreliterary culture Ancient truths may yet be recovered as new archaeo-logical evidence and new theoretical perspectives join with philologicalrigor in pursuit of that distant past, but their progress is not likely to

be quick Consider Livy’s famous digression on the origin of the ludi

scaenici, which may stand as a sobering example of the difficulties such

(Herder 1982: 286) For the concepts of Volk and Nation in Herder, see Barnard 1965:

73–76.

17 Niebuhr 1870: 661 Cf Lessing 1962 (1766) 96–97, contrasting the artificiality of Aeneas’ shield (“ein fremdes B¨achlein”) and the naturalness of Achilles’ (“Zuwachs des eigenen fruchtbaren Bodens”) Then again, Horace too had some hesitation about

the Eclogues or at least about the preciosity they might encourage See Zetzel 2002.

18 Zorzetti 1990b: 294 Habinek 1998: 54 reads early Roman literature as “an agent of

aristocratic acculturation.” For Niebuhr’s view of the carmina as the voice of the plebs,

see Momigliano 1957: 107–9.

inquiries face Livy’s account undoubtedly contains important evidencefor the history of Roman drama, but it has defied a century and more ofintense scrutiny.19 Nothing about the passage is clear Its association of

the early ludi with an outbreak of plague in 364 B.C is unusual, perhaps

unhistorical, and almost certainly colored by Livy’s own antitheatricalbias.20 The central role he assigns the Roman iuventus for motivating

change is vague and problematic, while the story of Andronicus

mim-ing cantica when his voice failed is scarcely credible.21 New finds from

Etruria or Latium may someday cast light on the Etruscan ludiones at the

center of these developments, and a better understanding of what Livycalled musical medleys (“impletae modis saturae”) may yet help us explainhow Andronicus could find actors in third-century Rome equal to thetask of performing his new Latin scripts, but good luck and great effortwill be needed to produce what may even then be only a small gain inknowledge

More yielding to immediate inquiry, and equally relevant to the lem of Rome’s literary origins, is the reception of archaic traditions bythe later Romans who first constructed a literary history – and indeed,defined a literature – out of the earlier remains Because the literary his-tory of the Republic as we tell it today is largely a first-century story, it

prob-is worth paying more attention than prob-is customary to how and why century Romans told it as they did This means understanding Romans ofthe late Republic as both users and shapers of their literary heritage That

first-is itself a complex task since the textual evidence of early times inevitablycomes wrapped in the arguments of later ones, and not every source of

later distortion is as easily recognized as Livy’s bias against the ludi (“ab

sano initio in hanc vix opulentis regnis tolerabilem insaniam”) Wework with secondhand and synthetic evidence and must constantly beaware that the more we build upon it, the more likely we are to magnify

19 Liv 7.2.3–13 Important recent discussions include Bernstein 1998: 119–29, Feldherr 1998: 178–87, and Oakley 1998: 40–58, with extensive bibliography provided by Suerbaum 2002: 51–57.

20 Liv 7.2.3 says only “dicuntur,” followed a little later by “dicitur.” Feldherr 1998:

183–85 notes the inefficacy of the ludi as a response to plague Livy’s source is widely,

though not universally, thought to be Varro, an uncertainty that makes his integration

of the antiquarian excursus and historical narrative especially problematic.

21 Jory 1981: 152–55 suspects, not without reason, the influence of pantomime in fostering this idea The tradition that Andronicus was himself an actor is much less incredible.

Leo 1913: 56–57 remains basic For the problematic iuventus of Livy’s story, see Morel

1969.

its inherent distortions.22 The resulting dilemma is well known to ologists, as Pierre Bourdieu observes (1990b: 102):

soci-However far one goes back in a scholarly tradition, there is nothing thatcan be treated as a pure document for ethnology It’s well known thatthe corpus which the ethnologist constitutes, merely by virtue of thefact that it is systematically recorded, totalized and synchronized isalready, in itself, an artefact: no native masters as such the complete sys-tem of relations that the interpreter has to constitute for the purposes

of decipherment But that is even truer of the recording carried out bythe story told in a literate culture, not to mention those sociologicallymonstrous corpora that are constituted by drawing on works from alto-gether different periods The temporal gap is not the only thing at stake:indeed, one may have to deal, in one and the same work, with semanticstrata from different ages and levels, which the text synchronizes eventhough they correspond to different generations and different usages ofthe original material

The carmina convivalia become precisely such a “sociologically

mon-strous corpus” when their reconstruction fails to distinguish sufficientlybetween the content and the context of the testimony used and to con-sider how the context influences its content The methodological issue

is important and worth a closer look, since no evidence of Rome’s earlycultural heritage comes to us independent of later filters A famous scrap

of testimony illustrates the point quite well It comes, as so often inmatters of early literary history, from Cicero

First-century Romans accepted as a matter of fact that the Greeks’literary achievement had long outstripped their own That concession

followed comfortably, as Cicero says in his introduction to the Tusculan

Disputations, from the belief that early Romans, with so many other

achievements to their credit, had never tried to rival the Greeks in thisarea.23There was therefore no serious poetry at Rome until the time of

22 Contrast the quality of the evidence available to Zorzetti 1990b with what is available

to Ford 2002: 24–45 in discussing the Greek symposium and its cultural impact A Roman equivalent to Ford’s kind of analysis thus seems beyond our capabilities.

23 Cic Tusc 1.3: “Doctrina Graecia nos et omni litterarum genere superabat, in quo erat

facile vincere non repugnantes.” The catalogue includes an ample range of endeavors

in which Roman efforts more than equaled the Greeks Cf the famously enigmatic

injunction of Aen 6.847–53, from which any litterarum genus is conspicuously absent.

The idea that literary culture came late to the Romans is attested first for Porcius

Licinus (Courtney 1993: 82–86), echoed famously by Hor Ep 2.1.156–9, as well as Liv 7.2.3 and eventually Suet Gram 1.1.

Andronicus, and even then it was not valued highly, as Cato is once morecalled upon to witness:

Sero igitur a nostris poetae vel cogniti vel recepti quamquam est in

Originibus solitos esse in epulis canere convivas ad tibicinem de clarorum

hominum virtutibus, honorem tamen huic generi non fuisse declaratoratio Catonis, in qua obiecit ut probrum M Nobiliori, quod is inprovinciam poetas duxisset; duxerat autem consul ille in Aetoliam, utscimus, Ennium quo minus igitur honoris erat poetis, eo minora studiafuerunt, nec tamen, si qui magnis ingeniis in eo genere exstiterunt, nonsatis Graecorum gloriae responderunt

Poets thus received late recognition or welcome from our

country-men Although we find in the Origines that guests around the table

were accustomed to sing to the pipe about the deeds of famous men,Cato’s speech in which he criticized M Nobilior for taking poets to hisprovince (the consul had in fact, as we know, taken Ennius to Aetolia)nevertheless declares that there was no honor in this sort of activity.And so the less poets were honored, the less attention was paid to them,although those whose great talent enabled them to stand out in thatactivity nevertheless matched the glory of the Greeks (Tusc 1.3)

Although ostensibly straightforward, Cicero’s argument here – and it is anargument, not an exposition – actually conflates and distorts three distinctlevels of witness There is the state of poetry in early Rome, what Cato in

the second century said in his Origines about banquet songs and what he

said in a speech attacking Fulvius Nobilior, and finally there is Cicero’scombination of Cato’s statements for his own purpose a century andmore after their original articulation Though some of the words in thepassage are certainly Cato’s, the association of ideas is Cicero’s, whichmeans that these relics of second-century polemic are preserved in amatrix of first-century argument They are all too well integrated intothat argument, which means that as evidence of earlier times, Cicero’saccount is seriously jumbled and unhistorical This becomes obvious assoon as we begin separating its levels of testimony

Cicero himself certainly has Ennius’ Annales in mind when thinking

here about poetry: the activity in question seems to embrace both the

archaic carmina and the epic It was a natural association for Cicero.24The

24 And perhaps for Cato J E G Zetzel points out to me that Tusc 1.3 could be taken to mean that Cato found no honor in performing the archaic carmina either His approval

of them, though widely assumed in modern scholarship, is not explicitly attested in any ancient source.

more detailed version on this argument about literary progress at Brutus 71–76, for example, explicitly evokes the archaic carmina of Cato’s Origines

as the first step in epic’s rise, and Cicero knew perfectly well that Annales

15 celebrated Nobilior’s Aetolian campaign and climaxed the first edition

of the poem with his restoration of the Aedes Herculis Musarum usingAmbracian spoils It was therefore logical for him to assume that Cato,whose hostility to Fulvius was well known, objected on these grounds

to his patronage of Ennius The problem with this line of association is

that the encomiastic tendencies of Annales 15 were probably not at issue

in Cato’s speech attacking Fulvius Nobilior Cato did not scruple there

to recall the contested Aetolian triumph of 187, but his immediate targetwas Fulvius’ censorship of 179.25 The speech is therefore dated to 178

The Annales project probably began about 184, after the poet’s return

from Ambracia, but it was never the sole claim to his attention Enniuscontinued to write plays and satires into the 170s, as well as a hexameter

poem about fish (the Hedyphagetica) He also had to research some five

hundred years of Roman history and develop a technique for creatingviable epic hexameters in Latin If, as seems likely, Ennius wrote his epic

in chronological sequence, with a significant break after Book 6, Book

15, which marked the end of the sequence, probably did not circulateuntil the late 170s If this is right, the action that aroused Cato’s disdain in

178 was not the writing of an epic poem glorifying Fulvius Nobilior.26

The provocation more likely came from the production of a play,

Ennius’ praetexta drama Ambracia, which was staged either in

conjunc-tion with Fulvius’ triumph or at the votive games he held the following

year The Scipio in honor of Africanus had already presented an unsettling precedent for Latin encomiastic verse, and early books of the Annales may

have further raised Ennius’ profile and stoked the fires of Cato’s tion, but the play would have attracted his particular attention because ofits conspicuous public role in the controversy of 187.27 He would have

indigna-25 So Malcovati 1953: 57 and now widely accepted, though the possibility of an earlier speech attacking the consulship and/or triumph of Fulvius cannot be excluded See Astin 1978: 110 n 22, Sblendorio Cugusi 1982: 294–96.

26 The dating of Annales 1–15 is problematic, with dates of composition well into the

170s most commonly favored, since it is difficult to imagine fifteen hexameter books researched, written, and circulated in little more than five years Gell 17.21.43 reports that Ennius wrote Book 12 in his sixty-seventh year (i.e., 173), but the information is not necessarily reliable See Suerbaum 1968: 114–20 and Skutsch 1985: 2–5.

27 Flower 1995: 184–86, Manuwald 2001: 163–66, and for the oddity of the play in this

context, Zorzetti 1980: 78–81 and Gildenhard 2003: 109–11 The laudatory Scipio is

thought it a particularly outrageous and unprecedented display of sanship, an artistic intervention in what was still in the 170s one of themost notoriously contested triumphs of the age “Who has seen anyonegranted a victor’s crown,” he asked in that same speech against Fulvius,

parti-“when a city had not been captured or an enemy camp not burned?”28The suborning of Ennius to tip the balance of public opinion in Nobilior’sfavor, the kind of ploy better suited to a Hellenistic dynast than a Romanconsul, must have been particularly galling since Ennius probably suc-ceeded in this effort: when Cicero eventually hailed the dedication ofthe Aedes Herculis Musarum with the remark that Nobilior “did nothesitate to dedicate Mars’ spoils to the Muses” he may be echoing not

just the sentiment but even the words of Ennius’ Ambracia.29 WhateverCato’s motives, however, a partisan debate of the 170s will not providereliable evidence for literary history

Though Cicero’s conflation of epic and play, speech and history may

be an unreliable guide to second-century attitudes, its implicit contrastbetween the songs of banqueters and the works of poets may nevertheless

go back to Cato, though not to his statement in the Origines Other erences to that passage make clear that Cato had understood the carmina

ref-to be a cusref-tom of the distant past, not a fact of his own second-centuryculture.30More explicit testimony about the status of poets in the secondcentury has been culled from another work, where he declared in lan-guage quite similar to what Cicero reports that the poets’ art originallyreceived no honor and its practitioners were dismissed as flatterers Both

almost certainly later than 187 but predates the Annales See the judicious discussion

by Courtney 1993: 26–30.

28 Cato 148M: “iam principio quis vidit corona dari quemquam, cum oppidum captum non esset aut castra hostium non incensa essent?” The story of the triumph and its resentments is told at Liv 38.43–44, 39.4–6.

29 Cic Arch 27: “iam vero ille qui cum Aetolis Ennio comite bellavit Fulvius non

dubitavit Martis manubias Musis consecrare.” For the possible echo here of Ennius’ play (scansion precludes an epic origin), see Manuwald 2001: 162–63 Cato would not have considered what Fulvius took from Ambracia legitimate “Martis manubias.” Gildenhard 2003: 110 notes the Hellenistic precedent for Fulvius’ use of Ennius.

30 Cic Brut 75 (“multis saeculis ante suam aetatem”) is explicit, confirmed by “apud maiores” at Tusc 4.3 The teleological argument at Brutus 75 required this more meticulous chronology Plut Cat 25.4 (



  
) suggests that Cato reintroduced the custom to his own banquets,˘

but that may simply be a misinterpretation of evidence like Tusc 1.3 R ¨upke 2001:

49–58 gets around this problem by suggesting that historical epic was also written for recitation at banquets in the second century, but the evidence for that otherwise appealing suggestion is not strong.

the text and the context of that statement remain problematic, although

it is reasonably clear that Cato did not have heroic verse in mind.The immediate source for it is Aulus Gellius, who illustrates the mean-

ing of elegans by quoting from a work he calls Cato’s Carmen de moribus.

He then continues in his rambling way with some further, seeminglyrandom excerpts from that book (11.2.5-6):

Praeterea ex eodem libro Catonis haec etiam sparsim et intercisecommeminimus: “Vestiri” inquit “in foro honeste mos erat, domi quodsatis erat equos carius quam coquos emebant poeticae artis honos nonerat siquis in ea re studebat aut sese ad convivia adplicabat, ‘grassator’vocabatur.”

I recall these other sayings random and piecemeal from the same book

by Cato: “It used to be the custom,” he says, “to dress becomingly inpublic, modestly at home They paid more for horses than for cooks.Poetic art was not respected Anyone who applied himself to that activ-ity or attached himself to parties was called a ‘grassator’.”

Grassator, ‘vagabond’ or ‘bandit’ in common usage, is often given a more

specific sense here with the help of Festus, who glosses grassari, the verb

behind the noun, as ‘to flatter.’ This would suggest that in Cato’s viewpoetry was at some point in Rome’s past considered little better thanflattery and poets therefore little more than fawners or parasites.31 Howshould we understand such a remark, and what may have been its basis

in fact?

Context may, despite appearances, provide a clue The Carmen de

moribus, known only from this one chapter in Gellius, was probably not

an original work at all but a collection of dicta drawn from other sources,

a carmen in the sense of a ‘prescription’ or a ‘refrain’.32 This particular set

31 Fest 86L: “grassari antiqui ponebant pro adulari grassari autem dicuntur latrones vias obsidentes; gradi siquidem ambulare est, unde tractum grassari, videlicet ab impetu gradiendi.” Thus R ¨upke 2001: 57, “nicht als ‘Wegelagerer,’ sondern als ‘Schmeich- ler.’ ” Peruzzi 1998: 159–60 prefers a specific sense, “(poeta) itinerante,” which seems

like special pleading Festus’ autem clearly acknowledges the more usual meaning, but

‘mugger’ (so Habinek 1998: 37–38) makes little sense in Cato’s context and would not motivate Festus’ comment, though Habinek, following Zorzetti 1990b: 294, is

probably right to equate Cato’s “poetica ars” with Greek techne Gruen 1992: 71–72

suggests, less probably, that Cato’s distinction is between types of poetry.

32 So Liv 3.64.10 rogationis carmen ‘electoral rule,’ Cic de Or 1.2.45: magistri carmen ‘a schoolmaster’s refrain.’ The model would have been the so-called carmen of Appius Claudius Caecus (Cic Tusc 4.4, cf Val Max 7.2.1) Scholarship has been silent on this

obvious possibility See Astin 1978: 185–86 for the standard view Gellius’ quotations

of dicta presents three subjects (dress, food, entertainment) united by acommon theme: archaic austerity is implicitly compared with somethingelse, no doubt with modern extravagance The moral values and the style

of presentation are familiar from Cato’s many speeches and ments concerning the conspicuous consumption of his contemporaries

pronounce-He was an active participant in the sumptuary debates of the day, famousfor complaining, among other things, that it was hard to save a city where

a fish cost more than an ox (ap Plut Cat mai 8.2).

That sort of complaint was hardly new Demea made it in Terence’s

Adelphoe, and the calls to convivia in Plautine comedy often suggest

invita-tions to license and immorality.33 The causes of Cato’s particular ance are recorded by Polybius in a moralizing passage of his own con-cerning the extravagant banqueting customs that came into vogue amongRoman aristocrats after Pydna The young Scipio, he says, found it rela-tively easy to win a reputation for moderation (

annoy-there were so few rivals among his peers Of these,34

some gave themselves up to affairs with boys, others to hetairai, andmany to musical entertainments, drinking parties and the extrava-gance they involve (

of the war with Perseus with Greek license in these things In fact theincontinence that had broken out among the young men grew so greatthat many paid a talent for a favored boy and many paid three hundreddrachmas for a jar of preserved fish from Pontus Marcus Cato became

so indignant at this that he said in a public speech that he recognized inthese matter the surest sign of decline in the state when pretty boys soldfor more than fields and jars of preserved fish for more than plowmen.35

preclude the a priori assumption of (most recently) Zorzetti 1991: 313–15 that carmen

in archaic contexts must refer to poetry.

33 Ter Ad 60–63 Plaut Most 933–4 alludes to this sort of party, while Stich 707 suggests

singing in Greek.

34 Polyb 31.25.5, with another version at D.S 37.3.5–6 Cf Cato’s attack on M Lepidus (fr 96M) for erecting statues to two Greek cooks (worth four talents each, according to

Diodorus) and his own claim to modest living in the speech De sumptu suo (fr 174M).

For his role in the sumptuary debates of the day, see Astin 1978: 91–97 and Gruen 1992: 69–72 The dubious morality associated with aristocratic banquets lingers in Livy’s description of Sex Tarquinius’ ill-fated dinner party at 1.57.6–9.

35 This was not strictly true A plowmen in second-century Italy cost more than three hundred drachmas (denarii?): the Roman slaves manumitted in honor of Flamininus

in 195 were ransomed for five minae each, i.e., five hundred dr (Plut Flam 13.4–5).

Rhetoric of course transcends economics.

Such comparisons eventually become a commonplace of Roman moraldiscourse: Sallust’s Marius will sound much the same note – no doubt

by design – when he proudly acknowledges that his dinner parties wereaustere and his cook less expensive than his bailiff.36 The obvious infer-ence to be drawn from the moral litany Gellius quotes is that in Cato’spresent the suppressed counter to each statement was true Poetry, we

must conclude, was receiving respect and poets were not called flatterers.

Cato may have liked that state of affairs no more than he liked theprice of fish from Pontus, but the confirmation of Cicero’s argument that

poetry came late to the maiores also confirms its status in Cato’s time.

Thus Ennius won the respect and benefited from the approval of a verywide range of prominent Romans, as Cicero himself had acknowledgedwhen defending Archias nearly twenty years before:37

Omnes denique illi Maximi, Marcelli, Fulvii non sine communiomnium nostrum laude decorantur ergo illum qui haec fecerat, Rud-inum hominem, maiores nostri in civitatem receperunt

And so all those Maximi, Marcelli, and Fulvii were honored with apraise that encompassed us all Therefore our ancestors bestowed citi-zenship on him who did those things, the man from Rudiae

We must conclude that neither poetry in general nor Ennius in ular was the target of Cato’s speech of 178 The attack was on Fulvius’wealth, the praise his wealth could secure, and the image he sought tocultivate It was good politics to be sure but therefore a dubious witness

partic-to contemporary attitudes and an even less reliable source for the culturalpractices of still earlier generations

This brief excursion into source criticism confirms an inconvenientbut inescapable fact Ancient sources sometimes say more than they actu-ally know and have a strong tendency to tailor whatever they say to theirparticular requirements.38 What Cato and then Cicero after him knew,

36 Sall Iurg 85.39: “sordidum me et incultis moribus aiunt, quia parum scite convivium

exorno neque histrionem illum neque pluris preti coquom quam vilicum habeo.”

37 Cic Arch 22 Brut 79 claims that Ennius received Roman citizenship in 184 through

the sponsorship of Nobilior’s son, Quintus, but the chronology is problematic (Badian 1972: 183–85) Ennius’ attested association with various Cornelii, Fulvii, Sulpicii, and Caecilii in any case transcends the partisan politics of the early second century See Badian 1972 and Gruen 1990: 106–16.

38 The tendentious nature of late Roman sources is noted by Cole 1991: 377–78 and Gabba 1984, who emphasizes their persistent “idealisation of the past as an avenue to the interpretation of the present” (86).

thought he knew, or is now thought to have known about the archaic

carmina convivalia are not necessarily all the same thing Their testimony

may not be coherent, nor can the philological analysis that advancesunderstanding of our informants and their world overcome the limits oftheir own knowledge The archaic phenomenon may easily have involvedmore, less, or simply something different from what the sources preserve,and until we are able to add to those sources, the historical reality behindthe banquet songs is likely to remain, like the details of Livy’s “dramaticsatura,” at a distance We must in any case resist the temptation to readour limited sources synoptically, as if they all understood the same phe-nomenon the same way and all had the same purpose in recalling it.Before constructing one of Bourdieu’s “monstrous corpora,” we need totake the evidence to pieces and evaluate its constituent parts separately.Only then can we assess their cumulative value for reconstructing archaicpractice.39

Happily for the present inquiry, however, the secondary and tertiarysources that provide such problematic evidence for archaic practice arethemselves primary evidence for the first-century attitudes toward earlyRoman literature and its reception that are the subject of this book.That does not necessarily mean that they are any more straightforward

Cicero’s fixation on the Annales, for example, when calling up the

mem-ory of Cato’s remarks on poetry recalls an important fact of literary

his-tory The epic poem dominates Cicero’s thinking as if the play Ambracia

did not exist, and in an important sense this was probably the case Notthat the genre was unimportant Plays on Roman themes, the so-called

fabulae praetextae, were said to be Naevius’ invention, and in the course

of the second century, great moments of history, legend, and cult were

reenacted on the stage at the regular ludi scaenici, as well as at individual

temple dedications and triumphs About a dozen such Republican playsare known There may have been dozens more Naevius, Ennius, Pacu-

vius, and Accius all wrote praetextae, and such pageants may have played

a significant role in disseminating the facts of Roman history, developingthe Romans’ sense of community, and enlivening the political discourse

39 This synoptic flaw is particularly marked in Zorzetti 1991: 312–18 on the Romans’

“ancient music” and Peruzzi 1998: 139–47 on the “evolution” of banquet songs from amateur to quasi-professional performances The alternative procedure is not what Zorzetti disparages as “hypercritical philology” but simply controlling for context

when evaluating secondary sources Thus, for example, Cic de Or 3.197, where Cicero’s point is not, pace Zorzetti, “the music of Numa,” but the practical effect of

rhythm on audiences Details in Goldberg 2006.

of the time As late as the Floralia of 57, a revival of Accius’ Brutus, a play

ostensibly about the last Tarquin, caused a major commotion when theactor Aesopus gave the line “Tullius, who secured the citizens’ liberty” acontemporary spin in Cicero’s direction.40

The genre actually outlived the Republic.41 Plays on Roman themescontinued to be written under the emperors – our one complete example,

the Octavia, survives in the Senecan corpus – but Republican praetextae

are known only from very meager fragments that are preserved almostentirely in grammatical rather than literary contexts Only once doesCicero, generally so fond of illustrating literary or philosophical points

with quotations from tragedy or epic, cite a praetexta for its content.42More typical in their path to survival are our four lines of Ambracia,

each cited for a lexical oddity by the fourth-century antiquarian NoniusMarcellus As a topical exercise without the cachet of a Greek pedigree,

the fabula praetexta evidently lacked the status of other genres and was

less likely to figure in later literary discussions Thus, when Cicero thinks

about Ennius, his memory of the praetexta easily becomes a casualty of the Annales’ greater prominence and the prestige that poem eventually

bestowed on the epic genre

This eclipse of the play introduces a final point of significance, which isthe definition of “literature.” In emphasizing what authors do in produc-ing texts, traditional accounts of Roman literary history pay considerablyless attention to the fact that literature requires readers as well as writers It

is not just the creation and collection of certain texts but an attitude towardthose texts that mark them as literature One of the safer inferences fromour all-too-problematic story about Livius Andronicus is that something

40 Cic Sest 123: “Tullius, qui libertatem civibus stabiliverat.” For the definition of the praetexta, see Ussani 1968 and Manuwald 2001: 14–52, and for its putative role in

disseminating Roman traditions, Zorzetti 1980: 53–73, Wiseman 1994: 1–22, and 1998: 1–16 Flower 1995 and Kragelund 2002: 17–27 review the occasions for its performance in the Republic.

41 The continuity of the genre is well argued by Kragelund 2002, though Accius may have introduced a significant turn toward tragedy that eventually made the imperial

praetextae significantly different from their Republican predecessors See Zorzetti 1980:

93–107.

42 That one clear exception is Accius’ Brutus, a play with some vogue in the late Republic, quoted for its dream narrative at Cic Div 1.43–45 and the political twist reported at Sest 123 The only other plays cited by a Republican author are Naevius’ Clastidium and Romulus, both quoted by Varro The pattern of citation is clear from the chart at Kragelund 2002: 12 If Cic Arch 27 conceals an echo of Ambracia, it is more likely a

reflection of Cicero’s research into Ennius’ career than an explicit allusion.

happened in connection with the ludi Romani in or about the year 240

that both defined what later generations would call their literature anddiminished their interest in whatever had preceded it Seen this way, thefocus of literary history starts to shift from matters of who wrote whatand when and under whose influence to points of connection betweenauthors and audiences: how and when did the Romans come to valuewhat Andronicus and his successors created, so that scripts for second-century actors became part of greater Rome’s cultural heritage? Whywould one genre – epic is the most obvious example – quickly become

a cultural benchmark while another, for example, praetexta drama, would

eventually hold only the grammarians’ interest? I shall be arguing here

that when Cicero refers to litterae, he often means “literature” in

some-thing very like the modern sense of texts marked with a certain socialstatus, whose “literary” quality denotes not simply an inherent aestheticvalue but a value accorded them and the work they do by the societythat receives them.43Literature is thus the result not just of creation, but

of reception

Distinguishing the creation of literature from the creation of texts,valuing the work of readers along with the work of writers, has impor-tant advantages It becomes much easier to understand how the traditional

“history” of early Roman literature took shape through a process of sight and back projection as men like Varro and Cicero sought amongthird- and second-century texts what they required to meet their ownfirst-century needs By reading the story they created for Roman litera-ture from the inside out rather than forward from its putative beginning

hind-in the conventional chronological sequence, the hind-inevitable first-centurydistortions become part of the story rather than obstacles to its telling.The changing role of drama on the cultural scene becomes more apparentand a little easier to understand, as does the range of influences dramaexerted on later literary and social discourse So too does the impetus tocreate the new genre of satire And above all, it becomes possible to seemore clearly how Romans, writers and readers alike, came to use literarytexts as they did, and why they found it advantageous to do so

What follows here is therefore not a traditional literary history, though

it is certainly an exercise in the history of literature It begins with epic

as the genre that first aroused literary sensibilities at Rome but puts less

43 So “omni litterarum genere” at Tusc 1.3 and litterae at S Rosc 46, Att 4.10.1, Fin 1.4 The references to litterae and deficiencies of education at Div Caec 39 and 2 Verr.

1.47 are similar The point is developed in Chapter 3.

emphasis on what early authors intended than on what later ones made

of their intentions How the genre was read changed with time, as didthe company it kept in the Romans’ bookcases Drama comes second

in this account because it became “literature” only in retrospect, andChapters 2–4 will examine how scripts written for the early dramaticfestivals were eventually reclaimed for the emerging high culture of thelater Republic and what work those scripts came to perform in that newrole The emphasis must of necessity be on comedy (and to a lesser degree

on Plautus over Terence) because that is where the preponderance of theevidence lies, but tragedy too will play a part in the discussion Thoughthe work they eventually did was significant, however, the generic con-ventions of both epic and drama limited their ability to explore socialissues, and we will then have to consider how that limitation combinedwith the growing appreciation of poetry’s power to stimulate formation

of a new genre that was capable of more direct social criticism Theresult is the subject of Chapter 5, centering on the distinctly aristocraticgenre we know as “satire,” which did much to solidify poetry’s place inthe Roman cultural landscape Chapter 6 will consider how in the latefirst century poetry achieved, both literally and figuratively, monumentalproportions at Rome just as the Republic was becoming history A briefrestrospective then reviews the justification and considers the method-ological implications of understanding Republican literary history inthis way

c h a p t e r o n e

THE MUSE ARRIVES

Acertain fabius, who affected the imposing cognomenUlulitremulus (‘Owl-quaker’), ran a cleaning establishment atPompeii just off the street we know as the Via dell’ Abbondanza Hemust have made some claim to education and experience On the rightdoorpost of his shop was a large picture of a meticulously patrician Aeneas

in high-laced boots and cuirass leading Anchises and Ascanius out from

Troy, and opposite it was a similarly dressed Romulus, with the first spolia

opima on his left shoulder These paintings were not original creations:

they recall the statues of Aeneas and Romulus that faced each otherfrom two large exedrae in the Forum of Augustus at Rome As such, thepaintings are a nice example of Augustan iconography and of its enduringappeal even beyond the city.1 Yet Fabius’ grandeur also set him up for

a tease Among the graffiti scrawled beneath the pictures is a hexameterverse, “Fullones ululamque cano, non arma virumque” (‘I sing of cleanersand owls, not arms and the man’).2 That joke at the expense of Fabius’

pretensions is an equally nice reminder of how deeply the Aeneid rooted

itself in the Roman consciousness and became inseparable in the Roman

1 The fullonica of Fabius is Reg IX ins 13.5 The pictures are reproduced as Fig 156

in Zanker 1988: 202, who implies a connection with the Forum Augusti This must

be right The Pompeian figures, somewhat illogically, faced away from each other, but these poses in the Forum, where the two heroes were reversed, would have them both looking toward the Temple of Mars See also Galinsky 1996: 204–6 on the iconography A parody of the Aeneas pose – the figures are monkeys – was also found

at Pompeii See Fuchs 1973: 57 and Galinsky 1969: 30–32.

2 The text (CIL 4.9131=CLE 1936) is no 60 in Courtney 1995 The association of fullers and owls is well documented, though badly understood Good discussion by Courtney, 280–81 The parodist might have mistaken Romulus for another Aeneas, bearing the arms of the defeated Turnus The faces in the two pictures are very similar.

mind from Augustus’ renovation of the Roman material and literaryheritage.3

Vergil’s success, however, should not obscure another fact of literaryhistory, which is the surprise his poem first generated among his peers.Roman poets of the 20s had learned to keep epic at a distance Someearlier epic projects were stillborn Others, like the poems on Caesar’sGallic campaigns by Varro of Atax and Furius Bibaculus had broughtRepublican epic to the brink of panegyric The common responses tothe resulting crisis in taste were either to withdraw, like Catullus andCinna, to the library or to make the very refusal to write epic a literarytopos.4 The discovery that Vergil, the model poet who had himself oncegracefully declined to write of kings and battles, was at work on an epictherefore caused a considerable stir.5Propertius bears witness to the shock(2.34.61–66):

Actia Vergilium custodis litora Phoebi,

Caesaris et fortis dicere posse ratis,

qui nunc Aeneae Troiani suscitat arma

iactaque Lavinis moenia litoribus

cedite Romani scriptores, cedite Grai!

nescio quid maius nascitur Iliade

The Actian shore in Phoebus’ charge and Caesar’s

brave fleet – Vergil can tell of these,

who now calls up the arms of Trojan Aeneas

and the walls he built on the Lavinian shore

Give way, Roman writers! Give way, Greeks!

A thing greater than the Iliad is being born.

The Vergilian echoes in Troiani, arma, and Lavinis litoribus suggest that

Propertius has heard at least the opening of the emerging poem, but he

3 So at Oxyrhynchus by the late first century a scribe practiced his letter forms by

copying over lines of the Aeneid: see Cockle 1979 The text is now P Oxy 50.3554.

For Vergil’s rapid dissemination throughout the Roman world, see Horsfall 1995: 249–55.

4 On the so-called recusatio, see Williams 1968: 102–3, Nisbet and Hubbard 1970: 81–

83 and 1978: 179–83, and Lyne 1995: 31–39 Cic Q fr 3.7.6 mentions an epic ad Caesarem that was never released (Allen 1955); Att 1.16.15 reports his (waning) hopes

for a poem by Archias.

5 Vergil always had detractors – Marcus Agrippa may have been one of them (Suet Vita Verg 44) – but Vell 2.36.3 had ample reason to call him princeps carminum Atticus’

learned freedman Caecilius Epirota was teaching Vergil’s bucolics to his privileged

charges by the early 20s See Suet Gram 16.3, with Kaster 1995: 188–89.

has of course gotten quite a lot wrong The Aeneid will have little to

say about Actium or, at least directly, about any contemporary event.6Propertius reveals less about the poem Vergil was writing than about what

he himself expected a contemporary epic to contain And with reason

The success of Ennius’ Annales had so codified and canonized the early

history of Rome and established history as the subject of Latin epic thatlater poets could imagine little more than a continuation of its story Byseizing upon the relatively obscure story of Aeneas in Italy, Vergil was able

to solve one of the great literary problems of the day His combination ofmythological and historical tendencies proved both artistically valid andideologically respectable and thus restored epic to a prominence it wouldnot again soon lose.7

Yet even when the practice of epic was at its lowest ebb, the idea of epic

never lost its status It was always the most prestigious, however achieving, poetic genre of Roman antiquity and by a kind of scholarlymetonymy became the very symbol of literature itself Some writers evenworked from the assumption that Rome did not have a literature at alluntil it had epic The earliest surviving fragment of a Roman literary his-tory, remnants of a didactic poem in trochaic septenarii by an aristocratnamed Porcius Licinus, makes precisely that claim:8

under-Poenico bello secundo Musa pinnato gradu

intulit se bellicosam in Romuli gentem feram

6 For the Battle of Actium on Aeneas’ shield (Aen 8.671–713), see Gurval 1995: 230–40, and for the association of the heroic parade that ends Aeneid 6 with the images of the

Forum Augusti, see Degrassi 1945 and Zanker 1988: 210–15 Servius thus had good reason to think that Vergil’s intention was “to praise Augustus through his ancestors,” though he was not necessarily right in thinking so How much of the poem Propertius heard and his thoughts on hearing it remain matters of debate See Tr¨ankle 1971 and the rejoinder of Stahl 1985: 350–52, and for Vergil’s recitation of his work in progress, see Horsfall 1995: 19 He was said to have been a very effective reader: the poet Julius

Montanus envied his voice and delivery (Vit Verg 29 = Suet Rhet fr 3).

7 So Horsfall 1995: 249: “The Aeneas-legend was, prior to Virgil, a political plaything

of the Iulii Caesares It was the Aeneid which transformed it into a truly national story.”

For the development of that story, see Gruen 1992: 6–51 and for Aeneas’ eventual prominence in Augustan art, see Zanker 1988: 201–10 Thomas 2001: 34–54 traces the (posthumous) development of this “Augustan” Vergil On Aeneas in earlier Roman

epic, see also Goldberg 1995: 54–55 (Naevius), 95–101 (Ennius) Thus Serv ad Aen.

1.273: “Naevius et Ennius Aeneae ex filia nepotem Romulum conditorem urbis tradunt.”

8 Licinus ap Gell 17.21.44 The poem probably dates to the later second century, but

precision is impossible See Leo 1912: 66–69, Courtney 1993: 82–86, Schwindt 2002: 64–70, and for the poet’s identity, Badian 1972: 163–64.

At the time of the Second Punic War, the Muse with wingedstep introduced her warlike self to Romulus’ savage race.

The Muse that reveals herself to be bellicosa must be the epic Muse: Licinus

is thus associating the beginning of Latin poetry with the rise of epic Hispoint of inception, the Second Punic War, was the time of Naevius, and

so the probability is that he identified the introduction of epic specifically

with the appearance of Naevius’ Bellum Punicum.9

Lucretius reveals a similar sense of epic’s importance, though by hisreckoning it was not Naevius but Ennius, “who first brought the ever-green crown from pleasant Helicon” (‘qui primus amoeno/detulit exHelicone perenni fronde coronam’, 1.117–18) Horace, perhaps echoing

Porcius Licinus, famously agreed (Ep 2.1.156–59):

Graecia capta ferum victorem cepit et artes

intulit agresti Latio sic horridus ille

defluxit numerus Saturnius, et grave virus

munditiae pepulere

Captive Greece captured the savage victor and brought

the arts to rude Latium Thus that crude Saturnian verse

drained away, and refinement drove off

the fetid smell

As hexameter poets themselves, Lucretius and Horace naturally seeEnnius’ metrical innovation as the decisive step in the history of Latinverse: the new, flexible hexameter not just enhanced the technical andaesthetic possibilities of Latin epic but helped it escape the echoes of rit-ual and superstition that inevitably clung to the old Saturnian cadence.10

The sentiments and the very language of their claim nevertheless lookback to Porcius Licinus.11

9 Licinus’ “Poenico bello” may play on the title The epic dealt with the First Punic War (264–241 B.C.), in which Naevius had himself served (Varr ap Gell 17.21.45), but as

a work of his old age (Cic Sen 50), it is dated to the last years of the century Schwindt

2002: 67–69 revives the old idea that Licinus’ reference is to Ennius, who came to Rome in 204, but his epic was then some twenty years in the future Courtney 1993: 84–85 makes a more convincing case for Naevius (and for the agreement of Licinus’

bellicosam with se).

10 Cf Cic Div 1.114, quoting Ennius’ famous line about the meter of “Faunei vatesque.”

Not that the Saturnian was entirely forgotten Catullus’ deliberate recollection of Saturnian cola in his hymn to Diana (c 34) suggests the abiding potency and ritual connotation of its rhythms even in the mid-first century.

11 Horace later dates the coming of tragedy “post Punica bella” (616–18), also a likely

echo of Licinus Similar language appears in Liv 7.2.3, who calls the ludi scaenici

Even Cicero, whose famous digression in the Brutus reflects the

anti-quarians’ tendency to trace Roman literary history back to the

third-century ludi scaenici, reveals a similar emphasis The passage from which

we extract that chronology does not itself begin chronologically with

Andronicus’ production in 240 B.C but in medias res by paraphrasing –

and praising – Ennius’ claim to preeminence in poetry Just as Homereclipsed the work of his predecessors, says Cicero, so Ennius left all rivalsfar behind:12

“nec doctis dictis studiosus quisquam erat ante hunc”

ait ipse de se nec mentitur in gloriando: sic enim sese res habet

“Nor was anyone careful over educated speech before him”

He says that about himself, nor does he lie in his pride: that is how it was

The emphasis is significant By beginning his discussion with Ennius,Cicero not just avoids a purely chronological approach to literary history

but obscures that history’s origin in the ludi scaenici, which is what a strict

chronology would emphasize The Ennius of this passage is specifically

the poet of the Annales, not the tragic dramatist, and the argument soon

makes clear that Cicero’s measure of the Roman achievement in poetry is

its accomplishment in epic Consular years and the staging of ludi scaenici

provide dates for poetic careers and gave his friends Atticus and Varroample grist for their antiquarian mills,13 but the poems Cicero values

most here in the Brutus are Naevius’ Bellum Punicum and the Annales.

Their prominence seems natural, even self-evident, but things couldhave turned out otherwise Porcius Licinus, Lucretius, Cicero, and Horacehad these texts to think about not because they were in continuous cir-culation from their first appearance on the cultural scene on into theAugustan age, but because the first Romans to take the editing and

(an Etruscan import) a “res nova bellicoso populo.” A century later, Suet Gram 1.1

represents literary study as something brought to a “rudis scilicet ac bellicosa etiamtum civitas.”

12 Cic Brut 71, cf Or 171 From this testimony Skutsch 1985: 373 restores exempli gratia, “nec dicti studiosus fuit Romanus homo ante hunc” (209) For the cultural

significance of the phrase “dicti studiosus” (= philologos), see Skutsch 1968: 6–7 and Barchiesi 1993: 119–20.

13 Cic Brut 60 and 73 make clear that the researches of Varro and Atticus included

veteres commentarii that integrated literary events into the framework of consular dating,

but none of these sources was sufficiently official to settle problems of chronology Thus

the famous conflict with Accius As Brut 72 observes, “est enim inter scriptores de numero annorum controversia.” The authority of these commentarii remains a matter

of debate See M ¨unzer 1905: 55–61 and Suerbaum 1968: 299–300.

dissemination of poetry seriously also looked first to epic When tonius, over two centuries after the fact, surveyed the history of literaryexegesis at Rome, he located the first signs of professional activity – thelegacy, however indirect, of a famous visit by the great Pergamene scholarCrates of Mallos around 167 – in the treatment of epic texts According

Sue-to SueSue-tonius, the poems of Naevius and Ennius were falling increasinglyinto obscurity until they were rescued for a new generation of readers by

a combination of popularization and education C Octavius Lampadio,

whose efforts made him famous, read and explicated the Bellum Punicum

and even gave the poem a new look by dividing it into books.14 A

little later, Q Vargunteius performed a similar service for the Annales

through attention that included readings before large, appreciative ences What these men found as obscure works (“carmina parum adhucdivulgata”) they thus set on their way to becoming the classics of Horace’sgeneration, eventually earning for themselves the authority he would so

audi-famously lament (“adeo sanctum est omne vetus poema,” Ep 2.1.54).

The limited circulation of Naevius’ epic by the later second centurysurprises nobody – Ennius may himself have hastened its decline – but

it is harder to grasp the implications of Suetonius’ report that the est poem of the Republic also had to be rescued from relative obscurity

great-by a grammatically minded dilettante Reception of the Annales is not

generally thought so problematic It surely enjoyed an immediate cess Ennius praised the great men of Rome and earned their praise

suc-in turn (Cic Arch 22), and the three books he added to the

origi-nal poem in old age surely responded to the demands of contemporaryacclaim The confidence with which Ennius assumed Homer’s mantleand replaced the Saturnian cadence with the Muses’ foot aroused anadmiration in later generations that is easily read back not just into his

14 Suet Gram 2.2 with Kaster 1995: 61–67, Christes 1979: 7–8 Book division has an Alexandrian ring (cf Heiden 1998), but the eighteen books of Ennius’ Annales would

have provided Lampadio with a model closer to home For the book divisions of that poem (original to Ennius), see Skutsch 1985: 5–6, and for early scholarly interest in it, 8–10 Gell 18.5.11 tells how the rhetor Antonius Julianus went to considerable trouble and expense to consult a text of Ennius with Lampadio’s autograph emendations, but that “liber summae atque reverendae vetustatis” was probably a forgery So Zetzel

1973: 239–41 Whether Lampadio edited Ennius as well as Naevius (cf Fronto ad

M Caes 1.7.4) remains uncertain Livius Andronicus’ Odusia must have undergone a

similar process, though no source deigns to take it seriously (“tamquam opus aliquod

Daedali,” Cic Brut 71) It was nevertheless available for the distinguished teacher Orbilius to pound into his students: Hor Ep 2.1.69–71, Suet Gram 9 See Brink

1982: 118–20, Kaster 1995: 128–34.

own time but continuously thereafter The Annales’ path to fame,

how-ever, was probably not so direct because there were also inherent limits

to its appeal Its innovations were certainly resisted in some quarters The

Bellum Punicum continued to attract readers into Horace’s lifetime, and

resentment of Ennius’ Hellenizing efforts in diction and meter may linger

in the claim that after Naevius’ death the Romans forgot how to speakLatin.15Enough reactionary bravado endured into the post-Ennian world

to produce the so-called Carmen Priami, which appealed not to Ennius’

Muses but to the “veteres Casmenas” in what purports to be Saturnianverse.16 In the 130s, Decimus Brutus Callaicus dedicated a grand newtemple to Mars, which featured a monumental statue by Scopas and adedicatory inscription by the famous Accius – in Saturnians Trochaicrather than dactylic rhythms remained common in didactic poetry.17

Countertendencies like these suggest that Ennius’ replacement of theold aesthetic was neither complete nor immediate among readers with aliterary turn

Less technically committed readers would also have had a quite practicalreason to lose interest in the poem: its content was quickly overtaken

by events Memory of the Hannibalic war dimmed with the years andwith the fading reputation of the Scipios.18 The poem’s conclusion waseven less appealing The chastisement of Aetolians and Istrians, which

the Annales celebrated so earnestly, soon paled before Aemilius Paullus’

victory in Maecedonia, which opened a new chapter in the politicaland cultural life of Rome by securing Roman dominance in the east,while vastly increasing the westward flow of Greek material and literaryculture The developments that Ennius in the 170s saw as the pinnacle ofRoman achievement thus turned out to be little more than its prelude.Glorification of what so quickly became old news could easily have meant

15 So Norden 1915: 145, citing the Naevian epitaph of Gell 1.24, “obliti sunt Romae

loquier lingua Latina.” Cf Hor Ep 2.1.53–54: “Naevius in manibus non est et

men-tibus haeret/ paene recens?”

16 Its single surviving line is quoted by Var L 7.28: “veteres Casmenas cascam rem volo profarei.” The false archaism of Casmena and the lack of word boundary after the fifth syllable (the so-called caesura Korschiana) indicate a late imitation of the old epic style.

See Cole 1969: 19–21, Timpanaro 1978.

17 Porcius Licinus’ literary history and Accius’ Pragmatica were trochaic poems So was Ennius’ own Scipio, though it probably predates the Annales For Brutus’ temple, Plin Nat 36.26, Cic Arch 27 with Schol Bob (Stangl 1912: 179), Val Max 8.14.2.

18 Ennius’ Scipio no doubt enhanced Africanus’ original reputation, but the Scipionic

legend that eventually restored his fame after the trials of the 180s was the work of Polybius’ generation See Walbank 1967.

oblivion for the Annales It certainly helps explain why Q Vargunteius

would have found the poem “parum divulgatum.”

Why, then, did he take it up? A Greek precedent may have played apart Homer had long been the archetypical poet of the Greek world,and epic was the genre of first reference in Hellenistic culture Crates’lectures at Rome doubtless centered on Homeric exegesis, and theymay well have inspired men like Vargunteius to search for equivalentLatin texts on which to perform comparable exercises.19 A Latin lit-erature would require a Latin Homer, which meant locating poemsthat could bear the burden of national identity while withstanding thescrutiny of grammarians and poets Lampadio and then Vargunteius

found such texts in the Bellum Punicum and Annales.20 These lar epics had the further attraction of conveying a strong sense of personalauthorship: both Naevius and Ennius expressly wrote themselves intotheir poems Ennius even proclaimed himself to be Homer reborn.21Var-gunteius’ newly found literature thus acquired at a stroke both poems andpoets

particu-Why, though, were audiences attracted to his readings?22 Epic might

by its very nature appeal to the people of education and privilege mostreachable by scholarly efforts, but it is worth asking what this new audi-ence heard in Ennius’ poem that their fathers apparently did not hear.Some of the new appeal may have been simply the result of Romantaste and experience catching up with Ennian innovation: the soundplay, metrical tricks, neologisms, and Homeric echoes so characteristic

of the Annales would have been particularly amenable to Crates’ style of

criticism and to the literary interests it would have aroused.23There was

19 Cf Aristot Poet 1448b27-49a15 and Aeschylus’ characterization of his plays (ap.

Athen 8.347e) as “slices from the great banquet of Homer.” For Homer’s prestige

in the Hellenistic world, see Brink 1972: 548–56, Cameron 1995: 273–77, and for Crates’ Homeric scholarship, Pfeiffer 1968: 234–46, Garbarino 1973: 2.356–62, Nagy 1998: 215–28 The fragments are now gathered and edited with commentary by Brog- giato 2001: 13–77 and 140–239.

20 For epic at Rome as the “highest” genre and thus Ennius as summus poeta, see Dahlmann

1963: 17–19 Keith 2000: 8–18 notes the special prominence of epic in Roman cation.

edu-21 Naev BP fr 44 (Gell 17.21.45), Enn Ann fr iii–x, with Skutsch’s notes.

22 Suet Gram 2.2 is explicit about Vargunteius’ (popular) public readings: “quos [Annales

Enni] certis diebus in magna frequentia pronuntiabat.”

23 The reemergence of Philodemus’ polemic monograph On Poems has vastly increased

our knowledge of Crates’ doctrines and their possible influence on the Romans See Asmis 1992, Janko 2000: 120–34, Broggiato 2001: xxvii–xli.