Virgil recomposed the mythological and secular centos in antiquity

The remaining four contain Christian material: the Cento Probae ofFaltonia Betitia Proba; Pomponius’s Versus ad Gratiam Domini; the anonymous De Verbi Incarnatione; and the De Ecclesia,

Virgil Recomposed

AMERICAN CLASSICAL STUDIES

The Augustan Succession

An Historical Commentary on Cassius Dio’s ‘‘Roman History’’Books 55–56 (9B.C.–A.D 14)

PETERMICHAEL SWAN

Virgil Recomposed

The Mythological and Secular Centos in Antiquity

SCOTT MCGILL

Oxford University Press, Inc., publishes works that further

Oxford University’s objective of excellence

in research, scholarship, and education.

Oxford New York

Auckland Cape Town Dar es Salaam Hong Kong Karachi

Kuala Lumpur Madrid Melbourne Mexico City Nairobi

New Delhi Shanghai Taipei Toronto

With offices in

Argentina Austria Brazil Chile Czech Republic France Greece

Guatemala Hungary Italy Japan Poland Portugal Singapore

South Korea Switzerland Thailand Turkey Ukraine Vietnam

Copyright # 2005 by The American Philological Association

Published by Oxford University Press, Inc.

198 Madison Avenue, New York, New York, 10016

www.oup.com

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 prior permission of Oxford University Press.

Library of Congress Cataloging-in-Publication Data

McGill, Scott, 1968–

Virgil recomposed: the mythological and secular centos in antiquity / Scott McGill.

p cm.—(American classical studies; no 48)

Includes bibliographical references and index.

ISBN-13 978-0-19-517564-6

ISBN 0-19-517564-6

1 Virgil—Adaptations—History and criticism 2 Virgil—Parodies, imitations, etc.—History and criticism 3 Epic poetry, Latin—Adaptations—History and criticism 4 Centos—History and criticism 5 Mythology, Roman, in literature 6 Virgil—Appreciation—Rome I Title II Series PA6825.M395 2005

871'.01—dc22 2004022887

9 8 7 6 5 4 3 2 1

Printed in the United States of America

To My Son, Charlie

This book began when I decided to take Ausonius outside with me on a lazysummer day and read the Moselle Opening Green’s edition at random, I insteadencountered the Cento Nuptialis, and a dissertation topic was born Desidiosumiuvat Fortuna

Several years have passed since then I completed and many times revised thedissertation; finishing (or better, abandoning) it now as a monograph, I feelsomewhat wistful, since the project is so closely associated with a remarkable time

in my life, and with many remarkable people I learned much as a graduate studentfrom my professors, particularly Michael Anderson, Bob Babcock, SusannaBraund, and Gordon Williams, who were all models of mentoring, prodding mepatiently and amiably to think harder and with more clarity As an advisor sinetitulo and a reader of the dissertation, Michael Roberts helped me to realize thisproject in more ways than I can recount Finally, John Matthews and EllenOliensis were as generous, supportive, and rigorous advisors as I could have hoped

to have

Since arriving at Rice University, I have benefited from the healthy and turing environment that the university and the Classical Studies Departmentcreate for its junior faculty In more concrete terms, I appreciate the editorialwork of Cyndy Brown, which certainly sped my progress My colleagues, CoulterGeorge, Christopher Kelty, Michael Maas, Hilary Mackie, Don Morrison, Car-oline Quenemoen, and Harvey Yunis also facilitated the preparation of mymanuscript Conversation with them, teaching alongside them, and having them

nur-as editors have been truly enjoyable and productive experiences

When this book needed a final round of scrubbing, Donald Mastronarde and theanonymous readers at the APA provided me with both general and specific assis-tance Their criticism allowed me to avoid many errors and escape many pitfalls—though fallibility is stubborn, and I am sure that mistakes and infelicities remain, forwhich of course I am alone responsible I must also thank Eve Bachrach, JessicaRyan, and Gwen Colvin at Oxford University Press for their guidance

XVIII Cento Nuptialis from The Works of Ausonius, by R.P.H Green (1991),was reprinted by permission of Oxford University Press The translation ofAusonius’s epistle to Paulus was reprinted by permission of the publishers and

the Trustees of the Loeb Classical Library from Ausonius: Volume 1, LoebClassical Library vol 96, translated by H G Evelyn-White (Cambridge, Mass.:Harvard University Press, 1919) The Loeb Classical Library1 is a registeredtrademark of the President and Fellows of Harvard College The De Alea wasreprinted by permission of Loffredo Editore Napoli SpA Finally, the Epitha-lamium Fridi and Medea were reprinted by permission of K.G Saur Verlag.Throughout the entire process of writing this book, my family has been ananchor I particularly want to thank my brother Sean and my parents, who taught

me by example how to be disciplined and to stick to a task until it is done Indifferent ways, I am indebted to old friends in the Northeast (though theacademic diaspora has taken us to far-flung locations) and new ones in Houston,and especially to Joseph Luzzi Finally, Sarah Ellenzweig makes everythingworthwhile and better than I deserve

At the risk of being precious, let me end by saying what a pleasure it hasbeen these past years to read and think about not only some of the wildest texts

in antiquity but also Virgil, who as a poet has no superior and just a few equals.non, mihi si linguae centum sint oraque centum (A 6.625),

caelicolae magni (A 10.6) possim superare labores (A 3.368)

carmina qui (G 4.565) matrisque dedit tibi, Mantua, nomen (A 10.200)

Abbreviations xi

Text Editions Used xiii

Introduction xv

1 Playing with Poetry: Writing and Reading

the Virgilian Centos 1

2 Tragic Virgil: The Medea 31

3 Virgil and the Everyday: The De Panificio and De Alea 53

4 Omnia Iam Vulgata? Approaches to the

BOOKS

AL R Anthologia Latina, Alexander Riese, ed Leipzig: Teubner, 1894

AL SB Anthologia Latina, D R Shackleton Bailey, ed Stuttgart: Teubner,

EV Enciclopedia Virgiliana Ed Francesco Della Corte Rome: Instituto

della Enciclopedia Italiana, 1996

Keil Heinrich Keil, Grammatici Latini, 7 vols; Heinrich Keil, ed Leipzig:

Teubner, 1855–80

OCD Oxford Classical Dictionary 3rd ed Simon Hornblower and Antony

Spawforth, eds Oxford: Oxford University Press, 1996

OLD Oxford Latin Dictionary P.G.W Glare, ed Oxford: Oxford University

Press, 1982

PLRE Jones, A.H.M., J.R Martindale, and J Morris Prosopography of the

Later Roman Empire Cambridge: Cambridge University Press, 1971–1992

RE Paulys Real-Encyclopa¨die der Klassischen Altertumswissenschaft

Stutt-gart: A Druckenmu¨ller, 1893–1972

ThLL Thesaurus Linguae Latinae Leipzig: Teubner, 1900–

VSD Vita Suetonii/Donati, Vitae Vergilianae Antiquae, ed Georgius

Brugnoli and Fabius Stock Rome: Typis Officinae Polygraphicae,1997

JOURNALS

AJAH American Journal of Ancient History

AJP American Journal of Philology

BMCR Bryn Mawr Classical Review

CJ Classical Journal

CP Classical Philology

CQ Classical Quarterly

CSCA California Studies in Classical Antiquity

HSCP Harvard Studies in Classical Philology

MD Material e Discussioni

MP Medieval Philology

PVS Proceedings of the Vergil Society

RLM Rheinisches Museum fu¨r Philologie

RLAC Reallexikon fu¨r Antike und Christentum

SIFC Studi italiani di filologia classica

TAPA Transactions of the American Philological Association

YCGL Yearbook of Comparative and General Literature

ZPE Zeitschrift fu¨r Papyrologie und Epigraphik

Text Editions Used

Nar-Cento Nuptialis, in The Works of Ausonius, ed R.P.H Green Oxford: Clarendon Press,1991

De Alea, in Il centone De Alea, Studi Latini 44, ed Gabriella Carbone Naples: Loffredo,2002

Epithalamium Fridi, in Luxurius, ed Heinz Happ 2 vols Stuttgart: Teubner, 1986.Medea, in Hosidius Geta: Medea Cento Vergilianus, ed Rosa Lamacchia Leipzig: Teubner, 1981

OTHER FREQUENTLY CITED EDITIONS

ANTHOLOGIA LATINA: Anthologia Latina I.1, ed D R Shackleton Bailey Stuttgart:Teubner, 1982

CATULLUS: Catullus, ed C J Fordyce Oxford: Clarendon Press, 1961

CLAUDIAN: Claudii Claudiani Carmina, ed John Barrie Hall Leipzig: Teubner, 1985.DRACONTIUS: Oeuvres, ed E´tienne Wolff Vol 4 Paris: Les Belles Lettres, 1996.MENANDER RHETOR: Menander Rhetor, ed D A Russell and N G Wilson Oxford:Clarendon Press, 1981

OVID: Heroides XII, ed Theodor Heinze Leiden: Brill, 1997 Metamorphoses, ed W S.Anderson Leipzig: Teubner, 1977

QUINTILIAN: Institutionis Oratoriae Libri Duodecim, ed Michael Winterbottom ford: Clarendon Press, 1970

Ox-SENECA: Tragoediae, ed Otto Zwierlein Oxford: Oxford University Press, 1986.SERVIUS: Servii Grammatici Qui Feruntur in Vergilii Carmina Commentarii, ed GeorgThilo and Herman Hagen 3 vols Hildesheim: Olms, 1961

SIDONIUS: Sidoine Apollinaire, ed Andre´ Loyen 3 vols Paris: Socie´te´ d’E´dition ‘‘LesBelles Lettres,’’ 1960–70

STATIUS: Silvae, ed E Courtney Oxford: Clarendon Press, 1990

VENANTIUS FORTUNATUS: Venance Fortunat Poe`mes, ed Marc Reydellet Vol 1.Paris: Les Belles Lettres, 1994

VITA SUETONII/DONATI: Vitae Vergilianae Antiquae, ed Georgius Brugnoli andFabius Stock Rome: Typis Officinae Polygraphicae, 1997

A note on my method of citing lines and passages in the centos I have chosen

to include in parentheses the Virgilian provenance (with E standing for theEclogues, G for the Georgics, and A for the Aeneid) for each verse segment ineach line that I cite This, I recognize, interrupts the flow of the line, with Virgilbreaking into the experience of reading the passages in the centos I believe thatsuch intrusions are appropriate The centos demand to be read not as trans-parent texts, but as works having a Virgilian basis

The Virgilian centos are some of the more striking texts to survive from Latinantiquity A cento—a word that in literature has the meaning ‘‘patchworktext’’1—is comprised of unconnected verse units taken from the Eclogues,Georgics, and Aeneid and pieced together to create narratives that differ fromVirgil’s own.2These units may consist of a segment of a hexameter line; an entireline; a line and some section of the following line; and rarely two or three entirelines.3Sixteen Virgilian centos remain from antiquity, ranging in date from ca

200 to ca 534.4Twelve are on mythological or secular subjects: Hosidius Geta’sMedea; Ausonius’s Cento Nuptialis; Luxurius’s Epithalamium Fridi;5Mavortius’sIudicium Paridis; and eight anonymous works, the De Panificio, De Alea, Nar-cissus, Hippodamia, Hercules et Antaeus, Progne et Philomela, Europa, and Al-cesta The remaining four contain Christian material: the Cento Probae ofFaltonia Betitia Proba; Pomponius’s Versus ad Gratiam Domini; the anonymous

De Verbi Incarnatione; and the De Ecclesia, perhaps written by Mavortius.6The mythological and secular centos are very different texts from theChristian variety The settings in which and for which the former works werecomposed, the ways their authors rewrote Virgil, and many of the interpretiveissues the texts raise all distinguish them from the Christian pieces In light ofthese disparities, my book isolates the mythological and secular centos A study

of these works will contribute to the growing field of scholarship on Christian Latin poetry in late antiquity (i.e., texts without Christian content andusually with classical prototypes and themes).7 The mythological and secularcentos especially help us explore the enthusiasm for light and playful versecomposition that abided in that era In addition, an examination of the centosadvances the current scholarship on Virgil’s reception Of particular value is theattention that the book gives to the late antique world Regarding Virgil’s re-ception in that period, there has been a great amount of work done on howChristian writers, and particularly the Church Fathers, responded to him.8While this subject is an important one, there remains much to be said about howaudiences not viewing Virgil through a Christian lens—for example, poetsworking with pagan and secular material, grammarians and other late antiquecritics, and students—treated him The centos help to illuminate these matters,

non-and at the same time allow us to revisit pertinent responses to Virgil that curred earlier in antiquity and to explore relevant moments in the interpretation

oc-of Virgil since that period, up to today So too, the centos provide insights intoseveral formal and thematic elements in Virgil’s poetry itself.9

Aspects of the mythological and secular centos bear upon a wide range ofother subjects of general interest in Latin poetry, which are in turn important inliterary studies as a whole These include questions related to reception theory(a topic vitally connected but not limited in this book to Virgil’s Nachleben) andgenre theory.10An issue of vital importance in the study of the centos, more-over, is how those radically intertextual works engage with their Virgiliansources allusively and speak to ideas and problems in allusion studies Thebroad hermeneutic reach and value of the centos are yet another reason why theworks are worthy of exclusive attention.11

The origin of the Virgilian cento lies in the Homeric cento, of which logical, secular, and Christian examples survive.12Such a binary view does nottake into account all ancient centos It excludes evidence for Greek examplesthat reuse Pindar and Anacreon,13as well as a Latin cento composed from apoet other than Virgil, Ovid’s work in malos poetas comprised of the verses ofMacer, a lost piece to which Quintilian refers.14Even so, the contention thatthe Virgilian cento arose as a counterpart to the Homeric cento is a sound one,based as it is on the irrefutable fact that Homer and Virgil are the principalsources for such texts in antiquity.15This cannot be coincidental AssociatingVirgil with Homer serves as one of the dominant gestures in Latin literaryculture from Propertius (2.34.65–66) to Macrobius (Sat 5) and beyond.16Amid this literary landscape, it would have been natural to take a poetic formlinked to Homer and apply it to Virgil, the poet of equal stature in the Romanworld Centonists are drawn to such canonical authors To present a cento isalways on one level to trade in cultural capital and to affirm one’s highbrowcredentials Moreover, the loftier the rank of the poet being rewritten, thegreater the effect of a cento Readers will be more likely to be familiar withsource poetry that resides at or near the top of the canon, and so will be morelikely to feel more strongly the frisson that centos, as the reconstituted poetry of

mytho-an eminent author, are designed to elicit

Not everyone has responded or will respond to the Virgilian centos withappreciative wonder or even neutral surprise Some ancient observers, for in-stance, raised objections to the texts None of these disapproving notices,however, should be seen to condemn the cento form as such; for upon closerexamination, they simply reflect the particular concerns of the figures thatvoiced them, Tertullian (De Praescr Haer 39), Jerome (Ep 53.7), and Ausonius(Cento Nuptialis, praef esp 1–5) Tertullian and Jerome were interested inestablishing the cento—for Tertullian, the mythological and secular type, and forJerome, the Christian17—as a parallel to how certain people misread the Bible,fitting scripture to their own purposes and so changing the original meaning of

the source material.18Jerome was also troubled by the alteration of Virgil so thathis verses related the story of the Bible, which caused some to posit Virgil as aChristian sine Christo; but for Jerome the act of altering Virgil itself was not atissue Tertullian and Jerome’s reactions are thus tailored to their specific con-cerns and interpretive and cultural climates While they taint the cento throughassociation with the misinterpretation of scripture, and while Jerome is un-comfortable with the Christianizing of Virgil’s poetry and of Virgil himself, theircritiques, filtered through a Christian lens, do not function as general literarycriticism and should not be taken as authoritative denunciations of the form’spoetic and aesthetic traits.19 In the prefatory epistle attached to his CentoNuptialis, meanwhile, Ausonius disparages cento composition as part of hisstrategy of modest self-presentation, and so for rhetorical ends (More on Au-sonius’s stance in chapters 1 and 5.) Like the comments of Tertullian andJerome, Ausonius’s are not definitive statements on the lack of merit of thecento per se.20

In the modern age, several scholars have also been appalled by the centoand, pursuing slash-and-burn literary criticism, have sharply condemned theform.21The majority of these negative reviews can be attributed to a classicizingprejudice that considers High Literature and the Great Author sacrosanct andscorns odd and secondary works that encroach on those monuments.22Suchreactions are a reminder that appropriative works of all kinds are prone to elicitaesthetic disapproval and even moral outrage from some quarters Though none

to my knowledge does so explicitly, perhaps in their minds the disapprovingcritics also conflate centos and plagiarism, or view cento composition as a type

of theft That would be a mistake, since the kind of open, reconstitutive propriation that occurs in the centos is far from plagiarism’s furta.23

ap-In this study, I wish to provide a counterweight to the often harsh responses

to the cento, responses that are inadequate in their proprietary and closed vision

of texts (and not unimportantly, canonical texts), if sometimes entertaining intheir Housmanian vitriol The reflexive condemnation of the patchwork texts forbeing curiosities rather than high literary art, and still worse, for turning highliterary art into a curiosity, misses the point of the works Centonists themselveswould no doubt agree that their works are strange and parasitic, and that thetexts fail to measure up to the aesthetic standards of great literature Indeed, bytheir very nature the centos are and do very different things from what con-ventional high poetry is and does Critics should bear this in mind and approachthe works on their own terms I fully recognize that, even when this injunction

is followed, the patchwork technique and texts will not be to everyone’s tastes.Yet this book aims to demonstrate that the twelve mythological and secularcentos can provide audiences with one of the more intricate and exciting readingexperiences of any poetry in antiquity.24

Once the cento form had been imported from the Greeks, it became part of aliterary world that in various ways treated Virgil as an open work, or as a body of

material that could be reworked to yield fresh texts.25 Most of the pursuitsthrough which certain members of Virgil’s ancient audience at certainpoints recast his poetry and made it anew have parallels in the ways Greekaudiences treated Homer, and indeed result from the application of the formula

‘‘As Homer, so Virgil.’’ My area of focus, however, is strictly the Roman contextand how the writing of mythological and secular centos relates to practices thatarose around Virgil In this arena, we find a wide range of works showing thatVirgil’s poetry was not only canonical and monumental but also a rich source forderivative or secondary composition

Conventional imitation offers one example of how ancient authors recast gil.26Yet there were also practices involving a more direct and insistent reworking

Vir-of Virgilian material The schools were an important setting for these pursuits.Virgil’s poetry, and especially the Aeneid, held a central position in the schools ofgrammar and an important one in the schools of rhetoric from the time CaeciliusEpirota made him a school text in or around 26 BC through late antiquity wherevertraditional secular education survived.27One of the things that students at bothlevels were sometimes called on to do was to rewrite passages of his poetry.Ethopoeiae, or exercises in which students composed a speech for a literary ormythological character,28serve as one example of how Virgilian poetry lay open toyoung authorial hands A notable reference to an impersonation of a Virgiliancharacter comes from Augustine The Church Father relates that as a student in aschool of grammar, he wrote a prose passage in which the Juno of Aeneid 1expresses her anger at being unable to keep the Trojans from reaching Italy Forthis exercise Augustine received a prize, the recollection of which brought him nosatisfaction later in life (Conf 1.17) Another Virgilian ethopoeia comes from En-nodius (473/4–521), who taught rhetoric before becoming bishop of Rome ca 513.Ennodius’s life as a teacher is reflected in his collected Dictiones, among which arepieces that served as Ennodius’s models of school exercises One of the Dictiones is

a work that modern editors have entitled Verba Didonis Cum Abeuntem VideretAeneam This piece, which demonstrates that Virgil has a place in the rhetoricalschools, takes A 4.365 (nec tibi diva parens generis) as its starting point and recastsDido’s speech that follows (A 4.365–387; Dict 28 [CSEL 6, 505–506]).Still more evidence for school exercises that take their cue from Virgil ap-pears in Servius.29In his note ad Aen 10.18, Servius mentions that Titianusand Calvus devised themata, which would appear to mean situations derivedfrom specific passages in Virgil’s poetry, that students might utilize ad dicendiusum.30In the same entry, Servius mentions controversiae written in conjunc-tion with A 10.18–95 Later in his commentary, Servius links Virgil further tothe schools of rhetoric by calling attention to one qui in Vergilium scripsitdeclamationes (ad A 10.532).31Presumably, these various exercises appeared inprose, the usual medium for such material

It may be that students were also educated in verse composition, despiteQuintilian’s assertion that poetry should be only a respite from study (car-mine ludere, studiorum secessus, Inst Orat 10.5.15–16).32 If such instruction

occurred in the grammatical schools, it may have involved recasting Virgil’spoetry, given his importance in the curriculum Whether this also could havehappened in the Latin rhetorical schools is a bit more questionable While verseexercises arose in the rhetorical schools of Egypt,33 the Western curriculumfocused more on the practice of declamation.34 Even so, it is possible thatstudents at that upper level also composed Virgilian ethopoeiae and paraphrases

in verse, or even hexameter declamations derived from the Aeneid 35

Examples of poems deriving from Virgilian school exercises appear in thecodex Salmasianus, a manuscript dating anywhere from the seventh to the earlyninth century.36The Salmasianus preserves a collection of poems put together inAfrica circa 534,37whose compiler is unknown.38This collection, which forms

an important part of the Anthologia Latina,39includes many of the mythologicaland secular centos The first of the Virgilian poems with links to school exercises

is the Locus Vergilianus (AL 214 SB) written by the late fifth- to early century poet Coronatus, who may have been identical to the author, calledCoronatus scholasticus, of a grammatical treatise on final syllables.40(Scholas-ticus could mean that Coronatus was a grammarian or that he simply waslearned.)41Whether or not Coronatus scholasticus was our Coronatus, the author

sixth-of the Locus, being able to write a poem on a Virgilian theme, was in all hood a highly educated adult (for there is no reason to think that the Locus is thework of Coronatus as a schoolboy) The title of the Locus as given in the Sal-masianus derives from A 3.315, where Aeneas encounters Andromache andasks what she has suffered since the fall of Troy Yet it seems that whoever gavethe poem its title was in error, since Coronatus’s work appears to derive from thesection in Aeneid 5 where the Trojan women have set fire to Aeneas’s fleet.42Inthis reading, the term Locus Vergilianus denotes simply a passage with a Virgilianpedigree; but the work is in fact a versified ethopoeia presenting Aeneas’s emo-tional reaction to the arson The other two Virgilian pieces in the Salmasianusare the anonymous Themata Vergiliana (AL 237, 249 SB), which recast

likeli-A 12.653–658 and 4.385–387 respectively As themata, the poems would seem

to be versified versions of the exercise that Servius mentions ad Aen 10.18.While we cannot know if the anonymous authors were students or adults, it ismore plausible that they, like Coronatus, were adults writing poems stemmingfrom pursuits they had known in the schools

Another group of second-degree Virgilian texts consists of several hexametersummaries of his poetry, and mainly the Aeneid The author of one set (AL 1SB), which is preserved in the late antique codex Romanus (R, VergilianusVaticanus Lat 3867), assumes the identity of Ovid in a preface to ten-linesummaries of each book of the Aeneid The figure of Ovid also looms behind thesynopsis whose author greatly increases the degree of difficulty of his under-taking by describing the content of all twelve books of the Aeneid in six lines(AL 672a R) This time, it is a manuscript that attributes the summaries toOvid;43the anonymous author himself offers no ‘‘first-person’’ preface and nomaterial in his poem to support that attribution

Other examples of bravura compression are two anonymous works, one thatsummarizes the entire Virgilian corpus in eleven lines (AL 717R) and the other

in seventeen lines (AL 720a R) The dates of AL 672a 717, and 720a R areuncertain; but it is quite possible that they belong to late antiquity In a lessvirtuoso performance, an anonymous author writes four-line argumenta thatShackleton Bailey presents alongside accounts (also tetrastich) of the Ecloguesand Georgics (AL 2 and 2a SB).44

Still another set of argumenta appears under the name of Sulpicius thaginiensis, who produces six-line summaries of the Aeneid (AL 653 R) Thisfigure is probably not the same Sulpicius who composed an epigram cited inVSD 38 on how Varius and Tucca thwarted Virgil’s dying wish and preservedthe Aeneid from immolation.45 Of that poem, the epitomizer offers a feebleimitation in a preface to his summaries,46perhaps in order to try to pass himselfoff as the Sulpicius Carthaginiensis of Virgil’s biography These two groups oftext probably date again to late antiquity Finally, twelve five-line summaries ofeach book of the Aeneid survive from the so-called Twelve Wise Men (AL 591–

Car-602 R); but it has been convincingly argued that the group is actually tantius, writing under twelve assumed names.47One of the Twelve Wise Men,

Lac-‘‘Basilius,’’ also writes a twelve-line synopsis of the entire Aeneid, with each linedevoted to a book of the epic (AL 634 R).48

The hexameter argumenta in all likelihood derive from the schools ofgrammar, where teachers probably gave students verbal summaries of sectionsand books of the Aeneid before embarking on deeper analyses of grammar andcontent.49Summaries in written form are also quite feasible; these would havebeen in prose, though the possibility that grammarians sometimes composedthem in verse cannot be ruled out It may also be the case that students wouldhave been called on to recite spontaneously synopses of passages or books of theAeneid, as well as to write them, and then in prose, and just maybe in verse.The authors of the hexameter argumenta were probably adults who had beenformally educated in the schools; some may have also been grammatici ShouldLactantius lie behind the Twelve Wise Men, moreover, one of the summarizerswould have been a teacher of rhetoric, assuming Lactantius wrote under theguise of that coterie while a teacher and before his conversion to Christianity

ca 303 These figures may have considered the Virgilian summaries they countered in the schools to be the pursuits upon which they were elaborating asthey developed various approaches to versifying synopses of Virgil’s epic, as well

en-as occen-asionally of his other works Such poetic efforts have the markings ofpastimes undertaken during the authors’ otium and as light entertainment,rather than of pieces intended for practical use in the schools

The mythological and secular Virgilian centonists almost certainly received atraditional education in the schools of grammar and rhetoric,50 where theywould have been relentlessly exposed to Virgil This would have enabled them

to acquire the sort of familiarity with Virgilian poetry necessary to pursue centocomposition when they were adults—for there is no evidence that any of them

was a child-centonist In addition, some of the centonists besides Ausonius, theprofessor of Bordeaux, may have been teachers, and so may have constantlybrushed up their Virgil in their professional lives.51

Links to the schools may also explain why the centonists were acculturated

to recasting Virgil Like the authors of Virgilian scholastic poetry, the patchworkpoets would have learned in the school setting that they could do things withVirgil’s poetry, which stood as a body of material open to recasting.52 Havingcome to understand in the schools that there was no barrier between them-selves and Virgil’s poetry, the centonists may have viewed patchwork compo-sition, being an act of secondary authorship, as an extension of the principlethat they had encountered in the curriculum.53

Of course, cento composition is a very different pursuit from creating Virgilianschool texts and from writing versified Virgilian ethopoeiae, themata, and sum-maries In fact, the processes of recasting Virgil in the mythological and secularcentos have more in common with another method of rewriting his poetry inantiquity This practice stems from Virgil’s vast popularity in the West, a si-tuation that owed much, but not everything, to his place in the scholasticcurriculum Virgil’s verses were something of a lingua franca in Roman society,though of course individuals had varying levels of command of that poeticlanguage One of the results of the renown of Virgil’s poetry was the directquotation of that material in a wide array of settings This could involve usingVirgil’s verses proverbially, as a sort of footnote supporting a particular state-ment or argument (and the belief that Virgil was a master in every branch oflearning contributed to this phenomenon),54or citing it as material to be re-futed.55More relevant to the cento is the practice of quoting Virgil’s verba only

to transform his content or res by adapting his tags to fit new situations andsubjects.56 While such activity was not exclusive to Virgil, audiences trans-formed his verses in this manner most frequently of all poetry in the Latintradition (Greek authors, especially Homer, were also reused in this way byLatin writers.) The productive quotation of Virgil could have comic ends, withhis language applied to low material, and so deflated humorously, or couldoccur in serious contexts

Directly quoting Virgil’s lines and adapting their meaning in new contexts curs in graffiti, and in the process shows that people of all stripes knew someVirgil.57Citing a line from the Eclogues, Georgics, or Aeneid also happened ineveryday conversations, though records of such ephemeral quotations naturallyappear in written sources The literary evidence, which consists of prose workscontaining the transformed line or lines of Virgilian hexameter, also demonstratesthat writers themselves often quoted and adapted the content of Virgilian lines intheir texts

oc-Seneca the Elder gives early examples of citations of Virgil that alter hiscontent In Suas 4.5, Seneca reports that Arellius Fuscus chastised a pupil forquoting A 4.379–380 (scilicet is superis labor est, ea cura quietos / sollicitat) in

a declamation about Alexander the Great when he could have cited moreaptly A 2.553, capulo tenus abdidit ensem, for ornamental purposes In thesame Suasoria (4.4), Seneca shows that Fuscus himself quoted A 4.379–380,applying Dido’s sarcastic (and, as it turns out, wrong) assessment of the gods’interest in Aeneas’s affairs in a speech refuting claims that the gods care aboutchildbirth (Suas 4.4) Seneca adds that Fuscus quoted the line summis cla-moribus, to very boisterous approval.58

Petronius provides further glimpses into the practice of transforming lian verses.59 In the Cena Trimalchionis, Petronius has Trimalchio quote

Virgi-A 2.44, sic notus Ulixes? to refer to his own heroic gourmandizing (Sat 39).Later, when describing the lady of Ephesus, Petronius has her nurse, playingthe role of Anna, quote A 4.34 and 4.38 (Sat 111, 112) Here parody of Aeneid

4 specifically is a goal, with the story of Dido recalled but comically adapted andlowered in the account of the bereaved lady of Ephesus who, despite hersorrow, succumbs to the advances of another man

Much of the rest of the non-Christian literary evidence for the tive quotation of Virgil—and the examples I give are meant to be representative,not exhaustive—is connected to emperors (Far from a sign that the practicewas largely an imperial phenomenon, the cluster of material simply shows that agood amount of the extant Latin prose literature after Virgil was concerned withimperial politics and those in power.) Seneca the Younger provides an example

transforma-of how one could change Virgil for comic purposes with his biting statement thatLivius Geminius will claim to have seen Claudius walking non passibus aequis—aphrase taken from A 2.724, describing Ascanius, and applied to the lame em-peror (Apocol 1.1) Later in the work, Seneca has Mercury cite G 4.90, dedeneci, melior vacua sine regnet in aula, in reference to Claudius (Apocol 3.2).Suetonius notes other instances of such citations of Virgil by or in relation tothe emperors Upon encountering men in dark cloaks rather than traditionalRoman dress at a contio, Augustus cries: Romanos, rerum dominos gentemquetogatam (A 1.282) (Suet Div Aug 40) In doing so, Augustus gives the Vir-gilian line not only a different referent but also a sardonic tone, since hedisapproves of the men’s clothing and is compelling them to remember andadopt the traditional Roman ways of dressing A freedman of Nero, meanwhile,reuses the Virgilian usque adeone mori miserum est? (A 12.646) when he seesthe emperor trying to flee from the perils that surround him (see Suet Ner 47).The freedman, emboldened by what he rightly sees as Nero’s imminent demise,delivers the line in disgust, and he wishes to draw a contrast between theemperor’s cowardice and the behavior of Turnus, exhorted to battle by Juturnadisguised as the charioteer Metiscus

Virgilian lines continued to be quoted and adapted in connection with lateremperors So Quintilian, in the slavering mode of panegyric, ends his praise ofthe poetic achievements of Domitian by citing E 8.13, inter victrices hederamtibi serpere laurus (Inst Orat 10.1.91–92) A later example appears when thepraetorian tribune Julius Crispus expresses his displeasure to Alexander Severus

at the poor progress of the siege of Hatra in 199 by quoting A 11.371 (scilicet utTurno contingat regia coniunx) The point is that Severus’s soldiers, likeTurnus’s in the Aeneid, are suffering in a war waged for no real reason Theemperor seems not to have appreciated the clever way that this criticism wasoffered, as he had Crispus killed.60

Further evidence for such alterations of Virgil appears in the Historia gusta While the historical accuracy of this material may be questioned, it atleast shows that the author of the Historia Augusta, or the sources that he may

Au-be following, is familiar with the act of modifying Virgilian lines Thus Hadrian

is reported to have quoted A 6.869–872, which refer to Marcellus, and to haveapplied the lines to his presumptive heir Verus (see HA Ael Spart., Ael 4.1-3).61Another example appears in conjunction with Diocletian, who is said tohave cited Aeneae magni dextra cadis (A 10.830) at an assembly when he killedAper, himself the assassin of the emperor Numerian Vopiscus, the nominalauthor of the entry in the Historia Augusta in which the anecdote appears, issurprised that a soldier should have such command of Virgil, but adds thatmany are accustomed to quoting passages from comedians and other poets(HA Flav Vop., Num 13.3–5) Vopiscus’s wonder seems misplaced, since Virgilcould have been known in army barracks as well as in imperial palaces.62Transforming the meaning of quoted Virgilian material also occurred outside ofimperial contexts and continued well into late antiquity, as is clear from theepistles of the fifth-century bishop, man of letters, and court figure SidoniusApollinaris Writing to the otherwise unknown Turnus in Ep 4.24.1, Sidoniuscites a line in the Aeneid containing a reference to Virgil’s own Turnus (A 9.6–7).Immediately before doing so, Sidonius says explicitly that he is adapting Virgilianmaterial appropriate to his addressee’s situation: bene nomini, bene negotio tuocongruit Mantuani illud: ‘‘Turne, optime optanti divum promittere nemo / auderet,volvenda dies en attulit ultro.’’ In Ep 5.17.7, moreover, Sidonius describes toEriphius a game of ball in which an enthusiastic Philomathius participated byciting A 5.499: hic vir inlustris Philomathius, ut est illud Mantuani poetae, ‘‘ausus etipse manu iuvenum temptare laborem’’ sphaeristarum se turmalibus constanter im-miscuit At still another point in his collection of epistles, Sidonius alters thereferent of a quoted line from the Eclogues, as he opens a letter to Constantius byapplying E 8.11 to him (a te principium, tibi desinet) (Ep 7.18.1)

Epitaphs constitute another significant body of material in which Virgilianlines are quoted and their meanings modified.63 In the inscriptions, Virgilianmaterial often appears as clausulae; yet there are also instances when wholelines of Virgil are reused.64Notable in this regard is an epitaph found in B 1786(CIL 6.9685), from Rome and inscribed under an image of a butcher’s wifeselling a goose, which goes so far as to reproduce three entire lines of Virgil(A 1.607–609): dum montibus umbrae/lustrabunt, [c]onvexa polus dum siderapascet/ semper honos nomenq tuum laudesque manebunt.65

Cento composition is closely linked to the semantic modification of quotedVirgilian lines.66Patchwork texts exist as just such altered verse units pieced

together to create new, coherent narratives—a more difficult enterprise, to besure Literary and epigraphical examples appear as steps along the way fromciting and adapting isolated Virgilian lines to writing full-fledged centos Inthose examples, we find what might be called inchoate centos, or very shortpassages made up of Virgil’s verses A comic manifestation of this practice, andone that provides the earliest evidence for any type of Virgilian cento, appears inPetronius (Sat 132.11) There Encolpius addresses his unresponsive mentulathrough Virgilian lines strung together to create a brief cento:

illa solo fixos oculos aversa tenebat

nec magis incepto vultum sermone movetur (A 6.469–470)

quam lentae salices (E 5.16) lassove papavera collo (A 9.436)

A very different example appears in Capitolinus’s account in the Historia gusta of Macrinus According to Capitolinus, Macrinus, a praetorian prefectwho assassinated Caracalla and became emperor, had bloodthirsty ways thatincluded reviving the punishment inflicted by Mezentius of tying a living person

Au-to a corpse and forcing him Au-to die a slow and smelly death (see A 8.485–488).Capitolinus says that someone composed a salute to Diadumenus, a rival ofMacrinus, by linking two lines of Virgil (but with no regard for meter): egregiusforma iuvenis (A 6.861 or 12.275, which read egregium forma iuvenem) cuipater haud Mezentius esset (A 7.654) (HA Jul Cap., Opil Macr 12.9).67Finally,certain inscriptions also consist of Virgilian verse units recomposed in centoform Examples are the epitaphs reading concordes animae (A 6.827) quondam,cum vita maneret (A 5.724, with slight alterations: the Aeneid reads dum vitamanebat) (L 1969, 1), and hic pietatis honos: (A 1.253) veteris stat gratia facti(A 4.539) (B 817).68

Petronius’s obscene passage, the political slogan, and the epitaphs are lessvirtuoso literary performances than are the twelve longer mythological and secularcentos Even so, their authors have moved from the semantic alteration of a singleVirgilian verse unit to reassembling discrete units in order to create a new nar-rative This shows that there were instances when Virgil was recomposed in a waythat mirrored in miniature the practices of the Virgilian centonists.69The shortpatchwork texts thus stand at a conceptual and formal midpoint between thequotation of individual Virgilian verses in new narrative settings and the creation

of longer literary texts completely comprised of reconnected Virgilian lines.The existence of inchoate patchwork texts demonstrates further that thosewho wrote the twelve mythological and secular Virgilian centos, while certainlyproducing startling works, were not totally anomalous in the Roman context.The parallels are exact, though different in scale, between the corpus of centosand the short, stitched-together pieces in Petronius, the Historia Augusta, andthe epigraphical material; but the centos also link up with the act of quotingand transforming discrete Virgilian lines Along with perhaps acquiring in theschools a sense of the openness of Virgilian poetry to acts of secondary

authorship, then, the centonists adopted a method of composition with aconnection to how a wide range of Virgil’s ancient audience remade units of hispoetry The available evidence does not reveal whether any of the centonistssaw a link between his pursuit and the widespread adaptation of isolated Vir-gilian verses or the brief cento passages What a broad view of Virgil’s ancientreception does show is that the patchwork authors resembled others whoproductively quoted Virgil, as well as those who in different ways rewrote Virgil

in school exercises or scholastic poems, in considering his poetry far frominviolate and capable of being recast For the centonists, as for various members

of the Roman world engaging in various pursuits, Virgil’s poetry remained lessthe domain of an isolated artistic genius than open and reusable material.70While the Virgilian corpus stood lofty and marmoreal, pieces taken from it,whether passages or individual lines, could, like Deucalion and Pyrrha’s lapides,acquire new formae and new life

Virgil Recomposed

Playing with Poetry

Writing and Reading the Virgilian Centos

An examination of the mythological and secular Virgilian centos requires aninitial overview of the cento form itself A thorough sketch of how authorscompose patchwork poems and how audiences can interpret them will begin todemonstrate that the centos, while eccentric, are complex and rich texts.Moreover, we will begin to see that the centos serve as valuable witnesses toVirgil’s reception in antiquity and beyond, and that they bear upon larger issues

in Latin literature and in literary studies as a whole, particularly those related tointertextuality and allusion

To pursue this inquiry, I turn to Decimus Magnus Ausonius, the prolific poetand important political figure of the later fourth century.1A centonist himself,Ausonius is the only author in antiquity to discuss in detail his own patchworkpoem and the Virgilian cento as a whole These reflections appear in a prefatoryepistle written in prose to the rhetor Axius Paulus,2which Ausonius attaches tohis Cento Nuptialis:3

perlege hoc etiam, si operae est, frivolum et nullius pretii opusculum, quod neclabor excudit nec cura limavit, sine ingenii acumine et morae maturitate cen-tonem vocant qui primi hac concinnatione luserunt solae memoriae negotiumsparsa colligere et integrare lacerata, quod ridere magis quam laudare possis proquo, si per Sigillaria in auctione veniret, neque Afranius naucum daret nequeciccum suum Plautus offerret piget equidem Vergiliani carminis dignitatem tamioculari dehonestasse materia sed quid facerem? iussum erat, quodque est po-tentissimum imperandi genus, rogabat qui iubere poterat imperator Valentinia-nus, vir meo iudicio eruditus, nuptias quondam eiusmodi ludo descripserat, aptisequidem versibus et compositione festiva experiri deinde volens quantum nostracontentione praecelleret, simile nos de eodem concinnare praecepit quam scru-pulosum hoc mihi fuerit intellege neque anteferri volebam neque posthaberi, cumaliorum quoque iudicio detegenda esset adulatio inepta, si cederem, insolentia, si

ut aemulus eminerem suscepi igitur similis recusanti feliciterque et obnoxiusgratiam tenui nec victor offendi hoc tum die uno et addita lucubratione prop-eratum modo inter liturarios meos cum repperissem, tanta mihi candoris tui etamoris fiducia est ut severitati tuae nec ridenda subtraherem accipe igitur opus-culum de inconexis continuum de diversis unum, de seriis ludicrum, de alieno

5

10

15

nostrum, ne in sacris et fabulis aut Thyonianum mireris aut Virbium, illum deDionyso, hunc de Hippolyto reformatum.

et si pateris ut doceam docendus ipse, cento quid sit absolvam variis de locissensibusque diversis quaedam carminis structura solidatur, in unum versum utcoeant aut caesi duo aut unus <et unus> sequenti cum medio nam duos iunctimlocare ineptum est et tres una serie merae nugae diffinduntur autem per caesurasomnes, quas recipit versus heroicus, convenire ut possit aut penthemimeres cumreliquo anapaestico aut trochaice cum posteriore segmento aut septem semipedescum anapaestico chorico aut * * post dactylum atque semipedem quicquid restathexametro, simile ut dicas ludicro, quod Graeci stom
aawion vocavere ossicula easunt: ad summam quattuordecim figuras geometricas habent sunt enim quad-rilatera vel triquetra extentis lineis aut <eiusdem> frontis, <vel aequicruria vel ae-quilatera, vel rectis> angulis vel obliquis: isoscele ipsi vel isopleura vocant, orthogoniaquoque et scalena harum verticularum variis coagmentis simulantur species milleformarum: elephantus belua aut aper bestia, anser volans et mirmillo in armis,subsidens venator et latrans canis, quin et turris et cantharus et alia eiusmodiinnumerabilium figurarum, quae alius alio scientius variegant sed peritorum con-cinnatio miraculum est, imperitorum iunctura ridiculum quo praedicto scies quodego posteriorem imitatus sum hoc ergo centonis opusculum ut ille ludus tractatur,pari modo sensus diversi ut congruant, adoptiva quae sunt ut cognata videantur,aliena ne interluceant, arcessita ne vim redarguant, densa ne supra modum pro-tuberent, hiulca ne pateant quae si omnia ita tibi videbuntur ut praeceptum est,dices me composuisse centonem et, quia sub imperatore tum merui, procedere mihiinter frequentes stipendium iubebis; sin aliter, aere dirutum facies, ut cumulocarminis in fiscum suum redacto redeant versus unde venerunt vale

Read through this also, if it is worthwhile—a trifling and worthless little book,which no pains have shaped nor care polished, without a spark of wit and thatripeness which deliberation gives

They who first trifled with this form of compilation call it a ‘‘cento.’’ ’Tis atask for the memory only, which has to gather up scattered tags and fit thesemangled scraps together into a whole, and so is more likely to provoke yourlaughter than your praise If it were put up for auction at a fair, Afranius wouldnot give his straw, nor Plautus bid his husk For it is vexing to have Virgil’smajestic verse degraded with such a comic theme But what was I to do? It waswritten by command, and at the request (which is the most pressing kind oforder!) of one who was able to command—the Emperor Valentinian, a man, in

my opinion, of deep learning He had once described a wedding in a jeu d’esprit

of this kind, wherein the verses were to the point and their connections amusing.Then, wishing to show by means of a competition with me the great superiority

of his production, he bade me compile a similar poem on the same subject Justpicture how delicate a task this was for me! I did not wish to leave him nowhere,nor yet to be left behind myself; since my foolish flattery was bound to be patent

to the eyes of other critics as well, if I gave way, or my presumption, if I rivaledand surpassed him I undertook the task, therefore, with an air of reluctance andwith happy results, and, as obedient, kept in favor and, as successful, gave nooffense

This book, then hurriedly composed in a single day with some lamp-lit hoursthrown in, I lately found among my rough drafts; and so great is my confidence inyour sincerity and affection, that for all your gravity I could not withhold even aludicrous production So take a little work, continuous, though made of disjointedtags; one, though of various scraps; absurd, though of grave materials; mine,though the elements are another’s; lest you should wonder at the accounts given

by priests or poets of the Son of Thyone or of Virbius—the first reshaped out ofDionysus, the second out of Hippolytus

And if you will suffer me, who need instruction myself, to instruct you, I willexpound what a cento is It is a poem compactly built out of a variety of passagesand different meanings, in such a way that either two half-lines are joined together

to form one, or one line and one accompanied by the following half-line For toplace two (whole) lines side by side is weak, and three in succession is meretrifling But the lines are divided at any of the caesurae which heroic verse admits,

so that either a penthemimeris can be linked with an anapaestic continuation, or athird-foot trochaic break with a complementary section, or at the seventh half-foot

needed to complete the hexameter: so that you may say it is like the puzzle whichthe Greeks have called stomachion There you have little pieces of bone, fourteen

in number and representing geometrical figures For they are quadrilateral ortriangular, some with sides of various lengths, some symmetrical, either of equallegs or equilateral, with either right or oblique angles: the same people call themisosceles or equal-sided triangles, and also right-angled and scalene By fittingthese pieces together in various ways, pictures of countless objects are produced: amonstrous elephant, a brutal boar, a goose in flight, and a gladiator in armor, ahuntsman crouching down, and a dog barking—even a tower and a tankard andnumberless other things of this sort, whose variety depends on the skill of theplayer But while the harmonious arrangement of the skillful player is marvelous,the jumble made by the unskilled is grotesque This prefaced, you will know that

I am like the second kind of player

And so this little work, the Cento, is handled in the same way as the gamedescribed, so as to harmonize different meanings, to make pieces arbitrarilyconnected seem naturally related, to let foreign elements show no chink of lightbetween, to prevent the far-fetched from proclaiming the force which unitedthem, the closely packed from bulging unduly, the loosely knit from gaping If youfind all these conditions duly fulfilled according to rule, you will say that I havecompiled a cento And because I served at the time under my commanding officer,you will direct ‘‘that pay be issued to me as for regular service’’; but if otherwise,you will sentence me ‘‘to forfeit pay,’’ so that this ‘‘lump sum’’ of verse may be

‘‘returned to its proper pay-chest,’’ and the verses go back to the source from which

Through this explanatory epistle, Ausonius not only communicates with Paulusbut also with a wider audience; for Ausonius wrote the letter with an eye tobroader dissemination alongside the cento Ausonius includes the letter in order

to ensure that Paulus—the explicit addressee of the epistle—and readers in thefourth century and perhaps even posterity, Ausonius’s implicit addressees, would

be equipped to read the cento knowledgeably In concerning himself with histext’s reception, Ausonius’s interests no doubt were more than pedagogical Bydescribing the origin and character of his own patchwork text and relating what acento is, Ausonius would have enabled his different readers to appreciate betterhis accomplishment in composing his work.5At the end of the epistle, the poetdemonstrates his desire to have Paulus, and by extension the tacit larger audi-ence, judge the Cento Nuptialis (40–43) Despite some patently insincere self-deprecation in the letter (more on this hereafter), Ausonius certainly would havewanted his work received favorably.

Ausonius’s efforts to secure the sort of readership he wants for the CentoNuptialis lead him to articulate a cento poetics.6 The account that Ausoniusoffers will serve as a point of departure for much of the discussion in thischapter This will involve not only analyzing Ausonius’s poetics closely but alsorelating it to the twelve extant mythological and secular centos, especially byscrutinizing how all the centonists put into practice the ideas and methods thatAusonius discusses In addition, I will investigate aspects of cento compositionthat Ausonius overlooks.7Such a survey will explain thoroughly what a cento is,and along the way will connect the Virgilian centos to topics of wider concern,including the nature of literary ludism, ancient mnemotechnics, and the roles ofthe author and the reader in allusion

Readers of Ausonius’s prefatory epistle to the Cento Nuptialis would have toagree with Erasmus that its author was one who, regarding the cento, legemetiam eius carminis tradit (‘‘Adagia,’’ Opera Omnia 2.542D) This assessmentstems from lines 21–28 of the letter to Paulus,8in which Ausonius presentsthe technical rules of cento composition The poet says that citations were toconsist either of two half-lines or of one line and the following half attached toanother half (praef 22) Although Ausonius himself sometimes connects longerunits,9he remarks that generally the gesture is cheap child’s play (praef 22–23).Ausonius proceeds to delineate the different metrical sections that the cen-tonist could conjoin The segments result from cuts at the strong caesurae

in the second, third, or fourth feet, or after a weak caesura in the third foot(praef 24–28).10(In practice, centonists sometimes make other cuts, includingoccasionally at diereses.)11 For many, this information constitutes what is ofvalue and interest in the prefatory epistle Ausonius’s codifications are oftenseen as his primary, if not his sole contribution to our understanding of thecento form

Overlooked in such a reading of the epistle to Paulus is the information aboutthe cento that Ausonius imparts in less explicit ways In addition to his statementsabout the proper length of citations and the varied metrical divisions of Virgil thatcan occur in the cento, Ausonius presents a discursive poetics, or an explanation ofthe cento form in more expressive and often figurative language This occurs both

in the sections of the epistle where he discusses the Cento Nuptialis alone (praef.6–20, 40–43) and where he examines the patchwork form generally (1–6, 21–40)

Ausonius’s narrative sketch provides deeper insights into the cento than histechnical (and somewhat corrupt) summary in lines 21–28.12

The initial Ausonian terms that I will discuss are ludus and ludere Ausoniususes the word ‘‘play’’ as a verb in line 3 (centonem vocant qui primi hac con-cinnatione luserunt) and as a noun in lines 9 ([Valentinianus nuptias quon-dam eiusdem] ludo descripserat) and 37 (hoc ergo centonis opusculum ut illeludus tractatur) These comments provide early examples of an important term

in cento criticism, where the patchwork poems have been described as a sort ofgame, sometimes pejoratively, sometimes incompletely.13 My aim is to useAusonius’s epistle as a starting point for defining more precisely what makescento composition a form of literary play

Ludere and ludus/lusus are regular, and almost technical, terms for differentkinds of verse in the Latin tradition.14They can denote poems of different kindsproduced in leisure hours;15 youthful works;16 light poetry as distinguishedfrom serious;17texts belonging to minor genres;18and poems in which authorstreat the verbal surface as game pieces that they fit into patterns—that is,carmina figurata, reciprocal verses, and the like.19The words can also be an in-sult or a means of self-deprecation.20To complicate matters, these differentcategories can sometimes overlap

In his prefatory epistle to Paulus, Ausonius defines the Cento Nuptialis andthe cento form as literary play in accordance with several of these measuringsticks One is the classification of poetic ludism as a product of otium, or leisure.21While there was some persistent suspicion attached to it in Roman culture,22leisure for the majority of Latin authors and the general public was usually aninterval in, and a preparation for, work (labor), business dealings (negotium), theperformance of duties (officia), or political, administrative, or military service.23Among the economic, political, and social elite and those of lower status whopossessed some cultural capital—for example, grammarians and rhetors—one ofthe ways to pass one’s relaxation was to write poetry (whether alone or withothers) Though some resisted and criticized this activity, the sources who discuss

it tend to represent it as a productive use of otium, or a means of refreshing one’sintellectual and creative faculties, or as a benignly frivolous passing of time.24Often writers mention the kind of work being produced in leisure hours, with epicand tragedy considered worthwhile and edifying Epigram, satire, and other lightgenres, meanwhile, were deemed inconsequential and flighty,25though still ac-ceptable as cultured play An example of someone who wrote poems belonging tothis second class of works, Pliny, describes his penchant for writing light verses asharmless fun (Ep 5.3) This attitude was no doubt common from the late republicthrough late antiquity, though it must be added that Pliny had to defend hiswriting and reciting versiculos severos parum (Ep 5.3.2) In cultural centersthroughout antiquity, recitation halls and dinner parties would have been im-portant loci for sharing ludic pieces.26

In late antiquity, evidence for such cultivated play during otium comesfrom the fifth-century Sidonius Apollinaris (e.g., Ep 9.13.2–5 and 9.15.1)

Especially notable is Ep 9.13, where Sidonius says that he is sending Tonantius

a lyric poem to be recited inter bibendum (9.13.2), and where he describesbygone convivial parties in which he and others chose meters by lots and wrotepoems on the same subject matter in playful competition (9.13.4–5) Criticshave argued that there was in Sidonius’s Gaul a decline in the number of lite-rati who partook of such ludic activities.27Even so, a vivid picture emerges inSidonius’s letters of how the learned could approach poetry as entertainment

in fifth-century Gaul, and so of how guilt-free leisured literary ludism was aliveand well in circles that valued and wanted to preserve their ties to Romanclassical culture In the codex Salmasianus, moreover, which (as noted in theintroduction to this book) contains a sixth-century collection of poems put to-gether in Africa and probably represents to a large degree the poetry of Africanwriters,28many works have the appearance of dilettantish products of leisure

In one, which modern editors have entitled the Epistula Didonis ad Aeneam (AL

71 SB), an anonymous author explicitly links his poem to otium (quid carminisotia ludant, / cerne bonus mentisque fidem probus indue iudex [2–3]).29Ausonius offers further evidence for such leisurely composition in the preface

to the Griphus (14–27), where he relates that he began composing the riddlingpoem while drinking with others during the Alamannic campaign of 368–369.30Itmay be, however, that Ausonius only got the idea for his poem at that point Amore secure connection to leisure, and more significant for my purposes, marksthe Cento Nuptialis, which arose in the otium at Valentinian’s court In his letter toPaulus, Ausonius reports that his patchwork poem began as a potentially incen-diary diversion, after Valentinian, having himself written such a ludus, challengedhim to a literary contest (praef 8–11).31This placed the centonist on the razor’sedge While he did not want to appear to have thrown the contest, he also couldnot beat the emperor too handily, lest he be charged with insolence (11–14).Fortunately, Ausonius devised a happy solution Taking up the task with seemingreluctance, he both stayed in favor by being obedient and, as an unwilling winner,avoided offending Valentinian (14–15) Despite the rhetorical nature of Auso-nius’s description of Valentinian’s order and the author’s dilemma, there is noreason to doubt that this competition occurred Notable in Ausonius’s account of

it is the language of conflict (contentione praecelleret [11], anteferri/posthaberi [12],aemulus eminerem [14], and victor [15]) These terms point to the place of thecompetitive impulse that is a main spur to play in the exchange of dueling cen-tos.32References to victory also indicate that playing at the cento could conferstatus on the successful competitor This ‘‘battle,’’ however, occurred in the de-marcated zone of otium, a parareality in which events took place that, while theycould have consequences in the larger world, were set off from that world.Whether the other mythological and secular Virgilian centos arose in asimilar background of cultivated leisure is difficult to determine The anony-mous centos in the codex Salmasianus as well as Mavortius’s Iudicium Paridis,which appears in the same manuscript, are candidates for such otium TheVirgil-saturated authors may have composed the patchwork poems either

for refined entertainment, and perhaps even for ludic literary competitions at

a banquet, or as diversions in their own spare time Whatever their initial formance contexts, the centos, like other Virgilian poems in the Salmasianus,Coronatus’s Locus Vergilianus (AL 214 SB) and the two anonymous ThemataVergiliana (AL 237 SB, 249 SB), have more than a whiff of cultured play, whichwould have occurred during leisure hours.33

per-Of the other mythological and secular centos, there is some evidence linkingHosidius Geta’s Medea to otium This comes from Tertullian, who mentions acento Medea in the De Praescriptione Haereticorum 39.3–4 Based on parallels

in name, form, and subject matter, it is extremely probable that this text is thevery Medea that survives in the codex Salmasianus After alluding to the Medea(denique Hosidius Geta Medeam tragoediam ex Virgilio plenissime exsuxit),Tertullian proceeds to refer immediately to a neighbor or relative who usedVirgilian verses to offer a new version of Cebes’s Pinax This text emergedamong other compositions written during the author’s leisure hours: meusquidam propinquus ex eodem poeta inter cetera stili sui otia Pinacem Cebetisexplicuit (39.4) While Tertullian fails to link the Medea explicitly to leisure, itmay be that the tragic cento arose in a setting similar to the centonized Pinax ofCebes—that is, in the time that cultured adults devoted to otium

The final cento to consider is Luxurius’s Epithalamium Fridi The centonistprobably did not perform this work at the wedding that occasioned it Instead,Fridus and his bride were in all likelihood meant to enjoy the poem during theirrelaxation Luxurius, moreover, may have composed the Epithalamium Frididuring his otium as a gift for the bride and groom At the same time, the couplemay have solicited the work, which means that it would have been a patronizedcommission rather than a pastime Thus there is some question as to whetherthe Epithalamium Fridi was the product of Luxurius’s otium; or something hewrote more by necessity (I will return to these matters in chapter 5.)

A second common way to define poetic ludism that I noted earlier is to tinguish literary play from serious poetry, broadly defined.34 Throughout hisepistle to Paulus, Ausonius does precisely this in describing the Cento Nuptialis.Admittedly, Ausonius pursues such an approach largely as part of a captatiobenevolentiae, which was recommended for the exordia of speeches and was found

dis-in the prefaces of literary works Designed to secure the sympathy of an audience,

a captatio usually contained self-effacing assertions of the inadequacies of anauthor and his text.35Ausonius includes such affected modesty in programmaticpassages preceding several pieces,36although not always to the desired effect; forthe poet’s protestations of humility have elicited critical wrath.37

Many statements in the epistle to Paulus contribute to the captatio WhenAusonius calls his cento a frivolum et nullius pretii opusculum (praef 1), claimsthat it is the type of work quod ridere magis quam laudare possis (4), and admitspiget equidem Vergiliani carminis dignitatem tam ioculari dehonestasse materia(6–7), he does not offer literary criticism of the Cento Nuptialis, let alone of thecento as such Instead, Ausonius seeks to win the goodwill of his audience.38

There is a moment, however, when Ausonius distinguishes his cento fromserious literary composition in a way that moves beyond the narrow purposes of acaptatio: accipe igitur opusculum de inconexis continuum, de diversis unum, deseriis ludicrum, de alieno nostrum (praef 17–19) While the word opusculumowes much to the modesty topos, as it does in line 1 of the epistle,39the passage

is not just a self-deprecating moment as the foregoing examples are, but contains

an accurate appraisal of the Cento Nuptialis and of the patchwork form as awhole The important clause here is de seriis ludicrum Ausonius says that centocomposition is something an author does to serious poetry in order to produce aplayful piece—ludicrum, a word obviously connected to ludus/ludere, which heredoes more than convey modesty and in fact describes the cento accurately.40Because the cento exists as a product of Virgilian poetry, it necessarily stands at adistance from the canonical, and so seria, Eclogues, Georgics, and Aeneid.41Theact of reconstituting source material, which Ausonius emphasizes when he usesthe verb ‘‘play’’ earlier in the epistle (centonem vocant, qui primi hac concin-natione luserunt, 2–3), comes between the Virgilian centos and their canonicalsource texts This gives the patchwork poems a different status from those grandworks, as they become examples of light, ludic poetry derived de seriis

That a cento is fundamentally a playful reworking of Virgilian poetry alsomeans that a patchwork text cannot belong in any simple way to a high genre.42Here the relevant work is Hosidius Geta’s Medea This cento is presented as atragedy, a genre that stands alongside epic atop the generic hierarchy In a cento,however, the intercession of Virgil causes the patchwork text to be somethingother than merely a representative of that particular genre While an individualpatchwork poem may take the form of a tragedy, it is first a cento, a text derivedfrom the manipulation of another author’s poetry Hence Geta’s Medea is whatAusonius describes the Cento Nuptialis to be: an opusculum ludicrum de seriis

No matter how lofty the genre it replicates, a cento is always at bottom a cento, or

a text adapting Virgil to a particular generic setting, and so standing at a removefrom that genre

Another category of ludic poetry that I noted earlier consists of texts in whichauthors isolate and reify the verbal surface, treating words as game pieces whosephysical existence they can manipulate.43Ausonius aligns the cento form withsuch ludism when he compares patchwork composition to the stom
aawion(praef 28–37) Within that passage, Ausonius uses the terms ludicrum (simile

ut dicas ludicro, quod Graeci stom
aawion vocavere [28]) and, for the final time inthe epistle, ludus (hoc ergo centonis opusculum ut ille ludus [the stom
aawion]tractatur [37])

The trifle or game to which Ausonius likens the cento is first attested in alargely lost work of Archimedes and in antiquity acquired the name loculusArchimedius Some have explained the word stom
aawion as meaning Neckspiel,

or brain teaser,44while others read ostom
aawion, a ‘‘battle of bones’’ puzzle that;takes its name presumably from the material first used to create the pieces of thatpuzzle.45In any case, the game consists of fourteen triangular or quadrilateral

figures,46which Ausonius labels ossicula (praef 28) The object of the game asAusonius presents it is to fit those geometric shapes together in different ways inorder to make countless objects: harum verticularum variis coagmentiis simulanturspecies mille formarum (32–33).47

The basic purpose of Ausonius’s comparing the cento to the stom
aawion is toshow how cento composition admits of various configurations of Virgil’s verseunits, which serve as verbal and metrical ossiculae, and how centonists use thoseunits to create new literary objects.48Extrapolating from this, it can be said thateach patchwork text exists because an author has imposed ‘‘play conditions’’ onVirgil’s verbal surface,49abstracting its constituent verse units and treating them

as though they were manipulable game pieces

Handling Virgil in this way compels writers to accept and abide by a set ofstringent ad hoc laws In similar ludic literature such as palindromic poetry,acrostics, or leipograms, these establish the particular boundaries within which

an author can pursue his game In the case of the cento, the rule that Virgil’spoetry must constitute a patchwork text, as well as the strictures governing themetrical incisions that the poet can make, set the limits for how a centonist canmanipulate or play with Virgil’s reified language Of course, poetry generallyimposes rules on a writer, from metrical and other formal constraints to genericcustoms and expectations.50 What distinguishes the cento from conventionalverse composition—and this point holds for other works whose authors treatverbal surfaces similarly—is how extremely circumscribed the space of material

is within which the centonist works, and how extremely tight the laws aregoverning his methods of composition The centonist severely delimits the lin-guistic possibilities available to him—that is, he confines the verbal area of hispoetry and the ways that he can handle that area much more than nonludicauthors do The centonist thus creates a ‘‘closed field’’51out of Virgil’s verbalsurface, which he rearranges according to the specific and conventional rulescontrolling his play

That centonists give ludic materiality to Virgil’s verse units and handle themaccording to a set of rules affects the reception of their texts As with all games,including literary examples, there is a strong aspect of spectatorship to readingcentos If they are to appreciate the works fully, audiences need to be awarethat the centos are Virgilian texts, created through a peculiar technique anddisplaying how writers handle that technique Indeed, centos are fundamentallyauthorial demonstrations of skill in creating a new composition out of Virgil’sverse units.52Though there will conceivably cases be where a reader does notknow that he is reading a patchwork poem,53a cento should not be met by suchnaı¨ve reading and instead calls upon its audience to exercise its critical faculty

in scrutinizing how the author negotiates the rules of his game and produces atext.54Because the cento is the kind of ludus it is, the processes that lie behindits linguistic surface intrude more forcefully on the reading act than do theprocesses underlying the production of conventional poetry, even as these are ofcourse also a central part of the interpretive experience

Ausonius’s use of ludere and ludus, then, aligns the Cento Nuptialis andcento composition generally with a wide range of criteria defining literary ludism

in antiquity, and so shows that the cento is a form of play The next step towardascertaining cento quid sit is to examine the specific ways that centonists pursuetheir Virgilian games To uncover how patchwork poets ‘‘investigate works fromthe past in order to find possibilities that often exceed those their authors hadanticipated,’’55I return to Ausonius’s prefatory epistle and other aspects of thefigurative poetics he offers

Early in the letter to Paulus, Ausonius uses a metaphor that vividly portrayswhat centonists do to Virgil’s poetry: solae memoriae negotium sparsa colligere etintegrare lacerata (praef 3–4) Ausonius here draws on a ‘‘conventional literaryvocabulary that figures texts and parts of texts as their authors’ bodies andlimbs’’56to describe the task of collecting and fitting together Virgil’s ‘‘mangledand strewn’’ verse units or membra, a word whose multivalence the centonistexploits.57For Ausonius, cento composition is a violent enterprise; yet instead

of only rending Virgil’s ‘‘limbs,’’ centonists put them back together While thecentonist tears Virgil’s original verbal surface apart, Ausonius relates, his pur-poses are ultimately creative, not destructive

Whether or not it was his intention, Ausonius’s figurative language in lines3–4 of his epistle to Paulus also adumbrates a connection between his and allcentonists’ negotium memoriae and the techniques prescribed in handbooks onmemory Specifically, the reference to Virgil’s sparsa et lacerata [membra] sug-gests that a centonist applies to his literary performance a version of divisio, orthe act of memorizing a long text in parts According to writers on mnemo-technics, this piecemeal approach was a necessary first step to committing alengthy work to memory.58Such division would have helped readers throughoutantiquity to remember Virgil, the poet who most occupied the memories of theeducated from early childhood onward, as I noted in the introduction.59Through their constant exposure to Virgil, educated Romans came to have Virgilhard-wired within themselves.60This would lead to the memorization of largeswaths of Virgil, if not the entire corpus of his canonical works, just as Greekreaders knew large sections of Homer’s Iliad and Odyssey or the entire poems byheart.61

The centonists develop a new version of the practice of divisio Having nodoubt memorized most or all the Eclogues, Georgics, and Aeneid by learningthose texts line-by-line and even dactylic segment-by-dactylic segment, thecentonists divide the units of Virgilian poetry anew Cento composition is amemory act requiring that authors be able to scan Virgil in their minds, isolatinghis verse units in order to find an appropriate membrum.62

Notable for his form of mnemonic composition is the author of the Christiancento De Ecclesia (a figure possibly named Mavortius).63 At a public perfor-mance of his work, this author recites a six-line extemporaneous cento as acoda to the De Ecclesia (AL 16a R).64Lines 4–6 of this passage are remarkably

similar to a passage in Hosidius Geta’s cento Medea, so much so that theysuggest a deliberate act of imitation of Geta.65If he was indeed responding tothe Medea, the De Ecclesia poet would demonstrate that he has committed tomemory Virgil and a section of the reconstructed Virgil of an earlier cento, andthat he can reproduce both of them on the spot.66

The only other centonist who provides any information about his method ofcentonizing is Ausonius Once again in the strains of the captatio benevolentiae,Ausonius claims that he dashed off his work in a day and revised it slightly (hoc,tum die uno et addita lucubratione properatum [praef 15–16).67However longAusonius in fact took to compose and polish his Cento Nuptialis, it is easy toimagine the poet consulting a written text of Virgil when he made some re-visions to his cento Yet there remains no reason to doubt Ausonius’s commentthat cento composition is fundamentally a negotium memoriae, with him andevery centonist having the fragmented membra of Virgil’s poetry in their heads

A written text of Virgil could be a reference tool for a centonist; but relyingentirely on a roll or a codex would not allow centonists to operate with theefficiency that the cento form demands.68

Along with divisio, a second important aspect of ancient memory instructionwas compositio, or the recomposition of the elements of a long text in theircorrect order Quintilian confirms this point, insisting that limits be placed onthe subdivision of a text so that one can connect words in their proper relation—

a task of great difficulty—and unite the various sections.69Late antique writers,Fortunatianus and Martianus Capella, restate Quintilian’s injunction.70 Theprinciple of segmented recollection would be central to memorizing either prose

or poetry, on the latter of which Quintilian (Inst Orat 1.1.36 and 11.2.41) andFortunatianus (Ars Rhet 3.13) claim students ought first to train their memo-ries Poetry was also the material on which the author of the Rhetorica adHerennium suggests children and adults should practice memoria verborum, orword-for-word memorization (ad Her 3.19.34) Hence memorizing Virgil wouldhave naturally involved dividing his verse units into workable segments and thenreconnecting them in their proper order

Such recollection of texts does not obtain in the Virgilian centos Rather thantrying to reproduce the Eclogues, Georgics, and Aeneid as they were written, thecentonists disarticulate Virgil’s poetry Yet in doing so, they of course do notmisremember their source material or avoid the very difficult gesture of re-connecting the memorized parts into a unity Having already memorized anoriginal order of verses, the centonists instead rearrange that material to create anew textual coherence Dismembering and reconstructing Virgil throughmemory becomes a means of remaking rather than restoring that model Whatremains of Virgil in a cento are his individual verse units, the membra that come

to constitute a new poetic harmony

Certain tangible things help the centonists perform their idiosyncraticmemory acts of isolating Virgilian lines anew and reconnecting them in originalways.71The first aid is situational agreement.72A cento will sometimes contain