Horace's Ars Poetica: Family, Friendship, and the Art of Living 9780691197432

Citation preview

hor ace’s a r s poet ic a

Horace’s Ars Poetica ­fa m i ly, f r i e n dsh i p, a n d t h e a rt of l i v i ng

j e n n i f e r f e r r is s-­h i l l

pr i nce­t on u n i v e r sit y pr e ss pr i nce­t on & ox for d

Copyright © 2019 by Prince­ton University Press Published by Prince­ton University Press 41 William Street, Prince­ton, New Jersey 08540 6 Oxford Street, Woodstock, Oxfordshire OX20 1TR press​.­princeton​.­edu All Rights Reserved

Prince­ton University Press gratefully acknowledges the assistance of the National Endowment for the Humanities in support of publication of this book. Any views, findings, conclusions, or recommendations expressed in this book do not necessarily represent ­those of the National Endowment for the Humanities. ISBN 9780691195025 LCCN 2019944690 Editorial: Rob Tempio and Matt Rohal Production Editorial: Debbie Tegarden Jacket/Cover Design: Layla Mac Rory Production: Merli Guerra Publicity: Alyssa Sanford Copyeditor: Aimee Anderson This book has been composed in Classic Arno Printed on acid-­free paper. ∞ Printed in the United States of Amer­i­ca 10 ​9 ​8 ​7 ​6 ​ 5 ​4 ​3 ​2 ​

For Robert, Beatrice, and Julian, again.

c on t e n t s

Acknowl­edgments  ix Preface: Horace, Ars Poetica: Text and Translation  xi Introduction: Becoming the Ars Poetica 1 The Name of the Poem 5 The Genre of the Poem 13 The Date of the Poem 17 The Standing of the Poem—­A Story 22 The Unity of the Poem 30

R eading the Ars Poetica

37

1 Humano 39 ­Human Nature 46 Living Language 62 Appropriateness and the Satires 69 Horatius senex 80 Callida iunctura 90 2 Pisones 100 The Identity of the Pisones 100 The Pisones in the Ars Poetica 106 Ages and Aging, Young and Old 129 Horatius pater 138 Conclusions 149

vii

viii  con t e n ts

3 Amici, risum 154 Risum 154 Amici 157 Laughter and Friendship 159 Roman Friendship and Horace as Amicus Pisonum 173 Horatius iudex 183 From Laughter to Madness 189 ­Toward the End of the Poem 197 4 The End of the Poem The Art of ­Doing Every­thing Ars Poetica, Ars Poiētikē (ποιητική) A Conclusion: Ars Poetica, Ars Vivendi

200 211 221 237

Epilogue: Receiving the Ars Poetica 244 Bibliography  273 Index  295

ac k now l ­e d g m e n t s

Nonumque prematur in annum / membranis intus positis (a p 388 –8 9)

while ­t hese lines from the Ars Poetica might well be used to preface any work, in par­tic­u­lar one devoted wholly to the poem, they are particularly apt in the pre­sent case since I almost wrote my PhD dissertation on the Ars Poetica. The prob­lem, however, was that I had not quite figured out what to say about the work, though I found it intriguing. It took writing my dissertation and first book on Old Comedy and Roman Satire (Roman Satire and the Old Comic Tradition, Cambridge University Press, 2015) for the arguments and readings presented h­ ere to begin to coalesce. The pre­sent proj­ect has accordingly under­gone a long gestation—­one that exceeds even Horace’s recommended nine years. Throughout t­ hese years I have been fortunate to enjoy the support and wise counsel of many friends and colleagues who have variously attended talks, corresponded by email, offered their expertise (especially on the reception of the Ars Poetica in ­later periods), recommended further readings and ave­nues of exploration, and graciously read portions of the manuscript: Jafari Allen, David Armstrong, Jaswinder Bolina, Ava Brillat, Anne Cruz, Caleb Dance, Mike Fontaine, Ernesto Fundora, Aaron Kachuck, Rebecca Katz, John Kirby, Jeanne Neumann, Sarah Nooter, Bijan Omrani, Frank Palmeri, John Paul Russo, Wilson Shearin, Mihoko Suzuki, Hugh Thomas, Richard Thomas, Han Tran, and Robyn Walsh. To all t­ hese and more I owe an inexpressibly huge debt of gratitude. It goes without saying that any remaining errors are entirely my own responsibility. I am also grateful to Joshua Katz for encouraging me to pursue publication with Prince­ton University Press, where I was fortunate to work with Jessica Yao and Rob Tempio. Jessica was a model editor, encouraging and involved in e­ very detail from the start while also seeing the book through two rigorous rounds ix

x 

ack now l­e d gm e nt s

of review; Rob Tempio expertly shepherded the book through to completion, with Matt Rohal and Debbie Tegarden. The anonymous readers selected by the press ­were exceptionally helpful, and this book would be far weaker w ­ ere it not for their suggestions. I have conveyed my thanks to them several times already and am pleased to be able to do so h­ ere again. I could not have hoped for a better editorial team and pro­cess. This proj­ect has been generously funded by the National Endowment for the Humanities through a summer stipend, and by the University of Miami in the form of two Provost Research Awards, a sabbatical leave, and a semester of associate professor post-­tenure teaching release. I am also grateful to all in attendance on ­those occasions at which I presented portions of the proj­ect: several meetings of both the Classical Association of the Middle-­West and South and the Antiquities Interdisciplinary Research Group convened at the Center for the Humanities at the University of Miami. Fi­nally, to my f­ amily: thank you for indulging me once again as I wrote another book; I hope that I have gotten at least a ­little better at ­doing it without causing undue disruption to every­one around me.

p r e fac e

Horace, Ars Poetica: Text and Translation

the full Latin text of Horace’s Ars Poetica (from Shackleton Bailey’s 2001 edition, except where indicated)1 appears on pages xii–xli. The original text is on the left-hand pages and my own translation appears on the right. Readers less familiar with the poem ­w ill want to begin by reading it in its entirety; ­others may find it useful to refer back to this translation as needed. In addition to offering readability of a fairly literal sort, the translation aims to reflect the reading of the poem offered in the pre­sent study.

1. I have removed Shackleton Bailey’s paragraph markings, for reasons detailed in my introduction. xi

xii  P r e face

Text Humano capiti cervicem pictor equinam iungere si velit et varias inducere plumas undique collatis membris, ut turpiter atrum desinat in piscem mulier formosa superne, spectatum admissi risum teneatis, amici? credite, Pisones, isti tabulae fore librum persimilem cuius, velut aegri somnia, vanae fingentur species, ut nec pes nec caput uni reddatur formae. ‘pictoribus atque poetis quidlibet audendi semper fuit aequa potestas.’ scimus, et hanc veniam petimusque damusque vicissim; sed non ut placidis coeant immitia, non ut serpentes avibus geminentur, tigribus agni. inceptis gravibus plerumque et magna professis purpureus, late qui splendeat, unus et alter assuitur pannus, cum lucus et ara Dianae et properantis aquae per amoenos ambitus agros, aut flumen Rhenum aut pluvius describitur arcus. sed nunc non erat his locus. et fortasse cupressum scis simulare: quid hoc, si fractis enatat exspes navibus aere dato qui pingitur? amphora coepit institui: currente rota cur urceus exit? denique sit quidvis, simplex dumtaxat et unum. maxima pars vatum, pater et iuvenes patre digni, decipimur specie recti. brevis esse laboro, obscurus fio; sectantem levia nervi deficiunt animique; professus grandia turget; serpit humi tutus nimium timidusque procellae. qui variare cupit rem prodigialiter unam, delphinum silvis appingit, fluctibus aprum: in vitium ducit culpae fuga, si caret arte. Aemilium circa Ludum faber unus et unguis exprimet et mollis imitabitur aere capillos, infelix operis summa, quia ponere totum

5

10

15

20

25

30

Hor ace , A r s p oet ic a , te x t a n d t r a nsl at ion  xiii

Author’s Translation ­ uman—to a ­human head if a painter should wish H to join a h­ orse’s neck, and make multi-­colored feathers grow on limbs gathered from all over, so that horribly in a black fish-­tail she ends, a beautiful ­woman up above, would you, friends let in for a viewing, hold back your laughter? Believe, Pisos, that just like that painting would be a book, whose images ­will be fashioned to be empty, like a sick man’s dreams, such that neither head nor foot can be reconciled with a single form. “Paint­ers and poets have always had the same opportunity to dare what­ever they wish.” We know, and we ask for and we grant this indulgence by turns; but not with the result that harsh ­things combine with gentle ones, not with the result that serpents be paired with birds, lambs with tigers. Onto weighty beginnings and ones that announce ­great t­ hings often one purple patch and another is sewn, which may gleam widely, when the grove and altar of Diana is described and the course of ­water rushing through pleasant fields, or the river Rhine or a rainbow. But now it was not the place for ­these. And perhaps you know how to reproduce a cypress: what good is this, if he who is painted swims hopeless away from broken ships, with money given? An amphora begins to be set up: why does a jug come off the ­running wheel? In short, let it be what you wish, provided that it is ­simple and unified. The greatest portion of poet-­priests, o ­father and young men worthy of the f­ ather, [we] are deceived by the appearance of right. I work to be concise, I become obscure; sinews and mind fail the one who chases smooth ­things; the one who has professed ­great ­things becomes inflated; the one who plays too safe and is wary of the storm creeps on the ground. He who desires to vary a single theme unnaturally paints a dolphin in the woods, a boar on the waves: the avoidance of [one] fault leads to [another] vice, if done artlessly. One workman around the Aemilian School ­will both press out fingernails and imitate soft hair with bronze, unhappy with the totality of the work, ­because he does not know how to set forth

5

10

15

20

25

30

xiv  P r e face

nesciet. hunc ego me, si quid componere curem, non magis esse velim quam naso vivere pravo, spectandum nigris oculis nigroque capillo. sumite materiam vestris, qui scribitis, aequam viribus et versate diu quid ferre recusent, quid valeant umeri; cui lecta pudenter erit res, nec facundia deseret hunc nec lucidus ordo. ordinis haec virtus erit et Venus, aut ego fallor, ut iam nunc dicat iam nunc debentia dici, pleraque differat et praesens in tempus omittat. hoc amet, hoc spernat promissi carminis auctor. in verbis etiam tenuis cautusque serendis dixeris egregie, notum si callida verbum reddiderit iunctura novum. si forte necesse est indiciis monstrare recentibus abdita rerum, fingere cinctutis non exaudita Cethegis continget dabiturque licentia sumpta pudenter; et nova fictaque nuper habebunt verba fidem si Graeco fonte cadent, parce detorta. quid autem Caecilio Plautoque dabit Romanus ademptum Vergilio Varioque? ego cur, acquirere pauca si possum, invideor, cum lingua Catonis et Enni sermonem patrium ditaverit et nova rerum nomina protulerit? licuit semperque licebit signatum praesente nota procudere nummum. ut silvae foliis privos mutantur in annos, prima cadunt * * * * * * * * * * * * * * ita verborum vetus interit aetas et iuvenum ritu florent modo nata vigentque. debemur morti nos nostraque; sive receptus terra Neptunus classis Aquilonibus arcet, regis opus, sterilisve †diu palus† aptaque remis vicinas urbis alit et grave sentit aratrum,

35

40

45

50

55

60 61a 61b

65

Hor ace , A r s p oet ic a , t e x t a n d tr a nsl at ion  xv

the w ­ hole. I myself would not wish for me to be this man, if I should care to put something together, any more than [I would wish] to live with a crooked nose, [though] admired for my black eyes and black hair. Take up material that is well-­matched to your strengths, you who write, and keep turning over in your mind at length what your shoulders would refuse to bear, what they would be able to; the one who has chosen his subject ­matter sensibly, him neither eloquence nor clear arrangement ­will abandon. This ­will be the manly excellence and feminine charm of arrangement, ­unless I am mistaken, that he already now says ­things deserving to be said already now, and puts off many ­things and leaves them out for the pre­sent time. Let the author of a promised song-­poem love this ­thing, reject that one. You, slender and cautious also in stringing together words, ­will have spoken outstandingly if a clever join gives back a familiar word as new. If it is perhaps necessary to show with recent symbols the hidden ones of ­things, it ­will fall to you to craft ones not heard by the girded Cethegi and a license taken up prudently ­will be granted; and words new and recently created ­will have trustworthiness if they fall from the Greek font, sparingly opened. What indeed ­will the Roman grant to Caecilius and Plautus that has been denied to Virgil and Varius? Why am I begrudged if I am able to add a few ­things, when the tongue of Cato and Ennius enriched the paternal conversation and brought forth new names of ­things? It has been permitted and always ­will be permitted to strike a coin marked with the pre­sent’s symbol. Just as the woods are transformed in their leaves each year, the first [sc. leaves] fall * * * * * * * * * * * * * * so the ancient age of words perishes and ones just born flourish in the manner of young men and are vigorous. We are owed to death, we and our ­things; w ­ hether Neptune, received on land, guards the fleets against the North winds, the work of a king, or ­whether a †marsh, for a long time† sterile and suitable for oars, nourishes the neighboring cities and feels the heavy plow,

35

40

45

50

55

60 61a 61b

65

xvi  P r e face

seu cursum mutavit iniquum frugibus amnis doctus iter melius: mortalia facta peribunt, nedum sermonum stet honos et gratia vivax. multa renascentur quae iam cecidere, cadentque quae nunc sunt in honore vocabula, si volet usus, quem penes arbitrium est et ius et norma loquendi. res gestae regumque ducumque et tristia bella quo scribi possent numero, monstravit Homerus. versibus impariter iunctis querimonia primum, post etiam inclusa est voti sententia compos; quis tamen exiguos elegos emiserit auctor, grammatici certant et adhuc sub iudice lis est. Archilochum proprio rabies armavit iambo; hunc socci cepere pedem grandesque cothurni alternis aptum sermonibus et popularis vincentem strepitus et natum rebus agendis. Musa dedit fidibus divos puerosque deorum et pugilem victorem et equum certamine primum et iuvenum curas et libera vina referre. descriptas servare vices operumque colores cur ego si nequeo ignoroque poeta salutor? cur nescire pudens prave quam discere malo? versibus exponi tragicis res comica non vult; indignatur item privatis ac prope socco dignis carminibus narrari cena Thyestae. singula quaeque locum teneant sortita decentem. interdum tamen et vocem comoedia tollit iratusque Chremes tumido delitigat ore et tragicus plerumque dolet sermone pedestri Telephus et Peleus, cum pauper et exsul uterque proicit ampullas et sesquipedalia verba, si curat cor spectantis tetigisse querela. non satis est pulchra esse poemata; dulcia sunto, et quocumque volent animum auditoris agunto. ut ridentibus arrident, ita flentibus afflent humani vultus: si vis me flere, dolendum est primum ipsi tibi. tum tua me infortunia laedent, Telephe vel Peleu; male si mandata loqueris,

70

75

80

85

90

95

100

Hor ace , A r s p oet ic a , te x t a n d t r a nsl at ion  xvii

or ­whether a river has changed its course, unfair to crops, having been taught to go a better way: mortal deeds ­will perish, still less could the honor and lively charm of conversations stand. Many ­things ­will be reborn which have already fallen, and the words that are now [held] in honor ­will fall, if usage ­wills it, in whose hands are judgment and the code and the norms of speech. The deeds of kings and generals and sad wars— in what meter t­ hese could be written, Homer showed. First lamentation was enclosed within verses joined unevenly, [and] afterwards also the expression of [thanks for] a granted prayer; which author, however, [first] sent forth slender elegies, the grammarians dispute and the case is still ­under [the consideration of] a judge. Rage armed Archilochus with their own iamb; the socks [of comedy] and the ­grand tragic boots took this [iamb] up, well-­suited for back-­and-­forth conversation and able to overcome the popu­lar uproar and born for ­doing t­ hings. The Muse granted it to the lyre to tell of gods and the boys of gods and the victorious boxer and the ­horse first in the contest and the concerns of young men and the freedom of wine. Why, if I am incapable of and ignorant at preserving the described turns and colors of works, am I greeted as a poet? Why do I wrongly prefer to be modestly unknowing than to learn? Comic material does not wish to be set forth in tragic verses; likewise, Thyestes’s banquet takes offense at being narrated in song-­poems that are private and almost worthy of the [comic] sock. Let individual ­things keep to the proper place they have been allotted. Sometimes, however, comedy also raises her voice and angry Chremes disputes with a swollen mouth and often tragic Telephus and Peleus express their pain in pedestrian conversation, when each, a pauper and an exile, flings forth bombast and sesquipedalian words, if he is concerned with touching the spectator’s heart with his complaint. It is not enough for poems to be beautiful; they ­shall also be sweet, and they ­shall lead the listener’s spirit wherever they wish. Just as ­human ­faces laugh at laughing ­people, so they weep at weeping ones: if you want me to weep, you yourself must first feel pain. Then your misfortunes ­will hurt me, Telephus or Peleus; if you speak ­things poorly entrusted to you,

70

75

80

85

90

95

100

xviii  P r e face

aut dormitabo aut ridebo. tristia maestum vultum verba decent, iratum plena minarum, ludentem lasciva, severum seria dictu. format enim Natura prius nos intus ad omnem fortunarum habitum; iuvat aut impellit ad iram aut ad humum maerore gravi deducit et angit; post effert animi motus interprete lingua. si dicentis erunt fortunis absona dicta, Romani tollent equites peditesque cachinnum. intererit multum divusne loquatur an heros, maturusne senex an adhuc florente iuventa fervidus, et matrona potens an sedula nutrix, mercatorne vagus cultorne virentis agelli, Colchus an Assyrius, Thebis nutritus an Argis. aut famam sequere aut sibi con­ve­nientia finge. scriptor †honoratum† si forte reponis Achillem, impiger, iracundus, inexorabilis, acer iura neget sibi nata, nihil non arroget armis. sit Medea ferox invictaque, flebilis Ino, perfidus Ixion, Io vaga, tristis Orestes. si quid inexpertum scaenae committis et audes personam formare novam, servetur ad imum qualis ab incepto pro­cesserit et sibi constet †difficile est† proprie communia dicere, tuque rectius Iliacum carmen deducis in actus quam si proferres ignota indictaque primus. publica materies privati iuris erit, si non circa vilem patulumque moraberis orbem, nec verbo verbum curabis reddere fidus interpres, nec desilies imitator in artum, unde pedem proferre pudor vetet aut operis lex. nec sic incipies ut scriptor cyclicus olim: ‘fortunam Priami cantabo et nobile bellum.’

105

110

115

120

125

130

135

Hor ace , A r s p oet ic a , t e x t a n d tr a nsl at ion  xix

I ­will ­either sleep or laugh. Gloomy words are fitting 105 for a sad expression, words full of threats for an angry one, playful ones for a joking one, ­those serious to say for a stern one. For Nature first shapes us on the inside for ­every circumstance of fortunes; she helps or drives us to anger, or leads us down to the ground with heavy grief and chokes us; 110 afterwards, she brings out the motions of our souls with her interpreting tongue. If the words of the one speaking are out of tune with his fortunes, the Roman knights and foot-­soldiers ­will raise a cackle. It ­will ­matter a ­great deal w ­ hether a divinity is speaking or a hero, ­whether a mature old man or one hot with still-­blooming 115 youth, and a power­ful married ­woman or a busy nurse, ­whether an itinerant merchant or the cultivator of a green ­little plot of land, a Colchian or an Assyrian, one reared at Thebes or at Argos. ­Either follow the received tradition or craft ­things that are in keeping with themselves. If as a writer you happen to put back on stage †honored† Achilles, 120 energetic, hot-­tempered, relentless, fierce, let him deny that laws ­were born for him, let ­there be nothing that he does not refer to arms. Let Medea be ferocious and unconquered, Ino tearful, Ixion deceitful, Io a wanderer, Orestes unhappy. If you entrust something untried before to the stage and dare 125 to craft a new character, let it be kept to its bottom such as it started from its top and let it be consistent with itself. †It is difficult† to say common ­things in one’s own way, and you ­will do better to render the song-­poem of Ilium into acts than if you are the first to bring forth unknown and unspoken ­things. 130 Public material ­will be of private owner­ship, if you do not linger around cheap and wide-­open areas, and you ­will take care not to give back one word for one word, a faithful interpreter, nor ­will you, an imitator, jump down into a narrow spot, from where modesty or the law of genre may forbid you from 135 bringing forth your foot. Nor ­will you begin thus as the cyclic writer once did: “The fortune of Priam I ­shall sing and the noble war.”

xx  P r e face

quid dignum tanto feret hic promissor hiatu? parturient montes, nascetur ridiculus mus. quanto rectius hic qui nil molitur inepte! ‘dic mihi, Musa, virum, captae post tempora Troiae qui mores hominum multorum vidit et urbis.’ non fumum ex fulgore, sed ex fumo dare lucem cogitat, ut speciosa dehinc miracula promat, Antiphaten, Scyllamque et cum Cyclope Charybdin. nec reditum Diomedis ab interitu Meleagri nec gemino bellum Troianum orditur ab ovo. semper ad eventum festinat et in medias res non secus ac notas auditorem rapit, et quae desperat tractata nitescere posse relinquit, atque ita mentitur, sic veris falsa remiscet, primo ne medium, medio ne discrepet imum. tu quid ego et populus mecum desideret audi: si plausoris eges aulaea manentis et usque sessuri donec cantor ‘vos plaudite’ dicat, aetatis cuiusque notandi sunt tibi mores mobilibusque decor naturis dandus et annis. reddere qui voces iam scit puer et pede certo signat humum, gestit paribus colludere, et iram concipit ac ponit temere et mutatur in horas. imberbis iuvenis, tandem custode remoto, gaudet equis canibusque et aprici gramine Campi, cereus in vitium flecti, monitoribus asper, utilium tardus provisor, prodigus aeris, sublimis cupidusque et amata relinquere pernix. conversis studiis aetas animusque virilis quaerit opes et amicitias, inservit honori, commisisse cavet quod mox mutare laboret. multa senem circumveniunt incommoda, vel quod

140

145

150

155

160

165

Hor ace , A r s p oet ic a , t e x t a n d tr a nsl at ion  xxi

What worthwhile ­thing could this promiser bring with such a gaping maw? The mountains w ­ ill give birth, a ridicu­lous mouse ­will be born. How much more correctly this one [does], who sets nothing 140 in motion ineptly! “Tell me, Muse, of the man, who ­after the days of captured Troy saw the customs of many men and their cities.” He intends to produce not smoke from a spark, but light from smoke, so that he may bring forth brilliant marvels, Antiphates, and Scylla, and Charybdis together with the Cyclops. 145 Nor is the return of Diomedes arranged from the death of Meleager nor the the Trojan war from the twin egg. It hastens always to the outcome and snatches the listener into the ­middle of ­things not other­wise than if they ­were known, and ­those ­things which he worries are not able to gleam through his ­handling of them, 150 he leaves aside, and he lies in such a way, he mixes false ­things with true thus, that the ­middle is not discordant with the beginning, nor the end with the ­middle. Listen to what I and, together with me, the ­people desire: if you have need of a clapper who waits for the curtains and who ­will sit all the way up ­until the singer says, “clap,” 155 the customs of ­every age must be noted by you, and appropriate qualities must be given to changeable natures and ages. The boy who already knows how to return speech and marks the ground with a sure foot is ­eager to play with his age-­mates, and he becomes angry and puts aside his anger rashly and he changes ­every hour. 160 The beardless youth, his guardian at last removed, rejoices in ­horses and dogs and the grass of the sunny Campus, like wax in being turned ­toward vice, harsh to ­those who watch over him, a slow foreseer of practicalities, extravagant with money, of lofty ideals and ­eager and speedy in leaving ­behind ­things he loves. 165 Once interests have changed, a man’s age and spirit seeks resources and friendships, is devoted to honor, is wary to begin something that he may soon ­labor to change. Many discomforts besiege the old man, ­either ­because

xxii  P r e face

quaerit et inventis miser abstinet ac timet uti vel quod res omnis timide gelideque ministrat, dilator, spe †longus†, iners

avidusque futuri, difficilis, querulus, laudator temporis acti se puero, castigator censorque minorum. multa ferunt anni venientes commoda secum, multa recedentes adimunt. ne forte seniles mandentur iuveni partes pueroque viriles, semper in adiunctis aevoque morabimur aptis. aut agitur res in scaenis aut acta refertur. segnius irritant animos demissa per aurem quam quae sunt oculis subiecta fidelibus et quae ipse sibi tradit spectator. non tamen intus digna geri promes in scaenam, multaque tolles ex oculis quae mox narret facundia praesens. ne pueros coram populo Medea trucidet aut humana palam coquat exta nefarius Atreus aut in avem Procne vertatur, Cadmus in anguem. quodcumque ostendis mihi sic, incredulus odi. neve minor neu sit quinto productior actu fa­bula, quae posci vult et spectanda reposci. nec deus intersit, nisi dignus vindice nodus inciderit; nec quarta loqui persona laboret. actoris partis chorus officiumque virile defendat; neu quid medios intercinat actus quod non proposito conducat et haereat apte. ille bonis faveatque et consilietur amice et regat iratos et amet peccare timentis; ille dapes laudet mensae brevis, ille salubrem iustitiam legesque et apertis otia portis; ille tegat commissa deosque precetur et oret ut redeat miseris, abeat Fortuna superbis. tibia non, ut nunc, orichalco vincta tubaeque aemula, sed tenuis simplexque foramine pauco aspirare et adesse choris erat utilis atque

170

175

180

185

190

195

200

Hor ace , A r s p oet ic a , t e x t a n d tr a nsl at ion  xxiii

he seeks [­things] and [yet], wretched, holds back from what he has found and fears to use it, or ­because he manages all ­things timidly and coldly, a delayer, †long† in hope, inactive and fearful of the ­future, a curmudgeon, given to complaining, a praiser of time past when he was a boy, a chastiser and censor of younger ones. The years as they come bring many comforts with them, [and] take many away as they recede. So that old men’s parts are not by chance entrusted to a young one and men’s parts to a boy, we ­will always take our time over traits associated with and fitting for an age. ­Either the ­matter is acted out on stage, or ­things done are reported. ­Things absorbed through the ear stimulate more sluggishly than ­those which have been placed before the faithful eyes and which the spectator himself hands to himself. Nevertheless, you should not bring onto the stage ­things that ­ought to be done inside, and you should take away many ­things from before the eyes that an in-­person eloquence ­will soon relate. Let Medea not slaughter her boys before the ­people or unspeakably wicked Atreus cook ­human innards openly, or Procne be turned into a bird, Cadmus into a snake. What­ever you show me thus I hate, not believing it. Let the tale be neither smaller nor more drawn out than a fifth act that wishes to be demanded to be watched and demanded again. And let a god not intervene, ­unless a knot worthy of a liberator has befallen; and let a fourth character not strive to speak. Let the chorus provide support for the actor’s parts and manly task; and it should not sing anything in between the ­middles of the acts that does not pertain to the ­matter at hand and relate to it fittingly. Let that one cherish good men and advise them in a friendly manner and rule the angry and love ­those fearing to sin; let that one praise the feasts of a sparing ­table, [let] that one [praise] healthful justice and laws and leisure with open gates; let that one conceal ­things that have been entrusted and pray to the gods and beg that Fortune return to the wretched, be absent from the arrogant. The double-­pipe was not, as now, bound with brass and a rival to the trumpet, but slender and ­simple to play without many an opening and useful at supporting choruses and

170

175

180

185

190

195

200

xxiv  P r e face

nondum spissa nimis complere sedilia flatu; quo sane populus numerabilis, utpote parvus, et frugi castusque verecundusque coibat. postquam coepit agros extendere victor et urbem latior amplecti murus vinoque diurno placari Genius festis impune diebus, accessit numerisque modisque licentia maior. indoctus quid enim saperet liberque laborum rusticus urbano confusus, turpis honesto? sic priscae motumque et luxuriam addidit arti tibicen traxitque vagus per pulpita vestem; sic etiam fidibus voces crevere severis et tulit eloquium insolitum facundia praeceps utiliumque sagax rerum et divina futuri sortilegis non discrepuit sententia Delphis. carmine qui tragico vilem certavit ob hircum, mox etiam agrestis Satyros nudavit et asper incolumi gravitate iocum temptavit, eo quod illecebris erat et grata novitate morandus spectator, functusque sacris et potus et exlex. verum ita risores, ita commendare dicaces conveniet Satyros, ita vertere seria ludo, ne quicumque deus, quicumque adhibebitur heros, regali conspectus in auro nuper et ostro, migret in obscuras humili sermone tabernas, aut, dum vitat humum, nubes et inania captet. effutire levis indigna tragoedia versus ut festis matrona moveri iussa diebus, intererit Satyris paulum pudibunda protervis. non ego inornata et dominantia nomina solum verbaque, Pisones, Satyrorum scriptor amabo; nec sic enitar tragico differre colori ut nihil intersit Davusne loquatur et audax Pythias emuncto lucrata Simone talentum, an custos famulusque dei Silenus alumni.

205

210

215

220

225

230

235

Hor ace , A r s p oet ic a , t e x t a n d tr a nsl at ion  xxv

at filling the seating area, not yet crowded, with its blast; ­there certainly the ­whole populace, countable since small, used to come together, honest[ly] and chaste[ly] and modest[ly]. ­After the victor began to extend his fields and a more extensive city-­wall began to embrace the city and the Genius began to be placated with daily wine on festal days with impunity, a greater license was added to rhythms and meters. Indeed what was the untaught man discerning about, and the one ­free from ­labors, a rustic man mixed with an urban one, a shameful one with an honorable one? Thus the pipe-­player added movement and luxury to the ancient art and dragged his robe wandering about across the platform; thus also voices grew onto stern lyre-­strings and a rushing fluency brought an unfamiliar eloquence and the way of thinking, keen in practical ­matters and foreseeing of the ­future, did not differ from lot-­drawing Delphi. He who competed with a tragic song-­poem for a cheap goat soon even took the clothes off the field-­Satyrs, and roughly attempted a joke with seriousness unharmed, ­because of the fact that the spectator had to be stayed with enticements and pleasing novelty, done with his sacred rites and drunk and exempt from the law. But in this way it is fitting to render attractive the laughing Satyrs, in this way the chatty ones, in this way to turn serious ­things to play, so that what­ever god, what­ever hero is brought on, con­spic­u­ous just now in his royal gold and purple, may not travel to dusky inns with ­humble conversation, or, while he avoids the ground, grasp at clouds and empty ­things. Tragedy, disdaining to babble light verses, just like a married ­woman ordered to dance on festal days, ­will be pre­sent a l­ ittle embarrassed among the impudent Satyrs. Not only unadorned and established names and words, Pisones, ­will I, a writer of Satyrs, love; nor ­will I strug­gle to be dif­fer­ent from a tragic color in such a way that it makes no difference ­whether Davus is speaking and bold Pythias, ­after she has profited a talent from swindled Simon, or Silenus, the guardian and attendant of a divine protégé.

205

210

215

220

225

230

235

xxvi  P r e face

ex noto fictum carmen sequar, ut sibi quivis speret idem, sudet multum frustraque laboret ausus idem: tantum series iuncturaque pollet, tantum de medio sumptis accedit honoris. silvis deducti caveant, me iudice, Fauni ne velut innati triviis ac paene forenses aut nimium teneris iuvenentur versibus umquam aut immunda crepent ignominiosaque dicta. offenduntur enim quibus est equus et pater et res, nec, si quid fricti ciceris probat et nucis emptor, aequis accipiunt animis donantve corona. syllaba longa brevi subiecta vocatur iambus, pes citus; unde etiam trimetris accrescere iussum nomen iambeis, cum senos redderet ictus primus ad extremum similis sibi. †non ita pridem† tardior ut paulo graviorque veniret ad auris, spondeos stabilis in iura paterna recepit commodus et patiens, non ut de sede secunda cederet aut quarta socialiter. hic et in Acci nobilibus trimetris apparet rarus et Enni in scaenam missos cum magno pondere versus aut operae celeris nimium curaque carentis aut ignoratae premit artis crimine turpi. non quivis videt immodulata poemata iudex et data Romanis venia est indigna poetis. idcircone vager scribamque licenter? an omnis visuros peccata putem mea, tutus et intra spem veniae cautus? vitavi denique culpam: non laudem merui. vos exemplaria Graeca nocturna versate manu, versate diurna. at vestri proavi Plautinos et numeros et laudavere sales, nimium patienter utrumque, ne dicam stulte, mirati, si modo ego et vos scimus inurbanum lepido seponere dicto legitimumque sonum digitis callemus et aure. ignotum tragicae genus invenisse Camenae

240

245

250

255

260

265

270

275

Hor ace , A r s p oet ic a , t e x t a n d tr a nsl at ion  xxvii

I ­shall trace my song-­poem crafted from something familiar, so that each man may hope for the same for himself, [and] may sweat a lot and ­labor in vain, having dared the same ­thing: so potent are linkage and combination, so much honor attaches to ­things taken from the ­middle. Let the Fauns led out from the woods be careful, with me as judge, that they not, as though born at the cross-­roads and almost men of the Forum, ­either ever behave like young men with excessively tender verses or rattle out unclean and ignominious utterances. For they ­will be offended who have a ­horse and a f­ ather and substance, nor, if the buyer of a broken chick-­pea and nut approves of something, do they receive him with equal spirits or reward him with a crown. A long syllable attached to a short one is called an iamb, a swift foot; from ­there also the name was ordered to grow onto iambic trimeters, since it was repeating six beats, like itself from beginning to end. †Not in this way previously† so that it might come slower and a ­little heavier to the ears, it received stable spondees into its paternal rights, obliging and tolerant, [but] not such that it would go away from its second or fourth place in the partnership. This both appears rarely in the noble trimeters of Accius, and it charges the verses of Ennius, sent onto the stage with their ­great bulk, with the foul crime ­either of being a work too swift or lacking care, or of ignorance of the art. Not just any judge sees unrhythmical poems, and an unmerited indulgence has been given to Roman poets. Should I therefore wander off and write without discipline? Or should I think that all men ­will see my faults, [and so be] safe and cautious, [staying] within hope of indulgence? In short, I avoided blame: I did not earn praise. You, keep turning over Greek models with your hand by night, keep turning them over by day. But your ancestors praised both Plautine meter and wit, having admired both too tolerantly, let me not say stupidly, if only I and you all know how to separate an inurbane utterance from an elegant one, and we have developed an understanding of a genuine sound with our fin­gers and ear. Thespis is said to have discovered the unknown genre

240

245

250

255

260

265

270

275

xxviii  P r e face

dicitur et plaustris vexisse poemata Thespis qui canerent agerentque peruncti faecibus ora. post hunc personae pallaeque repertor honestae Aeschylus et modicis instravit pulpita tignis et docuit magnumque loqui nitique cothurno. successit vetus his comoedia, non sine multa laude; sed in vitium libertas excidit et vim dignam lege regi. lex est accepta chorusque turpiter obticuit sublato iure nocendi. nil intemptatum nostri liquere poetae, nec minimum meruere decus vestigia Graeca ausi deserere et celebrare domestica facta vel qui praetextas vel qui docuere togatas. nec virtute foret clarisque potentius armis quam lingua Latium, si non offenderet unum quemque poetarum limae ­labor et mora. vos, o Pompilius sanguis, carmen reprehendite quod non multa dies et multa litura coercuit atque perfectum decies non castigavit ad unguem. ingenium misera quia fortunatius arte credit et excludit sanos Helicone poetas Democritus, bona pars non unguis ponere curat, non barbam, secreta petit loca, balnea vitat. nanciscetur enim pretium nomenque poeta si tribus Anticyris caput insanabile numquam tonsori Licino commiserit. o ego laevus, qui purgor bilem sub verni temporis horam! non alius faceret meliora poemata. verum nil tanti est. ergo fungar vice cotis, acutum reddere quae ferrum valet exsors ipsa secandi; munus et officium nil scribens ipse docebo, unde parentur opes, quid alat formetque poetam,

280

285

290

295

300

305

Hor ace , A r s p oet ic a , t e x t a n d tr a nsl at ion  xxix

of the tragic Camena and to have carried his poems on wagons, [poems] that men, their ­faces stained with wine-­lees, would sing and act out. ­After him Aeschylus, the discoverer of the mask and honorable mantle, both laid platforms with modest wooden beams and showed how to speak grandly and strive with the tragic boot. Old Comedy succeeded ­these, not without much praise; but freedom of speech degraded into vice and into a vio­lence that needed to be governed by law. The law was handed down and the chorus shamefully grew ­silent, its power to harm snatched away. Our poets left nothing unattempted, nor have they earned a small amount of distinction, having dared to abandon Greek traces and celebrate domestic deeds, ­either ­those who taught praetexta-­style plays or ­those who taught togate ones. Nor would Latium be more power­ful in virtue and in celebrated arms than in its tongue, if the effort of polishing and the time [taken over it] ­were not displeasing to ­every one of the poets. You, o blood of Pompilius, censure the song-­poem that many a day and many an erasure has not kept within bounds and corrected tenfold to a perfect fingernail. ­Because Democritus trusts that inborn talent is a greater blessing than wretched craftsmanship and he excludes sane poets from Helicon, a good portion do not care to offer their nails or their beard [to be cut], they seek out secret places, they avoid the baths. Indeed he w ­ ill obtain the prize and the name “poet” if he never entrusts his head, incurable with three doses of hellebore, to the barber Licinus. O, I am unlucky, I who purge myself of my bile just before the season of spring time! No one e­ lse would have made better poems. But nothing is of such ­great value. Therefore I s­ hall perform the role of a whetstone, which is able to make steel sharp, though itself incapable of cutting; writing nothing myself, I ­will teach the gift and duty, [namely] from where resources may be obtained, what may nourish and mold the poet,

280

285

290

295

300

305

xxx  P r e face

quid deceat, quid non, quo virtus, quo ferat error. scribendi recte sapere est et principium et fons. rem tibi Socraticae poterunt ostendere chartae verbaque provisam rem non invita sequentur. qui didicit patriae quid debeat et quid amicis, quo sit amore parens, quo frater amandus et hospes, quod sit conscripti, quod iudicis officium, quae partes in bellum missi ducis, ille profecto reddere personae scit con­ve­nientia cuique. respicere exemplar vitae morumque iubebo doctum imitatorem et vivas hinc ducere voces. interdum speciosa locis morataque recte fa­bula nullius Veneris sed pondere inerti valdius oblectat populum meliusque moratur quam versus inopes rerum nugaeque canorae. Grais ingenium, Grais dedit ore rotundo Musa loqui, praeter laudem nullius avaris. Romani pueri longis rationibus assem discunt in partis centum diducere. ‘dicat filius Albani: si de quincunce remota est uncia, quid superat? poteras dixisse.’ ‘triens.’ ‘eu! rem poteris servare tuam. redit uncia: quid fit?’ ‘semis.’ an, haec animos aerugo et cura peculi cum semel imbuerit, speremus carmina fingi posse linenda cedro et levi servanda cupresso? aut prodesse volunt aut delectare poetae aut simul et iucunda et idonea dicere vitae. quidquid praecipies, esto brevis, ut cito dicta percipiant animi dociles teneantque fideles. omne supervacuum pleno de pectore manat. ficta voluptatis causa sint proxima veris:

310

315

320

325

330

335

Hor ace , A r s p oet ic a , te x t a n d t r a nsl at ion  xxxi

what is fitting, what not, where virtue may lead, where error. Having taste/wisdom is the starting ­point and well­spring of writing properly. Socrates’s sheets of paper ­will be able to show the ­matter to you, 310 and words ­will follow not unwillingly a ­thing that has been planned ahead of time. He who has learned what he owes to his fatherland and what to his friends, with what love a parent ­ought to be loved, with what a ­brother and a guest, what a senator’s duty is, what a judge’s, what are the roles of the leader sent to war—that man undoubtedly 315 knows how to give back to each character the ­things that are fitting to it. I ­will order the learned imitator to look back at the example of life and of customs and to draw from ­there living voices. Sometimes a tale of no feminine charm, attractive in places and given manners correctly but with a clumsy weight, 320 delights the ­people more and stays them better than verses poor in resources and tuneful trifles. To the Greeks the Muse gave inborn talent, to the Greeks she granted [the ability] to speak with a rounded mouth, greedy for nothing besides praise. Roman boys learn how to divide a penny into 325 a hundred parts through long calculations. “Let Albanus’s son answer: if an ounce is taken away from a five-­ounce mea­sure, what is left? You could have answered [already].” “A third.” “Well done! You w ­ ill be able to preserve your wealth. An ounce comes back: what results?” “A half.” Can it ­really be that, when this rust-­canker and concern for 330 pocket money has tinged the soul once and for all, we may hope that song-­poems are able to be crafted that must be anointed with cedar-­oil and preserved in smooth cypress? Poets want ­either to be useful or to delight or to say ­things that are si­mul­ta­neously pleasing and suitable for life. What­ever advice you give, you ­shall be brief, so that teachable spirits 335 may swiftly catch hold of the sayings and faithfully hold onto them. ­Every superfluous ­thing drips out from a full breast. Let ­things made for the sake of plea­sure be very close to true ones:

xxxii  P r e face

ne quodcumque velit poscat sibi fa­bula credi, neu pransae Lamiae vivum puerum extrahat alvo. centuriae se­niorum agitant expertia frugis, celsi praetereunt austera poemata Ramnes. omne tulit punctum qui miscuit utile dulci, lectorem delectando pariterque monendo. hic meret aera liber Sosiis; hic et mare transit et longum noto scriptori prorogat aevum. sunt delicta tamen quibus ignovisse velimus; nam neque chorda sonum reddit quem vult manus et mens [poscentique gravem persaepe remittit acutum] nec semper feriet quodcumque minabitur arcus. verum ubi plura nitent in carmine, non ego paucis offendar maculis, quas aut incuria fudit aut humana parum cavit natura. quid ergo est? ut scriptor si peccat idem librarius usque, quamvis est monitus, venia caret; ut citharoedus ridetur chorda qui semper oberrat eadem: sic mihi qui multum cessat fit Choerilus ille, quem bis terve bonum cum risu miror; et idem indignor quandoque bonus dormitat Homerus; verum operi longo fas est obrepere somnum. ut pictura poesis: erit quae, si propius stes, te capiat magis, et quaedam, si longius abstes. haec amat obscurum, volet haec sub luce videri, iudicis argutum quae non formidat acumen; haec placuit semel, haec decies repetita placebit. o maior iuvenum, quamvis et voce paterna fingeris ad rectum et per te sapis, hoc tibi dictum

340

345

350

355

360

365

Hor ace , A r s p oet ic a , te x t a n d t r a nsl at ion  xxxiii

let not the tale demand for itself that what­ever it wishes be believed, nor let it draw out a living boy from the womb of Lamia who has dined on him. The century-­groups of elders decry poems that have no knowledge of morality, the lofty Ramnes-­tribe men pass over austere ones. He makes off with ­every reward, who intermingled what is useful with what is sweet, by delighting the reader and advising him in equal mea­sure. This book earns bronze from the Sosii; this one both crosses the sea and extends a long existence for its famous writer. ­There are, however, m ­ istakes that we would wish to forgive; for neither does the chord give back the sound that the hand and mind wishes [and it very often sends back a high-­pitched one for the one seeking a low-­pitched one] nor ­will the bow always strike what it is aiming at. But when many ­things gleam in a song-­poem, I ­will not be offended by a few blots that ­either carelessness caused or ­human nature took insufficient precautions over. What therefore is [the point]? Just as a writer of books, if he makes the same ­mistake continually, although he has been warned, lacks indulgence; just as a lyre-­player is laughed at who always makes a ­mistake on the same chord: so the one who very much fails to take action becomes to me [like] that Choerilus, whom I marvel at as good with a smile/laugh two or three times; and likewise I take offense whenever good Homer becomes drowsy; but it is divinely permitted for sleep to creep over a long work. Like a picture is poetry: ­there ­will be one that, if you stand closer, captures you more, and another [captures you more] if you stand further away. This one loves the darkness, that one, which does not dread the keen incisiveness of a judge, ­will wish to be seen ­under the light; this one has given plea­sure once, that one ­will give plea­sure [even if] sought out ten times. O elder of the youths, although you are both still being ­shaped ­toward the right by a ­father’s voice and you are wise in yourself, take up mindfully this

340

345

350

355

360

365

xxxiv  P r e face

tolle memor, certis medium et tolerabile rebus recte concedi. consultus iuris et actor causarum mediocris abest virtute diserti Messallae nec scit quantum Cascellius Aulus, sed tamen in pretio est: mediocribus esse poetis non homines, non di, non concessere columnae. ut gratas inter mensas symphonia discors et crassum unguentum et Sardo cum melle papaver offendunt, poterat duci quia cena sine istis, sic animis natum inventumque poema iuvandis, si paulum summo decessit, vergit ad imum. ludere qui nescit, campestribus abstinet armis indoctusque pilae discive trochive quiescit, ne spissae risum tollant impune coronae: qui nescit versus tamen audet fingere. quidni? liber et ingenuus, praesertim census equestrem summam nummorum vitioque remotus ab omni. tu nihil invita dices faciesve Minerva; id tibi iudicium est, ea mens. si quid tamen olim scripseris, in Maeci descendat iudicis auris et patris et nostras nonumque prematur in annum membranis intus positis. delere licebit quod non edideris; nescit vox missa reverti. silvestris homines sacer interpresque deorum caedibus et victu foedo deterruit Orpheus, dictus ob hoc lenire tigris rabidosque leones; dictus et Amphion, Thebanae conditor urbis, saxa movere sono testudinis et prece blanda ducere quo vellet. fuit haec sapientia quondam, publica privatis secernere, sacra profanis, concubitu prohibere vago, dare iura maritis, oppida moliri, leges incidere ligno. sic honor et nomen divinis vatibus atque

370

375

380

385

390

395

400

Hor ace , A r s p oet ic a , t e x t a n d tr a nsl at ion  xxxv

pronouncement for yourself, that mediocrity and bearableness are rightly allowed to certain ­things. The mediocre ­lawyer and pleader of cases is distant from the virtue of learned 370 Messalla and does not know as much as Cascellius Aulus, but nevertheless he is valuable: to mediocre men, [however,] not ­people, not gods, not the bookstores have granted that they be poets. Just as at a pleasing banquet a discordant group of musicians and thick perfume and poppy-­seed with Sardian honey 375 cause offense, b­ ecause the dinner could have been conducted without ­these, so a poem, born and discovered for pleasing spirits, if it comes down a ­little from the peak, it sinks ­toward the depths. The one who does not know how to play sports abstains from arms on the Campus, and untaught in ball[-­games] or discus or hoop he rests, 380 so that the assembled crowd may not raise a laugh with impunity: the one who does not know how to nevertheless dares to craft verses. Why not? [He is] ­free and nobly born, a member of the equestrian class as rated by his property holdings and far removed from ­every vice. You w ­ ill say or do nothing with Minerva unwilling; 385 you have that judgment, that [good] sense. But if at some point you write something, let it go down into the ears of the judge Maecius and of your f­ ather and ours, and let it be pressed ­until its ninth year within bindings placed upon it. It ­will be permitted to delete what you have not given forth; a voice once sent [out] does not 390 know how to be turned back. Orpheus, holy and an interpreter of the gods, scared woodland men away from slaughter and foul sustenance, [and he was] said on account of this to soothe tigers and rabid lions; Amphion also, the founder of the Theban city, was said to move rocks with the sound of his tortoise-­shell [lyre] and with flattering 395 prayer to lead them where he wished. This was wisdom once, to separate public ­things from private ones, sacred from profane, to prohibit [­people] from wandering sexual relations, to give rules to married ­people, to build towns, to cut laws into wood [tablets]. Thus honor and the name came to divine poet-­priests 400

xxxvi  P r e face

carminibus venit. post hos insignis Homerus Tyrtaeusque mares animos in Martia bella versibus exacuit; dictae per carmina sortes, et vitae monstrata via est et gratia regum Pieriis temptata modis ludusque repertus et longorum operum finis, ne forte pudori sit tibi Musa lyrae sollers et cantor Apollo. natura fieret laudabile carmen an arte quaesitum est. ego nec studium sine divite vena nec rude quid prosit video ingenium; alterius sic altera poscit opem res et coniurat amice. qui studet optatam cursu contingere metam multa tulit fecitque puer, sudavit et alsit, abstinuit Venere et vino; qui Pythia cantat2 tibicen, didicit prius extimuitque magistrum. an satis est dixisse ‘ego mira poemata pango. occupet extremum scabies; mihi turpe relinqui est et quod non didici sane nescire fateri?’ ut praeco, ad merces turbam qui cogit emendas, assentatores iubet ad lucrum ire poeta dives agris, dives positis in faenore nummis. si vero est unctum qui recte ponere possit et spondere levi pro paupere et eripere artis litibus implicitum, mirabor si sciet internoscere mendacem verumque beatus amicum. tu seu donaris seu quid donare voles cui, nolito ad versus tibi factos ducere plenum laetitiae: clamabit enim ‘pulchre! bene! recte!’ pallescet super his, etiam stillabit amicis

405

410

415

420

425

2. The codices’ cantat for Shackleton Bailey’s certat, an emendation suggested by Brink 1971: 399.

Hor ace , A r s p oet ic a , te x t a n d t r a nsl at ion  xxxvii

and to song-­poems. ­After ­these distinguished Homer and Tyrtaeus sharpened male spirits for Mars’s wars with their verses; lots ­were spoken through song-­poems and the road of life was shown and the grace of kings was sought with Pierian modes and play was discovered 405 and the end of long ­labors, so that the Muse, skilled in the lyre, and the singer Apollo might not by chance be a source of shame to you. ­W hether a praiseworthy song-­poem is made by nature or by art was the ­matter ­under investigation. I see neither what use it is to apply oneself without a rich vein [of talent], nor of what use unpolished talent could be; 410 in this way, the one demands the assistance of the other, and [each] conspires in a friendly manner. The one who is working to touch the desired turning-­post with his turn endured and did many ­things as a boy, he sweated and suffered, he abstained from Venus and wine; the double-­pipe player who sings the Pythian [songs] first learned and feared his teacher. 415 Is it ­really enough to say, “I bang out amazing poems. Let the skin-­disease take the furthest one; is it foul for me to be left ­behind and clearly confess that I do not know that which I have not learned?” Just like a herald, who collects a crowd for the purpose of buying wares, a poet, rich in fields, rich in money lent out 420 at interest, ­orders yes-­men to proceed ­toward profit. If in truth ­there is one who is able to serve up a rich feast properly and act as guarantor for a fickle poor man and snatch away the one tied up in tight lawsuits, I ­will be amazed if he, blessed, knows how to tell apart a lying and a true friend. 425 ­W hether you are given a gift or wish to give one to someone, do not lead a man full of gladness to verses made by you: for he w ­ ill shout, “Beautiful! Well done! Not one error!” He ­will grow pale over ­these ones, he ­will even make tear-­drops drip out

xxxviii  P r e face

ex oculis rorem, saliet, tundet pede terram. ut qui conducti plorant in funere dicunt et faciunt prope plura dolentibus ex animo, sic derisor vero plus laudatore movetur. reges dicuntur multis urgere culillis et torquere mero quem perspexisse laborent, an sit amicitia dignus: si carmina condes, numquam te fallant †animi sub vulpe† latentes. Quintilio si quid recitares, ‘corrige, sodes, hoc’ aiebat ‘et hoc.’ melius te posse negares, bis terque expertum frustra delere iubebat et male tornatos incudi reddere versus. si defendere delictum quam vertere malles, nullum ultra verbum aut operam insumebat inanem, quin sine rivali teque et tua solus amares. vir bonus et prudens versus reprehendet inertis, culpabit duros, incomptis allinet atrum traverso calamo signum, ambitiosa recidet ornamenta, parum claris lucem dare coget, arguet ambigue dictum, mutanda notabit, fiet Aristarchus; nec dicet ‘cur ego amicum offendam in nugis?’ hae nugae seria ducent in mala derisum semel exceptumque sinistre. ut mala quem scabies aut morbus regius urget aut fanaticus error et iracunda Diana, vesanum tetigisse timent fugiuntque poetam qui sapiunt: agitant pueri incautique sequuntur. hic, dum sublimis versus ructatur et errat, si veluti merulis intentus decidit auceps in puteum foveamve, licet ‘succurrite’ longum clamet, ‘io cives!’ non sit qui tollere curet.

430

435

440

445

450

455

460

Hor ace , A r s p oet ic a , te x t a n d t r a nsl at ion  xxxix

from his friendly eyes, he ­will leap, he ­will beat the ground with his foot. Just like ­those who hired to weep at a funeral say and do almost more than ­those who are grieving from the soul, so a laugher is in fact moved more than a praiser. Kings are said to press with many goblets and twist with unmixed wine the man whom they strive to have seen through, [to discern] ­whether he is worthy of friendship: if you found song-­poems, let †the souls† lying hidden †­under a fox† never deceive you. Whenever you ­were reciting something to Quintilius, he used to say “fix this, please, and this.” [Whenever] you said that you ­were not able to do better, he would order you in vain to delete something attempted twice and three times and to return the poorly rounded verses to the anvil. If you preferred to defend a ­mistake rather than transform it, he would take up no further word or empty effort, to prevent you from loving yourself and your ­things alone without a rival. The good and prudent man ­will check sluggish verses, he ­will find fault with the harsh ones, he ­will smear a black mark on the untidy with a reed turned sideways, he ­will cut back ambitious ornaments, he ­will force insufficiently clear verses to produce light, he ­will convict the one spoken ambiguously, he ­will note ­things that have to be changed, he ­will become Aristarchus; and ­he will not say “why should I offend a friend over trifles?” ­These trifles ­will lead into serious evils the one [who has been] laughed at once and poorly received.” Just as the one whom bad skin-­disease or the kingly illness press upon, or an orgiastic derangement and angry Diana, [so] ­people fear to touch the mad poet and flee him, if they are wise: boys harass him and incautious men follow him. If this one, while he sublimely burped up verses and wandered about, fell like a fowler intent on blackbirds into a well or a pit, even if he should shout “help, hey, citizens!” for a long time, ­there would not be anyone who would make the effort to raise him out.

430

435

440

445

450

455

460

xl  P r e face

si curet quis opem ferre et demittere funem, ‘qui scis an prudens huc se proiecerit atque servari nolit?’ dicam, Siculique poetae narrabo interitum. deus immortalis haberi dum cupit Empedocles, ardentem frigidus Aetnam insiluit. sit ius liceatque perire poetis. invitum qui servat, idem facit occidenti. nec semel hoc fecit, nec, si retractus erit, iam fiet homo et ponet famosae mortis amorem. nec satis apparet cur versus factitet, utrum minxerit in patrios cineres an triste bidental moverit incestus; certe furit ac velut ursus, obiectos caveae valuit si frangere clathros, indoctum doctumque fugat recitator acerbus. quem vero arripuit, tenet occiditque legendo, non missura cutem nisi plena cruoris hirudo.

465

470

475

Hor ace , A r s p oet ic a , te x t a n d t r a nsl at ion  xli

If someone should make the effort to bring help and let down a rope, I would say, “how do you know ­whether he threw himself into this place intentionally and does not wish to be saved?,” and I ­will relate the demise of the Sicilian poet. When Empedocles wanted to be considered an immortal god, he jumped, cold, into burning 465 Etna. Let ­there be the right and let it be permitted for poets to die. He who saves the unwilling man does the same as one who kills. Neither did he do this once, nor, if he is dragged back, ­will he now become a h­ uman and put aside the love of a storied death. Nor is it sufficiently apparent why he keeps making verses, ­whether 470 he urinated onto his ­father’s ashes or sacrilegiously disturbed a gloomy spot struck by lightning; certainly he rages and like a bear, if it has been strong enough to break the bars blocking its cage, the b­ itter reciter ­causes the taught and untaught man [alike] to flee. The one whom in fact he seizes, he keeps hold of him and kills him 475 by reading, not about to let go of the skin ­unless full of blood, a leech.

hor ace’s a r s poet ic a

I n t r oduc t ion

Becoming the Ars Poetica

the Ars Poetica stands in a lineage of ancient works conceived of as repositories for the essentials of poetics, extending back to the writings of Aristotle, Neoptolemus of Parium, and Philodemus, and forward to pseudo-­Longinus On the Sublime.1 Horace’s 476-­line poem was revered for over fifteen hundred years as the indispensable guide for practicing poets;2 it provided a blueprint for efforts at “updated” rules of literary composition; and it inspired numerous famous translators and imitators, among them Geoffrey of Vinsauf, Ben Jonson, Nicolas Boileau, Alexander Pope, and many other Eu­ro­pean and American writers. From the Ars Poetica have been quarried such oft-­quoted phrases as in medias res (“into the m ­ iddle of t­ hings”), ut pictura poesis (“poetry is like a painting”), and purpureus . . . ​pannus (“a purple patch”), or the dictum that poetry should be both pleasing and useful.3 The Norton Anthology of Theory and Criticism opens its entry on the work: “It would be impossible to overestimate the importance of Horace’s Ars Poetica (Art of Poetry) for the subsequent history of literary criticism.”4 And yet this poem has proven hard to love for recent readers: it is “unfashionable t­ oday: unfashionable even amongst classicists, and certainly so amongst non-­specialists.”5 As its ostensible value as “a kind of literary ‘Magna Carta’ ”6 receded and it ceased to be widely regarded as a document that could ever sincerely aid in literary composition, the Ars Poetica came to develop an entrenched reputation of being tedious and devoid of artistry, as if 1. Dated tentatively to the first ­century CE; see further Porter 2016 and my chapter 4. On the AP in relation to Aristotle and Neoptolemus, see especially Brink 1963: 43–150. 2. Overviews at Friis-­Jensen 2007: 300–302, Laird 2007: 132–33, McGann 2007: 305, Money 2007: 321. 3. The poem’s “invariably succinct and incisive” precepts (Campbell 1924: 238) are often noted, as at Brink 1981: 8, Armstrong 1993: 185, Hardison and Golden 1995: 41, Leitch 2010: 119. 4. Leitch 2010: 119. 5. Moul 2010: 175. 6. Laird 2007: 133; cf. Lowrie 2014: 121, “a virtual bible of poetics in its reception.” 1

2  In t rodu ction

in relentless perpetuation of Scaliger’s “ars sine arte tradita,” an “artless Art.”7 This duality is inevitably tied up with the understanding of it as modeled upon ­earlier Greek works, a relation that both granted the Ars Poetica a greater standing and yet doomed it to be seen as “an anthology of previous ideas, not a system of thought in which each idea has its place as a living and flexible member of an organically unified discourse.”8 The literary scholar Lionel Trilling, for example, branded the poem “a crashing bore and a disaster to lit­er­a­ture,” all while expressing surprise at “the anomaly of its having been written by a g­ reat poet” and confessing himself to be other­wise “engaged and delighted and quite often moved” by Horace.9 ­Others have pretended that the work hardly exists.10 As a result, despite being “the only classical essay on literary criticism that has been known with something like continuity from the date of its composition to the pre­sent day,”11 the Ars Poetica has been the subject, aside from the pre­ sent effort, of only one book in En­glish,12 having been more often considered alongside Horace’s two so-­called “literary epistles” (Epistles 2.1 and 2.2), lumped in with them to its detriment since its epistolary qualities are thus fronted, its many ­others obscured.13 My aim in this study is twofold. First, continuing the approach of Brink, though to rather dif­fer­ent ends, I consider the Ars Poetica as a complete and exceptional literary achievement in its own right.14 I work to show that the poem possesses an internal unity that supersedes its explicit topical breaks and that 7. Scaliger 1594: 878; see Frischer 1991: 10–11 on the per­sis­tence of Scaliger’s critique into the twentieth ­century (it is repeated at, e.g., Bate 1952: 46). 8. Armstrong 1968: 8. 9. Trilling 1970: 131 (in Brophy et al. 1970). 10. In his monumental Horace, for example, Fraenkel 1957 devotes a full chapter to ­every book of poems except the AP (he other­wise omits only Epist. 2.2). 11. Hardison and Golden 1995: 3. 12. I refer to Frischer 1991. This book may be characterized as one of the first instances of digital humanistic scholarship in the Classics, and it is concerned largely with determining the date of the AP, the identity of its addressees and other “interpretative implications of chronology,” and the poem’s genre; promised further volumes (xi) seem never to have been published. 13. E.g., Rudd 1989, Kilpatrick 1990. Armstrong’s 1993 forty-­seven-­page essay and Oliensis’s 1998 chapter in her monograph stand out for their consideration of the AP on its own terms. 14. See especially Brink 1963: 244–71, 1971: 445–523; while valuing the AP’s striking and often-­overlooked internal cohesion, his focus remains on the poem as a work of poetics and literary didaxis. ­Others have naturally also made the observation that the AP is itself a poem, and one worthy of study, e.g., Wimsatt 1970 (in Brophy et al. 1970), Williams 1972: 40–41, Segura Ramos 1989: 119–20, Armstrong 1993, Hardison and Golden 1995: 4, 239, Oliensis 1998, Russell 2006, Hösle 2009: 63–64, and Hajdu 2014, yet as recently as 2010 Moul was able to say that “the poem itself—­what it was for, or how it is meant to be read—is not often well discussed” (175).

be co m ing t he a r s p oet ic a  3

its coherence emerges through recursivity—­judicious echoes and re-­echoes of individual words, sounds, and images. Accordingly, the Ars Poetica’s “nutritive content”15 may be found to lie not in its prescriptive and often heavy-­handed advice but rather in certain subtle thematic strands that weave their way through the length of the poem: how best to teach, how to learn successfully, and what obligations t­ hose bound by ties of blood or friendship bear one another. Second, I elucidate the key place of the Ars Poetica in the Horatian corpus, in par­tic­u­lar its ties to Horace’s Satires, such that the two collections form complementary bookends to the poet’s c­ areer. The Ars Poetica’s thematic strands take up the overt concerns of the Satires, and while in the Satires discussions of ­human be­hav­ior often conceal reflections on lit­er­a­ture, in the Ars Poetica lit­er­a­ture can prove a cipher for ­human be­hav­ior. It becomes clear that in Horace’s worldview, all ­human activity, ­whether living or producing art, is to be governed by the same princi­ples: moderation and propriety are vital, as are self-­critique, caution, and deliberate care. Fi­nally, surveying the poem’s Nachleben and reception, I endeavor to show that generations of readers have understood the Ars Poetica on precisely the terms outlined ­here. The poem has suffered to excess from the notion that it once served as a manual or handbook, and I work to show that this surprisingly per­sis­tent and widespread view of its historical purpose, which lodges no doubt in its centuries-­long usage as a school text, is at odds with the consistent appreciation of it by many readers and writers as subtle, vibrant, and engaging—­a view reflected in its robust reception tradition. Under­lying my argument for the poem’s completeness and internal coherence is the fact that I see its entire material pre­sent in nuce in its opening ten lines. The first three chapters of the four that make up the body of the book, “Reading the Ars Poetica,” thus each take up a pivotal theme originating from a word or words judiciously placed in the opening vignette and follow it the length of the poem. Humano (“­human”), the first word and the title of chapter 1, heralds the Ars Poetica’s concern with all that living entails, thus casting the scope of the work far beyond poetry from the start. As the poem progresses this is borne out by Horace’s striking focus on h­ uman emotions, on life cycles (­whether of ­people or words), on nature and ­human nature, and on spoken language, all of which are given far greater prominence than seems justified in the ostensible context of creating believable characters for the stage. Horace’s concern, I argue, is with all h­ uman endeavor—­the ars vivendi (“art of living”). If the Ars Poetica is read for how it expresses itself, moreover, rather than merely for what it says, it emerges as an ideal exemplum of art, the w ­ hole proving seamless and lending itself to being remade in new ways by ­every reader and upon ­every reading. Fascinated and frustrated in equal mea­sure by the Ars Poetica, 15. To borrow the phrase applied by Gellar-­Goad 2012: 27 to Lucr. DRN.

4  In t rodu ction

Goethe wrote, “Dieses problematische Werk wird dem einen anders vorkommen als dem andern, und jedem alle zehn Jahre auch wieder anders.”16 “Pisones,” the ­family name of the addressees, provides the title of the second chapter, in which I contend that Horace subjects the Pisones to a far less gentle ­handling than has been generally acknowledged, and one in line with his aggressive treatment at times of addressees and other figures in his Satires and Epistles. I further explore how, in dedicating his poem to a unit consisting of a ­father and two sons, Horace is able to make the father-­son relationship a central narrative strand of the Ars Poetica and, with it, the theme of teaching. Ultimately, I see Horace presenting a studied evolution of his poetic persona from student-­son in the Satires, written at the beginning of his ­career, in which we witness him receiving teachings from his own f­ ather, to teacher-­father in the Ars Poetica, written at ­career’s end. From ­behind his masks of senex and pater, Horace instructs and helps to shape the Piso boys (and the general reader), as his own f­ ather had done for the poet’s youthful persona in Satires 1.4. Two key words, amici (“friends”) and risum (“laughter”), unite to form the focus of the third chapter, as Horace explores the paradox that the obligation of a true friend is to criticize, especially through laughter, even at the risk of causing pain (for in sparing a friend from short-­term pain, he runs the risk of exposing him to greater pain over the long term), and that criticisms issued by a friend are necessarily true. Within the framework of Roman and, especially, Epicurean amicitia (“friendship”), Horace boldly negotiates for himself the position of “friend” vis-­à-­vis the Pisones, although ­these figures are mentioned nowhere ­else in his corpus. This ruse of friendship is nevertheless what allows Horace to criticize his addressees’ literary talents and discourage them (along with perhaps ­every reader) from attempting to write poetry. The fourth chapter treats the end of the poem. In considering the Ars Poetica’s final lines, I return to its opening ones, showing how they are linked through a concern with the visual and with making and creating in numerous manifestations. I propose that the Ars Poetica be read as an ars poiētikē, “art of creating,” for Horace’s interests lie in the overlap of all h­ uman pursuits. The source of Latin poetica, and with it our “poet,” “poem,” and “poetry,” the Greek verb ποιεῖν (poiein) is rather more wide ranging in its senses, encompassing “make, produce, bring into existence, cause, and do,” that is, making and creating in a multitude of forms. In addition, in concluding the Ars Poetica by indulging himself in a flight of the sublime, I argue, Horace ends the poem’s conversation on creative endeavor by revealing definitively his superior and unmatchable mastery of the literary art. 16. At Keller 1916: 166, cited also at Hardie 2014: 54; see further Keller 1916: 165–69 on Goethe’s interest in Horace and in the AP.

be co m ing t he a r s p oet ic a  5

The book’s epilogue, “Receiving the Ars Poetica,” traces the themes and concerns of chapters 1 through 4 throughout the poem’s considerable reception history, as I show that practicing poets have long discerned what many literary scholars have not: that the poem’s value lies not so much in its stated contents as in its fine-­spun internal unity; in its interest in ­human nature and the onward march of time; in the importance of criticism—­both giving and receiving it— to the artistic pro­cess; and in the essential sameness of writing, of making art, and of living, loving, being, and even d­ ying. The argument made in this study for reading the Ars Poetica as a literary achievement in its own right may therefore be viewed as a return to the complex, nuanced ways in which it was already read in the ­Middle Ages, through the sixteenth ­century, and into the twenty-­first. The authors of the ­later works examined read the Ars Poetica as exemplifying and instantiating the sort of artistry that it opaquely commands, and they reflected this in turn through their own verses.

The Name of the Poem Horace’s poem has perhaps suffered most of all on account of its name. English-­ speaking readers, who know it as the Ars Poetica, take it up preconditioned by the expectation that we are ­going to encounter an “art of poetry” or The Art of Poetry.17 As the title for an ancient work, ars meant “something like ‘handbook’ or ‘statement of the princi­ples of,’ ” and could thus denote a straightforward textbook, such as the Ars Grammatica of Donatus, or a satirical treatment of the same, as in Ovid’s Ars Amatoria (an “art of love” in the sense that it purported to be a “handbook of seduction”).18 While titles may well be conceived of as descriptions, they are more precisely “names for a purpose,” and “the unique purpose of titling is hermeneutical: titles are names which function as guides to interpretation.”19 Illustrating the princi­ple, Wilsmore suggests that the reader reconsider T. S. Eliot’s The Wasteland as ­After the War (which would render the work historical and local), or Macbeth as The King’s Wife (which would have the 17. Frischer 1991: 5 likewise observes, “What we know, or think we know, about the background of a literary work can have a decisive effect on our interpretation. Clues provided by the author—­ particularly the title . . . ​create certain expectations in us even before we read the first words of a text.” 18. Hardison and Golden 1995: 26 (cf. Golden 2010: 392); Ovid repeatedly terms his work an ars (1.1, 1.3, 1.4 bis, ­etc.) and himself its artifex (1.7; on this “bogus teacher” see Watson 2007, as well as Volk 2002: 188–95, who notes also the work’s traditional exclusion from the genre of didactic on the grounds of being “not serious enough,” 157). On the tradition of the ancient ars, see Zetzel 2018, especially 162–82. 19. Fisher 1984: 288.

6  In trodu ct ion

effect of promoting her to protagonist).20 Titles may, accordingly, be “neutral”; they may be “underlining” or “reinforcing”; they may have the effect of “focusing” the reader as he encounters the work; they may be “undermining,” “opposing,” or “mystifying”; or “disambiguating,” “specifying,” or “allusive,” referring to other works or artists or to historical events.21 The pre­sent conundrum, then, is that it is not at all certain ­whether Horace himself dubbed his poem Ars Poetica; in fact, ­there is no secure evidence for what he called it, or even for ­whether he called it anything at all. We therefore find ourselves ­either approaching the work in blinkered fashion as an ars poetica (as which it has then so often been found wanting), or ­else attempting to determine how Horace intended for us to approach it. The designations ars poetica and (liber) de arte poetica are attested early,22 both first appearing in the first-­century CE rhetorician Quintilian’s Institutio Oratoria: usus deinde Horati consilio, qui in arte poetica suadet ne praecipitetur editio ‘nonumque prematur in annum,’ dabam his otium, ut refrigerato inventionis amore diligentius repetitos tamquam lector perpenderem. (Epistula 2) Then having made use of Horace’s counsel, who in his art of poetry (ars poetica) advises, in order that an edition may not be rushed out, that it “be set aside ­until the ninth year,” I was devoting some of my leisure to ­these ­things, in order that, with my love for my invention having been chilled, I might approach my revisited writings more carefully, like a reader. cui simile vitium est apud nos si quis sublimia humilibus, vetera novis, poetica vulgaribus misceat—id enim tale monstrum quale Horatius in prima parte libri de arte poetica fingit: Humano capiti cervicem pictor equinam iungere si velit et cetera ex diversis naturis subiciat. (8.3.60) ­ ere is among us [Romans] a vice similar to this one [sc. Greek Sardismos, Th a mixing of dialects], if someone should mix sublime ­things with ­humble ones, old with new, poetic with common—­for this is the same sort of monster as Horace creates in the first part of his book on the art of poetry (liber de arte poetica): 20. Wilsmore 1987: 403; see also Fisher 1984, Adams 1987: 7. 21. Levinson 1985: 34–37. For Genette 1997: 79–91 titles are e­ ither “thematic,” describing the work’s subject ­matter, or “rhematic,” stating its essential form (e.g., Epigrams, Dictionary, ­etc.). 22. Dilke 1958: 55, n2 also places value on Quintilian’s testimony.

be co m ing t he a r s p oet ic a  7

To a h­ uman head if a painter should wish to join a h­ orse’s neck —­and if he should add other ­things of dif­fer­ent natures. As secure as ­these two appellations might appear, ­whether they rise to the level of “titles” (rather than mere descriptions) must remain a m ­ atter of interpretation. Moreover, it is not clear ­whether the word liber (“book”) in the second passage is essential or incidental to the entitling phrase; that is, w ­ hether Quintilian meant to call the work “A Book Concerning the Art of Poetry” (liber de arte poetica), or w ­ hether he is speaking of “a book” (liber) entitled “Concerning the Art of Poetry” (de arte poetica) or of “a book concerning/ about/ on” (liber de) “The Art of Poetry” (in which case the work’s name would be ars poetica, once again).23 In both passages, however, Quintilian is clearly less con­ cerned with the Ars Poetica itself than with using some lines from it to illustrate a point of his own. Consequently, he provides the minimum amount of information necessary for his reader to identify the source of the quotation given. The question then becomes w ­ hether Quintilian is creating a name for the work (a ­simple, easily recognizable one), or w ­ hether he is referring to it by an exist­­­ ing (accepted, familiar) name—­one bestowed on it ­either by Horace or by an intermediary no longer extant. The fact that Quintilian was closer in time to Horace than we are does not by itself ensure that he possessed any insights inaccessible to us u­ nless we suppose, without any evidence that can be pointed to, that he was drawing his information from a line unbroken since the poet’s lifetime and now lost to us.24 My sense, nevertheless, is that in this instance a continuous tradition of calling the work ars poetica or (liber) de arte poetica since the time of Horace, and which is first found attested in Quintilian, is reasonably likely, for the following reasons. First, ­unless the work was unfinished at the time of Horace’s death—­which its complete state and coherence militate against25—­Horace himself, and his contemporaries, almost certainly called it something, for “Roman books did in general have titles” that “­were often demonstrably the author’s own and not ­those l­ater supplied by booksellers, librarians or purchasers.”26 While this 23. Daly 1943: 26 likewise observes that it is hard to say ­whether “the familiar type of prepositional phrase introduced by de” is “a formal title or simply a con­ve­nient manner of describing a work.” 24. Bowersock 1971: 73: “the mere antiquity of a testimony is no guarantee”; on this prob­lem more generally, see Lefkowitz 1981, rev. 2012. 25. Nevertheless, the AP’s perceived “want of structural completeness” has been taken by some as “proof that it was never finished,” Wilkins 1896: 330. 26. Horsfall 1981: 103; he is concerned with the period from Cicero to Suetonius (see his n1). On the state of pre-­Ciceronian titulature, see Daly 1943: 30–32 (who rightly points out that

8  In trodu ct ion

practice is easily comprehensible to us in that it closely resembles our con­ temporary one, the situation is somewhat complicated by the fact that, as Horsfall goes on to explain, “the ‘title’ of a work of Greek or Latin lit­er­a­ture can mean, in concrete and physical terms, one—or more than one—of four ­things:” “the tag or sillybos, which hung from the end of the roll as it lay on the shelf ”; “a title standing at the head of the work within the roll”; “the subscription . . . ​at the end of the work”; or a title “written on the verso of the [papyrus] roll along the outside edge.” What all this indicates in practice is that although Horace and his contemporaries, as stated above, likely called the work something, they may well have called it more than one something, such that the titular formulations Quintilian uses, ars poetica and (liber) de arte poetica, if we wish even to distinguish between them at all, may well have equal authority.27 Second, and more compelling, is that Quintilian, whose occupation as a teacher of rhe­toric indicates that he was writing a form of textbook to be used in schools,28 is evidently referring to a work that he expected to be familiar, perhaps even well known, to his readers, and he would have naturally and necessarily referred to it in a form that t­ hese relatively unsophisticated readers would have recognized (rather than employing an idiosyncratic or recherché name).29 The designations ars poetica and (liber) de arte poetica are therefore unlikely to have been in­ven­ted by Quintilian, especially in the short time since Horace’s death in 8 BCE. All of this strongly suggests (in the absence of any other evidence) that the poem was known as ars poetica or (liber) de arte poetica during or very shortly a­ fter Horace’s lifetime and thus that it may even have been termed particularly in the case of prose works, the title “was not then a ­matter of ­great concern to writers” since “­there was l­ittle or nothing” from which a par­tic­ul­ar work had to be distinguished). On authors naming their own works, see Horsfall 1981: 103, Wilsmore 1987: 404, Lowrie 1997: 62–63 (on Verg. Ecl. 6.11–12). 27. Of Ars Poetica and De Arte Poetica, Frischer 1991: 16 says, “­These titles are so similar that we need not expend any effort trying to choose between them.” On the existence of multiple titles/designations for ancient works, see further Daly 1943, Horsfall 1981: 105 (who notes that works could si­mul­ta­neously be known by both a title and a description of the subject ­matter, as in certain essays of Varro’s Logistorici, such as Marius de fortuna, Tubero de origine humana [OCD s.v. “Varro (Marcus Terentius Varro)”], and Horace’s own Odes, which he termed carmina, his Satires, which he referred to as sermones, and his Epodes, which he called iambi; cf. Horsfall 1979a). 28. Per Jerome (OCD s.v. “Quintilian [Marcus Fabius Quintilianus]”), he was the first rhetorician to receive a salary from the fiscus (imperial trea­sury), and his teaching ­career spanned twenty years. 29. As Laird 2007: 132 adds, Quintilian’s decision to preface his book with Horace’s dictum to lay aside a work for nine years before publishing it (AP 388–89) demonstrates “the stature of the Ars Poetica” for him (and his readers) already at this early date.

be co m ing t he a r s p oet ic a  9

this by the poet himself. Moreover, the existence of Quintilian’s two or three “titles,” if we indeed understand them to be distinct, might suggest that Horace was himself designating rather than precisely entitling. Conversely, it is hard to see why Quintilian (or his source) would have named the poem Ars Poetica, to the exclusion of all other possibilities, when the poem is so rich with further modes of reading, as I aim to show in this study. (By another school of interpretation, Quintilian has been blamed for leading astray generations of readers by erroneously dubbing the poem an ars).30 Quintilian’s testimony is also useful for another reason, if accidentally: in employing the singular term liber, ­whether following Horace or in­de­pen­dently, he indicates that for him, at any rate, a ­century ­after Horace’s death, the Ars Poetica was a stand-­alone work.31 In the centuries that immediately follow, the work is found referred to exclusively by Quintilian’s phrases as Ars Poetica or (Liber) De Arte Poetica. In the second c­ entury CE, for example, Terentius Scaurus wrote his Commentarii in Artem Poeticam, and in perhaps the third, the scholiast Helenius Acro introduced his commentary with “conposuit istum [sc. librum] de Arte Poetica ad Pisonem quendam poetam” (“he wrote this [book] concerning the ‘Art of Poetry’ [or, ‘Concerning the Art of Poetry’] to a certain Piso, a poet”).32 Pomponius Porphyrio, the other major Horatian scholiast, likewise termed it de Arte Poetica.33 Aside from ­these references, however, and a handful more in Servius’s commentaries on Virgil (where it is likewise termed ars poetica), no mentions of Horace’s poem by the names ars poetica or de arte poetica (or any other, for

30. See, e.g., Campbell 1924: 235 (Quintilian’s Ars Poetica is “misleading” ­because “it implies a complete and formal treatise”), Golden 2010: 392. 31. Frischer 1991: 16 takes Quintilian’s comment as meaning that ­there was only one book of Ars Poetica (i.e., that Horace did not write a second book, now lost), but it equally indicates that the AP did not belong to Epistles 2. Rather bizarrely, Pseudo-­Acro implies that the AP is a fifth book of sorts to the Odes: “with the preceding four books of Odes having been completed, he composed this fifth book concerning the Art of Poetry” (“terminatis [superioribus] libris IIII. Carminum conposuit istum quintum de Arte Poetica,” Hauthal 1866: 575). 32. Hauthal 1866: 575 (page references to the scholia are provided in this study only when a scholiast’s comment is not ad loc.). Acro’s commentary is referred to throughout as “Pseudo-­ Acro,” since it is thought to be a blend of Acro’s writings with ­those of o­ thers. 33. Hauthal 1866: 649: “hunc librum, qui inscribitur de Arte Poetica” (“this book, which is headed ‘Concerning the Art of Poetry’ ”; inscribere denotes the writing of any name, title, or dedication at the top of a work and thus can refer to any form of heading in a broad sense as well as to a title proper). The dates of t­ hese two major scholiasts on Horace are not securely established; see further Frischer 1991: 13 on them and their titles for Horace’s poem, and Zetzel 2018: 149–56 on the history and dating of the commentaries.

10  In t rodu ction

that ­matter) are to be found in the four centuries following its composition.34 In the ­Middle Ages, Poetria (“Art of a Poet,” “Poetry,” “Poesy”) is found in place of the titles evidenced through the third c­ entury CE, with the appropriation of a term that meant exclusively “poetess” in Classical Latin.35 The twelfth-­ century monk Conrad of Hirsau in his Dialogus super auctores plays upon this change in sense saying, as a teacher in conversation with an imaginary student, “poetria or poetrida is a w ­ oman enthusiastic about poetry; that poet [sc. Horace] is thought to have used this title [Poetria] for this reason, [namely] that he pre­sents the beginning itself of his work as though a beautiful w ­ oman up above” (“poetria vel poetrida est mulier carmini studens; quo titulo hac de causa usus putatur iste poeta, quod ipsum operis sui principium quasi mulierem superne formosam premonstrat”).36 It is hard to tell ­whether Conrad believed (erroneously) that Horace himself had named his poem Poetria or how seriously he meant his explanation to be taken, yet his comment is useful for showing continued engagement with the m ­ atter of the poem’s name, as well as being an early instance of confronting what titles of literary works are for and what they can do. Outside the anglophone classical world, the picture is muddier: the Ars Poetica has more commonly been termed the Epistula ad Pisones37 (continuing a practice seen also in the books of Epistles)38 or Epistles 2.3, as though it ­were the third poem of the second book,39 a view first attested in the Commentator 34. Results include ­those from a search of the Packard Humanities Institute database of Latin texts, which contains all Latin texts up to 200 CE, and select texts from ­later antiquity. The name Pisones (the title Epistula ad Pisones is discussed below) also does not occur in this period as a way of referring to the work. 35. Hauthal 1866: 574: “inde a saec. XI prodit in epigraphis et scholiis u. POETRIA” (“then, beginning in the eleventh c­ entury, it appears in the epigraphs and scholia as ‘Poetria’ ”). Poetria is first attested as a feminine abstract noun with this new sense in the seventh c­ entury in a scholium to Horace’s Epistles (Campbell 1924: 58, n1). 36. Huygens 1970: 113. 37. This is occasionally found in English-­language scholarship as well, e.g., Golden 2010: 391. Frischer 1991: 8–9 (cf. 73–74, and Mañas Nuñez 2012: 235) identifies Jason De Nores as the originator of the title Epistula ad Pisones in his 1553 edition (Mañas Nuñez 2015 suggests that Achilles Tatius called his own 1553 work In Q. Horatii Flacci poeticam commentarii precisely in order to avoid engaging with the debate about the work’s nature as liber, ars, or epistula), though Brink 1963: 233 suggests it may have been the AP’s original title. 38. Epist. 2.1 and 2.2 are known as the Epistle to Augustus and the Epistle to Florus, respectively, and the same convention is still applied to the epistles of book 1. 39. Wilkins 1896: 330, who notes that this “fashion,” although it “rests upon no ancient authority,” has “recently been revived” (it can be seen at, e.g., Reckford 1969, Williams 1972: 38–40, Lanternari 1974; cf. also Mañas Nuñez 2012: 225 on the tradition).

be co m ing t he a r s p oet ic a  11

Cruquianus (who therefore deviates from the understanding found in Quintilian of the Ars Poetica as a stand-­alone book).40 The generic classification of the Ars Poetica as an epistle is ancient, seen already in the late-­antique grammarian Charisius, who quotes verses from the Ars Poetica, attributing them to Horace’s epistulae/epistolae.41 Entitling the work an epistle naturally produces a dif­fer­ ent effect in the first-­time reader, leading him or her to believe that what s/he is about to read is a letter in verse form, even a companion piece to the two long poems of Epistles 2, as the common placement of the Ars Poetica directly ­after Epistles 2 aids in suggesting.42 Attempts have even been made to have one’s nomenclatorial cake and eat it, too, with the poem found entitled/described as Ars Poetica ad Pisones, De Arte Poetica ad Pisones,43 Epistula Tertia Libri Secundi ad Pisones De Arte Poetica,44 and Epistel über die Dichtkunst—­monstrous hybrid creations to rival that in the poem’s own opening lines, that by ingeniously combining two or three ele­ments of “epistle,” “ars,” and “Pisones” relieve their inventors from the responsibility of having to decide among t­ hese three titular building blocks.45 The discomfiting fact remains, however, that none of the titles in current use (Ars Poetica/ De Arte Poetica, Epistula ad Pisones, or Epistles 2.3), or ­those 40. The date of this Commentator is unknown, and he is named for Jacques de Crucque (Jacobus Cruquius), who in the sixteenth c­ entury published an edition containing notes, widely believed to be ancient, that he considered too insightful to have been the work of Porphyrio and Pseudo-­Acro. 41. Frischer 1991: 12–16 rejects Charisius’s designation on the grounds that it would “cast doubt on the other­wise unan­i­mous witness of ancient authors,” but Charisius’s choice of term is salvageable on two counts. He mistakenly attributes a quotation from the AP to Persius’s epistles (“Persius epistolarum”), which, though it has been corrected to “Horatius epistolarum,” indicates that satires such as Persius wrote could be termed epistulae; epistulae may thus be understood as a synonym for “hexameters” just as sermo is used freely by Horace, his contemporaries, and ­later ancient authorities to mean both the Satires and Epistles (all in hexameters); see further Mohr 1895: 302, Horsfall 1979a: 118, 1979b: 170, 1981: 108, Ferriss-­Hill 2015: 43–44. In addition, correspondences between the first poem of Lucilius’s twenty-­sixth book and Horace’s AP have long been recognized, again revealing the overlap of genres; see Fiske 1913, 1920: 446–75. 42. Frischer 1991: 12. For Brink 1963, 1971, the AP is firmly an epistle (cf. Ferenczi 2014b: 71, Geue 2014: 148), though the view of Reinhardt 2013: 502 that the two poems of Epistles 2 “complement AP irrespective of ­w hether we regard the latter as an epistle, too” seems more mea­sured. 43. ­These would be in line with, e.g., Cicero’s De oratore ad Quintum fratrem or Orator ad M. Brutum, which exploit features of personal correspondence without actually being such. 44. First in Cruquius’s 1578 Antwerp edition (Frischer 1991: 11). 45. Wimsatt 1970: 135 (in Brophy et al. 1970), recalling this tradition, ­settles humorously on “epistolary ars poetica.”

12  In t rodu ction

applied to the work throughout the past two millennia, is “supported by any evidence dating from Horace’s lifetime.”46 As a result, insofar as titles are “integral constituents of works of art,”47 the reader finds him-­or herself rather at a loss, for s/he cannot be sure that s/he is meeting the work in accordance with Horace’s authorial designs. This is true even if Horace did not title it anything, or entitled it more than one ­thing, for “an artist can keep track of and enable discourse about a work without actually titling it,” using instead “a pet description, derived from some key feature of the work, or . . . ​a completely neutral numerical or alphabetical catalog”—­options that would all imbue the work with a par­tic­u­lar and distinct significance, for “the title slot for a work of art is never devoid of aesthetic potential; how it is filled, or that it is not filled, is always aesthetically relevant.”48 It should also be noted that the desire to discover “what Horace ­really called his poem” and thus to force agreement on a single title for it, one ideally with ancient authority, may be a rather modern preoccupation: if the way of referring to a poem showed some variation in the ancient world, and if the titles we use ­today show a similar variation, it may be ­because such variation existed, absent of the need evidently felt now for one authoritative title to exclude all ­others. In ­these vari­ous ways, and w ­ hether by design or accident, the poem ended up becoming (even if not exclusively) the Ars Poetica. As the work must, however, be referred to by some (one) name in this study dedicated to it, I retain ­here the name Ars Poetica, first ­because this is ­today the most widespread and standard term for the work in the English-­speaking classical tradition and beyond. Second, and more importantly, I retain it on the strength of the evidence from Quintilian. Not only are his the earliest attested references to the work, but his ars poetica / (liber) de arte poetica are found in a work the aims and interests of which do not include the creation of or a preoccupation with the titles of the works he mentions, and accordingly any titles mentioned are less likely to show meddling by the author. Had it not been too radical, I would have introduced and used ­here yet another and a previously unattested title for the work: the Humano capiti, ­after the fashion of referring to individual Horatian odes by their opening words,49 and bearing the notable advantage that it preconditions the reader neither for an ars nor for an epistle.

46. Frischer 1991: 6. 47. Levinson 1985: 29. 48. Ibid.: 33–34, 29; emphasis his. 49. Feeney 1993: 44: “He still is often cited by the incipit, despite the introduction of a new numbering system of citation.” Similarly, Kenney 1970: 290 notes that “the Aeneid was commonly (and very early: Ov. Tr. ii 534) referred to as the Arma virumque.”

be co m ing t he a r s p oet ic a  13

The Genre of the Poem Bound up with its title is the fraught m ­ atter of the generic tradition within which the Ars Poetica may be located: is it an ars, a handbook of didactic intent; or a letter, to be considered alongside Horace’s books of epistles? My aim is not to reprise the vari­ous arguments for the work as didactic (mini) epic rather than verse epistle, or vice versa, or even to champion it as a tertium quid, though this would be closest to the mark,50 since such reductive approaches contribute ­little to our understanding of the work on its own terms.51 Rather, having elucidated how unavoidably hampered we are by the very title we use simply to acknowledge the real­ity of the poem’s existence,52 I trace the place of the Ars Poetica in vari­ous interwoven strands of literary genealogy with a view t­ oward interpreting it as what I suggest it is above all—­Horatian hexameter. Tradition long had it that ancient Greek (and with it Western) literary criticism began with Aristophanes’s Frogs, the text in which are first found a number of stylistic distinctions and judgments that would prove enduring.53 While Aristophanes’s Frogs is, like Horace’s Ars Poetica, in verse, both are commonly considered alongside Aristotle’s Poetics, the earliest extant prose work devoted to literary theory, especially drama.54 Aristotle also touches on literary ­matters in the Rhe­toric, while Plato famously considered ­whether t­ here was a role for poets in the ideal city of his Republic. Some de­cades a­ fter Aristotle, Neoptolemus of Parium during the third ­century BCE wrote poems

50. Readers interested in t­ hese ­matters can find an overview at Frischer 1991: 87–100 (who narrows down the choices to “epistle, didactic poem, or tertium quid”), Ferenczi 2014a, and Hardie 2014. Discussion of the AP’s genre shows no signs of abating, as may be seen at, e.g., Seeck 1995, Hösle 2009: 68. 51. Cf. Wimsatt 1970: 136 (in Brophy et al. 1970), Horsfall 1979a: 118, 1979b: 171 (on the extent to which the Epistles are r­ eally letters). 52. Perhaps the most obvious function of titles, a­ fter all, is “to provide the reader with a ­handle by which to make reference to his work” (Wilsmore 1987: 403); cf. Fisher 1984: 289 (“titling permits discourse”), Levinson 1985: 37–38 (titles “denote their b­ earers and facilitate reference to them”). 53. E.g., the g­ rand versus the slender style, and poetry of effort versus poetry of inspiration. On the place of Frogs in the ancient Greek literary critical tradition, see O’­Sullivan 1992: 15 (who rightly notes the “a priori unlikelihood that Aristophanes was solely responsible for his critical terms in an age of ­great interest in literary ­matters”), Silk 2000, Rosen 2004 and 2008, Hunter 2009: 29, Ferriss-­Hill 2015: 190–91. 54. This is not to suggest that the Poetics is an uncomplicated text, or one without an impor­tant philosophical component, as the work of Else 1957, Halliwell 1986, Nussbaum 1986a, Janko 1987, and ­others has shown.

14  In trodu ct ion

as well as works on literary criticism and philology.55 Neoptolemus was long thought to be a key figure in the background to Horace’s Ars Poetica, since Porphyrio states that Horace “gathered the teachings of Neoptolemus of Parium—­ not all of them, but the most outstanding ones” (“congessit praecepta Neoptolemi τοῦ Παριάνου non omnia sed eminentissima”).56 The fact that we have only Neoptolemus as filtered through Philodemus’s On Poems, however, rather complicates our ability to evaluate properly Porphyrio’s claim.57 If we are to read the Ars Poetica as being in the tradition of didactic poetry, on the other hand, rather than of literary critical treatises (though, as is clear, the two overlap), it must be considered alongside the near-­contemporary works by Lucretius (De Rerum Natura) and Virgil (Georgics), themselves in a tradition extending back to Hesiod.58 While most would be uncomfortable classing ­these as mock-­didactic (though jokes about attempting to farm with the Georgics are well known to classicists), they are equally far from straightforwardly and plainly didactic.59 Alongside ­these runs the genre of epic parody, which had begun already in the seventh or sixth ­century BCE with the pseudo-­Homeric Margites60 and is widespread in the fifth-­century plays of Old Comedy as well.61 From the fourth c­ entury BCE are known several works that combine mock didactic and mock epic, Archestratus of Gela’s Hedypatheia being the best preserved:62 the poet begins in epic vein by invoking his muse (­there is no 55. OCD s.v. “Neoptolemus (2).” 56. Hauthal 1866: 649. 57. See especially Tate 1928, Brink 1963: 43–78 with Williams 1964, Laird 2007: 133–34, and Reinhardt 2013, as well as my chapter 4, for how the understanding of Horace’s “debt” to Neoptolemus has shifted over the past ­century. 58. Campbell 1924: 40 describes the AP, too, as “on one side at least in direct descent from the Works and Days,” and Williams 1968: 357, 1980: 266–82 discusses some correspondences between the AP and Georgics. The connection between Virgil’s Georgics and Hesiod is also frequently made, e.g., Thomas 1986, 1988: 3–6. Volk 2002: 44–68 is useful on the development of the didactic mode up to and through the Roman period, and Ferenczi 2014b, Hajdu 2014, and Hardie 2014 consider the didactic features of the AP. 59. On the didacticism of the Georgics, see further Thomas 1988: 3, who notes that while “Virgil himself invites the characterization,” nevertheless “a poem which is to be truly didactic in content as well as form . . . ​implies the existence of an audience which is to be instructed, and in spite of the long-­held view that the function of the Georgics was to restore an interest in Italian agriculture, the fact is that no Roman farmer would have read the poem for practical instruction.” 60. See Poet. 1448b38–9a2 with Olson and Sens 1999: 5–6. 61. Olson and Sens 1999: 7. 62. Henriksson 1956, Gowers 1993: 135 with n101 (“gastronomic parodies of epic”), Olson and Sens 1999 (an edition of the fragments of Matro of Pitane) and 2000 (esp. xxviii–­xliii).

be co m ing the a r s p oet ic a  15

personal addressee as in the Georgics, De Rerum Natura, and Ars Poetica)63 and the didactic impulse finds its outlet in the prominent focus on gastronomy. In Latin and slightly l­ ater in date, t­ here may be found Ennius’s Hedyphagetica64 and Varro’s Peri Edesmaton, both epic poems on food in the tradition of Archestratus. Notably, like Horace, both Ennius and Varro also wrote satire, pointing at a perhaps natu­ral connection between satire and didaxis, facilitated also by the fact that most Roman satire was, like epic and didactic poetry, written in hexameters, as are also Horace’s own Epistles.65 It is easily apparent that didactic is itself a multivalent designation, encompassing works that might also be described as epic, as satirical or humorous or parodic,66 as bucolic, as philosophical, and more. The same is true of literary epistle, particularly in the hands of Horace: the Epistles are literary critical, didactic, satirical, and also tantalizingly personal (as are his Satires). The m ­ atter of the Ars Poetica’s genre thus quickly becomes murky and too dense to navigate successfully. If Horace had in mind Virgil’s Georgics as he wrote, Hesiod necessarily made his way into the poem, as did Lucretius and Callimachus,67 though Horace’s connection with t­ hese Virgilian models is direct as well. If Horace was looking to the Greco-­Roman literary critical tradition, Neoptolemus of Parium brought Aristotle, who brought Aristophanes’s Frogs (familiar to Horace in its own right, too), and Philodemus also brought Plato (again, pre­sent without filtering). If Horace was at times thinking of his own Satires, as it is clear he was (and had the Ars Poetica come down to us u­ nder the title of Sermo or Satura, it would declare its manifest connections to t­ hese books in incontrovertible terms), then Greek and Roman comedy and Greek 63. On the importance of epic’s invocation of the Muse and the related “self-­consciousness” of didactic and epic, see Volk 2002: 6–24, 42. 64. While Courtney 1993: 24–25 is skeptical that Ennius’s (or Archestratus’s) poem is mock-­ didactic, he regards Varro’s “gastronomic cata­logues,” along with t­ hose in Hor. Sat. 2.4, as “plainly satirical in intent.” 65. Ennius’s Hedyphagetica is in hexameters, as are Lucretius’s DRN, Virgil’s Georgics, and Hesiod’s Works and Days. 66. As Hardie 2014: 44 notes, for example, “didactic is an impor­tant part of the mix of Horatian sermo from its inception.” See Olson and Sens 1999: 5 on the distinction between satire and parody as it pertains to the works of Archestratus, Matro, ­etc. 67. Lucretius’s “linguistic influence upon the Georgics is pervasive” (Thomas 1988: 4–11), while the presence of Callimachus may be seen already in such fundamental aspects of the poem as its four-­book structure (modeled on the four books of the Aetia) as well as its “learning and interest in recondite m ­ atters of scholarly concern.” As Feeney 1993: 44 notes, moreover, “the very act of composing in books is a feature of Hellenistic culture, for pre-­Hellenistic poets composed poems, not books”: like the Georgics, then, Horace’s Satires, Odes, Epistles, and even Ars Poetica (liber de arte poetica) are Callimachean in form.

16  In t rodu ction

philosophy and Lucilius intrude from ­there, as do Varro and Ennius (despite Horace’s excision of them from his satirical canon),68 who bring also their mock-­ didactic writings. Lucretius arrives accompanied by the full material of the De Rerum Natura that contains also Epicurus and with him Philodemus (in whom resides Neoptolemus, and so on).69 The territory of the Ars Poetica has become crowded, each generic strand that informs it jostling for preeminence of place. If Roman satire is an “anti-­genre,” as Kirk Freudenburg has described it,70 then the Ars Poetica, which defies each and ­every generic classification,71 is anti-­ genre on the scale of an individual poem. Richard Thomas describes Virgil’s Georgics as having “no single formal or generic model—­that is, no author and no work could lay claims, even on the surface, to Virgil’s allegiance.”72 Horace’s Ars Poetica is best understood in ­these same terms. While Horace may well have started with Philodemus and Neoptolemus, Aristophanes and Aristotle in mind, and looked aside t­ oward Lucretius and Virgil, his poem as a final product is sufficiently far from any one model that it stands generically alone—­the first, as Hösle 2009 terms them, of the “poetische Poetiken,”73 a form that would be taken up by Geoffrey of Vinsauf, Boileau, and Pope. Furthermore, what­ever its precise date, Horace was writing the Ars Poetica around the time that Virgil wrote his Georgics, Ovid his Ars Amatoria, and Varro his Peri Edesmaton. The intellectual milieu, then, was fertile for extending the form and conventions of didactic to plausible but ultimately perverse subject m ­ atter, giving rise to a set of works that might be termed paradidactic (to avoid the pejorative connotations of mock-­didactic). Seen in this light, Horace’s ploy becomes especially clever, for poetry is the least problematic of all t­ hese topics: while “Ovid does not, of course, seriously offer the Ars [Amatoria] as an a­ ctual manual of seduction, any more than Virgil ­really aims to teach farming in the Georgics,”74 poetry and writing are such natu­ ral topics for a work of didactic that the ways in which the Ars Poetica 68. Horace acknowledges Ennius only as an epic poet and dramatist; see Ferriss-­Hill 2015: 10 with nn36–38. 69. See Janko 2000: 9–10 on Philodemus and Lucretius’s DRN. 70. Freudenburg 2001: 1. 71. Reinhardt 2013 is useful on the ways in which the AP confounds the reader’s expectations of it as didactic poetry; as he notes, for example, the “absence of conventional didactic structuring devices is a pervasive feature in AP” (517). 72. Thomas 1986: 173–74; cf. Reinhardt’s 2013: 505 comment that while some scholars, with the poem’s content in mind, can say that “­there is no original thought in AP,” others, looking to the poem’s form, can say, “This is Horace’s most imaginative, adventurous work.” 73. Cf. Seeck 1995: 143 (“wie kann man in dichterischer Form über Dichtung reden?”), Russell 2006: 324 (“a poem on poetics”). 74. Miller 1993: 232; cf. Hardie 2014: 43 and my introduction, n59.

be co m ing the a r s p oet ic a  17

(deliberately) fails as and yet goes far beyond straightforward didactic has often eluded readers. Praeceptor amoris is a plainly amusing fiction; praeceptor artis poeticae, by contrast, beguiles the Pisones and us into believing, or at least wishing, that by following the dictates enumerated in Horace’s poem we, too, may become expert in literary composition. By redirecting the evident con­ temporary taste for paradidactic back to a plausibly real topic in a further broadening of this already slippery tradition, Horace succeeded in obscuring and obfuscating the genre of his work, and with this its tone and aims. Above all, however, we should not lose sight of the fact that the Ars Poetica is in hexameters—­the same hexameters of the Epistles and Satires.75 Just as Horace’s Satires, though viewed as second in a genre with four representatives (Lucilius, Horace, Persius, Juvenal), w ­ ere in fact the first exclusively hexameter collection of saturae/sermones,76 and with his first book of Epistles Horace had in­ven­ted an entirely new genre,77 so we may be best off understanding the Ars Poetica as another, unique, Horatian starting point, rather than attempting to locate it securely (without success) within any existing tradition. Porphyrio already distinguished the Satires and Epistles, yet revealed their fundamental unity, by describing the former as “a conversation with someone who is pre­ sent” and the latter as an interaction with someone who is far away, adding that he saw them as differing in title alone.78 As conversation (sermo), the Ars Poetica encompasses both possibilities. Moreover, I suspect that a reader, absent a familiarity with a par­tic­u­lar line or passage, would be hard pressed to assign it correctly to the Satires, Epistles, or Ars Poetica.

The Date of the Poem As uncertain and controversial as its name and generic affiliations has been the Ars Poetica’s date. While the poem reads well as the culmination of a successful ­career, in which the wizened poet dispenses advice to tyros in the art, it goes without saying that this impression does not constitute evidence for the 75. On the meter of the AP, see Ott 1970 with Greenberg 1970, Möhler 1989. 76. The Saturae of Ennius and Lucilius are in a variety of meters (including, but not l­ imited to, the hexameter). 77. Kilpatrick 1986: xiii, Mayer 1994: 1–5, and Reinhardt 2013: 501 do well to remind us of this: although ­earlier authors had written letters in verse (e.g., Lucilius in his fifth book, Catullus in poems 13 and 35), Horace was the first to compose a coherent book of poems in this form. 78. Ad Hor. Sat. 1.1: “in Sermonum autem libris vult intellegi, quasi apud praesentem se loqui, Epistulas vero quasi ad absentes missas”; ad Epist. 1.1.1: “Flacci Epistularum libri titulo tantum dissimiles a (libris) Sermonum sunt, nam et metri et verborum communis adsumptio eadem est.”

18  In trodu ct ion

veracity of such a state of affairs.79 This reading is made all the more enticing by the placement of the poem, fairly standard since Stephanus’s 1549 edition, as last among Horace’s works,80 as though the jewel in the poet’s crown, offering a retrospective upon all his ­earlier writings and finalizing the “architectonic completeness”81 of his vita. Unfortunately, however, t­ here is very l­ittle to go on in any attempt to date the Ars Poetica. The scraps of supposed evidence include the following: first, mention is made in the Ars Poetica of several perhaps con­temporary figures, most notably the Pisones, but including also Aulus Cascellius (371), Maecius (387), Messalla (371), Quintilius (438), Varius (55), and Virgil (55). Second, ­there is Horace’s odd description of himself, apparently at the moment of writing, as nil scribens ipse (“writing nothing myself,” 306), also problematized by the fact that it is internal to the poem. Third, Suetonius’s Vita Horatii reports that Augustus was apparently put out at not being a dialogic partner in any of Horace’s writings so far (post sermones vero quosdam lectos nullam sui mentionem habitam ita sit questus—­irasci me tibi scito quod non in plerisque eiusmodi scriptis mecum potissimum loquaris). And fourth, Porphyrio ventures an (early but not con­ temporary) identification of the Pisones. ­These have given rise to the following interpretations, arguments, and counterarguments. First, that the poem was written between 24 and 20 BCE, given the roles of Maecius (Tarpa), whom Nettleship supposes to be distinguished but not yet el­derly, and of Quintilius (Varus), whom he supposes to be dead;82 that it belongs to 28–27 if Quintilius’s death is placed in 28 rather than 24 and if the dates of Aulus Cascellius (born 110/104) and Messalla (Corvinus) are taken into account (at the “height of his power” in 27);83 and that it must be prior to 19 BCE, the date of Virgil’s death.84 Second, that nil scribens ipse means “not [currently] writing lyric,” in which case the Ars Poetica belongs ­either to the intervallum lyricum between the first three 79. Oliensis 1998: 5 similarly regards the AP as a “late masterwork of deferential authority” in terms of Horace’s persona, while clarifying that it “may not be Horace’s last work chronologically but is certainly the work of an established author” (16). 80. Brink 1963: 239. Frischer 1991: 6–16 suggests that Lambinus’s 1561 edition, which followed Stephanus’s in the placement of the AP as last, was responsible for cementing Stephanus’s apparently innovative arrangement (cf. Wilkins 1896: 330). 81. Reckford 1969: 11. 82. Nettleship 1883: 44. 83. Elmore 1935. 84. See especially Smith 1936, who finds a way around the majority opinion that Virgil (and Varius) must be alive at the time of the AP’s composition by suggesting that the poem was composed in two phases (23–22 and 13–8); he also disputes Elmore’s dating particularly in regard to Aulus Cascellius’s age. See also further Dilke 1958: 52–53 and Frischer 1991: 19–20.

be co m ing the a r s p oet ic a  19

books of Odes (published around 23 BCE) and the fourth (perhaps 13 BCE),85 or to the period ­after the publication of Odes 4.86 Third, that Suetonius’s term sermones refers to the Ars Poetica and Epistles 2.2, such that Epistles 2.1 (to Augustus, at last, in fulfilment of his wishes) postdates both of t­ hese.87 And fourth, that Porphyrio is right in asserting the Pisones to be Lucius Calpurnius Piso and his sons, in which case the poem dates to prob­ably 15 BCE (the year of this Piso’s consulship) or l­ater; or that Porphyrio is wrong, in which case, if a certain Gnaeus Calpurnius Piso is meant instead, the poem dates to the twenties BCE.88 On the basis of ­these conflicting and altogether unauthoritative pieces of evidence, which admit of vari­ous interpretations, “almost e­ very date from 28 B.C. to the poet’s death in 8 B.C. has been suggested.”89 To put this in its Horatian context, the Ars Poetica has been placed between Satires 2 and Odes 1–3 (28–27 BCE); between Odes 1–3 and Epistles 1 (23–20); between Epistles 1 and Epistles 2.2 (20–19); between Epistles 2.2 and the Carmen Saeculare (18); a­ fter the Carmen Saeculare and before Odes 4 (17–16); before Epistles 2.1 (15); and a­ fter Odes 4 (13–8).90 Many of t­ hese efforts at dating are further vitiated by their circularity: the arguments for an early date for the poem are bound up with ­those for the Gnaei Pisones as addressees, while t­ hose for a l­ ater date are bound up with the arguments for the Lucii Pisones.91 85. E.g., Brink 1963: 183, 204, n4, 239–43; dates as at Harrison 2007a: 347–48. 86. Orelli 1844: 769, on the other hand, reads the phrase to exclude the dramatic genres and epic but include satire and lyric, while Kiessling and Heinze 1914: 342 and many o­ thers since have taken it as ironically self-­deprecating (“ironisch: ‘da ich nichts, das der Rede wert wäre, zu schreiben vermag’ ”). Other suggestions as to the meaning of nil scribens ipse are ventured throughout the pre­sent study, and increasingly, as at Sedley 2014: 115, the feeling is that “it is not so easy to agree . . . ​that nil scribens ipse would be a natu­ral way for Horace to describe himself as writing one ­thing rather than another.” 87. To Fraenkel’s 1957: 383 “it is obvious that in this context sermones quosdam cannot refer to the epistles of the first book but only to the letters ad Pisones [sc. the Ars Poetica] and ad Florum [sc. Epist. 2.2],” Dilke 1958: 49–50 objects, “To some of us it is by no means obvious.” Dilke adds that since we have only Suetonius’s paraphrase of Augustus’s putative request, we cannot be sure that sermones was the emperor’s original lexical choice nor that his complaint to Horace was exactly as Suetonius relays it, and he rightly concludes, “Thus the Vita Horati tells us nothing about the date of the Ars Poetica.” 88. The identity of the Pisones is discussed in chapter 2. 89. Dilke 1958: 49 provides a thorough overview of t­ hese previous attempts, and the m ­ atter has since been revisited by, e.g., Duckworth 1965, Newman 1967: 75–81, d’Anna 1983, Frischer 1991, Brink 1963: 239–43. 90. See Frischer 1991: 17–18. 91. So, too, Syme 1980: 339. The same objection may be raised against Frischer’s novel 1991 hypothesis that the Pisones are yet a third group, namely, the preceding generation of Lucii Pisones.

20  In trodu ct ion

­ ere are, however, three additional pieces of evidence, external to the poem Th and not consisting of claims by ­later writers, that align to indicate a late date for the work, further supporting the recent consensus that the poem was published in or around 15 BCE and that it was Horace’s last work before his death in 8 BCE.92 The first of t­ hese involves metrical analy­sis, and this has taken several forms. The investigation of Duckworth 1965 found that the four patterns of dactylic and spondaic feet that occur most frequently in Horace’s hexameters “show a more or less steady decrease in frequency, with the Ars Poetica marking the end of the progression in each instance, an apparent indication of lateness of date” (86), such that the Ars Poetica comes a­ fter Satires 1, Satires 2, Epistles 1, Epistles 2.2, and Epistles 2.1 (in that order).93 The eight most frequent patterns display the same progression and yield the same results (87). Analy­sis of the relative proportions of dactyls (lower e­ arlier on, higher l­ ater) and spondees (higher ­earlier on, lower ­later) in the first four feet of the lines points to the same conclusion: the Ars Poetica was Horace’s final and culminating work in hexameters, and, moreover, closest in date and style to Epistles 2.1, the Letter to Augustus (87–88).94 The variety of metrical patterns in units of sixteen lines also 92. E.g., Russell 2006: 325 (“much to be said for a late” date), Harrison 2007a: 347–48 (­after 12 BCE), Laird 2007: 133 (“around 10 BCE”), Nisbet 2007: 20 (noting that ­there has been “much controversy” on the mea­sure prefers a late date with Lucius Piso as dedicatee: “When the consul of 15 BCE returned to Rome a­ fter crushing a major rebellion . . . ​a literary epistle would be an unpo­liti­cal tribute to his broad culture”), Leitch 2010: 120 (“perhaps as late as 10 B.C.E., although the date remains controversial”), Günther 2013: 48 (the AP’s date “cannot be determined with certainty, but it is prob­ably Horace’s last work”), Hardie 2014: 43 (“I ­shall assume, without arguing for it, a late date for the Ars”). 93. Of the six feet in a line of Latin hexameter verse, the sixth and final foot contains only two syllables, of which the second may be e­ ither long or short, and the fifth foot almost invariably scans as a dactyl. The first four feet, where metrical variability therefore exists, may each be dactylic (d) or spondaic (s). The four patterns of dactyls and spondees that are found most frequently in Horace’s Satires and Epistles (and somewhat less frequently in the AP) are identified by Duckworth 1965: 75 as dsss, sdss, dsds, and ddss. 94. Duckworth 1965 pre­sents eleven categories of metrical pattern in all, most of which support the theory that the AP must have been written ­later than Epistles 1 and 2; Duckworth is even prompted by some of the aberrant evidence to won­der, “Is it pos­si­ble that the strange discrepancy ­here between the Ars Poetica and the other hexameter works indicates that the poem was written so near Horace’s death that it was not completely revised?” (89). While Frischer 1991: 20–25 criticizes the results of Duckworth’s tests, employing a statistical approach that may seem foreboding, two further prob­lems should be noted (Frischer’s own methodology for dating the AP has already been critiqued by Keyser 1992, though defended by Clayman, Crane, and Guthrie 1993). First, Frischer’s criticism of Duckworth’s first two tests relies on a t­ able in Duckworth containing mathematical errors that neither Duckworth nor Frischer noticed (in the t­ able on

be co m ing t he a r s p oet ic a  21

shows the same. Duckworth’s metrical analy­sis was, however, not the first: Nettleship had already considered the m ­ atter, as had Michaelis, yet both discounted their findings b­ ecause t­ hese failed to support their starting premise that the Ars Poetica was written in or shortly before 20 BCE and, accordingly, before Epistles 1.95 Second, Newman 1967 sees the key Augustan term vates (“poet-­ priest,” as opposed to poeta, more plainly meaning “poet”) used in the Ars Poetica in a manner that suggests a fully worked-­out understanding of the term, as described especially at 391–407.96 And third, d’Anna 1983 argues for a date ­after 13 BCE from correspondences with the Carmen Saeculare (performed in 17 BCE), as well as from the position of the Ars Poetica in the manuscript tradition.97 That ­these three disparate forms of evidence—­none of which takes p. 86, 16 of the 24 percentages given have been shifted into the incorrect rows, i.e., the correct data have been arranged, for reasons irrecoverable, in descending order in each column, rendering them mostly worthless; on p. 87, the percentages given for the “second four” are simply incorrect, as are the resulting totals, provided that the raw data on p. 92 is correct). The second prob­lem lies in Frischer’s assertion that “Duckworth has failed to find a significant correlation between the poems and the distribution of metrical patterns” (23), based upon Frischer’s chi-­ square test of in­de­pen­dence performed on each of Duckworth’s sixteen patterns across all the hexameter collections and yielding the unsatisfactory p-­value of 0.270. If Duckworth’s data is considered in groups of four, however, as Duckworth himself does (group 1 being the four most frequently occurring patterns, e­ tc.; see my introduction, n93), the resulting p-­value of a chi-­square test is a respectable 0.0369, lending support to Duckworth’s original findings that Horatian hexameters exhibit significant differences in metrical patterning over time (with the AP generally as the culmination of the trends). Frischer also overemphasizes the degree to which Duckworth sees or needs to see Horace’s metrical patterning as deliberate (while Duckworth does speak of, e.g., Horace’s “desire for greater metrical variety,” 78, the data he provides consist of metrical patterns that largely could not result from deliberate design on the poet’s part). 95. Nettleship 1885: 171–72, Michaelis 1877: 428–29 (cf. Duckworth 1965: 85: “his figures actually f­ avor placing the Ars Poetica ­after the other two Epistles of Book II, but, since he dates the composition of the Ars in 20 B.C. . . . ​he disregards his own results”). 96. This insight aside, Newman 1967: 75–81, 127–30 dates the poem to the late 20s BCE on the grounds that the understanding of vates demonstrated in the poem indicates that the AP “must come in the ‘vatic’ period of Horace’s life,” i.e., close to Odes 1–3 and Epistles 2.1. Newman (79) considers a late date but rejects it for not altogether convincing reasons: lines 391–407 do in fact show that the vates concept has not been “demoted,” and it is entirely plausible that Horace, at a ­later point in life, retrospectively “veered round to restate e­ arlier doctrines” (something Newman says would have been done “suddenly, for no apparent reason”). 97. The AP’s position in the manuscript tradition and in the ancient scholia has been widely deployed in arguments for its dating, though this approach is naturally problematic. The Ξ tradition (manuscripts ABCK) places the AP immediately ­after the Carmen Saeculare (and ­after the Odes and Epodes, and before the Satires and/or Epistles / Epistles and/or Satires), while the Ψ group (comprised of Rδπλlφψ) puts it in second position overall, a­ fter the Odes, and before the

22  In t rodu ction

into account the vexed identity of the Pisones, not to mention the status as living or deceased of the other figures named—­all coincide to point ­toward a late date for the Ars Poetica, one that places it last in Horace’s life and writing ­career, is prob­ably as solid an argument as can be made. I therefore feel secure in treating the Ars Poetica ­here as Horace’s final poem and as therefore possessing by rights the retrospective qualities it so plainly evinces. What­ever its date, however, the arguments I pre­sent ­here for the Ars Poetica as a response or mirroring counterpart to the Satires (especially Satires 1, as the second book would not be published u­ ntil 30) would still hold true, but that response would be more immediate rather than issued from across the gulf of a lifetime of writing. The Ars Poetica also has a ­great deal of imagery and thought in common with the Odes, and again, ­whether this is the result of temporal proximity or a long-­held worldview expressing itself de­cades apart is of ­little relevance to my arguments. By way of a final thought on the m ­ atter, I would maintain that not only is the poem’s date not of the greatest relevance; rather, much like the poem’s vexed title or the sought-­after identity of the Pisones, it is the wrong question. Practicing poets may work on a piece over the course of several de­cades, such that the work is, in a fundamental and essential way, undatable—as well as unnamable. We may well won­der w ­ hether the same is true of the Ars Poetica, especially given Horace’s advice not to rush to publication (AP 388–89).

The Standing of the Poem—­A Story The standard view remains that the Ars Poetica was long regarded as straightforwardly didactic and as possessing sincerely intended pedagogical value and application.98 While Horace’s Satires and Odes ­were read for their Latinity, their grammatical and literary value,99 the Ars Poetica was, we are repeatedly informed, Epodes, Carmen Saeculare, and Epistles/Satires. While ­either placement would appear to lend weight to the hypothesis that the poem was written in the vicinity of the composition of the Odes (­whether ­after Odes 1–3 and before 4, or ­after Odes 4), it should be noted that neither of the two classes of manuscripts places the Satires and Epodes, known to be Horace’s earliest works, first. See further Vollmer 1907: 290, Reynolds 1983: 184, Frischer 1991: 6–7, Nisbet 2007: 20. 98. Cf. Friis-­Jensen 1995b: 360 (“anyone who desired to learn the rules of poetry-­writing would turn to Horace’s Art of Poetry, guided by a teacher or an up-­to-­date commentary”), 2007: 300 (“medieval interpreters of Horace’s Ars Poetica all shared the view that the poem is entirely didactic”). 99. On the comparatively large number of early medieval manuscripts containing Horace alongside Juvenal and Persius (common in collected volumes from the eleventh through fifteenth centuries), attesting to the evident popularity of satire, see Reynolds 1996: 13–14, Copeland 2016a, 2016b, 2016c. Horace’s Satires and Epistles are metrically and linguistically more straightforward than his lyr­ics (cf. Tarrant at Reynolds 1983: 182), and their moralizing content also lends itself well to an educational context.

be co m ing the a r s p oet ic a  23

treated solely as what its conventional title communicates: an ars, read alongside Cicero100 as a set of rules for “rhetorical theory and compositional teaching.”101 Continually pre­sent in early medieval schools, it was supposedly valued “not for its statements about art in any elevated sense, but for its advice on such basic issues as how to choose one’s material, how to achieve stylistic consistency, and how to maintain narrative continuity.”102 Yet this story, that the Ars Poetica was trea­sured during late antiquity, the ­Middle Ages, and even into the early modern period for its rules and instructions, used by monkish teachers and their earnest students as a handbook while they crafted their own compositions in verse and prose, is undermined by abundant evidence that few (if any) readers have ever found the poem easy to understand. Even as its practical value is stated and restated, ­there may be found occasional qualifying notes that the work required and requires “introductory texts and glosses” to render it intelligible and that Horace’s teachings are and have always been “elusive,” though such passing observations do not then proceed to influence the claims made about the poem and its history.103 What is often framed as a l­ ater realization that the rules Horace lays out are not terribly clear and his instructions not easily applicable in any direct fashion (they amount often to “just do it” or “try harder”) was, however, as the robust and lengthy commentary tradition attests, in fact always apparent to his educated readership. While ­later readers might congratulate themselves on casting aside this “handbook” of benighted inhabitants of centuries past, the straightforward application of the Ars Poetica to the mechanics of teaching writing by previous generations has been vastly overstated, ignoring as it does the fact that the poem was always hard in its Latin, always inscrutable in its meanings. The notion that Horace’s late antique and medieval readers ­were only interested in the poem’s stated contents, however, allowed their interest in the poem to be dismissed as self-­evidently silly, whereas the true fault may lie in the fact that it ­later ceased 100. The pseudo-­Ciceronian Rhetorica ad Herennium and several of Cicero’s oratorical works (especially De inventione) w ­ ere central to the instructional curriculum alongside the AP; Friis-­ Jensen 1990, 1995b. 101. Copeland 2016c: 59. 102. Copeland 2016c: 60; see also Bolgar 1954: 115–16, Copeland and Sluiter 2009: 546–48. 103. Quotations from Friis-­Jensen 1995b: 362 and Copeland and Sluiter 2009: 548, respectively. Hints at the difficulty identified ­here may be found in some of the secondary lit­er­a­ture: Friis-­Jensen 2008: 239–40, for example, says that “its readers actually considered it to be a didactic treatise on the art of writing poetry. But they also soon realised that they needed professional help to be able to extract its precepts,” and he notes, 1995b: 392–93, that the work was “insufficiently explicit in terms of prescriptive doctrine,” speaking of “Horace’s sophisticated, witty but very demanding advice” and his “bewildering universe.”

24  In t rodu ction

to be read holistically. The understanding that the Ars Poetica was valued for its didactic content rests upon a misconception of where this didactic content was thought to lie: although widely supposed to have resided in its precepts, the works analyzed in the epilogue indicate that the value was actually seen as residing in its form and, above all, in its marriage of form and precept. The distinction between pre­sen­ta­tion (or external qualities) and content (or internal qualities) was one commonly made since antiquity and throughout the ­Middle Ages: “Writing about the art is to teach precepts about it” (“de arte scribit qui de ipsa precepta tradit”), an anonymous twelfth-­century prologue to the Ars Poetica explains, but “working from the art is to follow the teacher’s precepts” (“ex arte tractat qui eius precepta imitatur”).104 Friis-­Jensen 1995b: 360 accordingly speaks of the “double message” that medieval students of the Ars Poetica would work to glean: “A close scrutiny of the literary perfection of a classical masterpiece was in itself rewarding, and in this case the classical poem was also a didactic text which, properly understood, would teach the rules of the art.” It is curious that so many readers of the Ars Poetica ­after the twelfth ­century and into the twenty-­first have maligned the poem as a failed manual on account of their own preference for reading it de arte rather than ex arte, as so many of their pre­de­ces­sors w ­ ere able to do. At the same time that it has been harmed by this, however, the Ars Poetica has also benefited from the misconception of its uses and purposes, for without the widespread sense that this poem was somehow separate from the rest of the canon of classical antiquity,105 it would likely not have come to possess its unusual standing, alongside Aristotle’s Poetics, as a “dif­fer­ent” sort of work,106 one with (as every­one supposedly agreed) a concrete applicability. The afterlife of the poem in late antiquity and the Dark Ages107 is difficult to reconstruct. It is clear that Horace in general became a school text very early, 104. See further Friis-­Jensen 1988: 137. Copeland and Sluiter 2009: 409 note that this idea appears also in Thierry of Chartres’s twelfth-­century commentary to Cic. Inv. rhet. 1.8; see with Fredborg 1988: 76. 105. As indeed it is, being “the only poetic text transmitted to us from antiquity that has the theory of lit­er­a­ture as its exclusive subject,” Ferenczi 2014a: 11. 106. Geue 2014: 144 also marvels at this treatment of the AP: “Any other poem written in the thick of the ‘Augustan Age’ would have accreted a thousand po­liti­cal readings round its core by now.” 107. Roughly 500–1000, or 487–800, if defined as the period when t­ here was no Holy Roman Emperor in the West. What­ever we wish to term this period, “nearly all classical Latin authors went through a period of hibernation between the mid-­sixth ­century, when the copying of classical texts slowed to a halt, and their rediscovery at some point during the ­Middle Ages” (Tarrant 2007: 285). On the decline of Classical lit­er­a­ture u­ ntil its revival beginning in the eighth ­century,

be co m ing t he a r s p oet ic a  25

within a generation or two of his death, as Juvenal 7.225–27 attests,108 and that his poems continued to form a significant component of the educational curriculum during the M ­ iddle Ages.109 His importance is shown by such evidence as the testimony of Aimeric, who in his Ars lectoria of 1086 lists him among the “gold” writers, alongside Terence, Virgil, Ovid, Sallust, Lucan, Statius, Juvenal, and Persius, or by Conrad’s Dialogus, where he is likewise placed in the category of maiores.110 The place of the Ars Poetica specifically, however, can be hard to discern, since most references name only Horace, rather than enumerating his individual works. Likewise, the existence in vari­ous locations around Europe of manuscripts containing the Ars Poetica indicates only that this poem was valued to more or less the same degree as the other works included in a given collection. The thirteen earliest extant manuscripts of Horace and t­ hose regarded as “indispensable”111 contain vari­ous arrangements of works from the Horatian corpus. Manuscripts A (ninth or tenth c­ entury) and V (date uncertain, now destroyed) do not include the Ars Poetica at all. K (eleventh ­century), contains the Ars Poetica but not all of the other works. Other manuscripts contain only portions of the Ars Poetica, such as B (ninth ­century), which gives lines 1–440. The majority, however, contain the complete works of Horace: R (ninth ­century, and perhaps the oldest manuscript), l (ninth ­century), δ (late ninth ­century), λ (ninth/tenth ­century), a (ninth/tenth ­century), π (ninth/ tenth ­century; with some pages lost from what was likely an other­wise complete collection of Horace’s works), φ (tenth ­century), ψ (tenth ­century), and C/E (eleventh ­century; again, some pages missing to account for the gap from Odes 4.7.21 to Epodes 1.23).112 Two additional early manuscripts known to have been produced in the British Isles before 1100, during the aetas Horatiana,113 see further Reynolds 1983: xiv–­x xvi, and on the Horatian manuscript tradition, 182–86 (contribution by Tarrant). 108. While Horace was a standard school author in late antiquity (see Tarrant 2007, Stadeler 2014: 51–53), nothing can be said for certain about the extent of the readership in that period for the AP specifically. 109. “Much read in medieval classrooms,” Copeland 2016a: 9; see also Vollmann 1996, Copeland 2016b: 23–28 (on the Liber catonianus as fundamental to elementary teaching). 110. Huygens 1970: 111–14; see Bolgar 1954: 423, Copeland 2016b: 24–25. 111. Reynolds 1983: 183. 112. The number of manuscripts containing at least one poem by Horace has been placed by Villa 1992, 1993, 1994 at 815 (cf. Friis-­Jensen 1995a: 229–30), and her exhaustive cata­log shows much the same patterns persisting into l­ater centuries as t­ hose discernible in t­ hese earliest thirteen manuscripts. 113. As the tenth and eleventh centuries have been dubbed for their preponderance of Horace manuscripts; the aetas Virgiliana occupies the eighth and ninth, the aetas Ovidiana the twelfth and thirteenth (Friis-­Jensen 2007: 293, Ziolkowski 2016: 172).

26  In trodu ct ion

are also extant: one contains all of Horace’s works except the Epistles; the other, all the works.114 What, if anything, may be concluded from this smattering of evidence is that the Ars Poetica was through the eleventh ­century considered impor­tant but perhaps not the most impor­tant work by Horace—­that honor would go to the Odes, which are omnipresent in the manuscript tradition except where very small portions have evidently fallen out.115 Although many of the mechanisms by which its ascendance happened are now irrecoverable,116 by the twelfth c­ entury the Ars Poetica had succeeded in attaining its standing as the established and paradigmatic Ars Poetica: a central text used for teaching at all levels and appreciated as a seminal yet somehow dif­fer­ent work from classical antiquity, and one that also inspired imitations, as demonstrated by the numerous extant artes poetriae discussed in the epilogue. The popularity of the Ars Poetica in this period may owe something to the fact that it was a unique document, Aristotle’s Poetics being still relatively unknown.117 Although evidence is somewhat scanty for the ­later thirteenth and ­fourteenth centuries in comparison to the preceding and subsequent ones, the Ars Poetica was clearly known and being read: Dante alludes to it twice in the Divine Comedy,118 and Petrarch owned a copy of the poem and refers to Horace’s Epistles throughout his own Horatian metrical epistles as well as in his Letter to

114. Gneuss and Lapidge 2014: 157 (entry 179.5), 516 (entry 681.5); cf. also 199 (entry 252). ­These and more are noted by Copeland 2016b. 115. Much has been written on the readership of Horace’s Odes; see, e.g., Friis-­Jensen 1988, 1993: 258, Griffin 1996, Most 1996, Harrison 2007d, Money 2007, Talbot 2009, Sowerby 2012, Moul 2015: 540. 116. Cf. Friis-­Jensen 1995b: 360. Russell 2006: 340 summarizes that “from the early ­Middle Ages, Horace was a curriculum author, and no part of him was more studied than the Ars.” 117. It was long assumed that Horace and other Romans of his period did not have direct access to the texts of Aristotle (as recently as Reinhardt 2013: 505), but Tarán and Gutas 2012 make the following impor­tant points: “Aristotle’s works ­were available during the Hellenistic age” (28); “beginning in the second half of the first c­ entury BCE ­there was a revival of interest in Aristotle’s scholarly works” (though “we have no evidence of any par­tic­u­lar interest in the Poetics,” 31); the archetype of the Poetics dates to somewhere in the period from the second c­ entury CE to the end of the sixth or first half of the seventh (32–37). This evidence shows that it should not be considered impossible for Horace to have known Aristotle’s Poetics directly, although in the ­Middle Ages it would be known primarily in the 1278 Latin translation of William of Moerbeke (Friis-­Jensen 1995b: 362, Tarán and Gutas 2012: 36, 39–40). 118. AP 394 at Inferno 32.10–11 and AP 60–62 at Paradiso 26.137–38; Dante also cites the AP in his Epístola a Can Grande della Scala (cf. Frischer 1996: 72, Mañas Nuñez 2012: 226), itself an epistle, though other­wise for Dante he is “Orazio satirico” (Griffin 1996: 182; cf. Mañas Nuñez 2012: 228–29)—­ yet “Orazio satirico” should not be thought incompatible with the Horace of the AP.

be co m ing t he a r s p oet ic a  27

Horace.119 Despite its evident continuing popularity,120 the Ars Poetica remained dense and confusing enough—­yet also impor­tant and worthwhile enough—to its many interested readers that, extending the tradition of the tenth, eleventh, and twelfth centuries,121 thoroughly annotated editions began to be produced and published in print in the fifteenth ­century—­almost as soon as the printing press was introduced—­with the editio princeps appearing in or around 1470.122 In the period 1400–1700, numerous commentaries of fundamental importance w ­ ere written, at first in neo-­Latin, on the Ars Poetica as on Horace’s other works and on classical and Biblical lit­er­a­ture more generally.123 A key work, in that it was the “first modern humanist commentary” to be printed on Horace, was by Cristoforo Landino, published in 1482.124 The edition, which contained the full text of Horace with Landino’s notes, was both an “instant success” and “as far as we know t­ oday . . . ​the only humanist commentary accessible in a printed edition ­until Badius published his commentary in Paris in 1500, that is, eigh­teen years ­after Landino was printed for the first time.”125 Landino’s commentary continued to be reprinted into the sixteenth c­ entury, often 119. See Friis-­Jensen 1997b, and further McGann 2007: 307–10 on Horace and Petrarch. 120. Friis-­Jensen 1995a: 229–30, working with the manuscript lists of Villa 1992, 1993, 1994, calculates that 233 fifteenth-­century manuscripts contained the AP, a figure that indicates the work was “far from being neglected in the fifteenth ­century.” In the fifteenth ­century 404 manuscripts containing any/all works of Horace ­were produced in total, i.e., slightly fewer than half as many as exist containing Virgil (around 1,000), who “is likely to have been the most widely read author.” 121. Though “editions or even transcriptions of commentaries from the thirteenth to the fifteenth ­century are lacking,” Friis-­Jensen 1995a: 231–32 is right to see that “the transition from medieval scholarship to Re­nais­sance humanism prob­ably took the form of development rather than break,” and he notes (2008) that medieval commentaries w ­ ere so thorough and useful they continued to be used by l­ater scholars. On the fifteenth-­century commentary tradition, see further Weinberg 1961. 122. Hardison and Golden 1995: 3. See also Friis-­Jensen 1995a: 230 on t­ hese early printed editions; he puts the number of them at sixty-­nine (with another six that contain at least one ode), of which forty-­four contain the Ars Poetica, concluding, “Again t­ here is no evidence to show that this work suffered any neglect in the fifteenth ­century.” 123. See further Enenkel and Nellen 2013 on this trend. Weinberg 1961: 179 notes as extraordinary in this period the 1566 commentary in Italian (not Latin) by Giovanni Fabrini. 124. See Friis-­Jensen 1995a: 234–35. On Landino’s life and work, see further Weinberg 1961: 79–81, Pieper 2013, Stadeler 2014. 125. Friis-­Jensen 1995a: 235, 236 (he is evidently not counting Antonio Mancinelli’s commentary of 1492, which seems to have been printed only in a collection, rather than on its own); on Badius’s commentary, see further also Weinberg 1961: 81–85, Hardison and Golden 1995: 160–63, Mañas Nuñez 2012: 231. Friis-­Jensen also sees as impor­tant in this period a commentary by a pupil of Guarino, as he identifies him (1995a: 239, 2008).

28  In t rodu ct ion

accompanied by the scholia of Porphyrio and Pseudo-­Acro and a commentary by Landino’s fellow humanist scholar Antonio Mancinelli, as Horatius cum quattuor commentariis (first published 1492).126 The result of this edition was, according to Weinberg, to make of the Ars Poetica “something very special for Quinquecento readers,” for “Horace’s work was no longer a theory of poetry, but the theory of poetry, the summum of all useful ideas about the art,” and the ever-­proliferating accretions of commentary upon the poem made such volumes “a repository for every­thing that was being thought about poetry in the humanistic period and during the Re­nais­sance up to about 1545.”127 Friis-­ Jensen 1995a128 argues from this evidence that Horace’s Ars Poetica was far more widely known and read, and considered far more impor­tant in the fifteenth ­century, than has previously been acknowledged. This is evident also in the “material growth” of its tradition, where, as “more t­ hings [are] added to the Ars Poetica” such as commentaries and essays, so it was itself “constantly applied to more ­things.”129 The sixteenth ­century saw a conflation of the Horatian and Aristotelian literary critical traditions. For the first time Horace’s poem did not stand alone: Aristotle’s Poetics, certainly neglected if not exactly lost, enjoyed a rebirth beginning in 1508 with the production of a Greek text by Aldus, the editio princeps. ­There are four primary witnesses to Aristotle’s text, two of which preserve the Poetics in Greek: Parisinus Graecus 1741 (A), written around the ­middle of the tenth ­century or its second half, and Riccardianus 46 (B), prob­ ably mid-­twelfth ­century.130 Tarán and Gutas attribute the revival of interest in the Poetics to the appearance of t­ hese two manuscripts in Italy during the fifteenth ­century, ­after which they ­were copied numerous times, culminating in the Aldine edition. They arrived to a climate of “lively interest in literary criticism and theory” for which the Ars Poetica formed much of the existing basis. The Poetics was quickly taken up “in the same light . . . ​as a welcome 126. Pieper 2013: 221–22; on the role of Porphyrio and Pseudo-­Acro in inspiring and shaping the ­later exegetical tradition, see Weinberg 1961: 72–85, Friis-­Jensen 1995a: 239, 1997a, 2008, Enenkel and Nellen 2013, Pieper 2013: 226. 127. Weinberg 1961: 85, 110. 128. Cf. also Pieper 2013. ­ ere is also “methodological growth,” whereby the nature and concerns of the com129. Th mentary itself begin to shift; on some key stages in this pro­cess see Weinberg 1961: 85–110, Hardison and Golden 1995: 237–40, Wilson 2009, Golden 2010: 392–93. 130. Tarán and Gutas 2012: 36; the other two manuscripts are in translation, one into Latin (see my introduction, n117 and further Friis-­Jensen 1995b: 362, Tarán and Gutas 2012: 39–40), the other into Arabic (by Abū Bišr, before 934; on the Syro-­Arabic tradition of Aristotle’s works, see Copeland and Sluiter 2009, Tarán and Gutas 2012: 77–128). On the Aldine edition, see Tarán and Gutas 2012: 38–48.

125-81019_FerrisHill_HoracesArsPoetica_5P.indd 28

8/8/19 8:40 AM

be co m ing t he a r s p oet ic a  29

supplement and complement”131 to Horace’s work, and the two traditions became “fused and confused,”132 as evidenced in Francesco Filippi Pedemonte’s 1546 commentary on the Ars Poetica: Pedemonte proceeded by citing passages from the Poetics to match sections of the Ars Poetica, the first of numerous efforts to do so. So, “Horace ceased to be Horace and Aristotle never became Aristotle; each grew, instead, into a vast monument containing all the multiform remains of the literary past.”133 Tarán and Gutas seem to view the tradition that existed around the Ars Poetica as regrettably responsible for the subsuming of Aristotle’s Poetics into the same, with the result that the latter did not come to be properly understood and appreciated as a work altogether distinct in its intent, concerns, and circumstances of composition.134 It is equally the case, however, that the arrival of the Poetics may be seen as consigning the Ars Poetica to the bookshelf of reference works on literary composition,135 for this poem, which had been regarded and valued as an exemplary work of art in its own right throughout the M ­ iddle Ages, began to be put to use as one of what was now a pair, for its contents and prescriptions (not its form). Yet even in an age where the Ars Poetica was “an object of research,” Boileau and Pope would nevertheless approach it as “a living document,”136 as their medieval pre­de­ces­sors had done, crafting poems that, 131. Tarán and Gutas 2012: 39–40. 132. Bogue 1975: 5. On Pedemonte’s commentary, see Weinberg 1961: 111–18; Pedemonte’s method of comparing Horace and Aristotle was quickly taken up by numerous other scholars (see Weinberg 1961: 118–35, 156–62). García Berrio 1978 notes also the reverse: that the AP entered the earliest, sixteenth-­century commentaries on Aristotle’s Poetics to a notable degree. 133. Weinberg 1961: 47. Weinberg’s two-­volume work, A History of Literary Criticism in the Italian Re­nais­sance, remains monumental on the Horatian tradition, the Aristotelian tradition, and the two traditions together; Herrick 1946 also remains useful on the latter. Weinberg rightly points out that prior to 1545, it was not that the Poetics was wholly unknown to scholars working on the AP; rather, “such knowledge did not fundamentally affect their general reading and interpretation of the Horatian text” (111). With the re-­arrival of Aristotle, however, a school of thought grew up, propounded by some of the scholars working in Pedemonte’s footsteps, including Francesco Robortello, Vincenzo Maggi, and Jason De Nores, as well as Scaliger, that Horace’s AP could not be properly understood in the absence of the Poetics and Hellenistic literary theory (Mañas Nuñez 2012: 242). 134. Cf. Weinberg 1961: 154–55: “­there was no slightest intimation” that the Poetics and the AP “address themselves to essentially dif­fer­ent prob­lems, that they use widely dif­fer­ent methods, and that they produce statements of a completely dif­fer­ent nature about poetry.” 135. Bogue 1975: 6 similarly suggests that “Re­nais­sance criticism” (that is, the period when Aristotle’s Poetics had been rediscovered, and Horace’s AP was being increasingly assimilated with it) “intensified the schism between form and content in the Ars.” 136. Hardison and Golden 1995: 173.

30  In t rodu ction

Horatian in spirit and in much of their detail,137 rendered the archetype apposite for a distinct time and place.

The Unity of the Poem Despite its incomparable importance, especially during the ­Middle Ages and the early modern period, the Ars Poetica has been unduly harmed by its longstanding treatment as a set of puzzles to be solved: What was its title? Its date? Its genre? Who are the Pisones? What is the poem about? And how do its disparate sections relate to one another and cohere? While I venture some solutions to t­ hese prob­lems in this introduction and in the chapters that follow, I espouse above all approaching the poem on its own terms. Since it is a work of literary art, we should be ready to see, hear, and feel the sights, sounds, and sensations it offers.138 Michèle Lowrie framed her monograph, Horace’s Narrative Odes, as an effort to answer “the central bothersome question of Horatian lyric: what keeps ­these poems together when they try so hard to drive themselves apart?”139 As Freudenburg 1999 summarizes, Lowrie identifies “not architectural ‘blocks’ neatly stacked, but interpenetrations of imagery, sense, and technique that carry us forward in our sense-­making work, binding the smattering of parts into a w ­ hole,” where any unity is “gathered up, asserted, and continually revised.” This study attempts to answer the same question as it pertains to the Ars Poetica140 and extends Lowrie’s approach to another Horatian work. Just as readers of the Odes discovered “triads, pentads, enneads, and decads, with the poems in each group and the groups themselves disposed in an abstract pattern such as concentric framing or chiasmus,”141 the Ars Poetica has long been seen as falling into two sections, 137. Of the 1,100 lines of Boileau’s poem, for example, over 100 may be identified as coming from the AP, and another 100 from other Horatian works; Golden 2010: 396. 138. Consider the quotation from Susan Sontag, “Against Interpretation,” given at Porter 1995: 97: “What is impor­tant is to recover our senses. We must learn to see more, to hear more, to feel more. Our task is not to find the maximum amount of content in a work of art, much less to squeeze more content out of the work than is already ­there. Our task is to cut back content so that we can see the ­thing at all.” 139. Lowrie 1997: 3. 140. Edgar Allan Poe’s remark, cited by Santirocco 1986: 175, that “what we term a long poem is, in fact, merely a succession of short ones” is illuminating in this context. 141. Santirocco 1986: 3; noting the exhaustive investigation since the second half of the nineteenth c­ entury into the m ­ atter, he adds that “by the turn of the c­ entury an investigator could justly complain that not a single possibility had been left untried.” Harrison 2007c: 272 identifies Collinge 1961, Commager 1962, and West 1967 as foremost among t­ hese efforts.

be co m ing t he a r s p oet ic a  31

the first on the topic of ars (1–294), the second on the artifex (295–476),142 with a further subdivision of the first half to produce a tripartite structure for which Horace has been supposed indebted to Neoptolemus of Parium: 1–152 on poiēsis (or poiēma), 153–294 on poiēma (or poiēsis), and 295–476 on poiētēs.143 Most jarringly, a sequence of editors from the late sixteenth through the nineteenth centuries took it upon themselves to hack up the poem and rearrange its pieces—as if enacting Horace’s own suggestion at Satires 1.4.56–63 that by this method a fundamental difference between verses by Ennius and ­those of Roman Satire could be made apparent144—to produce the treatise that they ­were quite sure it was meant to be in what they no doubt considered a superior and clearer form.145 In the second half of the twentieth ­century, approaches to the internal bonds of the Odes began to move away from numerology, yet the same generosity has not been seriously accorded to the Ars Poetica, and it continues to be the case that “commentators and translators of the Ars Poetica divide up the texts of their 142. First proposed by Norden 1905; see further Gantar 1954, 1964 (who sees the AP as having a symmetry governed by the golden ratio, since 294/182 = 1.615), Greenberg 1961: 264–66, Brink 1963: 3–14, 1971: 325–29, Hardison and Golden 1995: 33–40, Laird 2007: 135, Golden 2010: 393. The vari­ous structural divisions that have been proposed for the poem are overviewed by Bogue 1975: 16–18 and Sbordone 1981, and as Friis-­Jensen 1995b: 364–67 points out, the desire to discern the structures under­lying the AP was already a major critical approach to the poem in the M ­ iddle Ages. 143. This tripartite division was taken up by Jensen 1923 and Immisch 1932, and the transposition of the sections on poiēsis and poiēma proposed by Brink 1963; see further Tate 1928, Wimsatt 1970: 139–40 (in Brophy et al. 1970), Bogue 1975: 16–18, 30–78, Armstrong 1993: 185 (who rightly warns, 206, that “the ‘thirds’ of the poem ‘bleed’ into each other, prophesy or recall each other’s main thought, rather than each boxing up neatly into themselves what belongs to their par­tic­u­lar area”; compare Brink 1971: 457: “no division within a Horatian poem is fully self-­contained”). 144. The point is that if the words of Ennius w ­ ere reordered, they would still be recognizable as poetry, whereas ­those of Lucilius or of Horace himself would not be. Horace’s idea may be traced to Philodemus, who terms the pro­cess metathesis (see Janko 2000: 225 with n3, Armstrong 1995, Oberhelman and Armstrong 1995, Tsakiropoulou-­Summers 1995: 127–40) and “is militant that . . . ​the thought or the subject in dif­fer­ent words is simply not the same” (Armstrong 1995: 222), a view espoused also at Lucr. DRN 2.1004–22 (Armstrong 1995: 227). Armstrong 1995: 231 says in reference to the Odes that “clearly Horace thought of his poems as mosaics of unrearrangeable words and letters building up complex thought pictures,” and Oberhelman and Armstrong 1995 argue convincingly that he thought the same of his hexameter poetry, too. 145. On ­these efforts by Antonio Riccoboni (1591, 1599), Daniel Heinsius, Jean Bouhier, Peerlkamp, and o­ thers, see Viola 1901: 101–66, 1906: 74, Weinberg 1961: 231–42, Sbordone 1981: 1867, Segura Ramos 1989: 120, Hardison and Golden 1995: 165, 168 (who notes that Heinsius blamed the disorder on “the negligence and stupidity of the medieval scribes”), Frischer 1996 (who focuses on the quarrel between Riccoboni and Nicolò Cologno), Caballero 2007: 79, Ferenczi 2014a.

32  In t rodu ction

editions or renderings into constituent units”146—­a pre­sen­ta­tion avoided in the preface to this study. W. R. Johnson in The Idea of Lyric (1982), noting Horace’s “technical virtuosity” (123) and his “spectrum of lyric voices,” carefully selected and arranged (134), describes how “by mingling styles and shifting moods, by employing frequent ellipses and understatement, by working his images into patterns of dialectical modulations,147 by sometimes closing his poems with a vivid and haunting image that . . . ​suggests continuities and reverberations outside its limits,” Horace pre­sents us “not so much with a finished artifact, an artistic object, as with pictures that speak and pictures that move” (127–28). Santirocco 1986 sees connections among the poems taking the forms of “meter, theme, and verbal reminiscence” (171), “tone or atmosphere,” “imagery,” “structural similarity,” and “literary model.” In Polyhymnia, militating against “the grave charge of impulsive meandering that has been intermittently laid at Horace’s door over the centuries” (1991: 10), Gregson Davis aims to demonstrate how “Horatian lyric discourse typically ‘argues’ a coherent nexus of ideas through nuanced variations in form and pre­sen­ta­tion” using “building-­blocks . . . ​of motifs, topoi, recurrent meta­phors, and rhetorical conventions that, for the most part, are set forth paratactically” (3).148 I see the Ars Poetica, which Armstrong 1993: 185 has described as the Horatian work that “most baffles any effort to give a coherent account and argument of its structure,” instantiating on a smaller and more compressed scale, as well as in the course of a single (book-­length) poem,149 the puzzle that has preoccupied readers of the Odes. Accordingly, I adopt the strategies applied to the Odes by Johnson, Santirocco, G. Davis, Lowrie, and o­ thers, and venture an approach that pays less regard to the Ars Poetica’s explicit thematic breaks and topical transitions, which are perhaps most visually evident in the vari­ous paragraphing schemes that continue to be proposed,150 146. Laird 2007: 135; the result is that “dif­fer­ent systems of thematic division point to dif­ fer­ent ideas about the ‘overarching’ or ‘under­lying’ structure of the poem.” 147. H ­ ere Johnson 1982: 206, n28 instructs the reader to see further Campbell 1924: 5–18, 224–28. 148. On patterning within individual odes, especially as this relates to overall repeated structures, see, e.g., Syndikus 1995, Tarrant 1995, Nisbet 1999, Harrison 2004. Zetzel 1980 has examined Horace’s Satires in a similar light; see also Van Rooy 1968, and on Horace’s poetic textures more generally, Williams 1980: 181 (“metonymic and synecdochic connexions of ideas are fairly frequent . . . ​even in his Satires and Epistles”) and Harrison 2007c. 149. The length of the AP is comparable to that of a single book of Georgics. 150. See, e.g., Brink 1971: 468–517 (where he provides the skeleton of and summarizes his own structural divisions), Russell and Winterbottom 1972: 279–91 (reprinted and endorsed at Laird 2007: 136, with the caveat that “a text’s structure is never an intrinsic property waiting to be discovered, but something that readers attach to it themselves”), Bogue 1975: 30–78 and passim

be co m ing the a r s p oet ic a  33

and more to a series of reminiscences that run throughout it, giving rise to an associative and organic form of internal coherence.151 In crafting a work whose didactic value and poetic worth was of this covert type, Horace was, ­whether deliberately and directly or not, adhering to the artistic views espoused by Philodemus: “What makes a poem good is not the sound, but the combination of words and sense. Although the sound appeals to the ear, the most impor­tant plea­sure we obtain from poetry is in the mind, and such is therefore the function of verse.”152 Consideration of the Ars Poetica qua poem and, most particularly, qua Horatian poem, might have been championed sooner if Philodemus’s On Poems had not been lost, to be recovered beginning in the eigh­teenth ­century only as charred fragments from the Villa dei Papiri at Herculaneum, where it had been preserved by remarkable accident during the eruption of Mount Vesuvius in 79 CE.153 The villa’s library contained between 800 and 1,100 books,154 among them a g­ reat many works by (he compares Boileau’s and Pope’s poems with Horace’s Ars Poetica precisely in terms of what he perceives to be structural similarities), Hering 1979: 78–85, Armstrong 1993: 185–90 (he reviews vari­ous of the divisions that have been proposed), Ruiz Castellanos 1994 (who speaks of the poem’s “micro-­textos,” which he identifies in thematic terms as being comprised of propositions and their proofs or refutations), Hardison and Golden 1995: 42–88, Frischer 1996 (also a review), Hösle 2009 (seeking to restore unity to the poem, he does so in thematic terms). 151. Brink 1963: 248 is right to ask why, if other g­ reat Augustan poems have been appreciated as revealing a “multiplicity of patterns,” we should not also expect the same of the Ars Poetica; in his view (1971: 446) the poem “is so or­ga­nized as to keep the reader’s mind and emotions on the move continually”; and he also (450–62) makes an “attempt to illustrate from the Odes some of the structural devices . . . ​employed in H.’s Ars and his other long hexameter poems.” Wimsatt 1970: 136 (in Brophy et al. 1970) sees the poem’s parts “bound together and smoothed into one another by multiple simultaneous associations (sometimes called, in the scholarship, ‘gliding transitions’)” that may be “highly ambivalent,” linking “to their immediate neighbors with greater and lesser degrees of intensity or necessity,” with “sometimes local connections” and sometimes standing “in a kind of parallel co-­ordinate regional order.” Armstrong 1993: 222 likewise sees that “large ele­ments involve and dictate the smaller ones in a poem and are reciprocally built up from them in a relationship of interdependence, in the poetics Horace was taught. So, style and subject, style and thought, are not two dif­fer­ent questions.” 152. Janko 2000: 8. Following on the work of Asmis 1995a and 1995b, Obbink 1995b, Sider 1995a and 1995b, Tsakiropoulou-­Summers 1995, and Wigodsky 1995, Janko elucidates how, although “Epicurus has been generally thought to have rejected poetry outright” (9), for Philodemus Epicureanism and the composition of poetry ­were not incompatible. 153. The events that resulted in the preservation of the papyri are described at Sider 2005: 14–15. 154. Janko 2000: 4; for the holdings, see Gigante 1979, 1995: 15–48, Janko 2000: 7–8, Houston 2014: 280–86.

34  In t rodu ction

Philodemus,155 as well as writings by numerous other thinkers in ancient philosophy, predominantly of the Epicurean school. Although the first papyri of On Poems ­were uncovered in 1752–1754 and efforts to unroll them began just a few de­cades ­later, the recent editions by Richard Janko156 have contributed significant improvements to the Greek text and have also rendered Philodemus’s work more accessible to En­glish readers. As a result, it has become clear how formative was Philodemus’s thought for the Augustan poets. Although Armstrong (2004: 267–68) cautions against plugging Philodemus in to ­every remaining gap in our knowledge, he explains: I have always wished that a doctrine of poetics that corresponds more nearly to what I understand to be the compositional practice of a­ ctual ancient poets . . . ​could be extracted from the remains of antiquity. If my argument is right and not mere wish-­fulfillment, the Herculaneum papyri have given us this doctrine. (Armstrong 1995: 231–32) The speculative part of Armstrong’s “wish” lies in the fact that ­there is no evidence that Philodemus and Horace ­were personally acquainted.157 Nevertheless, Horace names and apparently quotes him at Satires 1.2.121,158 and crucially, several members of the Maecenean literary world are attested as having a personal connection with Philodemus, since at least three of the books in

155. Recent de­cades have seen a wealth of publications on Philodemus, both in terms of editions of his writings (e.g., Indelli and Tsouna-­McKirahan 1995, Sider 1997, Konstan et al. 1998, Janko 2000 and 2010, Henry 2009, Tsouna 2012, and numerous papyri published in Cronache Ercolanesi) and scholarly essays on ­these works (e.g., Greenberg 1955 and Grube 1965: 193–206 among the ­earlier treatments, then a second wave beginning with Gigante 1990 (trans. 1995), Armstrong 1993, Tsouna 2007, and the essays in Obbink 1995a, in Fitzgerald 1996, and in Armstrong et al. 2004). 156. He published book 1 in 2000 and books 3–4 in 2011; book 2 (discussed slightly in his 2000 volume) and 5 are still being edited. Janko explains in detail in his 1995 article and in the introductions to his 2000 and 2011 editions both the original efforts and his own approaches to unrolling and reading the papyri. 157. White 1993: 234; for Janko 2000: 6, “it is unclear . . . ​­whether Horace knew Philodemus. It is pos­si­ble that the phi­los­o­pher died before Horace joined Vergil’s group . . . ​but the timing is close, and Horace may well have met him.” 158. Cf. Hendrickson 1918, Sider 1997: 200, Gowers 2012: 114–15. Tsakiropoulou-­Summers 1998: 23–24 (cf. 1995: 18–41), noting that Horace “rarely misses the opportunity to stress his ties with Vergil and his circle of friends, which he used to assert his social status as member of the highest Roman elite,” says of Sat. 1.2.121 that “what­ever e­ lse this quotation may show, it cannot document the alleged student/teacher relationship between Horace and Philodemus”; cf. my introduction, n157.

be co m ing t he a r s p oet ic a  35

On Vices and Their Opposing Virtues159 ­were dedicated to members of this group: On Flattery names Varius and Quintilius (Varus); On Greed, Plotius (Tucca),160 Varius, and Quintilius; and On Slander, all three alongside Virgil (ὦ Πλώτιε καὶ Οὐάριε καὶ Οὐεργίλιε καὶ Κοιντίλιε).161 Horace may have imbibed from Philodemus, as Virgil did, what the phi­los­ o­pher understood to be the essence of poetry, that “style and content are reciprocals, ἴδια, of each other,”162 or perhaps this approach merely seemed self-­ evidently right to him. Fragmentary as the work’s five books are, the prob­lem tackled in On Poems clearly emerges as “what is poetry?”163 The shards of answers that may be collected show that “Philodemus was far more concerned with the relationship between form and content and other literary ­matters than he was with content alone.”164 The plea­sure derived from poetry resides in “the appreciation of arrangement and unity built up (as in the universe) from the tiniest ele­ments into a coherent whole”—­something that both Virgil and Horace succeeded in ­doing, both poets developing the “uncanny ability to arrange in indissolubly perfect order the letters, sounds, words and verses of their poems into a greater ­whole.”165 This is what Horace achieved in his Ars Poetica, and this is what has been overlooked in the work, for even as generation a­ fter generation of readers perceived it, the critique of the poem as disjointed and rambling has persisted. Assessed according to the standard of its contents, the poem fails as an ars and as didactic, as epistle and as satire (sermo). But if assessed according to its poetic form, it succeeds—as a new and profoundly Horatian form of hexameter composition that admits of all t­ hese classifications while adhering to no individual one.

159. Περὶ κακιῶν καὶ τῶν ἀντικειμένων ἀρετῶν is thought to have contained the books On Arrogance, On Property Management, On Greed, On Slander, and prob­ably three books On Flattery; see further Gargiulo 1981, Glad 1996: 22–24, Konstan et al. 1998: 2, Tsouna 2012. 160. It had been thought (Körte 1890; cf. Dilke 1958: 50, Tsakiropoulou-­Summers 1998: 21), when only the letters τιε ­were readable on PHerc. 253, that Horace (Horatie) himself rather than Plotius might have been meant, but see now Gigante and Capasso 1989 and Gigante 2004. 161. On what sort of relationship such dedications might imply, see Armstrong 1995: 224, Sider 1997: 19–21, Delattre 2004: 245, Blank 2014; on the absence of Horace’s name, Armstrong 2004: 268 (prob­ably “more to do with Horace’s wealth and social status . . . ​than with any lack on Horace’s part of experience of Philodemus and his teaching”), Gigante 2004: 86 (a sign of his “definitive exclusion”; so, too, Tsakiropoulou-­Summers 1998). 162. Armstrong 1995: 219. 163. See Janko 2011: 57, 223–24; Greenberg 1955 and Grube 1965 remain helpful on Philodemus’s theories of poetry. 164. Sider 1995b: 45–46; see also Porter 1995, Tsakiropoulou-­Summers 1995: 55–166. 165. Armstrong 1995: 223, 232.

1 Humano

Humano—­the first word of the Ars Poetica, its three long syllables forcing us to utter it slowly and portentously, as if beginning an incantation—­promises a concern with the h­ uman. Next comes capiti, such that it quickly transpires we are dealing with a ­human head, capiti mirroring humano in size and syllabification yet undoing its sonorousness with the arrival of harsh punctuating sounds.1 To this ­human head is attached a h­ orse’s neck in chiastic arrangement; then limbs sourced from all over and decorated with multicolored feathers; and fi­nally a black fish tail, ­after which Horace circles back to the head and deigns to tell us that it belongs to a beautiful w ­ oman.2 Within five lines we have thus found ourselves informed that humano signifies the head of a ­woman atop a hybrid creature that combines all four branches of the ancient animal kingdom,3 the appropriate response to which we are further informed is laughter. The apparition with which the poem begins may well go back to

1. Humano consists of three low vowels and soft consonants (one aspirate, two nasals), each able to be sustained for a length of time, while capiti, also three syllables and six letters long, half consonants, half vowels, consists of consonants that are all stops (velar, labial, dental) and vowels of which two of three are high in pitch. Horace was likely familiar with the theories of language attributed to the euphonists and attacked in Philodemus’s On Poems 1: ­these thinkers held that ­there is a “hierarchy of sounds” in which “vowels are best, followed by continuants, with stops the least euphonic” ( Janko 2000: 177; see further Porter 1995: 136–37, Tsakiropoulou-­ Summers 1995, esp. 167–93). Pausimachus of Miletus is even reported to have said that “the collocation of mu with alpha does something out of place . . . ​and if nu is added, even worse” (τὴν τοῦ μῦ πρὸς τὸ ἄλφα συναφὴν ἄτοπόν τι ποιεῖν . . . ​ἂν δὲ τὸ νῦ προσλάβῃ, καὶ χεῖρον, col. 91.3–10), i.e., precisely the combination of sounds/letters to be found in Humano. Thus Horace is perhaps emphasizing the failed nature of the work of art with which the AP opens by humorously employing euphonist prescriptions on language critiqued by Philodemus. 2. Humano capiti cervicem . . . ​equinam; Orelli 1844: 698, Rudd 1989: 150. 3. “Man, quadruped, bird, and fish” (Rudd 1989: 150).

39

40  ch a p te r 1

Plato,4 and evoke Virgil’s Scylla5 and Triton6 and Lucretius’s prohibition against “composite natures,”7 or even make reference to trends in con­ temporary wall-­painting,8 but Horace’s aims are all his own, for the senses of humano can be traced the length of the Ars Poetica, where they work and collaborate to engender much of the poem’s internal cohesion. The creature to which this ­human head is attached further portends a poem that, like the image, may be confusing or hard to reconcile with itself, but that seethes and pulses with vibrant life and is propelled forward by an animating energy. The figure appears to be a ­jumble of pieces, “all variety and no unity,”9 its disjointedness mirrored and accentuated by the tortuous syntax of the first five lines, where the conjunction si, which would regularly be the first word in its clause and thus of the entire poem, is forced to wait ­until the second spot in the second line to enter the sentence.10 Horace’s use of the phrase nec pes nec caput, proverbial for something that is not a uniform and unified ­whole, also seems especially perverse given that his creature does not, as far as we have been told, have a­ ctual feet, but rather a fish tail.11 Nevertheless, it seems pos­ si­ble and, moreover, necessary to assem­ble or at least attempt to assem­ble the 4. Esp. Resp. 6.488a (where such hybrids are denoted by the term τραγέλαφος, “goat-­stag”), Phdr. 264c. 5. Aen. 3.424–28; Gantar 1964: 91 compares also the Pseudo-­Virgilian Ciris, 490–507. Virgil, like Horace, delays the detail that the ­human is female, and the similarities between the two descriptions are notable (see further Caballero 2007: 74). 6. Aen. 10.209–12. Horace’s creation has in common with Scylla (named at AP 145) and with tritons/tritonesses its fish tail. 7. DRN 5.878–924, Wickham 1891: 389. Fiske 1920: 450 adduces also Lucil. fr. 604 Krenkel, while Meerwaldt 1936 discusses pre­ce­dents in Lucian and Plutarch. 8. The comparison with Vitr. 7.5.3–4 has long been noted (Frischer 1991: 74–76, Elsner 1995: 51–58, Tsakiropoulou-­Summers 1995: 212–14, Nichols 2009, Citroni 2009; see further Ehrhardt 1987: 152–62 on Vitruvius). Hardie 1993 discusses the relationship of Horace’s poetry (especially his Odes) to con­temporary visual art, and Frischer’s list (1991: 78–79 with fig. 2–3) of extant wall paintings containing hybrid or fantastical creatures is useful; cf. also Sharrock 1996: 108 and Platt 2009. 9. Brink 1971: 81. 10. Cf. ibid.: 86: “The reader is faced at once with the substance of the si clause—­the violation of the lack of unity.” Almost all translations into modern languages begin by rendering si, followed by pictor: “If (/Suppose) a painter,” “Si (a) un pintor,” “Si un (/Supposez qu’un) peintre,” “Se (ad) un pittore,” though some follow si with capiti (“Si a una cabeza humana,” “Se abbozzando una testa il pittore volesse”). ­Those very few that strive consciously to avoid this are forced to begin instead with the prepositions required to reflect the syntax of humano, e.g., “A una cabeza humana si juntarle quiere un pintor.” 11. Cf. Frischer 1991: 73.

hum a n o  41

pieces Horace casts before us into a coherent ­whole, since we are, ­after all, supposed to be looking at a painting. The creature suggests itself as a hybrid of hybrids12—­part centaur, part eros, and part tritoness—­but some components are missing, we begin to realize, o­ thers inadequately explained. That the head belongs to a ­woman is unexpected (in this Greco-­Roman milieu a h­ uman head is by default a man’s), and this ­causes us to go back to the beginning/top and start over, readjusting and refining our m ­ ental image. The figure’s femininity also renders it more monstrous, allying it with such terrifying creatures as Gorgo/Medusa, Hydra, Chimaera, Empusa, Lamia, or Mormo, as well as with Horace’s witches, Sagana and Canidia, and the ­women at whom the vitriolic invective of Epodes 8 and 12 is directed.13 Of the body, on the other hand, we are told nothing: is it that of a ­woman (although the neck already is equine), or a ­horse (by extension downwards from the neck), or a bird? This would depend in part on ­whether collatis membris is understood to be dative with inducere or an ablative absolute construction:14 in the former reading the body is left unmentioned while the limbs, culled from a variety of animals,15 are (alarmingly) feathered; in the latter, the body is by implication a bird’s,16 and the creature has a variety of limbs (though ­these are mercifully not then further covered in feathers). The need to choose between the pos­si­ble syntaxes of 12. The connection with hybrid creatures is noted by, e.g., Gantar 1964: 90, Sharrock 1996: 124, Oliensis 1998: 199 (“profiles, blurred but unmistakable, of thoroughly familiar monsters”), but eroses or cupids have not previously been identified among ­these. Human-­bird figures with colorful wings from the ancient world are widely attested (e.g., Fine Arts Museum of San Francisco 1925.338, 1925.339, 1925.340; Museum of Fine Arts, Boston, 01.7794, 01.7793, 1970.62; British Museum, 1905,1021.4, 1805,0703.292, 1888,0601.125) and colored wings are found in wall paintings, e.g., the House of Augustus on the Palatine (Carettoni 1983, color ­table 16), the Villa Romana della Farnesina (Bragantini and de Vos 1982: 32–35, 55; Platt 2009: 51 compares the Farnesina siren to Horace’s image). 13. See further Hopman 2013: 97–100 and Lowe 2015: 70–163 on the effects of gendering monsters as female, and Paule 2017 on Horace’s witches. 14. Rudd 1989: 151, like Orelli 1844: 698, says it must be an ablative absolute (other­wise “the ­horse’s neck and fish’s tail would be covered in feathers” and “clearly that is not what H. means,” though I would ­counter that Horace’s aim is precisely to make the creature as horrifying as pos­si­ble), while Wickham 1891: 389, Wilkins 1896: 336, and Brink 1971: 86 understand membris as dative, the better reading since inducere (TLL) routinely governs this construction. 15. Undique is not so specific as to mean all four divisions of the animal kingdom (so, e.g., Rudd, my chapter 1, n3) and the figure becomes more unsettling if the reader is left to imagine from where the limbs could have been gathered, as well as how the fish’s tail fits in with the creature’s (hind) legs. Nevertheless, it would be especially disturbing for membra to include a single ­human arm or leg. 16. So Rudd 1989: 150.

42  ch a p t e r 1

collatis membris must be part of the general illusion: Horace is a deft enough poet that when he creates an ambiguous grammatical construction, and with it an ambiguous image, he typically intends for us to ponder and, ultimately, even if this goes against our better instincts, to live with the uncertainty rather than resolve it. Horace has given us beginning (head), ­middle (torso), and end (tail), as Aristotle instructed,17 even as the Ars Poetica overall is often accused of not possessing clearly ­these three components, jumping into its subject without “preliminaries” and lacking as it does a “conventional conclusion.”18 If Horace has enacted his famous instruction to begin in medias res (AP 148), his creature surely violates another princi­ple articulated ­later in the poem—­“ let not the ­middle part be discordant from the first, the last from the ­middle” (primo ne medium, medio ne discrepet imum, 152)—as he begins by showing us what not to do. The implicit promise of an odd and monstrous but nevertheless coherent and cohesive creature therefore turns out to be a trick, as Horace exploits the ­human brain’s desire to see patterns in visual sources and to assem­ble ­things into ­wholes,19 especially in the case of something we are given to believe is meant to be ­whole. The creature is impossible—­not in the ­simple sense that no such ­thing could exist, but rather in the sense that no single, complete creature at all can arise from Horace’s description. In practical terms, each person attempting to visualize it would do so in his or her own way—­and very likely in a dif­fer­ent way each time. This—an unpicturable picture—is surely the definition of a (deliberately) bad ekphrasis.20 A further irony is pre­sent in the fact that, since we cannot compare our own ­mental image directly with anyone ­else’s, we must instead render what we see in our mind’s eye into words; yet this produces merely a description of a picture in words—­the very starting point that Horace provides us with and that proves insufficient.21 The result—­whether the creature 17. Poetics 1450b: ἀρχήν, μέσον, τελευτήν; see further Caballero 2007 as well as Brink 1971 on Horace’s philosophical sources for his opening twenty-­three lines. 18. Quotations from Brink 1971: 85 and Frischer 1991: 74. 19. The lit­er­a­ture on this phenomenon is substantial, and Michael Shermer’s work is especially accessible to the nonspecialist (e.g., Why P ­ eople Believe Weird Th ­ ings [1997, 2002; Henry Holt], How We Believe [2000, 2003; W. H. Freeman], The Believing Brain [2011; Macmillan], Ske?tic [2016; Henry Holt], “Patternicity” [Scientific American, 2008]). 20. Moreover, Horace has opened with a conventional descriptio, a device he mocks at 14–21; cf. Brink 1971: 85, Frischer 1991: 73. 21. Cf. Schwindt 2014: 59: “Someone is giving an account of something which only exists in his head. Nobody is able to check and confirm the credibility of this description.” On the prob­lem that “each reader poems his own poem” (Stanley Edgar Hyman) more generally, see Holland 1975.

hu m a n o  43

itself, or the pro­cess by which we arrive at it—is unnerving (turpiter),22 yet we, called “friends” (amici) no less, are told that our reaction is to be not alarm but laughter, even as we, let in for a private viewing (spectatum admissi), now find ourselves alone with the (possibly deranged) artist.23 So the ground has shifted ­under our feet repeatedly in a mere five lines: we begin by being asked to visualize a painting; information about this painting is provided piecemeal (the syntax itself of the Latin working to keep us in the dark as long as pos­si­ble), and the succession of surprising ele­ments culminates in the revelation that the head is that of a ­woman, compelling us to go back and correct the (likely) male head we had been picturing; we are then in the fifth line, ­whether willing partakers in the pro­cess or not, placed suddenly into the persons of the paint­er’s friends, invited for the first viewing (though we may realize that we are si­mul­ta­neously Horace’s own amici24 admitted to see his image/poem); and fi­nally, we are instructed, or rather forced, to respond with laughter or, more specifically, to respond by restraining the laughter Horace suggests we are struggling to hold in. We perform ­these actions u­ nder compulsion, moving from one to the next like automata, controlled wholly by the poet. Our only choice in the m ­ atter would be to lay aside the poem entirely out of fear or disgust and in ­doing so to reject the duties and functions placed upon us to which we have not quite consented, for t­ here is no way to be a passive reader of the Ars Poetica while engaging with the work on the terms it demands. The first five lines of Horace’s Ars Poetica thus provide a disarming and yet sensorily thick experience, but only if we read slowly and submit to being led by the nose to wherever Horace wishes.25 22. Bentley was sufficiently unnerved as to emend plumas to formas, having also proposed pennas (Brink 1971: 86 is rightly unconvinced). 23. Orelli 1844: 699: “ubi tabulam finierat Pictor, primum familiares suos spectatum advocabat.” ­There is no indication that a third character is to be ­imagined, i.e., that the owner of the ­house has invited his friends to view wall paintings he has commissioned from a professional paint­er; it is likewise socioeco­nom­ically improbable that the owner of the ­house is himself a professional wall painter (see further Jones 2016). 24. We are similarly included among his amici at Sat. 1.10.87; see Ferriss-­Hill 2015: 147–50. 25. This mode of reading borrows from reader-­response criticism, of which an overview may be found at Rabinowitz 1986; on the applicability of this approach to Epicurean poetry, see Asmis 1995a with Sider 1995a: 38, Solomon 2004: 264–65. I have found the following works helpful for considering how Horace treats, even manipulates, his reader in the AP: Holland 1975, Ong 1975, and the essays in Tompkins 1980 (especially Walker Gibson’s “Authors, Speakers, Readers, and Mock Readers”).

44  ch a p t e r 1

The poem’s opening image leads in its immediate context to an exhortation to make a work, any work, simplex . . . ​et unum (23).26 Connected by et, simplex and unum appear at first glance to be parallel and complementary ideas, but the precise relation between them repays further probing.27 Simplex, which we may naturally associate with and translate by the En­glish “­simple,” is a compound of sim-­/sem-­, “one; as one,”28 and plicare, “fold, roll up.”29 What Horace is proposing thus becomes oxymoronic, at least when applied to his own poem: while something with a single fold or rolled up as one (sim-­plex) is certainly “one t­ hing” (unum), the Ars Poetica instantiates something that is ­whole, a single work (unum), yet not without folds and certainly not ­simple. Neither “one-­fold” nor “consisting of one ele­ment or ingredient,” it is formed rather of disparate parts stitched together, proudly displaying its folds though always inviting the reader to find ways to smooth them out. It has been conventional since at least Badius’s edition30 to mark paragraph breaks in the Ars Poetica, even if editors differ in the exact locations of t­ hese breaks (contrast, for example, ­those of Brink 1971 with ­those of Shackleton Bailey 2001), yet even this innocent practice, on the one hand designed to make the poem more accessible and on the other suggesting that it possesses a par­tic­ul­ar form of organ­ization, serves to obscure the links across ­these neat subunits, for from the poem’s opening image we are given to understand what we may expect from the entire work: the painted assortment of stitched-­together parts does result in a form of ­whole, as ­will the apparently unrelated and dissonant parts of the Ars Poetica. A second maxim to complement that of line 23 follows shortly: “the avoidance of [one] fault leads to [another] vice, if done artlessly” (in vitium ducit culpae fuga, si caret arte, 31), Horace intones, playfully suggesting also that success cannot be achieved without an ars, “handbook,” perhaps even this very one.

26. Horace may have in mind Aristotle’s ἕν καὶ ὅλον (Brink 1971: 104), though the connection is unlikely to be straightforward since, e.g., simplex also echoes Horace’s own lyric use of the term at Carm. 1.5.5 and 1.38.5 and the simplex and duplex sauce ­recipes of Sat. 2.4.63–70 (Caballero 2007: 77–78). 27. Similar are Brink 1971: 104 (“simplex makes unum more concrete”), Tsakiropoulou-­ Summers 1995: 224–30, Lowrie 1997: 172–73, Caballero 2007: 77–78 (“no son términos equivalentes sino convergentes”). 28. See Watkins 2000: 75 on the root *sem1-­ (also the source of Latin semel, “once”) and Ernout and Meillet 2001: 514–15 on simplex. 29. Cf. Lowrie 1997: 164–65 ad Carm. 1.38: “etymologically, simplex means ‘folded only once,’ hence in the moral sense, ‘not complicated, ­simple without detour.’ ” 30. Hardison and Golden 1995: 166.

hum a n o  45

If the opening of the illustrious Ars Poetica leaves us feeling unsatisfied, even perplexed,31 surely this is by design, for Horace does not merely want to lecture his readers about the importance of harmony, long taken to be the “point” of t­ hese opening lines. Rather, in performing an opening that is disjointed and bewildering, and to which we have been admitted as viewers, he compels us to feel the effects of art that lacks the guiding princi­ples he ­will articulate ­here and ­there throughout his poem. Yet ­there is also richness and even harmony in the heterogeneity of Horace’s painted creature, and if we only search we may find it conceals something valuable—­a full repre­sen­ta­tion of the poem itself. Although Horace has tricked us, or at the very least let us know that he has complete control over us, his readers, the promise heralded by humano ­will be fulfilled in the course of the poem:32 the Ars Poetica is concerned above all with ­human be­hav­ior and endeavors—­the purview of the Satires. In this sense, despite the lack of unity displayed as well as embodied in the opening five lines, Horace has succeeded in achieving the sort of unity prescribed by Aristotle as a desideratum, though he has traditionally been criticized for failing to do so.33 Brink, adducing the “old Aristotelian demand for consistency of composition,” explains this to mean that “once the ‘tone’ of a poem is set by its first words the rest must be attuned to it.”34 Although the most widespread solution has proposed that Horace deliberately begins in a disjointed fashion to introduce a disjointed poem,35 I rather see each thematic strand traced in this study, as it runs throughout the Ars Poetica, pre­sent as a single word in the poem’s opening, which consequently contains in embryonic form the entirety of the poem that unfurls from it. A critical function of the opening, I contend, is to prime the reader to perceive throughout the Ars Poetica its currents of ­human activity; its ­fathers and sons, teachers and students, intergenerational relationships, and life cycles; the notion that true criticism is best delivered by one’s friends through laughter, even at the risk of pain; and the central role of painting, which ­will be 31. Frischer 1991: 72: “Even ­after two millennia of trying, we still cannot be certain we understand what the speaker intends to say in the opening lines of the poem.” 32. Oliensis 1998: 200 also sees the poem’s opening word as key to its interpretation, though for her it emphasizes “the natu­ral and manly superiority of the head in relation to the rest of the body,” as explored by Cicero in De officiis. 33. This critique has been tackled by, e.g., Norden 1905, Armstrong 1993: 186, Caballero 2007: 79. The words of Wimsatt 1970: 135 (in Brophy et al. 1970) remain instructive: “­There is no such ­thing as a famous and successful poem, a classic, which turns out on study to be disorderly.” 34. Brink 1971: 94 ad 14–19. 35. E.g., Frischer 1991: 74 (the opening monster is “exemplified by the Ars Poetica itself ”), Laird 2007: 138 (“Horace could be warning us, the Ars Poetica itself is ­going to look like the sort of chimaera he has been describing”).

46  ch a p te r 1

consistently privileged over poetry and which anticipates the idea of making and creating in the broadest terms. Above all, the poem’s opening makes the careful reader alert to the ways in which ­these vari­ous themes ­will be stitched together, as we are also encouraged to appreciate the poem not merely for what it says but for what and how it makes us feel.

­Human Nature Wedded to the poem’s first line through its similar line-­initial appearance of humanus is 102: non satis est pulchra esse poemata; dulcia sunto et quocumque volent animum auditoris agunto. ut ridentibus arrident, ita flentibus afflent humani vultus: si vis me flere, dolendum est primum ipsi tibi. tum tua me infortunia laedent, Telephe vel Peleu; male mandata si loqueris, aut dormitabo aut ridebo. (99–105) Orelli 1844: 722, with typical insight, explains, “Humani is not idle, but indicates . . . ​that we are ­humans and can feel pain together with ­others, something that is denied to brute animals” (“Humani non otiosum est, sed significat . . . ​ quod homines sumus et condolescere cum aliis possumus, id quod brutis animalibus negatum est”). Just as humani echoes humano, moreover, its accompanying noun vultus (itself repeated for emphasis only four lines l­ ater, 106, and once more in line-­initial position) both reiterates capiti and evokes the stunned expressions of the paint­er’s friends, scarcely able to stifle their guffaws. Words that denote or imply emotional states are piled up,36 with ridentibus arrident, flentibus afflent . . . ​flere,37 dolendum (picking up on dolet from 95–96, where its subject was also Telephus and Peleus), infortunia, and laedent 38 all occuring within three lines. Envisioning the feelings that form the appropriate response to successful poetry, Horace works to articulate the symbiosis ideally required between author and audience. In keeping with Peripatetic ideas, if an author wishes to engender a certain emotional response in his reader-­spectator (100; cf. si curat cor spectantis tetigisse querela, 98), he must feel the same and evince

36. Wickham 1891: 398–99 similarly explains humani vultus as “­faces, which are the ­faces of men, and therefore have ­human feelings ­behind them” (cf. also his comments ad 99, 108). 37. Afflent/adflent is an emendation for the codices’ adsunt/assunt; Brink 1971: 185 gives a summary of the arguments. 38. “A strong word,” ibid.: 186.

hum a n o  47

this.39 The emotional impulse set in motion by the artist is envisioned as affecting, even infecting, the audience to produce a response that precisely mirrors the original impulse. Successful psychagogy (99–100)40 of this sort requires proper planning and correct execution (male mandata / male . . . ​loqueris), other­ wise audience and author w ­ ill not be in sympathy. The referent of the adverb male is ambiguous, no doubt intentionally, allowing two readings to coexist: taken with mandata it describes the author’s failure to entrust lines correctly to an actor, while with loqueris it describes said actor’s poor delivery of his lines.41 We may even sense at 101–3 the idea that emotion is prior to any expression of it: h­ uman feelings are irrepressible and exist regardless of ­whether ­these are then translated into language and shared with ­others.42 Horace emphasizes again at 99–105 the ­human aspects of both producing and consuming art, and humani vultus: si vis me flere, dolendum est serves to supplement the poem’s opening vignette and to make this function explicit. The painter of the poem’s imperfect first artwork, we may conclude, has himself not agonized sufficiently over his work (tortured though it is in appearance) and as a result produces not the desired emotional response in his spectators but a wholly undesirable one, for laughter can be an unintended reaction, not simply one arising when the artist himself laughs, wishing and causing his reader-­spectator to do the same. The sense that Horace is interested in ­humans, not merely characters, and in emotions beyond the usefulness of Aristotelian pathos as a compositional and rhetorical device is evident throughout the Ars Poetica,43 and discerning this strand can help to smooth out some of what have been seen as the poem’s rougher transitions of topic. Horace had appeared to set off abruptly in a new direction at 73, the Ars Poetica giving the impression, as it so often does, of being one side of a conversation other­wise hidden from us—to whom is Horace responding, and what exactly are we to imagine that this interlocutor (not merely anonymous, but wholly absent) has said?44 He takes up the topic of “poetic 39. Feigned emotion in oratory is widely adduced as a comparandum ­here: Pseudo-­Acro compares Cicero (evidently a paraphrase of De or. 2.45) and Porphyrio declares Horace’s approach Demosthenic; see further Brink 1971: 186. 40. See chapter 1, n55. 41. The former reading is espoused by Wickham 1891: 399, Rostagni 1930: 33, Brink 1971: 187, Rudd 1989: 168; the latter by the “Materia” commentary (Friis-­Jensen 1990: 351), with Pseudo-­ Acro and Orelli 1844: 722 noting the possibility. 42. Cf. Rostagni 1930: 33, Antolín 2002: 193, who note that the idea is both Stoic and Epicurean. 43. So Brink 1971, especially 174, 182–83. 44. Cf. Campbell 1924: 241 (“as usual, the transition is quite abrupt, but [again as usual] the logical connection is given us in time, at 86–88”), Brink 1971: 162–63 (though the transition is done “brusquely and unconnectedly,” Horace ­later “reaps the poetic benefit of this procedure”).

48  ch a p t e r 1

genres defined by metre and subject,”45 leaving ­behind the opening scene (1–13), which led to a second on unharmonious painting (14–23), and to ones illustrating the dangers of ­running to extremes (24–31), the importance of knowing one’s limitations (32–44), and the circumstances ­under which new words may be created (45–72). Yet Armstrong 1993: 187 is right to see that Horace is at such moments deploying “the Epicurean poetics” he “learned from Philodemus,” which are predicated on an “acute appreciation of the inter-­relation of atomic and macroscopic levels.” As each genre is listed in conventional order (73–85), then, I see them acting as a cipher for Horace’s far greater and sustained concern with h­ uman emotions, which are given preeminence of place: Homer is credited with establishing that the hexameter would be used for epic, defined as “the deeds of kings and generals” (res gestae regumque ducumque)46 and “sad wars” (tristia bella, 73); elegiac couplets ­were first used for “lamentation” (querimonia, 75)47 and “the expression of [thanks for] a granted prayer” (voti sententia compos, 76);48 the iambic meter was handed over to Archilochus by Rage herself, rabies (79);49 and the lyre is for, among other ­things, “the concerns of young men” (iuvenum curas, 85). Each genre is thus itself conceived of as containing, even feeling, its assigned emotions, or as producing ­these in ­others, or both, acting as a conduit that allows a character’s emotions (ultimately stemming from the author, 101–3) to pass into the reader. 45. Brink’s 1971: 161 summation. Brink 160–63 is useful on the structure of 73–85 and on Horace’s models; as he notes, however, despite t­ hese lines, Horace generally avoids in the AP the sort of terminology that would mark a work as an ars/technē (161, 170). 46. Despite some skepticism (Brink 1971: 164; Rudd 1989: 163 is s­ ilent), the resonance of Augustus’s res gestae (cf. Epist. 1.3.7: res gestas Augusti) would have likely been obvious to an ancient reader (see OCD s.v. res gestae on the availability of the work prior to its being inscribed onto bronze tablets in 14 CE). 47. An “archaism” (Brink 1971: 166), capable in Horace of meaning “lament” (Carm. 2.20.22), “quarrel” (Carm. 1.13.19), and “complaint” (Carm. 3.24.33). 48. Rudd 1989: 163 pinpoints “votive epigram” as the referent ­here. Wickham 1891: 396–97, Rudd, and Brink 1971: 165–67 all state that love elegy is not included within the category of elegy at line 76 (contra Orelli 1844: 715, who assumes that elegos includes love elegy) and does not appear ­until 85. 49. While proprio (“one’s own”) in Archilochum proprio rabies armavit iambo is generally taken to mean “his own,” i.e., Archilochus’s (so the anonymous twelfth-­century commentary at Hajdú 1993: 252–53, Orelli 1844: 716, Wickham 1891: 397), Rudd 1989: 164 assigns the notion of owner­ship to Rage instead (and indeed proprius ­ought to refer to the subject of armavit). The reading that emerges is appealingly proleptic, for once “Rage armed Archilochus with her own iambus,” this meter most suitable for expressing anger would also naturally become “his” and thus “their” shared property. Iambus ­w ill reappear, and twice in close succession, at 251 and 253, drawing further attention to the term and serving to tie the latter passage back to this e­ arlier one.

hum a n o  49

The emphasis on feelings is notable throughout 73–118. Following the selective parade of genres (73–85), ­others are next personified and endowed with the capacity to experience emotion: while tragedy generally regards with indignity (indignatur) the comic mode of speaking (90–91),50 comedy may at times raise her voice (93). The comic character Chremes may be tragically “angry” (iratus),51 while the tragic figures Telephus and Peleus may use the everyday speech (sermone pedestri) more typical of comedy to express their pain (dolet, 95). The point is manifold: Horace prescriptively delineates what each genre is for; at the same time, he manages to issue a directive on the importance of ignoring directives (comedy and tragedy exhibit some flexibility in register);52 and he communicates the emotional under­pinnings to lit­er­a­ture. Lines 99–118, where laughter is found interwoven with emotions and with speech, are notable for the repetition within them of the term animus.53 Indeed, Horace seems unusually concerned with the soul,54 first employing animus with agunto to translate and render the Greek term ψυχαγωγεῖν,55 as he envisions the personified poem physically leading the listener, then characterizing the tongue as the interpreter of the soul’s movements, a logical enough, if nevertheless unusual, conceptualization of its function. As in the opening vignette, I see Horace being particularly interested in the disconnect between a thought, feeling, or m ­ ental image, and its expression in words, ­whether written or spoken. The word animus had occurred once before as well (sectantem levia nervi / deficiunt animique, 26–27), but 99–118 marks the first of several passages in which it appears twice in close succession, the ­others being at 166 and 180; 330 and 336; and 432 and

50. Th ­ ere is also perhaps a hint that Horace is identifying Thyestes’s banquet as a poor topic for the genre of satire, which can be discerned in privatis ac prope socco dignis carminibus (on Roman Satire as both private and allied to comedy, see Ferriss-­Hill 2015: 29, 78–80, 112–20, 243). Persius may accordingly be read as having taken up Horace’s challenge when he inserts precisely this subject into his fifth satire (8–9). 51. Brink 1971: 168 rightly notes the unusual primacy accorded ­here to comedy over tragedy. 52. Laird 2007: 140 notes that this “makes sense in terms of modern thinking” about genre where “ ‘rules’ are certainly ­there, but t­ hose rules are ­there to be broken.” 53. Brink 1971: 174–94 is useful on the Aristotelian background (Rh. 3.7) to 98–118. 54. Perhaps a reflection, as I suggest also of Natura (see chapter 1, n69), of the influence of Lucretius’s DRN on the AP (on which more generally, and especially as undervalued, see Hardie 2009); on animus/anima in Lucretius, see Taylor 2007. 55. Brink 1971: 184, Antolín 2002: 192. The term ψυχαγωγία appears throughout Philodemus, On Poems 1 (Armstrong 1993: 209, where Janko 2000 translates it as “enthralment”; cf. also Tsakiropoulou-­Summers 1995: 242–46), and the history of psychagogicity in ancient poetry is traced well by Asmis 1995b.

50  ch a p t e r 1

437.56 Across t­ hese passages Horace describes: how t­ hings in tragedy that enter through the ear are more palatable than t­ hose seen with the eyes (segnius irritant animos demissa per aurem / quam quae sunt oculis subiecta fidelibus, 180–81;57 this comes shortly ­after the ­simple phrase aetas animusque virilis, 166, to denote a stage of life); the deficiencies in the Roman educational system, which fails in that it valorizes meaningless bean-­counting and a concern with money, said to be a rust or mold (aerugo) that infects or dyes the souls (animos) of young boys (330–31; the Greek, by contrast, values inspiration, creativity, and greatness); the need to express oneself concisely so that teachable souls may seize upon the words and hold them faithfully (esto brevis, ut cito dicta / percipiant animi dociles teneantque fideles, 335–36); how “poetry was born and in­ven­ted for helping souls” (sic animis natum inventumque poema iuvandis, 377); how Homer and Tyrtaeus “sharpened male souls for Mars's wars with their verses” (mares animos in Martia bella / versibus exacuit, 402–3); and the paradox that t­ hose who feel genuine grief are often outdone in their displays of anguish by hired (i.e., not truly grieving) mourners (ut qui conducti plorant in funere dicunt / et faciunt prope plura dolentibus ex animo, 431–32). Although Horace is naturally saying dif­fer­ent t­ hings at ­these moments, the sense that emerges—­and to which he makes us attuned through the repetition of the key word animus—is of an abiding interest in the internal life of ­people, just as the phrase animum auditoris (100), like cor spectantis (98), the syntax of which it repeats, identifies a seat of h­ uman emotion—­ heart and soul—­and the possibility of physically touching or causing this organ to move. Evident throughout is the intimacy Horace sees between the ­human soul and poetry: poetry, made by h­ umans, in return possesses the power to cause the souls of ­others to move, figuratively and actually. Perhaps most intriguing at 99–105, quoted above, is the contrast between pulchra and dulcia, for while the two would seem at first to be relatively synonymous in describing lit­er­a­ture, the context (namely, that the latter is required in addition to the former) necessitates that they be complementary and so in some fashion opposing traits.58 Following the scholiasts, modern 56. Animus also at 250, 377, and 402 (the latter two instances fairly close together but not obviously linked). The reading of line 437 (numquam te fallant †animi sub vulpe† latentes) is uncertain; see chapter 3, n16. 57. Blakeney 1928: 78 explains this as being b­ ecause “the use of sight is continual, that of hearing intermittent,” though interestingly the distinction between perception by the ear and mind is treated repeatedly by Philodemus in On Poems books 1 (e.g., col. 126, 128, 151, 171–72, 175) and 3 (fr. 6–7, 15–16, 26, ­etc.). 58. Brink 1971: 184: “dulcia: the opposite term to pulchra”; cf. also Campbell 1924: 64 (who speaks of “fusion”) and Rudd 1989: 167. Brink (183), a­ fter Orelli 1844: 721, compares καλόν and ἡδύ at Dion. Hal. Comp. 10.

hu m a n o  51

commentators emphasize the rhetorically charged character of Horace’s terms: pulchra is explained as “skillfully expressed,” as technically proficient but failing to touch the heart, and as “ ‘formally correct’, ‘well made’, even ‘noble’ ”;59 dulcia means “ethical,” that the art moves the soul and is lovely, and it “implies a direct influence on the emotions.”60 Yet even this does not capture the entire range of both terms. Beauty and sweetness, both said to be desirable in a poem, are united in appealing to the senses: beautiful ­things (pulchra) are seen with the eyes, while sweet ones (dulcia, even if the usage ­here is predominantly meta­phorical) are evocatively tasted with the tongue.61 Horace w ­ ill reprise this pair in a variety of forms throughout the Ars Poetica, not repeating but rather reworking the first iteration: the initial contrast of beauty and sweetness (pulchra, dulcia) becomes a more ethical one as it develops into the tension between providing plea­sure and being useful (aut prodesse volunt aut delectare poetae / aut simul et iucunda et idonea dicere vitae, 333–34), and fi­nally both are combined to illustrate the desirability of sweetness and plea­sure together (omne tulit punctum qui miscuit utile dulci, / lectorem delectando pariterque monendo, 343–44), the chiastic arrangements of both ­later formulations serving further to unite them.62 Also resonating outward from 99–105 to other points in the poem are Telephus and Peleus. They have reappeared almost immediately a­ fter their mention at 96 (Telephus et Peleus),63 as if to bond the section on “the deeds of kings and generals” (73–98, res gestae regumque ducumque . . . ), with its meandering parade of genres, to that beginning at 99. Although we understand that ­these figures are characters in literary works, Horace treats them as if they ­were far more substantial than that, as he pre­sents them speaking almost of their own ­will (loqueris; cf. loquatur, 114). Pre­sent at 104–5, as it ­will also be at 112–13, is thus an emphasis on what characters themselves say—­loqueris,64 dicentis—­rather 59. Pseudo-­Acro (“diserta”), Orelli 1844: 721 (“artis legibus . . . ​satisfacientia . . . ​sed cor non tangunt”), and Rudd 1989: 167, respectively (Rudd adds, “A work with this quality alone would be impressive in a rather cold and austere way”). 60. Pseudo–­Acro (“ethica”), Orelli 1844: 721 (“ψυχαγωγικά, lieblich, quae suavi etiam commiserationis voluptate afficiant animos”), Rudd 1989: 167. 61. Wickham 1891: 398 suggests the same in passing: “satisfying the taste.” 62. Aut prodesse volunt aut delectare poetae / aut simul et iucunda et idonea dicere vitae and omne tulit punctum qui miscuit utile dulci, / lectorem delectando pariterque monendo. On Philodemus’s similar comparison in On Poems of ὠφελεῖν and ψυχαγωγεῖν, see Armstrong 1993: 209, 219–25, Clay 1995: 7. 63. Brink 1971: 179 notes Telephus’s popularity as a dramatic character, though this does not fully satisfy as an explanation for Horace’s interest in him ­here. 64. The subject of loqueris is debated: Brink 1971: 187 sees Horace addressing “the dramatis persona” with the tragic poet involved “only at a remove . . . ​since the playwright is ultimately

52  ch a p t e r 1

than on what the author writes for them to say. This blurring of character, actor, and playwright has been noted,65 but its specific effects are worth identifying, for in endowing the poet’s creations with such a robustness and consistently treating them as though they ­were ­actual, living ­humans, Horace succeeds in shifting the focus of his poem from the crafting of lit­er­a­ture to the making of one’s way through of life. A few lines ­after the poem’s second deployment of the adjective humanus (102), Nature appears personified, Horace’s focus remaining on ­people as they actually are rather than on fictional characters created by a poet: format enim Natura prius nos intus ad omnem fortunarum habitum; iuvat aut impellit ad iram aut ad humum maerore gravi deducit et angit; post effert animi motus interprete lingua. (108–11) Natura, Horace explains, has thoroughly s­ haped each of us in advance in our own par­tic­u­lar way, such that ­whether our response to any chance occurrence is anger or sadness is presented as entirely beyond our individual control.66 The feelings produced in the course of this response are then communicated through speech.67 In place of a verb of speaking, however (though dicentis appears in

responsible for ‘assigning ill-­fitting speeches,’ ” but he notes that o­ thers have “connected male with loqueris, the addressee then being not the imaginary mythical hero but the actor playing the part”; cf. chapter 1, n41. 65. Rudd 1989: 168. 66. By the conventional reading, lines 109–10 are taken as describing the range of emotions that all ­people experience (so already Pseudo-­Acro, who explains that the two reactions named, anger and sadness, represent in compressed form “adfectus omnes”; cf. Orelli 1844: 722), but this reading somewhat overlooks the force of aut, which is used most commonly for “introducing two or more logically exclusive alternatives” (OLD). I instead see Horace suggesting that Nature makes some ­people predisposed to be melancholic, ­others quick to anger, what­ever happens, and indeed Philodemus also discussed individual dispositions in a number of works (see Tsouna 2007: 32–44). This reading would be in line with 105–7, where Horace speaks of “a sorrowful man” (maestum), “an angry man” (iratum), “a playful man” (ludentem), and “a stern man” (severum) apparently as fixed personality types, not descriptions of temporary emotional states. 67. Line 111 is difficult to construe (though the commentators, aside from Rudd 1989: 168, are largely ­silent): what is the import of post? Is Natura the subject of effert? What is the syntax of interprete lingua, and does this tongue belong to Nature or to the person experiencing the emotions? Brink 1971: 190 notes that interprete lingua is a Lucretian phrase (DRN 6.1149) “from a very dif­fer­ent context,” though it may be pointed out that ­there it likewise occurs in conjunction with animi.

hu m a n o  53

the next line),68 Horace has an arrestingly physical tongue, lingua, repeated from 56.69 Traces of the poem’s opening word and ancillary ones persist ­until its close in other ways as well. At 202–91 Horace pre­sents in two parts (an excursus on meter separates them) a narrative on the evolution of drama, in which history and prehistory, Greek and Roman,70 combine to accentuate at ­every turn the essence of all creative products (μουσική)71 as the result of h­ uman action and interaction through a pro­cess both unpredictable and inevitable. Horace begins with a description of the development of instrumentation (202–7), its strongly primitivizing flavor stressed by resonances of nature that increase in strength as the explanation goes on: “The double-­pipe was not, as now, bound with brass and a rival to the trumpet, but slender and s­ imple to play without many an opening” (tibia non, ut nunc, orichalco vincta tubaeque / aemula, sed tenuis simplexque foramine pauco / aspirare, 202–4).72 The opening word, tibia, is certainly the regular one for the double-­pipe, yet the name necessarily communicates the shin bone from which this instrument was originally made, and ­there is clear irony in the idea of this bone being embellished and altered to such

68. Cf. Brink’s 1971: 190 comment on how (as Kiessling and Heinze 1914: 308 observe) 112 “generalizes and sums up the lively incursion into emotive language. Once again it is only at the end of a context that its scope becomes fully apparent.” 69. Feeney 2009: 36 notes that “the power of Nature is an impor­tant meta­phor and concept in the Epistle to Florus passage as well as in the Ars.” Horace’s interest in Natura/natura in the AP almost certainly owes something to Lucretius’s De Rerum Natura, though the connection remains underexplored; among the substantial lit­er­a­ture on Natura in Lucretius, see most recently Reinhardt 2002, O’Keefe 2003, Solomon 2004, Asmis 2008. 70. Brink 1971: 263: “a composite picture of Greek and Roman.” Greek are the early Attic milieu of “goat-­song” (hircus explains the etymology of tragicus, since τραγῳδία is formed from τράγος, “goat,” and ἀοιδή/ᾠδή, “song”) and satyrs (220–21, 226, 233, 235), while Rome is suggested by the (Italian) fauns in the Forum (244–45) and a matrona (232). ­There is also a textual difficulty: the singular urbem, the reading of the codices deteriores and printed by Brink 1971 and Shackleton Bailey 2001, is suggestive of Rome, while the urbes of the more authoritative manuscripts would indicate Italy more generally or the cities of Greece (see further Williams 1968: 336–41, Brink 1971: 267–68, Rudd 1989: 185). 71. Horace’s conflation of lit­er­a­ture with m ­ usic suggests this Greek concept, which, “­until the l­ater fourth c­ entury, had embraced both the tune and the word performed to it” ( Janko 2000: 190, who is concerned especially with Philodemus’s position in the debate). 72. I understand aspirare as epexegetical with simplex (and tenuis right before it), as well as adesse with utilis and complere with nimis (contra Brink 1971: 266, Rudd 1989: 184); other­wise, my translation makes extensive use of Rudd’s glosses.

54  ch a p te r 1

a degree that it comes to resemble its original state very l­ ittle.73 Having moved on to discuss the lyre (216–19), Horace similarly speaks of voices “growing” (crevere) onto the strings. The standard term for growth of any kind, applicable to plants, animals, and more, the sense produced by crevere is evocative, suggesting that the addition of a sung ele­ment to what had previously been a purely instrumental per­for­mance was unavoidable, caused by forces outside ­human design. The result is a “rushing fluency” (facundia praeceps), where praeceps, which describes a headlong falling forward, even one that is uncontrolled, abrupt, impetuous, or destructive, suggests motion and the inexorable onward march of t­ hings, perhaps even in a sinister fashion, as do (more benignly) the verbs coibat (207), coepit (208), and accessit (211), and as morandus (223), vertere (226), and migret . . . ​vitat . . . ​captet (229–30) w ­ ill also do, themselves to be followed by effutire (231), moveri (233), protervis (233), and enitar (236). Alongside ­these verbs of movement, the resonances of nature and of natu­ral pro­cesses continue: “He who competed with a tragic song for a cheap goat, soon even took the clothes off the field-­Satyrs, and roughly attempted a joke with seriousness unharmed” (carmine qui tragico vilem certavit ob hircum, / mox etiam agrestis Satyros nudavit et asper / incolumi gravitate iocum temptavit, 220–22) is saturated with roughness, rusticity, ele­ments of the natu­ral world, and the absence of luxury and guile. Horace’s notable emphasis at 202–50 on ele­ments of the countryside (even if the prehistory and early history of drama is by convention rustic),74 continues in the second part of the narrative, 275–91. Fields, woods, and their inhabitants, w ­ hether ­human, animal, or mythological, find their counter­parts in 73. See Wellesz 1957: 380–81, 406–8 and Landels 1999: 24–46, 271–75 for details of the instrument’s construction. It is not helpful to speculate that Horace is relying on a now-­lost discussion of the same by Varro, who is as often conjured up as the con­ve­niently no-­longer-­extant archetype (Brink 1971: 262 repeats, though somewhat disavows, this suggestion). ­There are, however, some intriguing points of contact between Horace’s interest in ­music in the AP and that of Philodemus (On Poems 4, fr. 1–2, 10, 15, col. 119, ­etc., mention m ­ usic or instruments), while Delattre 2004: 249 (who discusses the overlap between Philodemus’s and Virgil’s interests in ­music more generally) sees Horace’s nostalgia for the rustic tibia aligned with Virgil’s “pastoral universe.” 74. AP 202–91 is often considered alongside Epist. 2.1.139–76, as well as Verg. G. 2.380–96, Livy 7.2, and Tib. 2.1.51–6. ­These accounts share such striking similarities that a common source for all five has been posited (Varro, as often; see chapter 1, n73), and among t­ hese similarities is a focus on the ­humble ancient “farmer” (agricola: Epist. 2.1.139, Tib. 2.1.51), resting ­after long and hard ­labor (Epist. 2.1.140, condita post frumenta; Tib. 2.1.51, adsiduo . . . ​satiatus aratro) and on a festive occasion (Epist. 2.1.140, tempore festo), and d­ oing so (especially singing) in a manner that is “rustic” (rusticus: Epist. 2.1.146, Tib. 2.1.52; cf. Livy 7.2, inconditis, Verg. G. 2.386, incomptis); see further Oakley 1998: 40–58.

hum a n o  55

the rustic wagon of Thespis, inventor of tragedy as Horace would have it,75 and his players, who have smeared their ­faces with wine lees (275–76). The scene pulses with motion, as vari­ous figures heave themselves about. So, too, does the intervening section (251–62) on meter, where iambus seems to come alive,76 and words such as citus, accrescere, ictus, tardior, veniret, recepit, missos, magno cum pondere, and celeris emphasize movement, both fast and slow, and organic pro­cesses. Yet beneath this all is a delicate symbiotic balance: we are invited to think of Thespis and his players not only travelling (vexisse, 276), but logically ­doing so to the locations in which they are likely to be best received; of Aeschylus’s introduction of the mask, robe, and speaking actor as having been done in response to audience desires and demands (278–80); and of the arrival of comedy, which we are told is well received (non sine multa laude), as the result of similar forces (281–82; cf. 221–30). In addition, the uncommon technical term pulpitum in 279 links this description back to 215, tibicen traxitque vagus per pulpita vestem, inviting us to construct from Horace’s account of the development of drama, which is spread over the course of almost 100 lines and broken up into vari­ous sections, a coherent and continuous ­whole. Rather than being interested in the history of drama for its own sake, then, Horace seems especially fascinated by it as the product of h­ umans and their actions, though its development is driven also by other forces that cannot be securely identified. Once set in motion, drama takes on a life of its own, propelled forward by its makers and enjoyers. The increased urbanism brought about by military victories leads to greater license in ­music (208–11), for example, while the facundia praeceps (the force of facundia as spoken language further heightened by eloquium) caused by the addition of a sung component to lyre playing (214–19) sounds alarmingly violent and unstoppable.77 Other social currents—­the populace craving novelty (220–26)—­occasion the introduction of humor, and so the differentiation of drama into tragedy and comedy occurs (226–39, 275–82). 75. Though with dicitur he disavows responsibility for the identification (Brink 1971: 312). Horace accords Thespis a somewhat more principal role than do other sources, which credit him with inventing the mask (Suda; Adler 1931: II 711, Pickard-­Cambridge 1962: 71, n14) or adding prologue and speech to what had been a purely choral per­for­mance (Them. Or. 26, 316d, who credits it to Aristotle; Pickard-­Cambridge 1962: 70, n12). The role of inventor of tragedy, on the other hand, was variously accorded to Epigenes of Sicyon (Suda s.v. Thespis; cf. Pickard-­ Cambridge 1962: 77, 104) and Arion of Methymna (OCD, Pickard-­Cambridge 1962: 77, 88–89); see Brink 1971: 312, and further Pickard-­Cambridge 1962: 69–101. 76. “Personification,” Rudd 1989: 192. 77. We hear the phrase as more insistent on account of e­ arlier facundia praesens, also in line-­ final position (184), where the sense of speaking is heightened by narret, “­will relate.”

56  ch a p t er 1

The vari­ous mentions of laws (sortilegis, 219; exlex, 224; legitimum, 274), rights (256), and judges and judging (244, 263) that dot the preceding section culminate in the following law of dubious historicity:78 sed in vitium libertas excidit et vim dignam lege regi. lex est accepta chorusqus turpiter obticuit sublato iure nocendi. (282–84) In vitium libertas excidit exhibits clear verbal and thematic resonances of in vitium ducit culpae fuga, si caret arte (31), with Horace now clarifying that vitium may denote a moral as well as an artistic failing, as the language of the Ars Poetica begins to admit of overlap with that of the Satires.79 The adverb turpiter (282) to describe the silencing of the chorus by a law is also not idle, for it takes us back to the opening vignette (turpiter, 3) and in d­ oing so suggests both the h­ uman ele­ment (humano) inherent to the development of drama, as well as the disturbing mess that results when t­ here are no laws that govern the creation of art, since all organic, natu­ral pro­cesses are potentially untidy and dangerous. The overall image is of a system that swells and wanes in response to h­ uman inclinations (­whether good or bad), yet exists outside them, too.80 While Horace’s account of the development of drama seems, from his viewpoint and ours, teleological (though not always to a positive end),81 it also reveals the power­ful and unpredictable forces of nature and of h­ uman actors, and of poetry and its changes over time as the direct result of ­human impulses.82 What emerges is the same undercurrent that runs throughout the Odes: that the onward march of time cannot be resisted and must be accepted, and that it brings with it natu­ral forms of change. In addition to 31, lines 282–84 above also recall 265–68, where Horace wondered aloud, “Should I therefore wander off and write without discipline?” (idcircone vager scribamque licenter?). The alternative, as he pre­sents it, is being “safe and cautious, [staying] within the hope of indulgence” (tutus et intra / spem veniae cautus?), lest ­mistakes be noticed (omnis / visuros peccata putem mea). 78. See Brink 1971: 316–18, Ferriss-­Hill 2015: 95–101 on how this law has been understood. 79. Vitium occurs at 1.2.24; ten times in 1.3; of Lucilius (personally and therefore stylistically) at 1.4.9; several more times in 1.4 in the vignette of Horace and his ­father; repeatedly in 2.2 and 2.7, ­etc. 80. Compare Hardie’s 2005: 39 summary of AP 46–74, where he sees Horace embracing “a radical notion of futurity as unending flux,” adducing Lucretian pre­ce­dents for this view and Horace’s innovations upon t­ hese. 81. As the resulting “strained” tragic style, “oracular” content, and general “excess” described in lines 218–19 indicate, Brink 1971: 273. 82. Rudd 1989: 16 also sees Horace making a connection between nature and art.

125-81019_FerrisHill_HoracesArsPoetica_5P.indd 56

8/8/19 8:40 AM

hum a n o  57

Yet, as at 31 and ­later at 378 (si paulum summo decessit, vergit ad imum, “if it comes down a l­ittle from the peak, it sinks t­ oward the depths”), neither course of action is correct. The point is not merely that a person should not commit one type of error through attempting to avoid its opposite, but Horace works to show that this ­middle course perhaps paradoxically requires the greatest expenditure of effort. The poet must sweat, and even at times ­labor in vain (sudet multum frustraque laboret, 241), and he must be willing to embrace “the effort of polishing and the time [taken over it]” (limae ­labor et mora, 291)—­eminently physical, even painful, practices. In his explanation of how the pursuits of poetry, ­music, and athletics are not dissimilar in their requirements for success, Horace likewise insists on sweat and pain (both directly, sudavit et alsit, and through abstinence from vari­ous pleasures, 412–14), as well as on the fear a student should feel for his teacher (extimuitque magistrum, 415).83 At its core, however, poetry is produced by sapientia. Though conventionally translated as “wisdom,” the noun is formed from the verb sapere, the primary meaning of which is “to taste, to have good taste” (perhaps even evoking lingua as the organ of taste, mentioned already at 56, 111, 290, or the ideal “sweetness” of poetry, 99, 343).84 Horace had planted the verb at line 212 (indoctus quid enim saperet), leaving its import unarticulated ­until, already attuned to it, we see it again at 309: scribendi recte sapere est et principium et fons (“having taste/wisdom is the starting point and wellspring of writing properly”). In this way Horace ­w ill make his position within the “inspiration versus effort” debate more and more explicit (if still never pellucid) as the Ars Poetica progresses; but for now, having reached line 284, we are left with the strong impression that writing verse and performing it is a h­ uman concern and creation, rather than one with a divine source that speaks through ­humans as if through vessels (a privilege reserved for the Greeks, 323–24). While it may seem facile to say that poetry is made by ­people, the distinction between (or rather, interplay of) divine inspiration and/or inborn talent (often termed ingenium in the Roman tradition) on the one hand, and effort and hard work on the other in the ser­v ice of creative endeavor is a seminal one to the Ars Poetica. If Horace has presented us with a broadly positive view of humanity so far, portraying poetry and art as (good) products of the h­ uman mind and of individual and collective effort (202–19, 268–88; even if global changes are sometimes compelled by ­people’s baser impulses), the final 150 lines of the Ars Poetica are permeated by an altogether dif­fer­ent atmosphere. Images of degradation, 83. Compare Hesiod’s insistence on sweat and pain at Op. 288–92. 84. The widespread sense of “to be intelligent, to show good sense” appears only as the sixth of its seven meanings in the OLD. See further Massaro 1974 on sapere and sapientia in Horace.

58  ch a p te r 1

filth, failure, and disgust surface repeatedly, yet even this content is not unremittingly negative; rather, moments that are intended to disturb break the flow of thought with increasing frequency. We accordingly encounter Democritus, unkempt bodies, and poison (297–302) alongside the teachings of the poet, wisdom, Socrates, and duty (304–22). We are told, as at 99–100, that poets should both instruct and delight, a section interrupted at its midpoint by the grotesque image of a still-­living boy being ripped from the womb of the monster Lamia, who has apparently devoured him (333–46),85 and we may well won­der what instruction or delight is to be found in this. We find artistic beauty juxtaposed with error (347–65), heavy perfume and cloying sweetness marring an other­ wise enjoyable banquet (374–78), and cannibalism (or perhaps merely carnivorism)86 alongside wild animals tamed by Orpheus, the first of several mythistorical examples of the civilizing power of poetry (391–407). Fi­nally, ­there appear false friends, false praisers, and false mourners (419–37); skin disease, jaundice, and madness (453–54); a burping poet (457); and a suicidal phi­los­o­ pher (463–66, Empedocles),87 as the poem hurtles ­toward its close with the aid of a nameless someone who urinates on his f­ ather’s ashes (470–71), a crazed and violent bear (472–73), and a leech glutted with ­human blood (475–76). Notable among ­these images of the poem’s closing third is the appearance of humana natura (353). We have in a sense been expecting this phrase and entity since we became alert, through humano/humani (1, 102), to the Ars Poetica’s interest in ­people, and since the appearance of Natura, personified at 108. Horace at last combines them, and it becomes clear that humana natura is something altogether dif­fer­ent from Natura: whereas Natura was presented as an external force, one separate from ­people yet able to mold them and indeed charged with d­ oing so, humana natura would seem to be a quality or force within each person. The context is h­ uman fallibility and error, specifically, the natu­ral tendency (humana natura) not to take quite as much care over something as one should (parum cavit, 353). Horace distinguishes between shortcomings that 85. Horace’s deliberate conflation of the sites of alimentation and procreation—­Lamia is said to have eaten (pransae) the child, yet he ends up in her “womb,” alvo—­both draws attention to Lamia’s proclivity for eating c­ hildren (­after her own w ­ ere killed by Hera) and heightens the horror of her actions. Lamia, queen of the cannibalistic Laestrygonians (cf. Wiseman 1988: 11), appears also as the counterpart to Antiphates, their king, at AP 145. 86. The phrase caedibus et victu foedo, “slaughter and foul sustenance,” is variously interpreted as the consumption of wild animals (Pseudo-­Acro) or of other food not generally eaten by ­people of Horace’s day (Orelli 1844: 785), of meat in general (see Brink 1971: 387), or even of other ­humans (Brink 1971: 387, Rudd 1989: 214, Kiessling and Heinze 1914: 355: “Menschenfresser”). 87. Though Horace terms him not a phi­los­o­pher but “the Sicilian poet” (Siculi . . . ​poetae, 463).

hum a n o  59

deserve indulgence and forgiveness (sunt delicta tamen quibus ignovisse velimus, 347) and t­ hose that do not. In the latter category are included ­mistakes resulting from a refusal to improve oneself even in response to advice and correction (354–55, 438–52) and mediocre per­for­mances stemming from an inflated sense of one’s own abilities, itself resulting from a lack of appropriate training and willingness to put in the necessary effort (379–82, 389–90, 409–18). So, a writer warned to avoid an error he has already made “lacks indulgence” (venia caret, 355), and a lyre player who makes the same ­mistake repeatedly is laughed at (355–56). A par­tic­u­lar prob­lem that emerges is the repetition of error: “So the one who very much fails to take action becomes to me [like] that Choerilus, whom I marvel at as good with a smile/laugh two or three times” (sic mihi qui multum cessat fit Choerilus ille, / quem bis terve bonum cum risu miror, 357–58).88 In being descriptive rather than prescriptive in this manner, however, Horace’s tone becomes harder to discern: are we supposed to condemn the reaction of the audience members as insufficiently understanding, or are t­ hese examples of unforgivable errors? He clarifies slightly by saying, “But when many ­things gleam in a song-­poem, I w ­ ill not be offended by a few blots that e­ ither carelessness caused or ­human nature”—­that force internal and par­tic­u­lar to each of us, yet also shared by all p­ eople—­“took insufficient precautions over” (verum ubi plura nitent in carmine, non ego paucis / offendar maculis, quas aut incuria fudit / aut humana parum cavit natura, 351–53). The Horace that emerges is invitingly ­human, much like the flawed persona that is the speaker of the Satires:89 although he admits, acknowledging his own unbecoming irritability, that he “take[s] offense whenever good Homer becomes drowsy” (indignor quandoque bonus dormitat Homerus, 359), Horace concludes by forcing himself to be charitable: “But it is divinely permitted for sleep to creep over a long work” (verum operi longo fas est obrepere somnum, 360). In this closing couplet Horace thus allows us to view him as he engages in the pro­cess of forgiveness that he exhorts as an ideal: we witness his very natu­ral annoyance at a defect where one is not wanted or expected (359), quickly followed by a more charitable reframing of the shortcoming as understandable, and thus forgivable. Any apparent contradiction within the lines (is Homer to be forgiven or not?) is resolved in this way.

88. The verb cessat is h­ ere in its third sense of “to do nothing when something is expected” (OLD), as Horace criticizes the lyre player for failing to respond appropriately, or at all, to correct an error that has been pointed out to him. 89. See, e.g., Zetzel 1980, Hubbard 1991: 88–112, Freudenburg 1993 and 2001, Gowers 2003: 85, Plaza 2006: 189–221.

60  ch a p t e r 1

Continuing the interleaving of humanus and natura ­after the u­ nion of the two terms at 353, natura appears alone again at 408 (as it had at 108). Horace summarizes the crux that has occupied much of 289–407: “­W hether a praiseworthy song-poem is made by nature or by art was the ­matter u­ nder investigation” (natura fieret laudabile carmen an arte / quaesitum est). In this instance, as the counterpart to ars (expertise acquired through effort), natura seems to stand in for ingenium (“inborn talent”) and also to suggest an external force beyond the artist’s control (akin to divine inspiration). Unlike at 108, then, where she appears as an in­de­pen­dent agent, natura at 408 encompasses both the external force and the internal qualities produced by this force. The Greek counterpart to natura, physis, which Horace may well have in mind ­here, similarly possesses this dual sense,90 and like natura is derived from a verb that means “to be born, to become, to come into existence.”91 Lucretius’s “emphatically pro­cessual understanding of natura as a ‘coming to birth’ ” where “Nature is viewed as something that happens over time, involving pro­cesses of change—­and continuity—­ over time” seems a clear precursor to Horace’s entity in the Ars Poetica.92 Horace’s Natura/natura, like that of Lucretius, is neither bad nor good; she simply is, acting over long time scales yet discernibly inexorable even over shorter ones.93 Humana natura, placed into each of us by Natura (108–11), is subject to time and change if conceived of as evolving over the course of the h­ uman life cycle that Horace details at 158–76, but it is other­wise the governing force of ­every individual that cannot be deliberately acted upon to change. As a free-­ standing external entity, on the other hand, and the one responsible for at least some portion of artistic success, natura’s cooperation cannot be forced—it is ­either pre­sent or not. 90. LSJ s.v. φύσις lists u­ nder definition II the vari­ous ways in which it may mean “the natu­ral form or constitution of a person or t­ hing,” and ­under III, ­those that pertain to “the regular order of nature.” ­Under IV, which gives the uses of the term in philosophy, ­there may be found “Nature,” personified. 91. [G]nasci and φύειν, a semantic but not etymologically related pair. The OLD gives the primary meaning of natura as “the conditions of birth,” encompassing and determining “physical characteristics,” “character, ability,” “status, relationships, e­ tc.,” all of which show the noun hewing closely to its verbal source. 92. Hardie 2005: 19. On a similar concept of nature in Virgil’s Georgics, see Gale 2000, especially 18–57, 58–112. 93. Contrast the views found in Plato and Aristotle: as Solmsen 1963: 485–87 explains, “Plato has no confidence in physis; nothing good—no order, form, or meaningful structure—­could ever emerge from its erratic ways” (he adds, although Plato “refuses to commit himself definitely on his own account, we may take it that an approach crediting nature itself with inherent rationality or teleology would be for him ‘less likely’ ”), while Aristotle “has full confidence in nature. By its own devices and through its own resources physis is able to produce the best.”

125-81019_FerrisHill_HoracesArsPoetica_5P.indd 60

8/8/19 8:40 AM

hum a n o  61

Horace seems keenly and empathetically aware of h­ umans as imperfect strivers and as given to a set of reactions and be­hav­iors, ­whether individually or collectively. Throughout the Ars Poetica error, sin, and forgiveness overlap with ­human nature and ­human impulses. Venia first appears at the close of the poem’s opening vignette (hanc veniam petimusque damusque vicissim, 11) to describe the allowance that ­ought to be given to paint­ers and poets for daring in their creativity, and again twice at 263–67, where t­ hese instances further sandwich a mention of peccata, the “sins” of a poet who has attempted a difficult undertaking (though this is perhaps preferable to daring nothing at all). Horace, it is clear, is framing creative endeavor in profoundly ­human terms: on the one hand, ­humans, with all their flaws (shared and individual), create artistic products, and t­ hese products may involve h­ uman actors impersonating (­whether well or not so well) h­ uman characters; on the other, t­ hese artistic products are received by other ­humans, who are counseled to moderate their worse impulses and receive the artistic product, flaws and all, with the appropriate mea­sure of kindness and understanding. This kindness and understanding, however, is not to be limitless: in the case of an artist who refuses to work on his flaws, the advice-­giver is exempted from his obligation. We may also find ourselves wondering how self-­referential Horace is being, asking for our indulgence t­ oward his poem, blots and all, since by lines 359–60 we are reasonably permitted to won­der how much longer it can possibly go on. The relationship of art-­creator and art-­receiver in ­these manifold iterations presented from the Ars Poetica’s beginning through its end is, then, perhaps what the poem is, at its core, most “about,”94 as ­will be explored at greater length in the coming chapters. The picture that emerges by grades from the Ars Poetica, beginning with its opening image, through its excursuses on the development of drama, and into its exploration of inborn talent or inspiration as against painstaking effort, is of the creation of art as an exacting and unbeautiful process—­a picture that w ­ ill culminate in the final scenes of deranged artists t­ oward which the poem is building. This view is both very Horatian (he exhorts the same via the example of Lucilius at Satires 1.10.64–76, for example) and somewhat at odds with the conventional Greco-­Roman narrative of poetry as produced e­ ither by divine inspiration or by hard work (or some combination of the two), since in this tradition, the specifics of what “hard work” r­ eally entails are often elided. Horace stands out as unusual in his insistence on revealing what ars truly demands of its creator.95 94. Cf. Armstrong 1968: 34: “audience reaction is very much part of the theory of the Ars Poetica: spectatum admissi is only the first of many appeals to this criterion in the poem.” 95. Kilpatrick 1990: 39 also notes Horace’s emphasis throughout the AP on “how much ars ­really entails of work and self-­sacrifice.”

62  ch a p te r 1

Living Language Although the topic of language appears only slightly at 73–118—­where Horace appeared to set off in a new direction, though one nevertheless very much tied to what came before it—­these lines are sandwiched by two passages that deal quite openly with the ­matter, as Horace’s interests in all that humano signals find yet another outlet. Brink (1971: 127) labels 42–118, his section 2, “The Arts of Arrangement and Diction in Poetry,”96 but the phrase “in Poetry” is somewhat misleading, as Horace’s precepts apply, and are applied, far beyond the confines of poetry or lit­er­a­ture. His concern is with life and be­hav­ior, not merely of ­people, but also of words, which are treated almost as p­ eople. Horace explains how it is that a writer w ­ ill exhibit “fluency” (facundia) and “lucid arrangement” (lucidus ordo, 41): ordinis haec virtus erit et Venus, aut ego fallor, ut iam nunc dicat iam nunc debentia dici, pleraque differat et praesens in tempus omittat. hoc amet, hoc spernat promissi carminis auctor. in verbis etiam tenuis cautusque serendis dixeris egregie, notum si callida verbum reddiderit iunctura novum. (42–48) With the phrases qui scribitis (“you who write,” 38) and promissi carminis auctor Horace might appear to have the writing of poetry foremost in his mind, yet the words that resurface insistently are dicere (its three instances above italicized) and verbum (in bold). Facundia may well be “more expressive than verba” and fit the meter, “which elocutio does not,”97 yet this cannot fully explain why ­Horace chose it (and he does not shy away from using verba ­either ­here or elsewhere). Rather, in combination with the rhetorical term ordo (“arrangement”)98—­repeated immediately (as the last word of 41, the first of 42) in an arresting example of ordo itself—­facundia, a characteristic more usually applied to speech rather than to writing or lit­er­a­ture,99 creates a feeling of 96. He identifies five sections in all, the ­others being “Poetic unity and ars,” “Subject-­matter and character in poetry, exemplified by drama and epic,” “Drama,” and “The poet.” Section 2 is further subdivided into “Arrangement” and “Diction in poetry,” consisting of “Vocabulary,” “Norms of diction in poetic genres,” and “Styles of diction exemplified from drama.” 97. Brink 1971: 126. 98. Ibid.: 126–29 calls it “the label of the rhetoricians.” 99. The TLL gives “facultas oratoria” as its primary meaning, with the first subusage (A) encompassing spoken language, the second (B) literary style.

hu m a n o  63

spoken rather than of written language. While it may seem conventional or a form of shorthand to use “say” where “write” is meant, Lowrie has rightly argued for the distinction between the two in Horace’s Odes, throughout which the language of speech creates a “poetics of presence” allowing “a speaker (not a writer)” to direct “an utterance to an addressee in the ­here and now.”100 Horace’s same judicious preference for “say” over “write” in the Ars Poetica enables him to treat words and language beginning at 42–48 as living entities with their own aims and desires and suggests further that sermo, “conversation,” with its attendant aims of moral and aesthetic teaching, ­will be an under­lying current of the poem. While the overall context and premise may be the writing of poetry, the atmosphere Horace creates is thus one of h­ uman speech, an uneasy coexistence ­going far beyond the s­ imple fact that “poetry is made of words.”101 While lit­ er­a­ture is pre­sent in the guises of Caecilius and Plautus, Virgil and Varius (54–55), the first name dropped ­after the pointed emphasis on speaking at 42–48 is that of the Cethegi (50)—­orators, not writers102—as Horace emphasizes the words they hear (exaudita, 50), not ­those they read or write. Words are entities to be created and crafted (fingere, 50; ficta, 52), especially for t­ hings and ideas as yet unnamed (48–51).103 Even as he gives permission for this, Horace counsels moderation: we must exercise the right “prudently” (pudenter, 51)104 and make use of Greek antecedents sparingly (Graeco fonte . . . ​parce detorta, 53), advice that echoes the worldview of the Satires.105 And yet, he ends with the 100. See Lowrie 1997: 56–57 and more generally her chapter 2, “The Time of Writing and of Song”; on the idea of a “poetics of presence,” as she instructs, cf. also Feeney 1993. As Lowrie further points out, scribere is rare in the Odes, occurring only twice, both times in 1.6. 101. Brink 1971: 131. 102. Wickham 1891: 374 (ad Epist. 2.2.117) and Brink 1971: 142 both explain that one M. Cornelius Cethegus was consul in 204 BCE (see further Enn. Ann. 303–8 with Hardie 2005: 37, 2009: 59–60), but Horace’s plural suggests that he has in mind several orators or p­ eople of Cethegus’s day more generally (cf. Pseudo-­Acro, Rudd 1989: 137, 158). 103. Cf. Orelli 1844: 709, Wickham 1891: 394, and Lucr. DRN 1.136 with Hardie 2009: 58–59 and Brink 1971: ad loc. 104. The repetition in fingere and ficta is matched by that of sumpta pudenter and sumite materiam . . . ​cui lecta pudenter erit res (38–40). The reading pudenter is preferred by Shackleton Bailey 2001 over the MSS’s potenter, to produce the unparalleled sense of “in a way that accords with one’s abilities,” though it is closely linked to the verb posse, “to be able” (cf. Wickham 1891: 706). 105. Moderation in general passim (e.g., Sat. 1.1.28–107); mea­sured use of Greek at 1.10.20–35. The sense of parce detorta is debated: it ­either refers, as taken ­here, to opening the meta­phorical tap in a careful and stinting manner (so Brink 1971: 144: “the lock is not opened fully as it w ­ ere,”

64  ch a p t e r 1

importance of innovation (55–59): “why am I begrudged?” (ego cur . . . ​invideor?), Horace demands, employing a judicious first-­person pronoun and verb and a rhetorical question that all create intimacy and invite us to side with him,106 when Cato and Ennius did it, and when “it has been permitted and always ­will be permitted to strike a coin marked with the pre­sent’s symbol” (licuit semperque licebit / signatum praesente nota procudere nummum, 58–59).107 Within what we sense was a robust debate in con­temporary culture concerning the invention of new words,108 the approach Horace espouses is remarkably unconservative, and fully in line with the dictum of modern linguistics that all (living) languages are changing all the time109 and eminently reflective of the liveliness of the Ars Poetica. Horace engages also with the title of Lucretius’s De Rerum Natura in speaking of how an e­ arlier generation of writers “brought forth new names of ­things” (nova rerum / nomina protulerit, 57–58), paralleling his own abdita rerum (49), “the hidden ones of t­ hings,”110 in the same position eight lines ­earlier. Evocatively, ­these new names for ­things ­were produced by the “tongue” of Cato and Ennius, lingua ­going beyond its plainer sense of “language” ­because it has also “enriched the paternal conversation” (sermonem patrium ditaverit). Deploying the buzzword for his own satire, sermo, moreover, with sermonem patrium ditaverit, Horace engages with, reverses, and

Rudd 1989: 158) or to the small degree of deformation that words should undergo in their transition from Greek into Latin (the less preferable, though more widespread, interpretation: e.g., the anonymous twelfth-­century commentary at Hajdú 1993: 251, Blakeney 1928: 63–64, Antolín 2002: 185). 106. Horace’s plea is embodied in the Graecism/neologism invideor, which translates φθονοῦμαι (Brink 1971: 145, Rudd 1989: 159, Armstrong 1993: 198). 107. Shackleton Bailey accepts Luisinus’s emendation (also ­adopted by Bentley) of the last two words, though Brink 1971: 146 argues that the reading should rather be the producere nomen of the manuscripts. The point that age should not be the determiner of worth is made also at Epist. 2.1.28–49 (see further chapter 2). 108. Brink 1971: 145 lists comparable remarks from Cicero and Varro; see also further Rawson 1985 (especially 117–31) and Feeney 2009: 36. Impor­tant h­ ere may be the theories of language associated with the vari­ous philosophical schools in Philodemus, On Poems 1; see e.g., col. 83–91, 120–26, 131–32 with Janko 1995, 2000: 165–89. 109. Campbell 1924: 241 and Hardie 2009: 57 also note Horace’s “judicious modernity” and “opposition to linguistic and literary conservatives.” Pseudo-­Acro already knew “cottidie locutiones mutantur,” and Brink 1971: 147 cites Varro, Ling. 5.3–5, 9.17 as evincing the same (see further Ferriss-­Hill 2014); on Horace’s own view of language as changing and inherently changeable, see Orelli 1844: 711–12, Wickham 1891: 395. 110. ­There is also an echo of Virgil’s lacrimae rerum (Aen. 1.462).

hu m a n o  65

rejects Lucretius’s patrii sermonis egestas, “poverty of paternal conversation” (DRN 1.832).111 The suggestion of birth inherent in protulerit (58),112 also nodded at with the adjective patrium, comes to full fruition in what follows: ut silvae foliis privos mutantur in annos, prima cadunt * * * * * * * * * * * * * * ita verborum vetus interit aetas et iuvenum ritu florent modo nata vigentque.113 (60–62) The life cycle Horace describes borrows a simile familiar from both Homer and Virgil,114 yet as in the poem’s opening image with its equally hoary literary ancestry, Horace remakes the trope for his own purposes to produce what Brink has called “poetically . . . ​perhaps the most remarkable piece of the Ars.”115 Horace interleaves ­human speech with the h­ uman life cycle, his focus on the life of words as they exist in speech, rather than in poetry. Using the language of “social standing”116 that aptly allows words and speech to live, not merely exist, Horace speaks of “the honor and lively charm of conversations” (sermonum . . . ​ honos et gratia vivax),117 as he explains that this is even less lasting (nedum . . . ​ stet)118 than “mortal deeds” (mortalia facta, 68–69). While the first order point 111. Brink 1971: 145 also notes the correspondence: sermo and patrius are repeated, ditaverit provides the cure for egestas (cf. Epist. 2.2.121), and lingua occurs in both passages as well (DRN 831). Horace’s view of naming, whereby he understands ­things as existing prior to their being given a name (evident in indiciis monstrare recentibus abdita rerum and nova rerum / nomina), is shared with Lucretius; see further Shearin 2015. 112. OLD s.v. profero 3. 113. Ribbeck proposed the lacuna dividing line 61 across two lines to mitigate the other­wise extremely compressed simile (ut . . . ​ita); he is followed in this by Brink 1971, Shackleton Bailey 2001. A second textual prob­lem lies in silvae foliis, variously transmitted as folia in silvis, foliis silve, or silvae flores; in addition, Bentley emended pronos to privos. 114. Brink 1971: 147 lists Il. 6.146–49 and Aen. 6.309–10, as well as instances from lyric, rightly noting that Horace’s version is “threefold—­leaves, words, ­humans” in contrast to his “two-­sided” models. 115. Ibid.: 146, referring to lines 60–72. 116. Ibid.: 157 on honos et gratia. 117. Ibid.: 157 notes also “the biological meta­phor” inherent in vivax, “retentive of life, enduring.” 118. As Brink 1971: 156 notes, nedum (which he accepts as the correct reading, though ­others have favored haud) “occurs only ­here in H. and is rare in all Latin verse.” Brink takes sermonum as meaning “diction,” i.e., the use of words (verba) rather than verba themselves, rejecting the pos­si­ble meaning “conversation.”

66  ch a p te r 1

is that no m ­ atter how lively and captivating our conversations may be, they are ephemeral (a sad fact emphasized by the end-­of-­line placement of vivax directly below peribunt), ­there is surely a wink h­ ere at his own Satires, which he termed sermones,119 especially through the emphasis on “lively charm,” the aim of writings in the Callimachean tradition, that manage, through obsessive honing, to appear unstudied. And yet, while reminding us that all living ­things die, Horace offers up in the very next line the possibility of rebirth, not for entire conversations, but for individual words: multa renascentur quae iam cecidere (70). Faint tendrils from humano may be discerned extending to t­ hese points in the poem, into h­ uman language, and into the birth and death of words and conversations. The rings of death and rebirth at 60–62 and 70–72 enclose within them the “Augustan building proj­ect” vignette, which illustrates the power and unpredictability of natu­ral forces, even without any explicit appearance by Natura: sive receptus terra Neptunus classis Aquilonibus arcet, regis opus, sterilisve †diu palus† aptaque remis vicinas urbis alit et grave sentit aratrum, seu cursum mutavit iniquum frugibus amnis doctus iter melius. (63–68) While scholarly discussion of this passage has concentrated on which specific con­temporary or recent feats of engineering (if any) Horace has in mind,120 the correlative idea pre­sent, though it seems to have gotten rather lost, is of the erratic and uncontrollable be­hav­ior of ­water.121 With echoes of the Odes,122 Horace explains that ­human efforts may for a time carve out a sea harbor,123 drain a swamp to yield farmland, or alter the course of a river, yet Nature merely needs to play the long game and wait since “mortal deeds ­will perish” (mortalia facta peribunt, 68) and she w ­ ill one day reclaim all of her rightful property. Th ­ ere 119. Cf. Sat. 1.4.42, 48, Epist. 1.4.2, 2.1.250, 2.2.60. Though Wickham 1891: 396 glosses sermonum rather blandly as “language,” he notes that the plural is “unusual,” which is precisely why it may be seen to recall Horace’s Sermones. 120. Brink 1971: 151 also thinks it is beside the point ­whether the three examples listed correspond to par­tic­u­lar building proj­ects as, e.g., Orelli 1844: 712–13, Wickham 1891: 395–96, Wilkins 1896: 347–48, Rostagni 1930: 19–20 seek to establish, following in the footsteps of the scholia and the “Materia” commentary; the tendency persists as late as Fedeli 1997: 1490–91, Antolín 2002: 187–88. 121. On the uncertain reading palus, however, see Brink 1971: 154. 122. E.g., Carm. 3.1.32–33, 3.30.1–5 (cf. Brink 1971: 151). 123. As sive receptus terra Neptunus has been understood since Pseudo-­Acro.

hu m a n o  67

is accordingly a strong sense of hubris and futility in the specific “­great h­ uman enterprises”124 listed. Whereas Horace had proclaimed his Odes if not exactly eternal, then certainly longer-­lasting than physical monuments, ­here he characterizes ­human speech—­composed, like poetry, of words—as naturally less long-­lived than any physical construction (nedum sermonum stet honos et gratia vivax, 69).125 ­These musings on how the conversations of a person’s lifetime, however gratifying and lively, are doomed to be lost to history conceal within them a glance at Horace’s own sermones, his Satires, equally conversational and ephemeral. Horace suggests that ­there is a hierarchy of longevity to ­human creations, with lyr­ics at the top, built objects in the m ­ iddle, and words, conversations, and the genre of satire at the bottom. The grandeur of the building proj­ ects described, the likes of which must have surrounded Horace in Rome, is undercut and subverted by mortalia facta peribunt (68)—­a sentiment that links the passage with Horace’s repeated injunctions to undertake literary proj­ ects with an awareness of one’s own talents and limitations. Nature, so pre­sent even if not named outright at 60–72, is given a curious counterpart in the form of another intangible personification—­usus. Horace concludes by explaining that the mechanisms of birth, death, and rebirth that he has outlined at 70–71 lie in the hands of this “usage.”126 At Epistles 2.2.109–19 Horace invokes the same entity, dubbing it “begetter” (genitor) of new words and viewing the ­whole system as governed by a single poet acting as censor, who in his creation of a literary work makes careful se­lections as to which words he wishes to include, which not.127 This aligns with the ways in which linguistic change is envisioned elsewhere, including during late antiquity, where the grammarian reprises the role of censor-­figure.128 In the Ars Poetica, however, and in keeping with the sense that permeates the work that ­there exist forces far larger than individuals, usus appears more as a sort of precursor to Adam Smith’s invisible hand. That Horace expects his reader to be surprised and confused by 124. Brink 1971: 151. 125. Ibid.: 146 is uncertain as to Horace’s overall point (it “seems to be simply to give poetic ‘body’ to the idea that language is as prone to change as other creations of man”), but this rather underrates the force of nedum and overlooks the nuances of sermonum. Horace may also have in mind at 69 how the language of conversation changes over generations (Hardie 2005: 37 somewhat similarly sees Horace illustrating “the unceasing change of fashion in linguistic usage”). 126. Brink 1971: 158–59 argues persuasively for this meaning (so also Hardie 2005: 39). 127. Cf. Reckford 1969: 139, Oliensis 1998: 212–13, Feeney 2009: 36 (“very broadly, in the Epistle to Florus Horace stresses individual control over language, while in the Ars the emphasis is much more on language as a natu­ral pro­cess which is in a state of drift beyond the power of any individual’s ultimate control”). 128. See Kaster 1988, especially 139–68, 193–95 on the fifth/sixth-­century CE grammarian Pompeius’s discussion of usus.

68  ch a p t e r 1

this character whom he has just introduced into his poem is clear from line 72, where he explicates usus further with the aid of three criteria: in its hands (the preposition penes being far more common with p­ eople than inanimate objects)129 lie arbitrium, ius, and norma loquendi, glossed by Rudd as “decision,” “authority,” and “rule.”130 A fuller sense of arbitrium, however, is “the action or pro­cess of deciding”; of ius, “that which is sanctioned or ordained; a ­legal system or code; that which is good and right, equity”; and of norma, “a standard/pattern (of practice or be­hav­ior),” as in En­glish “norm.” Th ­ ese three qualities, distinct yet complementary, together make up the fullness of usus and contribute to the sense of a dynamic, ever-­evolving, non-­static, self-­governing system that instantiates “the princi­ple of flux.”131 Usus in the Ars Poetica also comes as something of a surprise on the heels of 53–57, where the license to create new words is requested for individual, named literary figures (Caecilius, Plautus, Virgil, Varius, Horace himself, Cato, and Ennius), a mini-­canon of word-­ inventors that is itself an aberration in the context of spoken rather than written language, where Horace portrays the creation of new words and the dispensing with old ones as a continual pro­cess resulting naturally from the wear and tear caused by any and all who make use of Latin.132 Seminal, then, is the idea of interaction: one individual cannot control usus, which requires “habitual dealings between persons, intercourse,”133 just as speaking (loquendi, particularly with norma) implies dialogue between two or more parties, and arbitrium and ius an ongoing (and thus multiparty) pro­cess. In this par­tic­u­lar regard Horace’s usus in the Ars Poetica has been widely misunderstood: ­there is no justification other than the widespread assumption that he must be talking about poetry and written language (it is the “Ars Poetica” ­after all) for such explanations for usus as “the norms of educated speech” or the claim that “a poet can in a small way help to shape usage,”134 as is the case in Epistles 2.2. Rather, in the Ars Poetica Horace 129. Cf. also Brink’s 1971: 158 comment on si volet, where he notes that “usus—­like utilitas and consuetudo . . . ​—is often credited with an active force. This suits what H. wants to do in this passage: to lay stress on the impersonal or ‘natu­ral’ features of language” (cf. 175 ad 89 where he similarly notes of the phrase versibus exponi tragicis res comica non vult that “key terms easily acquire animate status,” or how at 202–5 the tibia is treated as “quasi-­personal,” p. 265). 130. Rudd 1989: 162–63. On ­these three terms, along with usus itself, see further Ruch 1963. 131. Campbell 1924: 240. 132. Porphyrio (cf. Kaster 1988: 195, n76) likewise views usus as governing spoken language (“nihil enim aliud est quam regula sermonis Latini”), and Orelli 1844: 715 also de-­emphasizes written language and stresses the influence of p­ eople at large, not merely poets, in shaping how p­ eople speak. 133. OLD s.v. usus 10. 134. Rudd 1989: 162, emphasis mine (though he adds that “in the main, as H. implies, the forces at work are large, imperfectly understood, and beyond the control of any individual,” 163). Brink 1971: 146, in saying that “language ­here is not the poet’s creation, but the common

hu m a n o  69

is using the example of language as one sphere among many to instantiate and reveal what may organically result from ­human interaction. The concerns of lines 42–70 are evident also at 128–34, and the passages embrace the intervening discussion of genres and their emotional effects: †difficile est† proprie communia dicere, tuque rectius Iliacum carmen diducis in actus quam si proferres ignota indictaque primus . . . nec verbo verbum curabis reddere fidus interpres. (128–30, 133–34) Both 42–70 and 128–34 are marked by an emphatic recurrence of words for speaking. In the latter are found: dicere and indicta, in almost the same line position and with diducis between them creating further assonance; the double repetition of verbum, as at 46–47 (and in 52 and 61), reinforcing the pervasive interest in speech and words; and interpres (134), last seen in conjunction with lingua, the instrument of speech (111), which forms a pointed “hook” connecting back to line 111, especially since fidus interpres is tautological, given the phrase verbo verbum just before.135 In the interval between 42–48 and 128–34, moreover, may be seen a crescendo of clustering, as verbum reappears insistently again and again.136 In ways such as t­ hese, as Horace explores emotions and the expression of them through words, all with humano in the background, he also begins crafting for his Ars Poetica a remarkable auditory texture through judicious repetition of specific terms.137

Appropriateness and the Satires The roles played by h­ uman be­hav­ior and emotions and by language throughout the Ars Poetica seem out of proportion to their ostensible usefulness for prospective writers, suggesting that Horace is interested in them for their own language which the poets share,” similarly overstates the extent to which Horace views language as belonging to poets rather than to ­people more generally. 135. So Rudd 1989: 172: “The fidelity has already been explained.” Horace’s formulation may owe something to Cicero, De optimo genere oratorum V (=14); cf. Orelli 1844: 730. 136. At 52, 61, 97, 106, and twice in 133—­instances all in close succession (with vocabulum also at 71). The remaining appearances at 235, 311, and 443 are rather more isolated, underscoring the clustering in the poem’s first quarter. 137. Pickering 2003 argues that both ancient Greek and Roman writers ­were sensitive to repetition, regarding it in general as undesirable (­because clumsy or careless), though prizing it if done deliberately and well; see further also ­Wills 1996.

70  ch a p t e r 1

sake, as he is in his Satires. The Ars Poetica evinces also an outsized concern with crafting characters that speak and behave in a believable manner given their age, social standing, or other characteristics. This theme of appropriateness is among ­those the poem is best known for, and it is seen as taking up Aristotle’s τὸ πρέπον, a concept it expresses predominantly with the verb decere (often rendered as proprietas in the secondary lit­er­a­ture).138 We see this concern first at 101–5, where the ostensible point, at least initially, is that a character burdened with a famous name and a specific legacy (namely, Telephus and Peleus) must be recognizable as that character. The topic appears in more extensive form at 114–18: intererit multum divusne loquatur an heros, maturusne senex an adhuc florente iuventa fervidus, et matrona potens an sedula nutrix, mercatorne vagus cultorne virentis agelli, Colchus an Assyrius, Thebis nutritus an Argis. (114–18) The scholiasts’ comments help to illuminate Horace’s point: being a Colchian or an Assyrian is not merely a happenstance of birth; rather, a character’s nationality contains and expresses to a broader audience certain essential aspects of his personality. What­ever the (dubious) merits of the specific traits he gives (that Colchians are fierce, Assyrians clever and rich, Thebans unlucky, Argives lucky), Pseudo-­Acro shows that for Horace and Romans generally, describing a man in terms of his place of birth was as indicative of his expected temperament, be­hav­ior, and speech as it was to specify that someone was a divinity or a hero, young or old. The section, which began with a warning that incongruence ­will produce laughter (112–13), culminates in breathless, asyndetic lists of the emotions appropriate to yet more named figures.139 Horace explains that Achilles must be “energetic, hot-­tempered, relentless, fierce” (impiger, iracundus, inexorabilis, acer, 121). Likewise, “let Medea be ferocious and unconquered, Ino tearful, Ixion deceitful, Io a wanderer, Orestes unhappy” (sit Medea ferox invictaque, flebilis Ino, perfidus Ixion, Io vaga, tristis Orestes, 123–24), a motley cast bound together by insistent assonance of “i” and “o” (Ino, Ixion, Io, Orestes) as well as the lighter alliteration of Medea, Ino, Ixion, along with the fact that each 138. Cf. Brink 1971: 174 on decentem (92; cf. also 106, decent); Horace does not in the AP use decorum, the ordinary ancient term for what he is describing. Brink 1971: 464 is right to note that “appropriateness” is no single ­thing for Horace in the AP, and this may have a pre­ce­dent in Philodemus (see, e.g., On Poems 1 col. 136 with Janko 2000: 355, n2 or col. 9 of book 3 with Janko 2011: 55). First used in the “Materia” commentary, proprietas would become pop­u­lar­ized by Landinus as the standard term for this Horatio-­Aristotelian idea (see further Friis-­Jensen 1995a: 236). 139. Cf. Brink 1971: 200, who notes that this is effective in an ethologia, as also at 163–65, 172–73.

hu m a n o  71

comes with a fixed and well-­known backstory.140 ­These figures are furthermore strongly linked to line 118 by the fact that Medea was from Colchis (Colchus; she reappears again at 185), Orestes from Mycenae/Argos, that is, Argive (though Argis is also auditorily suggestive of the Argo, the ship on which Medea’s Jason had sailed), Io also from Argos, and Ino a queen of Thebes (Thebis).141 What­ ever ­else they are d­ oing, then, the lists (series) of character-­types, names, and places at 114–18 and 120–24 constitute a iunctura, a “join”—­yet another instance of two separate points in the poem allied through sense and sound. ­These terms, series and iunctura, individually and collectively essential to appreciating the textural quality of the Ars Poetica, ­were already introduced at 46–48, and although the reader may need 242 to grasp that they form the basis of Horace’s poetics, as discussed below, they may be seen on display throughout. Horace’s three criteria for appropriateness in characters replicate t­ hose of Aristotle: appropriate be­hav­ior ­under the circumstances (AP 89–98, Rh. 1408a12–13); the appropriate emotional response, pathos, to stimuli (AP 99–111, Rh. 1408a16–25); and ethos (AP 114–18, Rh. 1408a25ff.), that is, be­hav­ior fitting to one’s demographic qualities.142 Yet t­ hese exist to illustrate the culminating tidbits of advice given at 119 and 127: what­ever you do, “create ­things that are in keeping with themselves” (sibi con­ve­nientia finge) and “let it be consistent with itself ” (sibi constet). Th ­ ese formulations, with their clear interdependence, and the worldview ­behind them, reflect in turn Horace’s Satires, especially the first three poems of the first book. It is commonly agreed in the case of Horace’s two books of Satires, as it is not for the Ars Poetica, that they have ­human be­hav­ior as their primary focus, what Juvenal would ­later describe as, “what­ever men do—­vows, fear, anger, plea­sure, joys, to-­ing and fro-­ing” (quidquid agunt homines, votum, timor, ira, voluptas, / gaudia, discursus, 1.85–86). In Satires 1.1, dubbed “The Race for Wealth and Position,”143 Horace criticizes the ­human bent for avarice (38–40, 41–42). He provides one illustration ­after another of the dangers of being too acquisitive, even too parsimonious, yet his interlocutor persists in misunderstanding the poet’s advice as an instruction to run to the opposite extreme (101–5). The point, Horace is at last forced to spell out, is this: “­There is moderation in ­things, t­ here are certain limits, in short, 140. The “Materia” commentary emphasizes the fama (AP 119) attached to each figure (Friis-­ Jensen 1990: 352–53). 141. Only Assyria (not Greek) remains unrepresented in the second iteration. Thebis prefigures Cadmus (187) since he was the founder of the city. 142. See Brink 1971: 174, 182, 190, and Prol. 97–99. 143. ­These titles attached to the Satires appear in the En­glish tradition of the past few centuries, and they are helpful in that they sum up each poem’s main topic. The following pages are an abridged version of Ferriss-­Hill 2018.

72  ch a p te r 1

beyond or before which Right does not wish to stand” (est modus in rebus, sunt certi denique fines / quos ultra citraque nequit consistere rectum, 106–7). All one needs, in fact, are ­those basic items without which “­human nature” (humana . . . ​ natura, 75) would feel pain—­one of only two instances in Horace of the phrase outside the Ars Poetica, and also in conjunction with an emotional response.144 Several pairs of “fools” (stulti) in Satires 1.2, “The Folly of ­Running to Extremes,” further exemplify the failing of r­ unning into other vices by attempting to avoid their opposites (dum vitant stulti vitia, in contraria currunt, 24), and ­here the maxim is, “­there is no mean” (nil medium est, 28). The third poem of book 1, “On Mutual Forbearance,” describes at some length a single individual who is internally inconsistent, a vignette Horace prefaces with “­there was never anything balanced in that man” (nil aequale homini fuit illi, 9) and rounds off with “nothing was ever so inconsistent with itself ” (nil fuit umquam / sic inpar sibi, 18–19). ­These failings of be­hav­ior, ethics, and morals that run throughout the first three satires of book 1 are the very same failings Horace warns against in his Ars Poetica as pertaining ostensibly to artistic and literary production, and the maxims that frame the vignettes and demarcate them as significant in Satires 1.1–1.3 are found in reconfigured form throughout the Ars Poetica. Nil aequale homini fuit illi (Sat. 1.3.9) and nil fuit umquam sic impar sibi (18–19) foreshadow sibi con­ ve­nientia (AP 119) and sibi constet (AP 127). Dum vitant stulti vitia, in contraria currunt (Sat. 1.2.24) becomes in vitium ducit culpae fuga, si caret arte (AP 31),145 with vitium repeated, the position of the preposition in shifted, and stulti reworked as si caret arte. And est modus in rebus (Sat. 1.1.106), further explicated through a personification of “Right” with its feet firmly planted within certain bounds (sunt certi denique fines / quos ultra citraque nequit consistere rectum, 106–7), and nil medium est (Sat. 1.2.28) anticipate the Ars Poetica’s continually stated need for moderation and the consistent presence of Right (25, 367; cf. 129, 140, 309, 319, 369, 422, 428). Horace also works in Satires 1.3 to make a distinction between minor and serious failings (19–20, 96–97), emphasizing the importance of forgiveness, especially in the context of familial relationships or friendships (23–28, 32–67, 74–75, 139–42), and this finds itself expressed in the Ars Poetica as “­there are, however, ­mistakes that we would wish to forgive” (sunt delicta tamen quibus ignovisse velimus, 347). Furthermore, the characters Horace parades before us at Ars Poetica 114–18 markedly resemble ­those who appear at the beginning of Satires 1.1: 144. The other is at Epist. 2.2.188, in a reference to “Genius, the divinity of h­ uman nature” (Genius . . . ​naturae deus humanae). The phrase occurs also in Cicero (e.g., Phil. 13.1, Rep. 1.22, Tusc. 4.62, Nat. D. 2.133). 145. Noted already by Pseudo-­Acro ad AP 31, as well as by Wickham 1891: 392 and Brink 1971: 115.

hu m a n o  73

‘o fortunati mercatores!’ gravis annis miles ait, multo iam fractus membra labore. contra mercator navem iactantibus Austris ‘militia est potior. quid enim? concurritur: horae momento cita mors venit aut victoria laeta.’ agricolam laudat iuris legumque peritus, sub galli cantum consultor ubi ostia pulsat. ille, datis vadibus qui rure extractus in urbem est, solos felicis viventis clamat in urbe. (Sat. 1.1.4–12) “O lucky merchants!,” a soldier heavy with years cries, his limbs now broken by much work. On the other hand the merchant, when the South winds are tossing about his ship, says, “Soldiering is better. Why? Th ­ ere is an engagement: in an instant of time e­ ither swift death comes or happy victory.” When a client beats on his door before the cock’s crow, the one skilled in justice and the laws praises the farmer. The latter, who has paid his surety and been dragged from the countryside into the city, exclaims that only ­those who live at Rome are happy. The mercator is transferred w ­ holesale to Ars Poetica 117, his maritime travels condensed into the single adjective vagus; the lengthy image of the farmer’s life becomes cultor . . . ​virentis agelli, the two ele­ments that make up agri-­cola reversed and refashioned (AP 117);146 while the aged soldier anticipates the senex that ­will appear shortly at Ars Poetica 169–76.147 We begin to sense the ways in which the Ars Poetica forms a counterpart to the Satires: in the former Horace repeatedly states the need for writers to craft characters that behave as they are expected to, as though this ­were the end in itself, yet it is comparison with the Satires where such types, already crafted by a writer (Horace himself), are set before our eyes, that allows us to see that the point may rather be the ser­vice in which such characters are deployed, namely, to illustrate notions such as sibi constet or the undesirability of nil medium est. The emphasis in the Satires is on 146. On the reading virentis see chapter 2, n144. 147. Notably, however, the characters in the first twenty lines of Sat. 1.1 fail to illustrate what Horace prescribes at AP 114–18: even as Horace pretends to be quoting directly their very words, the farmer sounds just like the l­ awyer, and the merchant just like the soldier. Even the divinity (divusne loquatur, AP 114) is not distinguished by his speech at Sat. 1.1.15–22: although the archaic form voltis seems appropriately portentous, Jupiter makes liberal use of colloquial interjections such as en and eia (the former being especially common in Plautus and Terence and in numerous other authors in direct speech and address), and he proves petulant and pedestrian in his conduct as well (Horace imagines him angrily puffing out his cheeks, described with the quotidian word buccas over the loftier os, and sulking).

74  ch a p te r 1

the proper be­hav­ior of p­ eople in the world (illustrated by characters created by a poet), while the Ars Poetica’s insistent focus remains on the minutiae of character-­creation, yet t­ hese apparent mirror images are not merely “the Aristotelian Mean” of the Satires “transferred from morals to poetry.”148 Rather, the Ars Poetica’s tenacious hold on its label of ars has obscured the poem’s overarching concern with the same themes that run throughout the Satires—­ quidquid agunt homines, defined as the feeling and expression of votum, timor, ira, voluptas, gaudia, discursus. Not content with having twice told his reader that characters must speak in the appropriate manner, Horace restates his point again at Ars Poetica 158–78, and sets up a still more explicit connection with the Satires. Expanding upon the life cycle of words described at 60–72, Horace now pre­sents a ­human life cycle, describing the traits and habits of each representative of its four stages.149 This passage describing the course of a man’s life is tied to the one describing words and leaves through the repetition of key words including aetas (61, 166; cf. 156) and annus (60, 175; cf. 157, 160, in horas)150—­natu­ral enough terms for the subject m ­ atter at hand, yet often employed in a slightly odd way that draws further attention to them:151 tu quid ego et populus mecum desideret audi: si plausoris eges aulaea manentis et usque sessuri donec cantor ‘vos plaudite’ dicat, aetatis cuiusque notandi sunt tibi mores mobilibusque decor naturis dandus et annis. reddere qui voces iam scit puer et pede certo signat humum, gestit paribus colludere, et iram concipit ac ponit temere et mutatur in horas. imberbis iuvenis, tandem custode remoto, gaudet equis canibusque et aprici gramine Campi, cereus in vitium flecti, monitoribus asper, utilium tardus provisor, prodigus aeris, 148. Brink 1971: 115. 149. As against Aristotle’s three, Rh. 1389a–90b; Brink 1971: 229 won­ders why, and notes further the difference that Horace’s treatment is “chronological” as opposed to Aristotle’s “logical schema: youth, old age, maturity,” while Williams 1968: 333 rightly sees that where Aristotle’s passage is “analytical and descriptive,” Horace’s is “evocative, impressionistic, poetic,” in keeping with Horace’s aims and desire to “pre­sent a picture of ­human life from beginning to end: at first, like the tide coming in, full of promise, at the end, the inexorable ebb.” 150. Brink 1971: 241: “175 anni resumes 157 annis.” 151. Consider the unusual or highly poetic aetas animusque virilis / quaerit opes (166–67), and the much-­debated phrase in annos (60; Brink 1971: 149).

hu m a n o  75

sublimis cupidusque et amata relinquere pernix. conversis studiis aetas animusque virilis quaerit opes et amicitias, inservit honori, commisisse cavet quod mox mutare laboret. multa senem circumveniunt incommoda, vel quod quaerit et inventis miser abstinet ac timet uti vel quod res omnis timide gelideque ministrat, dilator, spe †longus†, iners

avidusque futuri, difficilis, querulus, laudator temporis acti se puero, castigator censorque minorum. multa ferunt anni venientes commoda secum, multa recedentes adimunt. ne forte seniles mandentur iuveni partes pueroque viriles, semper in adiunctis aevoque morabimur aptis. (153–78) In the first place, the passage reveals yet again Horace’s fascination with ­human be­hav­ior (in a way that goes far beyond literary composition) and with emotions: a very young boy, now talking and walking, is prone to tantrums and outbursts of anger; the youth rejoices; and the old man is afflicted with fear and unhappiness. As Brink puts it, what could be “more Horatian, in Odes, Satires, and Epistles alike, than this perceptive eye for typical ­human features?”152 Whereas his emphasis previously was on the assigning of suitable dialogue to a variety of characters, Horace’s focus ­here appears to be on actions and be­hav­ ior rather than on speech (though he concludes with speech, 176–78, thus continuing and solidifying the verbal connections that have been forged so far throughout the poem). The emphasis on be­hav­ior at 153–78 is communicated through a notable clustering of agent nouns, words that embody the doer of an action and thus concretize and add vividness to a par­tic­u­lar be­hav­ior. Formed by adding -­or onto a verbal stem,153 ­these are markedly prevalent in the Ars Poetica, and no fewer than six are found in the passage above, with a further two just before (plausor and cantor, 154–55): monitor, provisor, dilator, laudator, castigator, and censor.154 Such nouns serve (logically) to emphasize the action each communicates, but whereas a relative clause describes a person who does a par­tic­u­lar action, t­ hese agent nouns define the person in question as a doer of that action, almost to the 152. Brink 1971: 226. 153. See Brink 1971: 214, 236 (“a fertile source of new words”). 154. By way of comparison, the entirety of Epistles 1 (1,005 lines) contains twenty-­eight agent nouns while the AP contains forty in a poem less than half as long. Brink 1971: 236 further notes that provisor and dilator are likely Horatian coinages.

76  ch a p t e r 1

exclusion of all ­else. Since Horace also uses relative clauses in the Ars Poetica to describe persons d­ oing actions, he seems to be laying the groundwork for a bipartite schema where some are dabblers, o­ thers genuine “doers.” A title that results from a consistent pattern of be­hav­ior may accordingly confer praise or blame. Amphion truly stands as the “founder,” conditor, of Thebes (394), Apollo as paradigmatic cantor, “singer” (407), and Aeschylus as “discoverer,” repertor, of the mask (278), just as Licinus is in fact a barber, tonsor (301; this seems relatively neutral, as are, e.g., actors and victors, 84, 193, 208, 369, or auditor, 100; cf. also 149, 154–55, 163, 173–74, 344). Alternatively, the effect may be negative, as in the case of the painter, pictor, from the poem’s opening; the “praiser” (laudatore, 433), who habitually provides the poor form of critical reception that gives him this designation; the “yes-­men,” assentatores, of line 420; or the “­bitter reciter” at the poem’s conclusion (474), who merits the title recitator acerbus through his habit of inflicting recitations of his work on ­others. This emphasis on mono-­competence, ­here and throughout the passage, suggests the stock characters typical of New and Roman Comedy, each with their assigned and often rather two-­dimensional attributes.155 Quite how all of this is meant to ensure that a writer avoids giving old men’s parts to a young one and men’s parts to a boy (176–77), however, remains unclear. Although the “beardless youth” who loves exercising on the Campus Martius appears in the Odes,156 child-­characters such as the young boy described in the preceding lines are all-­but-­nonexistent in Roman lit­er­a­ture generally. Similarly, while the statement that once this hy­po­thet­i­cal boy/youth has become a man he “seeks resources and friendships,” loves “honor,”157 and has developed a degree of impulse control (168) may well be accurate in broad strokes, it is hopelessly vague as advice on how actually to create such a character. Fi­nally, the old man who forms the culmination to this life cycle seems entirely a caricature: besieged by aches and pains (presumably what multa . . . ​incommoda means), unhappy, fearful (170–71), and ornery (173), he is defined by nostalgia for a bygone era that manifests itself in shouting at young ­people (173–74). Perhaps we are merely supposed to be grateful that Horace is at long last giving us a positive 155. Rostagni 1930: 51, e.g., notes such resonances of Roman comedy in equis canibusque and monitoribus asper. 156. See Nisbet and Hubbard 1970: 108–15 on Carm. 1.8, and on the Campus Martius as a place for equestrian and other athletic exercises, see also Carm. 3.7.25–26 and 3.12.7–9 with Armstrong 1993: 204. 157. With a nod at the cursus honorum (Rostagni 1930: 51) that an aristocratic male of this age would naturally begin to embark upon.

hu m a n o  77

instruction (morabimur, “we w ­ ill take our time over”)158 rather than telling us what not to do, in contrast to “the almost wholly negative value” of the notable majority of the advice dispensed in the Ars Poetica.159 Equally unhelpful yet detailed are the instructions on how properly to write a chorus given some twenty lines ­later: actoris partis chorus officiumque virile defendat; neu quid medios intercinat actus, quod non proposito conducat et haereat apte. ille bonis faveatque et consilietur amice et regat iratos et amet peccare timentis; ille dapes laudet mensae brevis, ille salubrem iustitiam legesque et apertis otia portis; ille tegat commissa deosque precetur et oret, ut redeat miseris, abeat Fortuna superbis. (193–201) The first princi­ple articulated, “let the chorus protect the actor’s parts and the manly task,” is less straightforward than it might appear, and we are perhaps beguiled by partis and virile, which echo partes . . . ​viriles (177), into feeling that we understand the words ­because we have seen a version of them before. Although defendat suggests that the chorus’s task is to “provide support” for the actor, the commentaries universally agree that Horace, with Aristotle’s Poetics 1456a25–27 in mind,160 regards the chorus as itself having an actor-­like role in the play. Horace accords the chorus a sequence of verbs, introducing them with three line-­initial emphatic deployments of ille that point at this actorly body (196, 198, 200; ille repeated mid-­line at 198). ­These verbs continue the elevation of the chorus to the status of not merely an actor but even a h­ uman, as Horace encourages its writer to make this entity “cherish good men” and “advise them 158. The codices’ morabitur and the emendation moraberis are more widely printed, but morabimur as a conspiratorial exhortation with the addressee is attractive (Brink 1971: 243 notes some parallels, e.g., 11–12, 24, 25–6, 153, 331, 347, though Rudd 1989: 179, following Kiessling and Heinze 1914: 321, objects, “H. never includes himself in precepts in the ­future tense”). The translation given ­here, “we ­w ill take our time over,” is not suggested by any commentaries (though Wickham’s 1891: 405 comments ad loc. are thoughtful), but it is supported by OLD 11b and c as well as by the usage of mora at 291 (limae l­ abor et mora) where it describes precisely the deliberate care that should be taken by poets over their work. 159. Campbell 1924: 252–53. 160. Brink 1971: 254–55, 256, Rudd 1989: 182. Officium, however, is a particularly Roman term, denoting “a helpful or beneficial act done . . . ​in fulfillment of an obligation” (OLD), as explored at length in Cicero’s De officiis; see further Saller 1982 (e.g., 15–22), and Sedley 2014: 110–11 on the term in the AP.

78  ch a p te r 1

in a friendly manner”; “rule the angry” and “love t­ hose fearing to sin”;161 praise the essentially Roman (and Horatian) virtues of the mensa brevis,162 laws and justice, and otium in times of peace;163 and even act as a keeper of secrets and one who is in a position to entreat the gods. The chorus is conceived of as the controlling agent that determines what events are to take place and manages them as they do; as a moderating force and a voice of temperate reason (which resembles rather l­ittle the hysteria-­prone, confused, and oblivious choruses often typical of Greek tragedy); and even as having the ability to compel the actors around it to certain courses of action (196–97; ­these courses of action, not too high, not too low, ensure that the actor ­will occupy the middling territory, as instructed also at 31 and elsewhere). ­These activities seem to go far beyond not merely the role of a chorus, however, but even of an actor, for Horace is describing the be­hav­ior of an ideal Roman citizen164—­one alluded to, if not quite ever modeled, also in his Satires. Above all, the emphasis on emotions remains,165 and not merely ­those of the unexpectedly and improbably sentient chorus: anger and fear, and with them friendliness and prayer. The section on the chorus leads to that on the development of drama (202–91), where another exploration of appropriateness is found (though still not the last). Preceded by instances of inappropriateness in the images of royally robed divinities and heroes entering lowly taverns (227–29), in tragedy rejecting light verses (231), and in a matron gyrating at a festival (232; Roman married ­women did not dance), Horace concludes by insisting that the comic characters Davus and Pythias166 ­ought to speak differently from the Silenus of Satyr-­play (236–39). With a brief detour to a version of the idea seen already at 119 (aut famam ­sequere aut sibi con­ve­nientia finge), Horace demonstrates the desired be­hav­ior by saying, “I s­ hall trace my song-poem crafted from something familiar” (ex noto fictum carmen sequar, 240). The explanation given is that “so much honor attaches to ­things taken from the ­middle” (tantum de medio sumptis accedit 161. See Brink 1971: 257–58 on the textual prob­lems of peccare timentis. 162. On slender sustenance as self-­referential for Horace’s poetry, e.g., Sat. 1.6.114–18, see Mette 1961, Gowers 1993: 126–61. 163. Like officium (chapter 1, n160), otium and apertis portis are characteristically Roman terms/phrases; on the Greek precursors to Horace’s discussion of the chorus, on the other hand, see Brink 1971: 258–59. 164. Brink 1971: 256 agrees that Horace’s advice can also “be read in Roman and Augustan terms, and as part of a Horatian poem should be so read.” 165. Noted also in passing by Rudd 1989: 182–83 ad 197. 166. On ­these names, see Pseudo-­Acro, Orelli 1844: 755, and Brink 1971: 287 with Sat. 1.10.40, 2.5.91. Pythias, ­here the name of a comic slave girl, aurally anticipates the Pythian games and song at AP 414.

hum a n o  79

honoris, 243) where medio (in the ­middle of what is very nearly the m ­ iddle line of the poem)167 denotes that which is held in common and familiar to all. The two formulations are, nevertheless, slightly distinct: first (119), making use of existing narratives is presented as one of two equally v­ iable options, whereas ­later, Horace seems to express a preference for making use of existing material (ex noto fictum carmen sequar; not quite the same as following the “received tradition,” fama) on the grounds that it brings more honor (presumably ­because “it is difficult to say common t­ hings in one’s own way,” †difficile est† proprie communia dicere, as Horace spells out at 128).168 It is not so much a question of Horace changing his mind but rather of refining and clarifying and filling out the edges of his ­earlier declarations in the course of restating them, and what Horace manages to convey through his restatement is that a good poet can make any material his own, but that this is excedingly difficult to accomplish (and accomplishable only by a good poet).169 ­After 236–39 ­there follows yet another pronouncement on the speech appropriate to certain characters (one linked through its goatish fauns to Silenus and the four mentions of Satyrs that immediately precede him): silvis deducti caveant, me iudice, Fauni ne velut innati triviis ac paene forenses aut nimium teneris iuvenentur versibus umquam aut immunda crepent ignominiosaque dicta. (244–47) As in the passage describing the chorus, Horace has turned his eye ­here to a fundamentally nonhuman entity, yet his emphasis manages to be on the be­hav­ ior of ­humans: fauns are not to “behave like young men” (iuvenentur), with all that this entails occupying the bulk of the description.170 As in the poem’s opening and at 105 and 112–13, Horace is concerned with the audience response to incongruity. That response, however, rather than laughter (or sleep), is ­here the refusal to award a prize, and, as at 112–13, where the “Roman knights and foot-­ soldiers” act in unison, at 248–50 the reaction is that of both ­those “who have a ­horse and a f­ ather and substance” (quibus est equus et pater et res) and “the 167. Horace would not be alone in such a deliberate placement of medius at a mid-­point: Kyriakidis 2004 shows the same at work in Lucretius and Thomas 2004 in Virgil. 168. See further chapter 2 on this line and its textual difficulties. 169. Both formulations also overlap with Philodemus’s view (PHerc. 1676, Tractatus Tertius) that each poet’s treatment of a topic, even an existing one, “makes the subject itself dif­fer­ent and idion to its creator” (Armstrong 1995: 219). 170. Wiseman 1988: 2 reads line 245 as a stricture “against satyrs behaving like comedy characters,” but I see the contrast being rather between country-­and city-­dwelling figures (forenses, for speaking in the context of the law courts, is especially marked as urban and Roman).

80  ch a p t e r 1

buyer of a broken chick-­pea and nut” (fricti ciceris . . . ​et nucis emptor), that is, the highest and lowest classes of Roman society.171 The mention of souls (animis), moreover, ties this passage explic­itly back to the ideas expressed e­ arlier: of poems as ideally being psychagogic (quocumque volent animum auditoris agunto, 100); of the tongue as the soul’s interpreter (111); and of ­things that enter through the ear as slower to move the soul than ­those perceived by the spectator’s eye (180–82). Horace is interested, it is clear, not merely in how fictional characters ­ought to behave but in how ­actual ­people do—specifically, how they may be affected by fictional characters. At the same time, however, as we understand most clearly throughout the Satires, the individuals whose be­hav­ ior we witness are, of course, Horace’s own fictional characters:172 he is, as he makes clear, the ultimate judge, me iudice (244).

Horatius senex If lines 158–78 prove flawed as advice for a would-be writer, on what terms do they succeed? I suggest that the answer is to be found in the unusually lengthy and detailed description of the senex (169–76) and the vari­ous sections that follow the passage surveyed above. Of the unusual concentration of six agent nouns, for example, four are given to the senex: he is “a delayer” (dilator),173 “a praiser” (laudator) of times past, and “a chastiser and censor” of younger ­people (castigator censorque minorum). Th ­ ese words do not merely describe, they concretize, defining the old man entirely as one who performs ­these and only ­these actions. Also curious is the emphasis on his timidity (notable even without the reading pavidus for avidus),174 as he is said to avoid and even fear using what he has amassed and to ­handle affairs apprehensively and without enthusiasm.175 ­These qualities go beyond such traditional epithets as difficilis, querulus, and miser, and the ste­reo­type of aches and pains, though ­these are also 171. Rudd 1989: 191 suggests a “lightly satirical” tone to 248, “as not infrequently when H. is talking of a class above that into which he was born.” 172. On the literary conundrum that readers of any work are required to participate in a world within which the fictional characters encountered are considered somehow “real,” see Rabinowitz 1977, 1986. 173. This despite the ­little that remains of his life span. The association with dilator can help to validate the reading spe †longus†: the objection has been that it is not reasonable for him to hope for a ­great deal more time in life (so Brink 1971: 239, Rudd 1989: 178), but his nature as “a delayer” is precisely in line with his being “long in hope” (where the ablative spe is to be preferred over the proposed emendation of Peerlkamp and Heinze, spei). 174. Shackleton Bailey 2001 prints Bentley’s emendation pavidus for avidus. 175. Rudd 1989: 177 explains gelide as “without warmth” and adduces Arist. Rh. 2.13.1389b31ff., ὁ φόβος κατάψυξίς τις ἐστίν, “fear is a kind of chill.”

hu m a n o  81

pre­sent. In addition, regarding the banal directives that follow in 179–205 about the proper length of a play, the correct number of actors and speaking parts, what may and may not be shown on stage, and the proper use of musical instruments, we would do well to ask not merely “why?” but “why h­ ere?” I see it as no coincidence that Horace describes the traits of the senex at the greatest length of all the four stages of life he outlines, and that this description immediately precedes some of the poem’s driest, most unloved, and most overtly pedantic and (thus, as the thinking has gone) didactic portions. I suggest that Horace may profitably be read as setting up and entering a role—­not some idea of a literary critic, as has been suggested,176 but rather that of the grumpy old man. The words that follow the life cycle vignette (158–76) and Horace’s brief summary of his advice (176–78) come abruptly:177 aut agitur res in scaenis aut acta refertur (“­either the m ­ atter is acted out on stage, or t­ hings done are reported,” 179). The reader may startle at this suddenly prescriptive tone and the banal content for which it has been reserved,178 especially as it turns out merely to be the first in a series of mostly negative directives: “you should not bring onto the stage ­things that ­ought to be done inside” (non tamen intus / digna geri promes in scaenam, 182–83); “let Medea not slaughter her boys before the ­people or unspeakably wicked Atreus cook h­ uman innards openly, or Procne be turned into a bird, Cadmus into a snake” (ne pueros coram populo Medea trucidet / aut humana palam coquat exta nefarius Atreus / aut in avem Procne vertatur, Cadmus in anguem, 185–87); “let the tale be neither smaller nor more drawn out than a 176. Frischer 1991 (cf. 1996: 110–13) reads the AP as “a parody of Peripatetic poetics” (1), “the inept ramble of an unreliable narrator” (85), and while Armstrong 1993 says that “the application of persona-­theory to this poem was long overdue” (191) he sees Frischer overstating the degree to which “Horace portrays the speaker of the poem as a pedant and an ignoramus” (52), concluding, “I cannot believe in Frischer’s joke-­figure, the Beckmesser-­like Aristotelian poetry-­ spoiler, as the speaker” (197). Ferenczi 2014b, having said that “Frischer’s general claim of the need for a reconsideration of critical approaches” has found more f­ avor than the specifics of the re-­reading proposed (2014a: 14), follows Frischer in showing (rightly) how the speaker of the AP is of weakened didactic authority. 177. “A brusque new beginning,” Brink 1971: 245, with “a change of style: prescription instead of description” (245). To the extent that commentators (including Brink) have seen a connection between 179ff. and the preceding passage, it is only in their shared overall topic of drama. 178. Horace’s statement seems facetious—­how and where could dramatic action take place ­either than on stage or off stage and be subsequently reported, as Landinus indeed blandly noted (“ita se habent fabulae,” at Bugada 2012: 116)?—­though cf. Sharrock 1996: 104 on a similar idea at Arist. Poet. 1459b. Although this has not been suggested, Horace may be hinting at an essential difference between telling and showing encapsulated within the AP itself (see my introduction on the distinction between ex arte and de arte), overtly enumerating certain rules to covertly numerating how poetry is best written.

82  ch a p te r 1

fifth act” (neve minor neu sit quinto productior actu / fa­bula, 189–90); “and let a god not intervene, ­unless a knot worthy of a liberator has befallen” (nec deus intersit, nisi dignus vindice nodus / inciderit, 191–92); “and let a fourth character not strive to speak” (nec quarta loqui persona laboret, 192). Resembling “mechanical trappings of received wisdom,”179 ­these grave yet vacuous prounouncements ring so tinny and rankle as so naïve that they should arrest the reader, compelling him to ponder what Horace might ­really have been up to. I suggest that this section makes the best sense in context, coming as it does on the heels of Horace’s description of the senex and the accompanying implication that Horace himself be understood as having taken up this role. The tone of it aligns precisely with the qualities attributed to the old man:180 the ­imagined speaker of lines 179–92 acts as a castigator censorque, and for whom could his advice be intended if not for younger writers, minorum? The advice itself is so unnecessary (what would-be poet setting out to write drama could be unaware of the five-­act law,181 the rule of three actors,182 or the prohibition against showing murders and other disturbing events on stage,183 or have failed to learn that the deus ex machina was regarded as a “suspect” device?184) and the examples used to illustrate it so hackneyed,185 that it may be seen precisely as exemplifying the anx­i­eties of any irritable old man. (One won­ders also why Horace is asking his reader to swallow the general premise that the Pisones, setting out to write drama yet wholly lacking in any knowledge regarding its most basic components, would have consulted, of all t­ hings, a verse epistle for instruction.) Moreover, some of the pronouncements that this voice of authority issues are confused or plainly wrong,186 ­others stricter than attested ancient 179. Rudd 1989: 34. Brink 1971: 226 speaks of the “miscellaneous rules” at 179–201. 180. The term senex was reserved for “Romans over 60 years of age” (Neue-­Pauly s.v. “Age[s]”), a number that Horace was likely nearing at the time he was writing the Ars Poetica (see introduction) though did not quite attain, living from December 8, 65 BCE, to November 27, 8 BCE. 181. Brink 1971: 248–50 reminds us that although we may think of this as an article of faith from the ancient world, ­there is ­little solid evidence for it prior to the Hellenistic period (see also Campbell 1924: 244–45); nevertheless, for Horace and his first readers, it would have already been dogma. 182. On which, see now Starkey 2018. 183. Standard, so far as we know, in Greek tragedy. 184. Brink 1971: 251–53. 185. Ibid.: 247 ad 185. 186. Regarding the three-­actor rule, Brink 1971: 253–54 points out that Horace appears to be conflating two related stipulations: the one that t­ here could be only three speaking actors in tragedy, the other that if a fourth actor appeared on stage his character was not permitted to speak (see further Pickard-­Cambridge 1953: 137–53, Williams 1968: 334–36). Similarly, Beare 1946, 1948, noting the context in which Horace stipulates that t­ here be five acts (i.e., for a play

hum a n o  83

Roman practice is known to have been,187 perhaps to illustrate a ste­reo­typical geriatric rigidness. That this cantankerous speaker is Horace himself is suggested by the fact that we have encountered two of ­these four examples (Medea, Atreus, Procne, Cadmus) before in the Ars Poetica: Horace instructed his scriptor, amid a series of directives, to make his Medea “ferocious and unconquered” (ferox invictaque, 123), and he mentioned the cena Thyestae at 91 (Thyestes being the ­brother of Atreus; the latter had tricked the former into eating his own sons). Telling, too, is the speaker’s reaction: “What­ever you show me thus I hate, not believing it” (quodcumque ostendis mihi sic, incredulus odi, 188). This prescriptive voice of authority, who forbids rather than guides or exhorts, experiences a strong and visceral reaction to murder onstage (Medea, Atreus), as well as to mythological transformations (Procne, Cadmus)—­the unbidden response of an individual and distinct from the more balanced advice a teacher might be expected to give. This old man’s fearfulness and bad humor give way to nostalgia of a particularly Roman, Augustan, and Horatian type, as the history of drama and lit­er­a­ ture in lines 202–84 takes the form of a praise of times past, the quality of the senex as laudator temporis acti now being permitted to play out. In the oddly technical discussion of instruments and the role of musical accompaniment with which it begins, a judiciously placed ut nunc is the first formal indicator that we are looking back to a more virtuous past when the tibia was pure bone, which stands in contrast with (what other than) a decayed and decaying pre­sent.188 The sense that ­things are not as they once ­were—­that they are, to be clear, much worse—is compounded by such details as: nondum (the seating area is “not yet” crowded); populus (the ­whole populace, still neologistically “countable,” numerabilis189 since “small,” parvus); and coibat (“used to come together,” the imper­ fect tense conveying that this would happen habitually), the action naturally done “honest(ly) and chaste(ly) and modest(ly)” (frugi castusque verecundusque, “that wishes to be demanded to be watched and demanded again,” quae posci vult et spectanda reposci), concludes Horace “cannot have meant to imply that all tragedies had five acts.” Williams is right to see Horace taking deductions made by Aristotle from experience and turning them (humorously) into “instructions.” 187. Cf. Brink 1971: 253. Reinhardt 2013: 500 further points out the “antiquarian” flavor of Horace’s instructions: they “derive from a Greek 5th ­century BC context; some even apply to that context only.” 188. See Brink 1971: 260–63 (“what underlies the musical decline is said to be a moral decline. On this topic H. often accepts the familiar Roman pattern of degeneration”; he notes the parallels with Pl. Resp. 424c), Rudd 1989: 183. 189. The antonym of innumerabilis, numerabilis was apparently coined by Horace for this moment (Brink 1971: 267).

84  ch a p te r 1

205–7).190 The image of the victor extending his fields (coepit agros extendere victor, 208) is eminently evocative of the soldier-­farmer, that paragon of ancient upright morals, and the mention of the growing city wall (208–9) suggests early Rome, as does the arrival of the Genius (210), a pointedly Italianate deity,191 who now began to be propitiated “on festal days” (festis . . . ​diebus) as the all-­ controlling Roman religious calendar begins to take shape. As the city grows, however, and its rites and rituals become increasingly regularized, the concomitant effect is of the execrable growth of licentia, (“lack of restraint,” even “unruliness, wantonness,” 211) and luxuria (“luxury,” 214). Accompanied by the contrasting pairs country-­city (rusticus urbano, 213) and shameful-­honorable (turpis honesto, 213),192 a view of the contrast between past and pre­sent has emerged that is typical of a senex, whose mantle as querulus, laudator temporis acti, and castigator censorque minorum Horace seems to have taken up and be indulging himself in performing ­here with gusto.193 The ideas introduced at 202–11 are repeated and further worked out through a series of illustrative examples and contexts over the next 80 lines, as Horace continues to speak in the persona of laudator temporis acti, ever fueled by that of the castigator censorque minorum. He bemoans the loss of honest, rustic morality (212–15) and ­simple aesthetics (216–19), and he dispenses all manner of advice, much of dubious practical value: fauns should not behave like young men from the city (244–50), for example, as the spectators ­will be annoyed by this. The same continues during the interlude that interrupts the two portions of the excursus on the development of drama (202–250, 275–91), which itself retreads the tone and some of the territory of 179–201: a short syllable followed by a long is termed an iambus (251), building block of the iambic trimeter (252–54); spondees have a slower beat (255–62)—­this should all come as news to no one. Prehistory and history intermingle, as Horace name-­drops and critiques 190. What appear to be plain adjectives, moreover, double as Roman cognomina: Castus, Verecundus, and Victor are well attested (Kajanto 1965: 30, 68, 251–52, 264, 278) but perhaps most remarkably Frugi (Kajanto 1965: 66–68) was a cognomen of vari­ous Calpurnii, Pisones, and Calpurnii Pisones (see Syme 1960 and 1986, genealogical ­table XXV; Forsythe 1990, 1994: 2; and my chapter 2). AP 206–7 thus suggests the coming together of specific named prominent gentes (Kajanto 1965: 20–21, 62–70), with the gens Calpurnia (Kajanto 1965: 239, 242) further and honorifically identified among them; compare Horace’s puns on cognomina at Sat. 1.3.44–48 with Gowers 2012: 129–30. 191. Brink 1971: 267–68: “Roman colouring.” This tutelary deity (a guardian spirit and deified concept closely allied with the notion of “self ”) is dated to early Rome; see further Neue-­ Pauly, OCD. 192. The terms are “both social and moral” (Rudd 1989: 185), as are rusticus and urbanus. 193. Geue 2014: 158 speaks more generally of how “it is the older generation’s job to be a laudator temporis acti and castigator minorum.”

hu m a n o  85

Accius and Ennius (258–62), Plautus (270–74), Thespis and Aeschylus (275–84), in no par­tic­u­lar order. All the while, he harangues his reader: vos exemplaria Graeca / nocturna versate manu, versate diurna (268–69); and ­later, vos, o / Pompilius sanguis, carmen reprehendite (291–92). While the section 251–74 might appear one of the driest in the Ars Poetica (as it surely is, in some ways), it becomes more compelling if Horace is understood as having taken up the voice expected of him in a work of poetic didaxis.194 If Horace wanted to succeed in his paradidactic experiment and pass off a long poem in hexameters as a serious literary treatise (rather than as verse satire, for example), he could not but include sections such as t­ hose at 179–201 and 251–62. Yet t­ hese are integrated into the poem in an in­ter­est­ing and organic fashion through Horace’s co-­opting of the role of senex described at 169–76. Accordingly, Horace is critical, portraying his pre­de­ces­sors as talented but undisciplined.195 It is not hard to picture the young ones to whom this senex figure addresses himself within the world of the poem nodding politely while rolling their eyes in exasperation at how the old man repeats himself and at the inanity of his “pearls of wisdom.” ­There is one further moment that may be seen to signal the presence in the Ars Poetica of the Horatian satirist (though he is not the poet exactly as he appears in the Satires, but rather an aged version of him): the much-­discussed phrase Satyrorum scriptor. While o­ thers have argued that Horace had entertained the possibility of writing such a drama or in fact wrote one that is now lost,196 or that a­ ctual Pisones w ­ ere planning a foray into this genre,197 or 194. Cf. Brink 1971: 296–97 ad 251: Horace “begins in lecturing style.” If 251 is understood to be a “memoria technica line” (Blakeney’s 1928: 82 suggestion, though Brink 1971: 297 is dismissive), Horace is embracing his teacherly role to the utmost by uttering ­actual pieces of canonical, conventional advice dispensed in schools. 195. The specific complaint directed at Accius and Ennius seems to be that each has too ­little iambic rhythm in their verses; see Brink 1971: 301. The charge that they are ponderous, clumsy, and perhaps underhoned overlaps with Horace’s critiques of Lucilius at Sat. 1.4.9–13, 1.10.1–3, 56–71. 196. Orelli 1844: 755: “nihil aliud significare potest nisi ‘fabularum Satyricarum scriptor’ ”; Wickham 1891: 412 (cf. Blakeney 1928: 11, Rostagni 1930: 68, Brink 1971: 286) suggests, “if I ever write satyr-­plays”; for Rudd 1989: 189, Horace “is merely using the first person for the sake of variety” (so, too, Campbell 1924: 246). 197. Wiseman 1988: 1–2. The viability of satyr-­play as a Roman genre has been much debated, as in turns by Brink 1963, Williams 1964: 195 (“it is not remotely conceivable that Horace thought of writing satyr-­plays: it is not more conceivable that satyr-­plays r­ eally ­were appearing on the Roman stage”; he regards their presence in the AP as a vestige of Horace’s Greek models for the poem, specifically, Aristotle), Brink 1971: 274 and 1995: 271–75 (“I overstated the case against Roman satyr-­drama,” referring to 1971: 274), and fi­nally Wiseman 1988 (he concludes, “Roman satyr-­play did exist ­after all—­largely, no doubt, in generically contaminated forms,” 10).

86  ch a p te r 1

perhaps that Horace simply had drama on the brain,198 ­these words serve also to remind the reader, in precisely that subconscious way Horace has of intimating ­things throughout all his writings (and as he has done already with sermonem . . . ​sermonum . . . ​sermonibus . . . ​sermone, 57, 69, 81, 95), that the creator of the Ars Poetica is the creator also of satires. It is often asserted199 that the Romans did not much exploit the connection between satyrs and satire (satura), a (false) etymology that was especially in vogue during the Re­nais­sance,200 yet our collective certainty on this point seems curious (pace Casaubon)201 given, for example, the Priapic satirist of Horace’s Satires 1.8, which incorporates unavoidable resonances of satyrs.202 In fact, the association between satyrs and satura is ancient, seen already in the first of Diomedes’s four etymologies for the genre-­term, where he suggests that it is named “from Satyrs, b­ ecause in such a poem ridicu­lous and scandalous t­ hings are said in a similar way, just like ­those which are uttered and done by Satyrs” (“a Satyris, quod similiter in hoc carmine ridiculae res pudendaeque dicuntur, velut quae a Satyris proferuntur et fiunt,” Ars Grammatica 3; Keil 1857: 1: 485.30– 486.14). The opposite direction of filiation (namely, that satyrs are named from satura) was also posited in antiquity, as among the scholia to Persius where, just as in Diomedes’s other three etymologies,203 satura is associated predominantly with the idea of fullness.204 Evanthius and Donatus, on the other hand, 198. Campbell 1924: 245–46 sees Horace’s interest in satyr-­play following “naturally” on the heels of his discussion of tragedy, as though enacting the Attic tetralogy. Reinhardt 2013: 517 suggests that Horace has in mind “the 5th ­century BC variety” of satyr-­play. 199. See, e.g., Van Rooy 1965: 124–39, 144–45 with my ac­cep­tance of his statement at Ferriss-­ Hill 2015: 102 (though cf. 54: “­these lustful creatures lurk in the genre”); contra Wiseman 1988: 13: “the first of Diomedes’ three definitions of Satura is generally waved away by modern theorists, but it was evidently taken seriously in the ancient world.” 200. See Campbell 1938: 27–35 on this confusion/conflation of satire (also spelled satyre) and satyrs. 201. Casaubon 1605, espressing his vehement disagreement (29) on the grounds that satyr-­play is drama and satire is not, that satyr-­play has a chorus of joking and dancing figures and satire does not, e­ tc. (280–81; cf. 311–15 for his discussion of the spellings satura / satira / satyra), is credited with fi­nally and definitively separating the “almost universally held” connection between satire and satyrs (Campbell 1938: 29, n27). 202. See Ferriss-­Hill 2015: 54–55. 203. ­These posit that the genre-­term comes from a plate stuffed with first fruits (the lanx satura), or from a type of sausage (farcimen) stuffed with many t­ hings and termed satura, or from a bill similarly constructed (lex satura). 204. “Moreover, satire is so named from its fullness, that is, abundance. From that, the attendants in the chorus of f­ ather Liber, full of wine and feasting, are called Satyrs” (“Dicta autem satira a saturitate i. habundatia. Unde in choro Liberi patris ministri vino atque epulis

hu m a n o  87

derive satire, which they term satyra, from Greek Old Comedy (vetus/archaia comoedia), and in ­doing so conceive in another, if more roundabout way, of the Latin genre as intimately bound up with satyrs and satyr-­play.205 Linguistic merits aside, the impression we rightly glean from t­ hese sources is of the term satura having multiple disparate and yet si­mul­ta­neously valid origins and associations, as if mirroring the genre’s own essential nature as a hodgepodge.206 Alongside this evidence from grammarians and scholiasts, ­there may be found in Horace and in the other writers of Roman verse satire numerous plays upon words that sound like satura but do not have any secure etymological connection to it. Among ­these are references to the god Saturn and his Saturnalia (known for its topsy-­turvy festive license), as at Satires 2.3.5 and throughout Satires 2.7 (set during the festival), or Persius 2.59 and 5.50.207 In Horace’s self-­ effacing description of himself as not having been carried around his estate on a fancy, Satureian h­ orse (non ego circum / me Satureiano vectari rura caballo, Sat. 1.6.58–59) ­there has also been seen a nod at satura.208 ­These occur alongside the repeated repre­sen­ta­tion by all four Roman satirists of their genre as “full” (satur), both explic­itly through strategic deployment of this adjective,209 and implicitly through lengthy expositions upon food and fullness (“satira a saturitate,” as the scholia to Persius also say).210 Given that writers of satire are known to allude to their satura in t­ hese vari­ ous ways, often showing l­ittle regard for etymological accuracy,211 it is pleni Saturi appellantur”; at Jahn 1843: CLII, n1 [Prolegomena], where he explains that he found the comments in both the codex Bernensis 665 and the Darmstadt codex). 205. Evanthius, de fa­bula: excerpta de comoedia 2.5 (Wessner 1902: 16–17) = Donatus, Terentii vita et de tragoedia ac comoedia non pauca, ex Aelio Donato (Erasmus and Rivius 1535: 12, cf. Campbell 1938: 24–25), with manifest conflation/confusion of satire and satyr-­play; see further Hendrickson 1894, Jensson 2002. Confusion of the two genres (or, perhaps, the understanding that Horace is speaking at AP 220–30 of satire) is evident also in the “Materia” commentary (e.g., nudavit is explained as “ad proprietatem satire respexit,” satiros [sic] agrestis as referring to the fact that “non enim in satira ornata uerba sunt, sed agrestia et inculta” and temptavit iocum to “levitatem satire,” and risores . . . ​dicaces . . . ​satiros [sic] with “proprietates sunt satire”; Friis-­ Jensen 1990: 361). The ancient sources on satura are discussed at length by Van Rooy 1965. 206. Keane 2002: 12–13: “inescapably multiple and ambiguous.” 207. See further Ferriss-­Hill 2015: 103–5. Naevius satura fr. 1 (Quianam Saturnium populum pepulisti?, Lindsay 1913: 306; though Courtney 1993: 3 is skeptical) may also number among ­these. 208. Freudenburg 1993: 192, 2001: 61, Gowers 2012: 234. 209. As at, e.g., Sat. 1.1.119, 1.1.120 (cf. Freudenburg 2001: 32), Pers. 1.31, 3.27, 3.78, Juv. 7.62, 8.118. 210. See Gowers 1993: 109–26, Ferriss-­Hill 2015: 101–18. 211. See Henriksson 1956: 77 with Jensson 2002: 88, n10 for the view that etymological accuracy was not seen as essential. If the Satyrs-­satura connection was underused compared to the

88  ch a p t e r 1

eminently plausible to read a wink at satire and at his own Satires in Horace’s Satyrorum scriptor.212 That the phrase is ensconced among three additional occurrences of the term Satyr (221, 226, 233), rounded out by a bonus appearance from Silenus (239), satyr extraordinaire,213 should only enhance our suspicions, for “Silenus is not just the f­ ather of the satyrs” but also “the source of arcane wisdom, if he can be caught and made to divulge it,” a strong hint that Horace is up to something devious in this passage. The reading argued for ­here is further supported by the unusual form Satyrorum: satyr-­play was referred to more commonly by the adjectival phrases satyricae scaenae (“the stages of satyr-­ play”) and satyrico more (“in the manner of satyr-­play”), or in the substantivized form satyrica (“satyr-­things”).214 Had Horace intended to communicate his status as a writer of satyr-­plays, any one of t­ hese adjectival forms would have allowed him to do so unambiguously.215 In this light the closeness of the unexpected form Satyrorum (rather than Satyricorum/Satyricarum) to Saturarum, Horace’s genre-­term for his own Satires in the corresponding case, becomes especially notable. Moreover, the context of the phrase Satyrorum scriptor is itself evocative of satire. Horace declares that he “­will love not only unadorned and established names and words” (non ego inornata et dominantia nomina solum / verbaque . . . ​amabo, 234–35), an allusion to the coining of new words and the resurrection of underused ones advocated for at Ars Poetica 46–62 that may be seen at work also in his Satires.216 Telling, also, are the ways in which satyrs are described in lines 220–33: though initially rustic (agrestis) and participants in a rough (asper) early effort at humor, they come to be funny (risores, a striking (linguistically accurate) idea that satura was “full,” it is more likely that this was b­ ecause ancient writers did not find it particularly in­ter­est­ing or useful, rather than b­ ecause they viewed it as etymologically problematic. 212. Gowers 1993: 116 puts it the other way around: “It would be naïve to think that, to a bilingual Roman, the word satura did not also conjure up satyrs.” 213. Wiseman 1988: 5. 214. Found at Vitr. 5.6.9, 7.5.2 and Diom. Ars Gramm. 3 (Keil 1857: 1: 482.27–29, p. 491.4); cf. also Porphyrio ad AP 221 with Wiseman 1988: 2. Moreover, the only attestation in the OLD of the Latin noun satyrus meaning “satyric drama” is, in fact, Horace AP 235, rendering the definition circular (Brink 1971: 286 is accordingly forced to provide a parallel from Greek: Σάτυροι at Ar. Thesm. 157). 215. The metrical difficulty that Satyric-­consists of three short syllables would be surmountable by placing the word in line-­initial position, where some flexibility would permit the first short syllable to scan long (as at, e.g., line 45: hoc amet). 216. On the language of Horace’s Satires, see further Ferriss-­Hill 2015: 228–33. At AP 233–37 ­there may be another illustration of the princi­ple that judicious placement can render a familiar word new (AP 47–48), since interesse at 233 (intererit) means “­will be pre­sent among” while at 237 (intersit) it means “it does [not] ­matter.”

hum a n o  89

agent noun) and chatty (dicaces), and to mingle seriousness and play (vertere seria ludo)—­all key terms with which Horace describes his own verses in the course of his Satires.217 In addition, the alternate genre-­term sermo, favored by Horace throughout his Satires over satura/satira, occurs for the fifth time at 229,218 tucked away in the description of a divine figure incongruously visiting a tavern: ne quicumque deus, quicumque adhibebitur heros, / . . . ​migret in obscuras humili sermone tabernas (“so that what­ever god, what­ever hero is brought on . . . ​ may not travel to dusky inns with ­humble conversation”). Especially tantalizing is the identification of sermo as “­humble” (humili; literally, “of the ground”), as satire is described also at Epistles 2.1.250–51: sermones . . . ​repentis per humum (“conversations/ satires that creep along the ground”).219 Within the Ars Poetica, too, we have already found the suggestive phrase sermone pedestri (95), apparently describing the occasional use of plainer language by tragic figures, yet inescapably evoking Horace’s famous Musa pedestris, “walking Muse,”220 from Satires 2.6. By alluding to his Satires through satyrs and satyr-­play in the manner suggested ­here, and describing ­these satyrs, which already aurally evoke satura, in the marked terms used of his own satire, Horace manages to have it both ways. Since the Ars Poetica is not overtly about satire, a direct reference to the genre might have seemed out of place (indeed, the statement that “­humble conversation/satire,” humili sermone, is undesirable for loftier characters and situations may itself metapoetically nod at Horace’s avoidance of the genre-­term satura at this very juncture). Horace therefore points to his earliest hexameter writings without confining himself to be thought of as a writer in that genre alone, and in the wider-­ranging scope of the Ars Poetica he hints at the satirical material that is nevertheless pre­sent in this poem—­all signaled, as suggested h­ ere, through the term humano with which it opens. It has perhaps also become clearer why Horace chose drama as the primary literary genre for his Ars Poetica. He famously did not work in this genre 217. Play and laughter at, e.g., Sat. 1.1.23–27, 1.4.139, 1.10.7, 1.10.14, 1.10.37, 2.1.73; in addition, dicaces may be seen alluding to the meaning of sermo, “conversation.” On Horace’s Satires in relation to the countryside, see further Braund 1989, Harrison 2007b. 218. Horace describes his writings as “closer to conversation” (sermoni propriora) at Sat. 1.4.42, and refers possibly also to his Satires as sermo/sermones at Epist. 1.4.1; see further Ferriss-­ Hill 2015: 43–44, 103–4. 219. The “Materia” commentary (see chapter 1, n205) again suggests that Horace has satire on the mind (or that satyr-­play and satire are one and the same), explaining humili sermone with “id est humilitate satire” (Friis-­Jensen 1990: 361; cf. 362–63). 220. The phrase occurs, furthermore, next to a mention of “satire”: saturis musaque pedestri (Sat. 2.6.17); cf. also the association of sermo and metrical feet at Sat. 1.1.47–48.

90  ch a p te r 1

himself, and scholars have ventured a number of explanations for his forefronting of drama as an area in which he could, as they understand it, pre­sent himself as sufficiently expert to give instruction.221 As Dilke summarizes, Horace has been said to be “writing in conformity with a Hellenistic school of literary criticism which was much indebted to Aristotle”; or he chose drama since epic (the other “high genre”) had already been taken up by Virgil with his Aeneid; or (by the most widely repeated justification) “prob­ably one or more of the Pisos was writing or translating a drama” (perhaps a “satyric drama, not for the stage but for recitatio in a circle of friends”).222 I suggest, however, that drama provides for Horace the ideal vehicle for connecting writing with living, and so the Ars Poetica with his Satires. Especially in ancient Greece and Rome the dramatic genres ­were public ones,223 where being viewed in the audience, occupying the par­tic­u­lar seat mandated by one’s social standing and the po­liti­cal and religious offices one held, rivaled the experience of viewing the play itself. Not only w ­ ere ­humans in the audience watching, in unison, h­ uman actions and events on stage, but throughout the event social hierarchies and mores ­were continually reinforced. Far more than satire, then, which remains private and closed-­off—­ though it has much in common with comedy, as Horace reminds us—­drama performs life, and does so in a public, communal fashion. In using drama to facilitate the connection between writing and living that he seeks to make, Horace emphasizes again that he regards poetry as the messy, ever-­evolving, and potentially flawed product of h­ umans and their interactions. I also see him suggesting through the manifest presence of the Satires within his Ars Poetica that the latter is at least as much an ars vivendi (“art of living”) as it is an “art of poetry.”

Callida iunctura While generations of readers have been beguiled, particularly on the basis of lines 201–91 and ­others like it, into regarding the Ars Poetica as a usable manual for writing, it remains incontrovertibly the case that the work is in large part a list of rules, both for what and (mostly) what not to do. Th ­ ese rules, however, 221. Cf. Lowrie 2014: 122: “it is striking that Horace keeps coming back to tragedy as a touchstone given that he did not write it.” 222. Dilke 1958: 54; cf. also, e.g., Blakeney 1928: 68. Dilke adds that Horace’s interest in drama has been seen as not unique to the AP, being evident in Epist. 2.1 as well (Campbell 1924: 121 similarly sees Horace as being “on the threshold of his dramatic period”). 223. Oliensis 1998: 203 makes a similar observation: “If Horace, himself no playwright, devotes so much of his Ars to dramatic proprieties, one reason is that drama, more directly than epic or lyric, offers embodied models for acting in public.”

hum a n o  91

go far beyond the technicalities of putting together a verse in mechanical fashion, and so it is that the Ars Poetica may be best regarded as a constitution of sorts. Far more than for what it says, the poem is impor­tant for how it says what it says—­the topic, in fact, of much of Philodemus’s On Poems—­and for the fact and form of its existence. The Ars Poetica thus becomes, in its entirety, a full and complete exemplum of an ideal poem:224 one who merely follows the rules stated within it ­will not produce beautiful verses; rather, through careful and continued study of the ­whole, we may begin to discern, first ­here and t­ here and gradually more and more, the qualities and habits of mind and of practice that make a preeminent poet and poem. The work’s didactic content lies not where it appears to, that is, in its dicta, but in the ways t­ hese and its other ideas are expressed and, especially, welded together. Readers have faulted the discussion on drama that occupies much of 202–91 for being apparently confused, particularly in its conflation of Greek and Roman, even attributing this lack of clarity and absence of linear progression to textual prob­lems,225 but I would venture that this alinearity and blurring is the point. First, by moving stealthily between the temporal and the atemporal, the par­ tic­u­lar and the universal, Horace is not only mediating between Greece and Rome, as he does elsewhere in declaring himself the first Roman poet in the Greek lyric tradition (Carm. 3.30.13–14; cf. Epist. 1.19.23–24), but even showing how such interactions can play out: Plautus’s meters and wit, a Camena (the Latin analogue to the Greek muses), and “our poets” (nostri poetae, 285) are found alongside Thespis and Aeschylus; Latium and the Pisones, “Pompilian blood,” are juxtaposed with Helicon and Democritus (289–97).226 Through this puzzling interlacing of Greek with Roman, in a manner that pays l­ ittle regard 224. Cf. Armstrong 1968: 104 (throughout the AP “Horace gives an example of how the ­thing is done in the very pro­cess of describing it. In fact, throughout the poem Horace does this wherever he can”), Brink 1971: viii (the AP is “a work of the imagination that makes a poetic symbol out of literary theory”), Williams 1980: 281 (Horace, in “purporting to be a teacher,” is in fact at moments “exemplifying the essence of poetic composition by practising it as he pretends to teach”), Newman 1986: 61–62 with Tsakiropoulou-­Summers 1995: 165–66 (Newman identifies among all Horace’s precepts in the AP “one precept so huge and obvious that it usually escapes attention,” i.e., that the lesson of the poem lies in its form), Hardie 2014: 50 (“another answer to the question of what is the Ars about, might be that it is a leçon par exemple of how to write successful poetry”). 225. Campbell 1924: 249 speaks of Horace’s “habit of apparently confusing features from Greek and Latin lit­er­a­ture in his semi-­theoretical and semi-­historical summaries”; see chapter 1, n70. Brink 1971: 267–68, in noting that “some commentators are as emphatic on the Greek colouring of this passage . . . ​as ­others are on the Roman,” reveals the historical certainty that has been sought in Horace’s description. 226. Williams 1968: 329–57, too, sees this “blending” as a key aspect of the AP.

92  ch a p t e r 1

to chronology, and certainly not to teleology, Horace goes beyond merely saying that Roman poets should look to Greek pre­ce­dents (vos exemplaria Graeca / nocturna versate manu, versate diurna, 268–69; cf. 286–87); rather, in the constitutive document that is the Ars Poetica, he shows how this may be done, both explic­itly in terms of content, and implicitly, through the crafting of a novel literary form. Second, if we succumb to Horace’s narrative authority, which he established in the poem’s opening scene, and accompany him as he meanders through his conjectural history,227 we may find ourselves experiencing and enjoying the passage for the sensations it offers about nature, ­humans, and time. Above all, we should read 202–91, as we should all of the Ars Poetica, not for what Horace is stating about artistic production, but for what he is suggesting and implying, for many of his points are too subtle to be spelled out, and indeed become weakened by being stated explic­itly. Within 202–91 may be found a passage that acts as a fulcrum about which the rest of the poem centers: lines 220–39, which, not coincidentally, encompass the Ars Poetica’s midpoint. Several terms repeat in the course of the passage, and ­these and ­others reach, through their judicious replication from elsewhere, to points ­earlier and l­ ater in the poem. The density and extent of ­these repetitions may best be appreciated through a visual illustration such as the following, in which key terms occuring more than once within the passage have been given in bold, italic, underlined, or combinations of ­these, and numbers in brackets indicate the lines in which previous or subsequent instances of a word are to be found:228 carmine qui [89, 95]tragico vilem [78, 84]certavit ob hircum, mox etiam [17, 208]agrestis Satyros nudavit et [163]asper incolumi gravitate iocum temptavit, eo quod illecebris erat et grata novitate morandus [ 182]spectator, functusque sacris et potus et [135, 199, 219]exlex[283 bis, 399]. verum ita [5, 101 bis, 105, 139]risores[356, 358, 381], ita commendare dicaces conveniet Satyros, ita vertere [107]seria[451] [107]ludo[379, 405], ne quicumque [83, 191, 200]deus, quicumque adhibebitur [114]heros, [ 65, 73]regali[404, 434, 453] conspectus in auro nuper et ostro, migret in [26]obscuras[363] [28, 110, 159]humili [57, 69, 81, 95]sermone tabernas, 227. Also known as histoire raisonnée, this narrative method describes what could plausibly have happened, rather than what actually happened as reconstructed from evidence; the phenomenon and its history have recently been treated by Palmeri 2016. 228. Among “previous or subsequent instances” are included ­those in other parts of speech, e.g., nouns and adjectives, nouns and verbs, e­ tc. Previous instances are noted before a word, subsequent ones ­after it. The inventory given is not exhaustive; rather, only ­those references are included that I consider most relevant or thought-­provoking.

hum a n o  93

aut, dum vitat[267, 298] humum, nubes et inania[443] captet. effutire levis[423] [24, 90, 91, 138, 183, 191]indigna[264, 283, 359, 436] tragoedia [

75, 89]versus[246, 260, 322, 382, 427, 441, 445, 457, 470],

ut [210]festis [116]matrona moveri iussa [210]diebus[293], [ 114]intererit Satyris paulum pudibunda protervis. non ego inornata et dominantia [58]nomina solum [ 46, 47, 52, 62, 97, 106, 133 bis] verbaque[311, 443], Pisones, Satyrorum [120, 136]scriptor[346, 354] amabo; nec sic enitar tragico[275] differre [86]colori ut nihil intersit Davusne [72, 104, 114, 192]loquatur[280, 324] et audax Pythias[414] emuncto lucrata[420] Simone talentum, an [161]custos famulusque dei[391, 464] Silenus229 alumni. (220–39) Such latticework that runs the length of the Ars Poetica, whereby apparently disparate sections are fastened together through verbal echoes, h­ ere becomes clearly evident, and in addition to contributing to the poem’s aesthetic texture and, in short, constituting its literary style, it serves further to provide a thematic scaffolding. ­Here are, accordingly, some ideas I see Horace alluding to in lines 220–39 through his signature recursivity: an interest in wordplay and etymologizing, since the derivation of tragedy (“goat-­song”) anticipates that at 275–77 of comedy, via its self-­referential genre term, τρυγῳδία (“wine-­lees-­ song”);230 a broader interest in the creation of new words and the novel reuse of existing ones; an association between poetic genre and the law, and the importance of adhering to such strictures; colors, alongside light and dark; young men and their training u­ nder guardians; the opposition of tragedy or the tragic style with straightforward conversation and the conversational style, sermo; and along with this, the abiding presence of the genre of sermo/satura, and naturally Horace’s own Sermones/ Satires, in the Ars Poetica, aurally heightened by the curious fourfold repetition of Satyr. Perhaps most evident in all this is Horace’s interest in polarities: seriousness and jest; work and leisure; rich and poor; high and low genres and modes of speech. ­These intersect clearly with two previously stated concerns: the need for appropriateness of vari­ous sorts (73–98, 103–7, 112–18, ­etc.), once again, and the danger of ­running into a second and opposite fault by attempting to avoid a first (25–31). Th ­ ese ­earlier passages are also connected to 220–39 through pointed use of unusual words that 229. In bold ­because Silenus was a satyr, see above, p. 88. 230. The term is favored by the poets of Old Comedy to refer to their own genre, and is formed on the model of τραγῳδία as a compound of τρύξ, “wine lees,” and ἀοιδή/ᾠδή; see chapter 1, n70 and further Ferriss-­Hill 2015: 25–28.

94  ch a p te r 1

recur in the latter, again reflecting the poem’s under­lying concerns: obscurus and humi at 24–31, for example, as at 229–30; at 89–98 a marked clustering of comica . . . ​tragicus, indignatur . . . ​dignis (digni also at 24), versibus . . . ​verba, and sermone; and at 114 (intererit multum divusne231 loquatur an heros) a close resemblance to 237, with a further reappearance of the matrona from 116 at 232. In ­these vari­ous ways, lines 220–39 exemplify much of what I see Horace d­ oing in the Ars Poetica: while making certain explicit, first-­order points, he manages si­mul­ta­neously to suggest an array of further thoughts through the meticulous interpenetration of words and ideas. While the sort of patterning by way of glue described above may seem unremarkable in a book of Odes designed to be a unified ­whole and less necessary within a single, complete poem, the effects and procedures are the same, and the Ars Poetica no less than a book of Odes requires something to hold it together while (to paraphrase Lowrie) its pieces try so hard to drive themselves apart. As Nietz­sche famously said of Horace’s Odes, ­there is a “mosaic of words, in which e­ very word radiates its strength as sound, as place, as concept, to the right and to the left and over the ­whole, this minimum in the range and number of its signs, the maximum which this attains in the energy of the signs.”232 In the Ars Poetica, too, “when we throw up our hands and declare anarchy, countless interwoven connections re-­establish dialogue” in a pro­ cess that is “extra-­rational.233 Horace’s Ars Poetica thus comes to possess a bewitching, almost lyric musicality all of its own,234 such as Isocrates saw to be pre­sent in all poetry and absent from prose,235 and which inspired Pope to say of Horace that he “still charms with graceful negligence, / And without methods talks us into sense.”236 Fi­nally, by making use of a device—­verbal rivets of varying sizes that reach out and bind—­employed to such acclaim in the Odes, the Ars Poetica manages also to be covertly about lyric, even while appearing in the first place to be about drama, and perhaps in the second place about epic.237 In fact, perhaps this is why Horace chose drama as his primary 231. Divus(ne), resonant of Davusne (237), connects also to deus (227). On the alternate readings Davusne/divusne at 237, see further Orelli 1844: 724. 232. Nietz­sche, trans. Large 1998: 76 (from the essay, “What I Owe the Ancients”). 233. Lowrie 1997: 137, 244 of the Odes; see my introduction. 234. Kilpatrick 1990: 32–33 similarly notes the work’s “structural iridescence.” 235. Evagoras 10–11; see further Johnson 1982: 96–97. 236. An Essay on Criticism, 655–58; see further my epilogue. Wimsatt 1970 (in Brophy et al. 1970) notes, moreover, how Pope’s description engages with Horace’s own faux quandaries over ­whether his Satires qualify as poetry. 237. Brink 1971: 445–67, in his section entitled “Poetic Patterns in the ‘Ars Poetica’ and the ‘Odes,’ ” considers the two collections alongside each other, while Tracy 1948 argues for Horace’s

hum a n o  95

literary exemplum for the poem: drama is, in a sense, the Greco-­Roman poetic genre par excellence, yet one that Horace did not himself write in. By writing about it, however, and by deploying the characteristic tools of his lyric, which also shows sustained engagement with the mythologies informing tragedy, Horace is able to include it in his Ars Poetica. Likewise, through signs and signals, he writes also about satire, another genre in which he did work. Horace twice describes in the Ars Poetica the technique of circling, layering, and fastening that I see governing and defining the poem and dictating its pacing. ­These two passages are, moreover, themselves connected in the very fashion they describe. The first, at 46–48, anticipates the second, which comes directly ­after 220–39, as if drawing attention to the proliferation of joins that emanate outward from 220–39: in verbis etiam tenuis cautusque serendis dixeris egregie, notum si callida verbum reddiderit iunctura novum. (46–48) Horace’s callida iunctura has been much discussed, both in itself and via its reinvention in Persius’s hands as iunctura acris (“harsh join”), the hallmark of the latter’s highly compressed satire.238 Although the phrase has been widely interpreted as referring to units of two or more words (since a iunctura presupposes a ­thing that is itself joined and a ­thing to which it is joined), Horace’s emphasis seems rather to be on how a single familiar word can be made into a single new one through judicious placement.239 This form of (re-)creating is illustrated even as it is expounded: the sense that iunctura has ­here of a “joining” together of words is previously unattested.240 Moreover, ­there was an existing word to

style in the AP as being essentially lyrical in that, rather than arguing logically and presenting evidence in the typical didactic manner, he pre­sents images that elicit a certain emotional and aesthetic response. 238. See in the first place Dessen 1968, Brink 1971: 131–40, Ferriss-­Hill 2015: 32, 83–84, 233. 239. Armstrong 1968: 45–46 (cf. Oberhelman and Armstrong 1995: 251 with n67) and Brink 1971: 139 get it right: “the joinery of all the ele­ments, large and small, of a work of art,” which makes a traditional story “something new and perfectly your own,” and “the ‘refreshment’ . . . ​ of ordinary words in a pointed context.” For Armstrong 1968, the term iunctura is rightly critical to understanding the structure of the AP as a ­whole. 240. It is found prior to Horace to denote a straightforward joining of objects, as in Lucretius (DRN 6.1086), Caesar (B Gall. 4.17.6), and Virgil (Aen. 2.464, 12.274); cf. TLL and Oberhelman and Armstrong 1995: 252 with n69, who all note that a pos­si­ble ­earlier usage by Varro is not intelligible, and Armstrong 1968: 47. ­Later (e.g., Quintilian Inst. 9.4.32; Ruch 1963: 254) iunctura is attested with the sense of “the arrangement of words within a phrase.”

96  ch a p t er 1

describe this, iunctio, which Horace does not use.241 Also noted has been “the play on novum and notum, a word changed into one of very dif­fer­ent meaning through the change of a single letter,” which Hardie sees as suggestive of “Lucretian atomology.”242 Similar is the pair virtus et Venus just a few lines ­earlier (42), in which the words yield sense only in relation to each other: virtus typically means “manliness, courage” and from ­there “worth, excellence,” while Venus is the goddess of love, yet ­here they seem to denote complementary male and female qualities that appeal in art, as Horace forces Venus to take on a wholly new sense (subsequently made use of at 320), and makes us perhaps think also of the feminine qualities pre­sent in the creature of lines 1–4.243 Three forms of creation in short succession, then: iunctura—­a new meaning for an existing word (and co-­opted from another existing word, iunctio); notum/novum—­a word turned into another through the change of a single letter; and Venus—­a word that, through the com­pany it is made to keep (the pro­cess of iunctura, in fact) is compelled to generate a new sense for itself.244 The phrase in verbis serendis (46) is also noteworthy, and Brink observes the “endless debate” about its meaning.245 Latin has two verbs that begin sero, serere: the first is serō1, serere, sēvī, satum (“plant [seeds] in the ground, sow; cause to be born, beget”); the second, serō2, serere, seruī, sertum (“link together, entwine, interlace; join in a series, string together”). While apparently distinct (in 241. Cic. De or. 3.191: iunctio verborum (cf. Ruch 1963: 254, Innes 2003: 24), though it should be noted that Horace was constrained by the inability of the Latin hexameter to accommodate a single short syllable surrounded by two long ones, as in any inflection of iūnctǐō. Lines 46–48 overlap with some concerns of Philodemus, On Poems 1, including old and new words (col. 49, 169, 178), and the se­lection of words (ἐκλογή, 167) and their arrangement (σύνθεσις, 37, 55–57, 193). 242. Hardie 2005: 35–36; cf. Armstrong 1993: 228–29 and 1995: 231, Freudenburg 1993: 144–45, Oberhelman and Armstrong 1995: 251–52, Hardie 2009: 58 (who, with Brink 1971 ad loc., points out that “notum is not in fact a common—­a notum—­verbum for usitatum”) and 2014: 51. Armstrong 1995 is a useful overview of ancient atomology, i.e., the theory that individual letters ­were imbued ­ atter (a view found in with a special quality that would allow them to act as building blocks of m Plato, Philodemus, and Lucretius), on which see further also Kyriakidis 2006, Shearin 2015. 243. It does not appear to be attested prior to Horace’s use of it in the AP (twice: 42, 320) in the sense “(without sexual connotation) charm, grace” (OLD 3b), as Brink 1971: 129 also glosses it, though apparently this was a natu­ral enough usage that it was subsequently taken up, e.g., Sen. Ben. 2.28.2 (also juxtaposed with virtus), Plin. HN 35.79, Quint. Inst. 4.2.116, 10.1.79, 10.1.100. 244. The AP is full of similar instances; cf. Laird’s 2007: 138 comment ad 79–80 on its “mischievous wordplay.” Horace revives archaisms, e.g., honos (69) and sunto . . . ​agunto (99–100); he Graecizes, e.g., invideor (chapter 1, n106), iuvenentur (chapter 2, n147); and he coins new words, e.g., cinctuti (50; Brink 1971: 142; cf. Armstrong 1968: 59), delitigat (94; Rudd 1989: 166), promissor (138; Brink 214), numerabilis (206; Rudd 185 and my chapter 1, n189). 245. Brink 1971: 135–36.

125-81019_FerrisHill_HoracesArsPoetica_5P.indd 96

8/8/19 8:40 AM

hum a n o  97

addition to having dif­fer­ent third and fourth principal parts, they are described as deriving from dif­fer­ent Proto-­Indo-­European roots: *sē-­1 and *ser-­3, respectively),246 they share a semantic feature, since the action of sowing crops involves arranging them in rows or in order.247 Nevertheless, the word in question at Ars Poetica 46 is accepted as being serō2, since its presence in an idiomatic expression with a word meaning “conversation” (colloquia, sermo) is, if not exactly common, well enough attested.248 Yet no attested instances of the idiom that I have been able to locate unambiguously make use of the second verb, serō2,249 suggesting that perhaps ancient speakers (not to mention modern readers), in not differentiating rigorously between the two verbs in this construction, ­were uncertain as to which was the verb in question—­yet another play on language at 46–48, then. If we consider the form sermonem/sermones serere, moreover, it turns out that a cognate accusative is at hand, for ancient sources (correctly) derived sermo, “conversation,” from serō2.250 By substituting verba for sermo in the idiom, Horace has therefore done something rather subtle: on the one hand, he has created another form of callida iunctura in the very line where he pre­sents this phrase, and on the other, precisely through not using the word sermo, he manages nevertheless to suggest it and insert its presence into the passage. This passage in turn can then reach out to the vari­ous points in the Ars Poetica where sermo is used: lines 57, 69, 81, 95, and 229, where the possibility of Horace’s own satire (sermo) is ever pre­sent. In discussing at 240–43 the effort that must go into poetry and the difficulty of being original, Horace makes the rather off-­topic statement, “so potent are linkage and combination” (tantum series iuncturaque pollet, 242). In openly discussing the importance of “linkage” and “combination,” Horace is deploying the very same, since both serere/series251 and iunctura appeared previously at 246. Watkins 2000: 73, 76. 247. Pseudo-­Acro’s gloss blurs the two verbs in precisely this way: “serendis. Propagandis, ordinandis, ponendis, inserendis.” 248. See OLD s.v. serō2 2b, colloquia serere (cum), “to engage in conference, parley (with),” where the illustrative examples involve especially sermonem/sermones serere, and also Norden 1957: 182, Austin 1977: 87–88, Armstrong 1968: 41–48, O’Hara 1996: 167. 249. Cf. Innes 2003: 24. The ambiguity lies in the fact that all attested instances involve forms from the first two principal parts, which are identical for both verbs. 250. Varro, Ling. 6.64: “Sermo (‘conversation’), I think, is from ‘series’ (series), from where comes the word ‘garland’ (serta). . . . ​For a conversation cannot exist in one man alone, but where [­there is] speech joined together with another person” (Sermo, opinor, est a serie, unde serta . . . ​ Sermo enim non potest in uno homine esse solo, sed ubi oratio cum altero coniuncta); cf. Riganti 1978, Flobert 2003: 142, Sharland 2009: 3. 251. The noun series derives from the verb serere, the source also of sermo. Brink 1971: 135–36 likewise connects series (242) with serendis (46).

98  ch a p te r 1

46–48, and even within the same sentence, as they do again at 242. If Horace’s callida . . . ​iunctura denoted in the first place a clever and unexpected usage or placement that succeeds in reinvigorating a well-­worn word, it can now also be seen alluding to the clever and unexpected manner in which ­these two moments in the poem have been joined to each other. Horace would thus seem to have shifted subtly from discussing the arrangement of individual words (46–48) to discussing that of passages within the poem as a ­whole (242). The relation between the two goes further, however, for series describes a collection of points and iunctura the pro­cess of joining t­ hese together through attachment in two or more places.252 Having thus appeared twice, wedded to each another, neither serere/series nor iunctura appears again in the Ars Poetica, compounding the effects not only of each term individually, but especially in unison. Where Horace appears heavy-­handed and reiterative, as he so often does in the Ars Poetica, employing tedious repetition that has bored so many readers and even duped them into regarding the poem as a more or less straightforward literary treatise, ­there may be seen instead a subliminal layering and interleaving whereby we are subtly yet insistently invited to discern and extract certain key ideas from the larger work. As asserted in the introduction, the Ars Poetica ­will never succeed as a teaching document or guide to composition if scrutinized for what it says.253 Rather, if it is read and reread for what it is—in form, texture, and distinct poetic style—we may succeed in gleaning an altogether dif­fer­ent set of other­wise incommunicable lessons. Brink 1971: 457 has also considered the Ars Poetica in ­these terms, concluding that “perhaps no Horatian poem is so daringly successful as the Ars in suggesting by structure and tone the philosophy of art which it does not, and indeed cannot, express by conceptual argument.”254

252. Armstrong 1968: 46 also sees iunctura and series as key to understanding Horace’s poetic and stylistic proj­ect in the AP: together, they act as “agents of . . . ​transformation.” In addition, for him iunctura similarly works by “creating a semantic gap between juxtapositions of words, phrases, ­whole segments of a poem, and enlisting the reader’s aid to fill it in.” 253. Brink 1971: 517 expresses a similar view: “in my Prolegomena [1963] I put forward a claim: that the Ars would reveal its secrets when we had ceased making H. responsible for what is not likely to be his—­the outlines of a trivially constructed piece of literary theory—­and had begun making him responsible for what without doubt is his—­the features of his Horatian poem.” Nevertheless, Brink (especially his Appendix 1, 1971: 445–523) still concentrates on the traditional themes of ars and decorum, unity and variety, Rome in competition with Greece, e­ tc., i.e., the topics of an ars poetica rendered into poetic form. 254. Brink 1981: 7 speaks ­later, as well, of “the poetic essence of the ­great poem in which Horace had poetically demonstrated” his strong and consistent framework of literary theory. Brink takes to task the Re­nais­sance critics for their failure to “look for the amalgam of content and form which makes poetry poetic” in their focus exclusively on the poem’s contents (1981: 8).

hum a n o  99

What Horace cannot say and so shows instead is on the one hand how a verbal texture makes a poem and on the other that within and through such texturing, certain ideas may be accorded preeminence of place. Overattention to the external, most easily vis­i­ble structures necessarily prioritizes what the poem says over how it is, and it is for this reason that I have worked to trace for it ­here a unity of an altogether dif­fer­ent kind. The Ars Poetica’s “glossy, pregnant units, or capsules,” as Wimsatt terms them,255 are stitched together such that the cracks are often left showing: the ­whole (unum) is not straightforward or onefold (simplex), yet its coherence may be seen originating paradoxically in this. The fruitful interaction of form and content that governs the poem may, in fact, be what Horace had in mind when he speaks in the Ars Poetica of the poet’s need to mingle usefulness and plea­sure, to both instruct and delight (333–34, 343–44).256 If poetry is constituted by “the tension and difference . . . ​between sound and sense, between the semiotic sphere and the semantic sphere,”257 then the failure of the Ars Poetica to have been sufficiently appreciated as poetry lies in the outsized attention that has been devoted to ­these ele­ments individually at the expense of the interplay among them. Above all, however, the poem’s nature and concerns are contained within its opening lines, even its opening word, from where they radiate outward throughout its entire length. Illuminating the reason that a ramble through the poem such as has been presented in this chapter is both necessary and challenging, Russell speaks of the poem’s “printing in two tones” where “lines and sections read quite differently according to what you hold in mind from the context, and w ­ hether you look backwards or forwards.”258 Accordingly, “anyone who undertakes to guide a party round the poem is likely to be pointing out ­things that are not ­there, and missing ­things that are.” I would add, however, that t­ here is in actuality very ­little that is not in the Ars Poetica or that cannot be fruitfully and legitimately discerned within it and, as Goethe suggests,259 it may even be in this that its enduring appeal resides.

255. Wimsatt 1970: 136–38 (in Brophy et al. 1970). 256. On the impor­tant Epicurean background to this synergy (especially Philodemus and Lucretius with his honeyed cup, DRN 4.11–22), see Clay 1995, who is concerned especially with the tension (the “ancient quarrel”) between poetry and philosophy (on which see further several other essays in Obbink 1995a, especially ­those of Asmis 1995a, Obbink 1995b, Sider 1995a and 1995b, and Wigodsky 1995, and also Ford 2002: 46–66 and my introduction, n152). 257. Agamben 1999: 109; cf. 110: “All poetic institutions participate in this noncoincidence, this schism of sound and sense.” 258. Russell 2006: 328. 259. See pp. 3–4.

2 Pisones

The Identity of the Pisones The Ars Poetica has also been termed the Epistula ad Pisones, ­after the persons to whom it is addressed and ­after the fashion of referring to vari­ous of the poems in Epistles 1 and 2 with similar nomenclature. We glean the scantest of details about ­these figures in the course of the poem itself, however. From their first appearance in line 6 as Pisones, plural (as also at 235), we know that they must number more than one. We are further informed in line 24 (pater et iuvenes patre digni, “­father and young men worthy of the f­ ather”) that this unit consists of (at least) three members: a ­father and (at least) two sons, the elder addressed alone at 366 (o maior iuvenum).1 Fi­nally, they are dubbed Pompilius sanguis at 291–92, in what is taken to be an honorific reference to the descent claimed by the gens Calpurnia, of which the Piso ­family formed one branch, from Calpus, son of King Numa Pompilius.2 Five direct addresses, then, in four dif­fer­ent forms. The questions to be considered in this chapter are: why the repeated ad­ dresses? Why the variation in the form ­these take? And, most critically, why the Pisones (and which ones) at all? Since antiquity the preferred candidates for the position of dedicatees have been Lucius Calpurnius Piso (Pontifex) and his sons, an identification first attested in Porphyrio, whose commentary on the work’s opening line begins with “he sent this book, which is entitled ‘Concerning the Art of Poetry,’ to Lucius Piso, who was ­later the guardian of the city, and to his sons. For Piso himself was also a poet and a champion of liberal learning” (“hunc librum, qui inscribitur de Arte Poetica ad Lucium Pisonem, qui postea urbis custos fuit 1. The sons are invariably spoken of as two in number, evidently on account of the comparative adjective maior. 2. E.g., Landinus (at Bugada 2012: 123), Kiessling and Heinze 1914: 340, Antolín 2002: 212; alternatively, the phrase has been read as generalizing, denoting the kings of Rome as forefathers to all Romans; so Brink 1971: 322 (following Rostagni 1930: 85). 100

pis on e s   101

eiusque filios misit. nam et ipse Piso poeta fuit et studiorum liberalium antistes”).3 This par­tic­u­lar Lucius Calpurnius Piso, who lived from 48 BCE to 32 CE, served as consul (15 BCE), pontifex (which became attached to his tria nomina as a fourth ele­ment, presumably to distinguish him from the very numerous other similarly named Calpurnii Pisones), and praefectus urbi (13 CE onward; evidently what Porphyrio intends by the phrase urbis custos).4 He worked to quell a rebellion in Thrace (13–11/10 BCE), and he was patron to Antipater of Thessalonica. With this litany of achievements he followed very closely in his own ­father’s footsteps, both po­liti­cal and literary: his ­father, Lucius Calpurnius Piso Caesoninus, had been consul in 58 BCE5 and patron to none other than Philodemus.6 In addition to being known as the father-­in-­law of Julius Caesar, to whom he had married his ­daughter, Calpurnia (the Pontifex’s older half-­sister),7 Lucius Calpurnius Piso Caesoninus also possesses a certain mea­sure of notoriety as the object of In Pisonem, a vitriolic and fictionalizing screed in which Cicero attacks him for, among other ­things, his philhellenism.8 ­W hether or not the ­family owned the Villa dei Papiri at Herculaneum 3. The value of Porphyrio’s identification has been thought buttressed by the fact that what immediately follows (the note that Horace was working from Neoptolemus of Parium) has been understood as at least somewhat valuable (though see further my chapter 4), and it is pos­si­ble that both pieces of information came “en bloc from the same e­ arlier commentator” (Dilke 1958: 50) rather than being Porphyrio’s own inferences from the text of the AP itself; Nisbet 2007: 20 also rates him as “knowledgeable elsewhere on prosopography.” 4. See Tac. Ann. 6.11, Plin. HN 14.145, Suet. Tib. 42 with Dilke 1958: 51 and n19 and Prosopographia Imperii Romani entry 249 (Klebs 1897: 286–87), and further Syme 1980: 333, 1986: 329–45 (who, while providing detailed and thorough overviews of Piso Pontifex’s life and exploits, notes that he remains “an isolated figure” with numerous relatives that “elude detection”). 5. Piso Caesoninus’s numerous and distinguished c­ areer exploits are detailed in the appendices of Nisbet 1961 and at Syme 1986: 329–45, 367–81. 6. This widely stated fact is inferred from the perhaps two epigrams Philodemus addressed to Piso (27 Sider [= Anth. Pal. 11.44] and 38 Sider [of dubious authorship, Sider 1997: 200–201]), as well as from the dedication to him of at least one additional poem (On the Good King According to Homer) and from Cicero’s In Pisonem (chapter 2, n8); see further Sider 2005: 2, 5–8, 81–83, 1997: 5, n11, Dilke 1958: 50, Gigante 1995. 7. Calpurnia was born around 74 BCE and betrothed to Julius Caesar in 59 BCE, nine years before her half-­brother was born. 8. As insinuated at, e.g., Pis. 22, 42, 68–75, though as Sider 2005: 5–6 notes, Philodemus is not named in the work, being referred to instead as the Graecus who lives with Piso. In Pisonem owes much to vari­ous forms of invective lit­er­a­ture, as Nisbet 1961: 192–97 and Zetzel 2009: 228–31 elucidate (the work is “a repertory of the ways in which Romans traditionally insulted their enemies” and “not a single ele­ment . . . ​is even close to veracity”), though Zetzel also explains why Cicero’s animosity might be genuine (Piso was consul in the year in which Cicero was forced into exile, a ­matter in which he had evidently not intervened to the latter’s satisfaction).

102  ch a p t e r 2

specifically with its wealth of works by Philodemus,9 Caesoninus’s son, Lucius Calpurnius Piso Pontifex, was evidently raised in a rarefied cultural milieu.10 No sons of his own, however, are securely identifiable for the Pontifex to be the iuvenes patre digni required by Ars Poetica 24, rather problematizing Porphyrio’s other­wise attractive identification.11 Although this might merely mean that, if ­these putative boys did in fact reach adulthood, their po­liti­cal or military accomplishments fell short of their ­father’s and grand­father’s, nevertheless ­these young men from a prominent ­family on the cusp of adulthood would be expected to be attested somewhere.12 The commonly proposed alternatives for the Ars Poetica’s Pisones are Gnaeus Calpurnius Piso and his two sons;13 curiously, the precise relation between ­these two families bearing the name Calpurnius Piso has not been resolved.14 Proponents of this theory often point to the scholium of Pseudo-­Acro, which leaves the praenomen of the Piso in question open: “he wrote this [book] . . . ​to a certain Piso, a poet” (“conposuit istum . . . ​ad Pisonem quendam poetam,” Hauthal 1866: 575). Gnaeus was a figure approximately as prominent as Lucius, 9. The Villa dei Papiri was even for a time known as the Villa dei Pisoni, and while ­there has been more skepticism of late (e.g., Houston 2014: 87–129) the ­family is still regarded as the likeliest candidate (so Gigante 1995: 13, Janko 2000: 3, 2011: 5, n2, Sider 2005: 4–8 [who further suggests that the wealth of Philodemean papyri in the villa was the result of its ­owners acquiring or being given Philodemus’s own library upon his death]). 10. Dilke 1958: 50, Nisbet 1961. 11. Syme 1980: 340 is among the majority in concluding, “All in all, one may incline ­towards the Pontifex—at the cost of a necessary hypothesis, namely the pair of unattested sons” (though cf. 1986: 380: “It need not ­matter. The poem speaks”); see also Tsakiropoulou-­Summers 1995: 42–54. 12. Dilke 1958: 52 and Syme 1980 discuss numerous pos­si­ble candidates, but in each case the identification is problematized by details of date or geography or lineage (the name Piso being common to several families). The term iuvenes is also problematic in that it is vaguer than it might appear (Neue-­Pauly, s.v. “Youth,” “Age[s],” Iuvenes): Armstrong 1993: 192, 202, 218 would fix it at “ephebic age,” i.e., “fourteen to (just) sixteen,” but see Konstan et al. 1998: 12 (cf. Glad 1996: 34) on Philodemus’s habit of referring to “members of the community as ‘­those in preparation’ or the ‘young,’ ” where “the ‘young’ are beginning students of philosophy generally, irrespective of their age.” If the boys ­were younger than the scanty evidence from the AP indicates, it is odd that they would be framed as interested in writing; if older, the application of the term iuvenes to them would be infantilizing (even given the “long de­pen­dency of sons on their f­ athers” at Rome, Neue-­Pauly). 13. Attributed by Brink 1963: 239 to J. H. van Reenen (1806) and A. Michaelis (1877); he adds at 1971: 322 that he regards Gnaeus as “a modern, and I think unlikely, candidate for the place of the pater of H.’s dedications.” 14. Nettleship 1885: 173.

pis on e s   103

serving as consul suffectus with Augustus in 23 BCE (though he had fought against Julius Caesar in Africa),15 and connected to Horace (perhaps even personally) in that they had fought on the same side at Philippi.16 Moreover, he is known to have had two sons: Gnaeus Calpurnius Piso and Lucius Calpurnius Piso (known as Augur, presumably to distinguish him from Lucius Calpurnius Piso Pontifex et al.).17 The debate over Gnaeus’s candidacy for the Piso pater of the Ars Poetica hinges in part on ­whether his sons’ ages are suitable for the poem’s date (and the vari­ous arguments for the poem’s dates are, naturally, in turn bound up with the identity of the Pisones addressed: if Lucius Calpurnius Piso Pontifex is the addressee, the poem’s likely date is around 10 BCE; if Gnaeus Calpurnius Piso, it can be placed 10 years before that).18 The elder son, Gnaeus, was consul in 7 BCE, having started his official ­career around 25 BCE, meaning that, barring the earliest pos­si­ble date of 28/27 for the poem, Horace’s description of him as “still being shaped toward the right by a father’s voice” (quamvis et voce paterna / fingeris ad rectum, 366–67) would be rather odd (even given the Roman practice whereby sons, who might be adult or even el­derly, remained ­under the l­egal and financial control of their paterfamilias ­until the latter’s death).19 The other problematic aspect to the candidacy of the Gnaei Pisones is that they just seem unsuitable as the addressees of the poem, certainly in comparison to the Lucii Pisones, ­grand patrons of the arts: not only is ­there no indication of an individual or hereditary interest in lit­er­a­ture, but the familial temperament was, according to several sources, marked by brutality and arrogance.20 Nevertheless, one further detail in ­favor of this identification may be noted: Gnaeus Piso was “the first Piso to put a portrait of Numa [Pompilius] on his coins,”21 a fact that Horace may have had in mind with his address to o Pompilius sanguis (291–92). 15. The details of Gnaeus Calpurnius Piso’s life and ­career are again helpfully summarized by Dilke 1958: 51 and Syme 1980: 336–38, as well as Wilkins 1896: 330–32. 16. White 1993: 227; see further Nisbet 2007 for details of Horace’s life. 17. See Syme 1986, genealogical ­table XXV, for his Stemma Pisonum. 18. See Rudd 1989: 19–21. 19. So, too, Armstrong 1993: 202; cf. White 1993: 227, and my chapter 2, n12. 20. Tac. Ann. 2.43, 4.21, Sen. De ira 1.18.3–6, with Dilke 1958: 51, Rudd 1989: 21. Syme 1980: 339 muses, “The character and comportment of t­ hese Pisones . . . ​fails to agree with the writing of poetry or with favour ­towards poets, so some may be impelled to object,” though he adds, “Who can tell?” Geue 2014 has recently revived the argument that the Pisones are Gnaeus and sons (and the poem therefore written around 20 BCE), reveling precisely in their unsuitability as addressees and reading the poem as a caution against their “virulent Republicanism” (146): “an aggressive overture to an insubordinate ­family” (171). 21. Brink 1971: 322.

104  ch a p te r 2

A third line of argumentation, proposed by Frischer some de­cades ago though not taken up much since then,22 would have the pater of the Ars Poetica be Lucius Calpurnius Piso Caesoninus, the aforementioned consul of 58 BCE and patron of Philodemus—an elegant solution in that it merely moves up the addressees by one generation within the line of the Lucii Pisones and thus allows Porphyrio’s identification to stand. While attractive on the grounds of the f­amily’s literary patronage generally and that of Philodemus specifically (whose importance to the Ars Poetica is manifest), however, this Piso is known only to have had one son, the consul of 15 BCE known as Pontifex.23 Moreover, even if the composition of the poem is placed in the 20s, this “youth” would already have been in the pro­cess of ascending the steps of the cursus honorum, on his way to this consulship, again rendering the description of him as “still being s­ haped by a f­ ather’s voice” rather inappropriate as it would be also of Gnaeus Calpurnius’s sons. The abundance of information available about vari­ous historical Pisones is seminal to the approach to their identity ­adopted in this study: despite the rich and secure biographies that scholars have been able to reconstruct, the identity of Horace’s Pisones remains contested. It is therefore worth asking ­whether Horace’s con­temporary or near-­contemporary reader would have been able to decide any more effectively than we can on a single set of f­ ather and sons for the poem’s Pisones: the wealth of pos­si­ble candidates suggests not (as does the fact that Armstrong 1993 and Geue 2014 have both produced compelling readings of the Ars Poetica in which the Pisones must be, respectively, Lucius and sons, and Gnaeus and sons). We would also do well to remember that often in Horace, when t­ here appears to be a strict dichotomy of pos­si­ble interpretations, the choice proves illusory, with e­ ither reading producing good sense for a given passage. The effort that has been expended on arguments for one alternative or another may therefore be part of the game Horace plays with (or the joke he plays on) his reader, and I see the Pisones forming another instance of a situation in which it works well to accept any and all pos­si­ble interpretations as coexisting.24 The personal qualities and ­career accomplishments of a par­tic­u­lar 22. The major introductions to Horace and/or the AP since Frischer’s 1991 study make no mention of it, e.g., Nisbet 2007, Laird 2007 (who mentions Frischer but not his conclusions, 134, n9), Reinhardt 2013. Armstrong 1993: 190 calls the identification “startling,” while Janko 2000: 6, n3 gives a lukewarm endorsement. 23. Frischer’s 1991: 54 suggestion that the Pontifex could be the maior iuvenum despite the existence of his older half-­sister fails to convince (though the term can encompass females as well as males) and his “one attested son is better than none” is weak. 24. Comparable is my argument that the Plato named at Sat. 2.3.11 is not ­either the phi­los­o­pher or the Old Comic poet, but can represent both si­mul­ta­neously and in equal mea­sure; see Ferriss-­ Hill 2015: 40–41.

pis on e s   105

Lucius Calpurnius Piso (­whether Pontifex or Caesoninus) or Gnaeus Calpurnius Piso thus become of secondary importance, despite the tantalizing implicit presence of Philodemus into the poem that the former (two) would introduce (in fact, that the Pisones of the Ars Poetica can almost but not quite be reconciled with t­ hese ideal, philo-­Philodemean candidates makes the conundrum all the more delicious). The fact remains that despite having attracted arguably the greatest attention of all Horace’s putative addressees a­ fter Maecenas and Augustus, the Pisones have not been adequately studied as personae or actors within the Ars Poetica, where the manifold qualities any Pisos bring to the poem may and should still inform our readings of it.25 ­These may emerge as multiple, coexisting, even contradictory, and we should resist the impulse to resolve once and for all the unresolvable and ultimately wrongheaded ­matter of “who they ­really are.” I see Horace addressing not so much a specific f­ ather Piso and his sons, identifiable with ­great difficulty and even then only tentatively so, but rather a par­tic­ul­ ar class of person: Rocke­fel­lers or Kennedys, Morgans or Mellons, Clintons or Bushes, to suggest some less and more con­temporary analogues.26 Such names conjure up power, wealth, po­liti­cal and social connections, influence, and above all, the reach of ­these beyond a single figure and one generation (though not ­every scion attains the standing of the primogenitor): as Earl notes, the name Piso was “in the late Republic and early Principate proverbial for antiquity of lineage.”27 The Pisones exist in the Ars Poetica as aristocrats 25. This despite Wickham 1891: 383 already seeing that “the address to the Pisones . . . ​is not conventional or complimentary, but has a vital relation to the course of the poem” (cf. Kilpatrick 1990: 34–35, who describes Wickham’s work on the poem as “perceptive” yet “disregarded”), and the subsequent similar observations of Armstrong 1993: 185 (“can giving them greater prominence, and a share in the dialogue about the poem’s meaning and structure, move us nearer to a satisfying reading?”) and Reinhardt 2013: 524. Armstrong’s approach differs from mine in that he understands the specific Pisones addressed as critical to the poem’s meaning qua historical personages (192–93), since “the addressee was part of a Horatian poem in detail, every­thing that is known about them in real life contributing in some way to the explanation of ‘their’ poems” (190), whereas I consider the Pisones as personae within the poem, just as Horace himself pre­ sents a persona in the AP. Likewise, Geue’s 2014 reading relies upon the Pisones being Gnaeus and sons, though he well elucidates “the addressees as producers of thematics” (146), comparing the essays of Feeney 2002 and Freudenburg 2002 on Epistles 2.1 and 2.2, respectively. Compare Zetzel’s 1982: 95 description of Maecenas in Satires 1 as “an ele­ment in a work of art” and White’s 2007: 204 observation that “it is now taken for granted that the addressee is integral to the argument” of a Horatian poem. 26. Gold 1992: 164, n8 contemplates much the same in reference to Sat. 1.1, dedicated to Maecenas. 27. Earl 1960: 283.

106  ch a p te r 2

drawn in broad strokes, and they function as placeholders for any person who is not Horace but think s/he can be, for “what Horace teaches the Piso b­ rothers is fi­nally not what to do or not to do but what he can do and they cannot.”28 Above all, like the opening word humano, the Pisones allow Horace to introduce into the Ars Poetica from its opening vignette a series of interwoven key themes that ­will run the length of the poem and reach out to other points in his corpus as well: the relationships between ­fathers and sons, and their respective proper conduct and roles in the world; life cycles, ­human and not; and the interdependence of friendship and criticism. ­These, it ­will be shown, make up the poem’s fabric, fixed ­there by the Pisones—­not as they existed in any historical real­ity, but as characters Horace has crafted.

The Pisones in the Ars Poetica The first group Horace turns to address in the Ars Poetica are not quite the Pisones, but the amici of line 5.29 The last word of the poem’s opening sentence, amici serves to create a ring composition with the first, humano, and yet ­these humane and friendly terms enclose within them a mostly nonhuman hybrid creature. The amici are in the first place the paint­er’s friends, let in for a viewing of his busily disjointed artwork, but we may feel as readers that the group includes us as well (invited to view Horace’s artwork-­in-­words, and with it his entire poem). We feel this dual valence of amici, if we feel it at all, for a mere two words, however, since line 6 begins credite, Pisones, and we are forced to shift our understanding once again,30 as we also ­were regarding the gender of the creature’s head. The “friends” and readers Horace has in mind, it turns out, are not us, but the Pisones. It is slightly curious, then, that a­ fter he has demoted the ­bearers of this noble name to second addressees, his first word to them is a 28. Oliensis 1998: 198. The observation that the AP is not aimed solely at the Pisones but rather at any prospective writer is naturally not new, though its antiquity may surprise: the author of the twelfth-­century “Materia” commentary (Friis-­Jensen 1990: 336) already saw that Horace’s AP was both “communal” in addressing itself to any poets experiencing difficulty in the art (“communis ut doceat quoslibet poetas in arte poetica aberrantes”) and “par­tic­u­lar” and “private” in that it responded to a supposed request from some Pisos specifically (“specialis, id est privata, ut doceat Pisones, quorum rogatu hoc opus incepit”); early scholars such as Klingner 1937: 14 also wrestled with the uneasy presence of “hohe Freunde” and the addresses by Horace that seem directed at general pupil-­figures, “der allgemeine Du.” 29. The prevailing view (scholia, Rostagni 1930: 4, Brink 1971: 88) is that amici is vocative and concludes the opening sentence (alternatively, it is nominative as the subject of teneatis or the first word in the vocative of a new sentence that continues into line 6). 30. Horace’s conflation of amici with Pisones perhaps evokes Philodemus, epigram 27 Sider: it is addressed to φίλτατε Πείσων, an address that Sider 1995b, 1997: 152–60 translates as “friend Piso.”

pis on e s   107

curt command: credite. Our discomfort may increase with the intrusion of a voice into the poem in the next sentence: pictoribus atque poetis / quidlibet audendi semper fuit aequa potestas (9–10). Although commonly termed the anonymous interlocutor, both ­here and elsewhere in Horace’s hexameters, the name is inapt, since this speaking figure is not easily disentangled from the named addressee, ­either ­here or elsewhere.31 In the pre­sent instance, too, the prevailing reading has been that lines 9–10 are uttered by a nameless, disembodied figure, one unrelated to the illustrious addressees, even though it appears immediately a­ fter the Pisones have been invoked.32 While this assumption preserves any Pisos’ intelligence and honor intact, it deprives Horace of the full credit for having crafted a coherent poem, implicit within which is the fact that the world of the poem should be, if not self-­contained, then certainly self-­ consistent (as indeed Horace commands: sibi constet, 127).33 The objection of lines 9–10 is a slightly dim t­ hing to won­der aloud, and attributing it to the Pisones would be faintly unflattering ­toward them.34 As such, it would be the harbinger of what I see as their treatment throughout the remainder of the poem, in which they function not so much as aristocratic recipients of a work they have commissioned (the implied premise),35 but as 31. Consider, e.g., the overlap that runs the length of Sat. 1.1 between the poem’s “you” and its addressee, the fabulously wealthy Maecenas, as Horace delivers a tirade about material greed. Nevertheless, t­ here has been widespread unwillingness to see the poem as critical of Maecenas (e.g., Gold 1992: 169 ad 38–42, “it is difficult to imagine that Horace would adopt so abusive a tone ­toward Maecenas or call him timidum. We can therefore assume that Horace is no longer addressing Maecenas ­here”), although Horace has, for example, been read as rather aggressive ­toward his patron in Epist. 1.7 (Pseudo-­Acro, Shackleton Bailey 1982: 55–60, Armstrong 2004: 287 [“theorists of friendship in Horace’s poems have been embarrassed to account for Horace’s irony and harshness”], Kemp 2010: 75). 32. So, e.g., Orelli 1844: 700 (“respondet ficto adversario”), Antolín 2002: 181 (“un adversario ficticio”); Brink 1971: 91 allows that this “objection by an interlocutor unnamed” may be uttered by “the Pisos or anyone.” 33. While Brink 1971: 90 is right to say that “the fashion of pinning on the Pisos ­every address in the poem is open to objection,” it remains reasonable to accord the AP a certain degree of continuity and coherence in this re­spect, and Vahlen 1911: 752–57 (to whom Brink refers the reader) traces how the advice that follows a given direct address is applicable to the person or persons most recently addressed. 34. It would also distinguish them somewhat from typical addressees of didactic, where “­after Hesiod, the ­silent presence of the addressee becomes a convention” (Strauss Clay 1993: 24). In Horace’s Epistles it is similarly difficult to align at ­every moment the interlocutors that pop into poems (e.g., 1.6.52–54, 1.7.15–19, ­etc.) with their addressees. 35. That the AP was written in response to a request from certain Pisones has been part of the poem’s lore almost since the start, being stated simply as fact in the medieval commentary

108  ch a p t e r 2

placeholders for any reader—­a widespread trope in both literary epistle and didactic lit­er­a­ture. As Servius explains of the latter genre, “It is necessary that they be written to someone; for teaching requires the persona of both teacher and student” (“necesse est ut ad aliquem scribantur; nam praeceptum et doctoris et discipuli personam requirit”).36 Such placeholders, moreover, are traditionally on the less perceptive side: this is both natu­ral—in that they are viewed as being in need of instruction—­and a clever ploy, in that the discerning reader is able to separate himself from and feel superior to the named addressee of the poem he is reading. As Konstan describes, “If you already knew, Reader, what the teacher intends to communicate, you would not stand in need of instruction. It follows, accordingly, that you are just a ­little dim, or at least relatively unenlightened, and this image of you ­w ill inevitably be represented in the text.” Alongside this, “inevitably I, as preceptor, am closer to the source of truth; other­w ise, how dare I presume to teach you?” and “still worse: you also know—­since the text you are reading has been published and was not consigned to your exclusive and private perusal—­that ­others are reading, so to speak, over your shoulder. What their state of knowledge may be you have no way of ascertaining. They may well be much smarter than you are. You do not even know their names, but they know yours, which adds to the humiliation.”37 The tradition began already with Hesiod, who in the Works and Days addresses himself to his b­ rother, Perses, whom he repeatedly terms νήπιε, “fool” (286, 397, 633). In De Rerum Natura, which Konstan 1993: 13 describes as a “notorious example of condescension ­toward the addressee of a didactic poem,” Memmius is portrayed as “superstitious, intellectually ­limited, and prone to infantile fears,”38 an attitude aided by the Lucretian persona’s complementary conviction in his own moral and intellectual superiority.39 The Pisones share features also with addressees of the Satires and Epistles: in t­ hose tradition (Friis-­Jensen 1990: 323 quotes supporting passages; cf. also the anonymous commentary at Hajdú 1993: 246), but it should be noted that t­ here is nothing in the poem itself that would confirm this (aside from the suggestions of commissioning an artwork perhaps contained in aere dato, 21, or promissi carminis auctor, 45; cf. Rudd 1989: 153, White 1978: 86, and my chapter 2, n185. 36. Ad G. proem. and in reference to the Georgics; see further Schiesaro 1993. 37. Konstan 1993: 21. 38. Mitsis 1993: 125; Memmius fails to learn any Epicurean teachings by the end of DRN (Fanti 2017: 77–78). 39. E.g., DRN 1.410–17. Despite the widely acknowledged influence of DRN on Virgil’s Georgics, the latter has no counterpart to Memmius (the work is addressed to Maecenas, Virgil’s patron whom he shared with Horace, merely in the sense that it is dedicated to him, and beyond the mentions at 1.2, 2.41, 3.41, and 4.2, “he seems other­wise to have ­little to do with the poem,” Thomas 1988: 69; cf. Volk 2002: 122–39).

pis on e s   109

collections the Horatian persona views his reader (both internal and external) as ever in need of guidance and correction, yet the reader who considers himself better and wiser than this hy­po­thet­i­cal average common man and derives plea­sure from this sense of superiority occupies merely another stratum in Horace’s finely striated readership.40 In t­ hese ways, the Ars Poetica succeeds in combining ele­ments from didactic, literary epistle, and satire in crafting the figure of its interlocutor-­addressee. While ­these genres exhibit overlap as it is (mock-­didactic may be considered a form of satire, while satires and epistles in hexameters may be indiscriminately termed sermo),41 the Ars Poetica nevertheless emerges as a wholly innovative melding of them, and one that draws also on the ancient literary critical tradition. The poem instantiates, in fact, the idea that t­ here is r­ eally no such t­ hing as genre; t­ here are only individual works, each of which remakes in its turn the rules and limits of what its putative genre allows. Returning to the text of the Ars Poetica, we see Horace in lines 9–13 ­doing what he w ­ ill do throughout the poem: softening a critique with a conciliatory gesture.42 The short-­sighted and banal comment impudently ascribed to the Pisones (9–10) is followed immediately with scimus (11), an intimate first-­ person verb. While it could pass for a poetic plural, designating only the author himself, it also envelops within its embrace the addressee(s), and with them the reader, allowing that although a somewhat simpleminded objection has been offered in response to Horace’s opening gambit, this is perhaps not the speaker’s own opinion, but rather something someone ­else might say, and that should be indulged briefly for the edification of certain ­others who might not understand the fine distinctions being made. Our and the Pisones’ sense of being once again included among t­ hose in the know and u­ nder the poet’s benign protection is continued and confirmed by the next words, et hanc veniam petimusque damusque vicissim (11), as Horace acknowledges that we all (himself included) are guilty of minor and forgivable foibles such as this one.43 Yet if we thought we could rest secure in our understanding that Horace is, a­ fter all, 40. On this elsewhere in Horace see Ferriss-­Hill 2015: 149–50, and on the same in other authors, Strauss Clay 1993 with Konstan 1993: 13, and Gellar-­Goad 2012: 15–19. 41. See introduction, n41. 42. Armstrong 1993: 213 sees a similar alternation operating in the poem (­after “covering him with compliments” Horace works to “correct” the Piso son’s/sons’ ­mistakes). 43. If Pseudo-­Acro is right to say that petimus treats the subjects of the verb as writers and damus as critics, Horace also creates a sense of unity by showing how he and the Pisones jointly occupy all the areas that make up literary activity (though ­there is perhaps also the suggestion that the pro­cess of recitation and critique has become a closed and self-­serving echo chamber).

110  ch a p t e r 2

well disposed ­toward the Pisones and his external readers, we ­were mistaken: he ­will not let up. “But” (sed, 12), he goes on, a lack of unity and harmony such as that described in the poem’s opening lines is not permitted, not even ­under the guise of artistic license, for the state of affairs should not be such that “harsh ­things combine with gentle ones” nor that “serpents be paired with birds, lambs with tigers” (12–13). The flow and ebb of reproach and retreat established in the poem’s opening thirteen lines persists the length of the Ars Poetica, and although the impression of the poem’s tone that is left resonating in our ears is generally one of gentle understanding, this covers and conceals a steely unwillingness on Horace’s part, beleaguered as he may become at times, to yield on certain points. The opening thirteen lines have thus taken the reader on quite a journey. The section begins with a description of an image that we assume we are supposed to picture, though Horace declines to provide sufficient details that would allow us to do so in a satisfying manner; it turns to address amici who prove to be Pisones; it shifts the par­ameters of the discussion by introducing writing as an activity parallel to painting (­those of us expecting an Ars Poetica remain still unsatisfied);44 it allows a voice (whose, exactly?) to intrude; and fi­nally, it reminds us (for we do, in fact, need reminding) that Horace is in charge, and more knowledgeable and experienced than we are (a group that includes but is not l­ imited to the Pisones). Above all, the effect of the poem’s opening is to instill in the careful reader a sense of profound disquiet that ­will only grow more marked as the poem progresses: Horace ­will repeatedly criticize the Pisones and/or the poem’s “you,” yet soften any potential criticisms, however faint, by balancing each with a conciliatory, conspiratorial moment. Yet the reverse effect is also operative, whereby intimacy is undermined by subsequent critique. The anxiety engendered by this oscillation ­will be heightened by inconsistent use of second-­person singular and plural forms, as well as first-­person ones, as Horace blurs the lines between his Pisones and a single “you,” and between “I” and “we.” In ­these ways Horace w ­ ill keep us and the Pisones teetering between feeling as though we and they are safe, part of the poet’s in-­group, and realizing uncomfortably that perhaps the joke is on us, or at the very least on the Pisones. The burden of being a satirist’s reader45 is active also, it would appear, for Horace’s Ars Poetica. It is worth examining how ­these subtle yet significant effects are built up over the course of the poem. The dynamic evident at 1–13 next finds itself repeated some ten lines l­ ater, as Horace explains that “the greatest portion of poet-­­priests . . . ​ 44. Brink 1971: 93 ad 14–23: “­there is still no straightforward literary argument”; this is discussed further in chapter 4. 45. See further Ferriss-­Hill 2015: 170.

pis on e s   111

[we] are deceived by the appearance of right” (maxima pars vatum . . . ​/ decipimur specie recti, 24–25). The first words, maxima pars vatum, would lead us to expect a verb in the third-­person singular (with pars as its subject), yet instead, and prominently as the first word of the next line, we are given decipimur, a first-­person plural form that can only be rendered accurately into En­glish by clumsily inserting the pronoun “we.” Adding to the grammatical disharmony, the counterintuitively paired noun and verb are separated by an address in the vocative to “(o) ­father and young men worthy of the ­father” (pater et iuvenes patre digni), which might have led us to expect a verb in the second-­person plural. The effect, as before, is to create a closeness between Horace and his addressees, as all are encompassed within the verb decipimur.46 Since Horace speaks of himself numerous times in the first-­person singular, t­ hose relatively fewer instances in which a first-­person plural verb or pronoun is employed, as ­here, become especially striking.47 Despite this flattering oneness with the poet, however, decipimur admits a weakness, one shared by Horace and (explic­itly) the Pisones, all of whom in unison “are deceived by the appearance of right.”48 To mute this, Horace includes them all among the vates.49 “A new word for their profession,” the term is one of high praise: previously used to denote straightforwardly a “priest,” vates was co-­opted by the Augustan poets to describe their role as writers of “national, civic poetry” informed by “Alexandrian technique,” all while implying that they enjoyed divine support.50 The vates-­ideal is, as Newman 1967: 78 notes, spelled out in the Ars Poetica more fully than anywhere ­else among the Augustan poets, as Horace employs the term for the second and final time in outlining the prehistorical and historical roles of poets (391–407): Orpheus instructs primitive man how to eat and also soothes wild animals; Amphion founds Thebes by luring inanimate rocks to move through song, 46. So, too, Rostagni 1930: 9, Brink 1971: 106, Armstrong 1993: 206–7 (who compares Disraeli’s words to Queen Victoria, “we writers, ­ma’am”), Seeck 1995: 157. 47. The AP contains thirty-­three first-­person singular verbs (plus sixteen instances of the personal pronoun ego variously inflected and one of the adjective meus), as compared to only ten in the first-­person plural (with only two instances of the pronoun nos and three of noster). Excluded from ­these data are instances that occur in the direct speech of an interlocutor (i.e., where “I” does not refer to Horace). See Williams 1964: 190 for an analy­sis of Horace’s use of the first-­person singular, with the criticisms of Brink 1971: 107, 284–85; both attempt to categorize the instances into more and less personal usages. 48. So Orelli 1844: 702. 49. H ­ ere Wickham 1891: 391, Newman 1967: 77, Brink 1971: 92, and Armstrong 1993: 206 all see Horace identifying himself with the Pisones, and all of them together honorifically as poets. Newman 1967: 76–77 (contra Brink 1971: 106) argues convincingly for vates at 24 as distinct from poeta. 50. Newman 1967: 9, 12.

112  ch a p t e r 2

paralleling and even outdoing Orpheus’s control over wild animals; and Homer and Tyrtaeus sharpen with their words men’s souls for war. On account of such mythical and mystical achievements, “honor and the name came to divine poet-­priests and to song-­poems” (honor et nomen divinis vatibus atque / carminibus venit), the force of vatibus heightened by the fact that Horace’s model for his description, Aristophanes’s Frogs 1030–36, to which he other­wise hews quite closely, contains no direct equivalent to the term.51 Newman 1967: 78–79 summarizes: “The doctrine of the vates in this passage is extraordinary. They are spokesmen of the gods, they found cities, give laws, establish public morality and religion,” all “colossal claims.” The importance of the term in line 24, designating a category within which Horace, as vates, envelops the Pisones, is thus heightened by the remarkable and full definition of the term l­ ater in the same poem; and yet, the vates of 24 are all in unison “deceived” (decipimur). In addition, the phrase iuvenes patre digni need not be unalloyedly positive:52 if the ­father is excellent, it follows that the sons are excellent, too, but what if he is not? Horace has thus again succeeded in marrying hints of congenial intimacy and inclusiveness with hints of reproach, shifting continually between the two and not resting long enough to allow his reader to decide which way we are meant to understand him as feeling ­toward the Pisones. The section introduced by maxima pars vatum . . . ​decipimur specie recti is followed by a string of statements with a generalizing, proverbial air about them (25–30), yet Horace deploys words and concepts that have remarkably h­ uman, even bodily, senses to illustrate his caution against artistic errors. Accordingly, while levia, “smooth ­things,” certainly alludes to the “Polished Style” (the other two being the “Plain” and the “High”),53 the image we are given is of a man whose “sinews” (nervi) and “mind” (animi, as the seat of consciousness and emotions, but also “anger, animosity,” or “pride”) are insufficient to grant him the success at which he physically grasps (sectantem). Likewise, the person who has uttered g­ reat t­ hings (grandia, the “High Style”) is said to himself swell up,54 51. The closest analogue in Aristophanes is θεῖος Ὅμηρος, “divine Homer”; on the two passages, see further Tate 1928, Ferriss-­Hill 2015: 65–67. Horace’s passage is in turn taken up by Boileau at L’Art poétique, Canto 4 (see epilogue). 52. For dignus as sarcastic or ironic or as indicating that a person is worthy of something bad, see TLL s.v. dignus, e.g., 5.1.1143.72–74, 5.1.1145.53, 5.1.1145.62–63 (= Hor. Sat. 2.2.40, reference incorrectly given as 2.3.40), 5.1.1148.72–73, many of which are marked “per ironiam.” Geue 2014: 154 likewise points out that the phrase “is not necessarily a tag of praise.” 53. Brink 1971: 109–10. 54. This is highly reminiscent, through the use of turget and the literary application, of Horace’s critique of Furius Bibaculus (Sat. 1.10.36–37, 2.5.39–41; cf. Brink 1971: 111 and Ferriss-­Hill 2015: 199–201), and thus engages with the vast ancient Greco-­Roman tradition whereby a man’s

pis on e s   113

while the one trying to attain the “Plain” style is described crawling on the ground. This assortment of h­ uman bodies in vari­ous poses is summed up: “the avoidance of [one] fault leads to [another] vice,55 if done artlessly” (in vitium ducit culpae fuga, si caret arte, 31), the overall physicality punctuated by the culminating idea of flight (fuga). The warning at line 31—­which foreshadows the need to know one’s limitations, a theme that w ­ ill shortly come more fully into its own (38–41)—is allied with the advice of line 23,56 the neat prescription that forms a coda to 1–22 but introduces a second central tenet of the Ars Poetica, the idea that in attempting to avoid one vice a person is liable to commit its opposite.57 The formulation of line 31 is rendered more barbed than it might appear at first by the fact that, about ten lines l­ ater, by way of conclusion to the vignette of a craftsman who excels at sculpting fingernails and hair but is frustrated by his inability to produce a work that is pleasing as a w ­ hole, Horace, with an unexpected address,58 makes explicit whom he has in mind: sumite materiam vestris, qui scribitis, aequam viribus et versate diu quid ferre recusent, quid valeant umeri; cui lecta pudenter erit res, nec facundia deseret hunc nec lucidus ordo. (38–41) The italicized words account for three of the eight second-­person plural verbs to be found in the entirety of the Ars Poetica,59 where for the most part instruction is given via an unnamed third person, paraded through a series of hy­po­ thet­i­cal scenarios with the aid of jussive subjunctives. With an implied shift to

style (literary, oratorical, or other­wise) was viewed as reflecting his inborn nature (see, e.g., Möller 2004). 55. Horace uses two dif­fer­ent words for “fault,” first culpa, then vitium. Both can have a moral sense (vitium is often translated “vice,” culpa with “blame”) but are also found in literary critical contexts. 56. Brink 1971: 77 (see further 75–85) explains how the two are connected: “Only he who has mastered the art can produce a work that is all of a piece, a varied and articulate unity”; see also Caballero 2007: 81 with Fedeli 1997: 1473–74. 57. Horace’s emphasis throughout the AP on the idea of one error stemming from an overzealous attempt to avoid its opposite perhaps owes something to the premise of Philodemus’s On Vices and Their Opposing Virtues, which, as the title suggests, conceives of actions in opposing pairs. 58. Brink 1971: 121. 59. Th ­ ese are: teneatis, credite (5–6); sumite, scribitis, versate (38–39); versate, versate (269); reprehendite (292). I do not count the verbs at 155 (plaudite) and 459 (succurrite) since ­these are spoken by interlocutors and directed at the internal audience of a scene.

114  ch a p te r 2

the third-­person singular (cui),60 Horace explains that t­ hose who select their subject ­matter well are blessed with eloquence and clarity of logic (40–41). While use of a hy­po­thet­i­cal “someone” can pre­sent the opportunity for diplomacy, in that an eminent person is not faulted directly, the effect h­ ere is quite the opposite, as Horace is willing to grant that a person exists who is able to choose his topic wisely, but he w ­ ill not include that person among the second-­ persons plural (who necessarily include the Pisones) whom he is addressing. Moreover, the mood of scribitis seems pointed in the context: in place of a subjunctive (“you who would/may write”), which would convey an aspiration or plans for a potential ­future activity, Horace uses the indicative to communicate that the persons addressed are in fact already writing and to suggest that this effort, which is underway, is being hampered by the failure to choose material suited to the writer’s abilities. The use of the relative clause, qui scribitis, in place of the agent noun, scriptor, also seems pointed and rude in light of the bipartite schema traced in chapter 1: reserved for t­ hose who are still immersed in a given pro­cess, from which they may or may not emerge as a “doer” (­whether for good or ill) of the action they are performing, relative clauses allow Horace to emphasize the pro­cess of becoming.61 This is particularly the case when it comes to writing and writers: Horace himself is scriptor (235);62 he speaks honorifically of a scriptor cyclicus (136; cf. auctor, 77) and of the scriptor hypothesized at 346 who has attained international acclaim (mare transit) and longevity beyond his h­ uman life span (longum . . . ​aevum) and has in turn earned a “reader” (lectorem);63 while the successful “author of a promised song-poem” (promissi 60. “Implied” in the sense that Horace (perhaps deliberately) avoids using an active third-­ person singular verb with “someone” as its subject, offering instead the periphrases cui lecta pudenter erit res and nec facundia deseret hunc. Nevertheless, the individual in question is distinguished from the second-­person plural subject(s) of versate and any implied second-­singular reader. Neuburg 1993 notes similar shifts between second-­and third-­person singular in Manilius’s addresses to his reader. 61. E.g., 220, qui certavit; 288, qui docuere; 356–57, qui oberrat, qui cessat; 414, qui certat; 431, qui plorant—­each notable in that an equivalent agent noun is attested. 62. The phrase scriptor librarius at 354, moreover, suggests Horace’s own post as scriba quaestorius, as he perhaps humorously recalls being chastised for his errors as such. 63. Only once, at 120, is scriptor conferred upon Horace’s addressee, a move unusual enough that Wickham 1891: 400 explains it as a substantive in place of a participle with the sense “when you write.” This substitution on Horace’s part is, however, unlikely to be idle, and the noun serves rather to elevate and reify (perhaps ironically) his addressee’s literary aspirations. The case of scriptor at 120 is debated (see Brink 1971: 199, Rudd 1989: 169), and Rudd (171) is again keen to play down any pos­si­ble connection between the scriptor addressed and the Pisones (though I see l­ ittle reason why they should be exempted, especially given the connection scriptor makes to qui scribitis).

pis on e s   115

carminis auctor, 45) knows what to include and what to reject. The description of the Pisones as qui scribitis, therefore, with its indicative mood, emphasizes that they (like the external reader, perhaps) are currently immersed in the activity of writing, while leaving open ­whether they ­will ever emerge from it successfully as fully formed scriptores. Despite the evident bifurcation, however, the boundary between the two categories is permeable, since “someone who writes” may yet become a “writer” (in fact, one cannot become a writer other than through writing), though of course not all ­will succeed. Lastly, t­ here is at 38–41 no recusatio that might redeem the writer in question, since he does not himself (even if this ­were done with faux humility) refuse to take on something too big; rather, it is his shoulders that are forced to perform this refusal on his behalf.64 If we are prepared to see Horace as more critical of the Pisones than has generally been recognized at lines 9–13, 24–31, and 38–41, numerous other moments in the poem begin also to seem potentially subversive. “It is difficult to say common ­things in one’s own way” (†difficile est† proprie communia dicere), Horace intones at 128. The sentiment seems banal and generalizing enough, and it sets up the advice, “and you ­will do better to render the song-poem of Ilium into acts than if you are the first to bring forth unknown and unspoken t­ hings” (tuque / rectius Iliacum carmen diducis in actus / quam si proferres ignota indictaque primus, 128–30). Brink says that line 128 “has been described as one of the hardest in Latin lit­er­a­ture” and is “certainly the most frequently discussed of the Ars,”65 but the gist of Horace’s point seems relatively unproblematic and in line with his advice elsewhere: “­either follow the received tradition or craft ­things that are in keeping with themselves” (aut famam sequere aut sibi con­ve­ nientia finge, 119), he had ­earlier said, ­here refining the two options to add that in fact, although following received tradition is harder than it might appear (128), it is still less hazardous than being an innovator. The strug­gle, as at 46–59, is how to make something new and one’s own, w ­ hether a story or the words used to tell it. Horace’s advice is pointedly the opposite of what we might have expected:66 far from counseling his addressee to make the attempt regardless of its difficulty, he warns that sometimes it is best not to try to be primus—­a 64. Lowrie 1997: 212–13 rightly sees recusent (39) evoking recusatio, a device Horace makes use of throughout his Odes. 65. Brink 1971: 204, who devotes to 128–30 an entire appendix (I), and a second to 129–31. 66. “All commentators assume that Horace, in saying that something is difficult is telling the reader not to try it,” Williams 1964: 190 summarizes, though the impertinence of this as directed ­toward the Pisones has nevertheless been undervalued (for Williams, tuque has a “slightly adversative force” and he therefore reads Horace as saying, “It is not easy . . . ​but you are better occupied ­doing that than . . .”).

116  ch a p te r 2

term of high poetic praise that Horace, like numerous other Greek and Roman poets, proudly claimed for himself.67 Horace’s instructions beginning at 128–30 are directed at a person addressed with a succession of second-­person singular verbs (sequere, finge, 119; reponis, 120; committis, audes, 125; diducis, 129; proferres, 130; moraberis, 132; curabis, 133; desilies, 134; incipies, 136), their singularity further accentuated by tuque, “and you” (128). This concatenation of second-­persons singular is noteworthy, since they have been rare in the poem up to this point: a­ fter only scis (20),68 dixeris (47), vis (102), and loqueris (104), we find at 119–36 eleven second-­person singular verbs.69 Why the shift in number, and who is this individual addressee upon whom Horace has now fixed his eye? The seeds have been sown already in line 24, where Horace divides up his Pisones into the components pater et iuvenes patre digni. Although they are still considered a group and addressed at first with second-­person plural verbs (sumite, scribitis, versate), Horace has nevertheless succeeded in inserting into his poem the notion of their individuality, perhaps anticipating this with scis in line 20 and reflecting back on it with dixeris in 47. Although it is often stated that the singular “you” of any poem is both the singular addressee and the solitary reader,70 this is complicated in the case of the Ars Poetica by the fact that it has a plural addressee—­a choice on Horace’s part, and this can be no minor observation, that makes his poem unique among all extant Greek and Latin works of didactic and of literary epistle.71 The addressees in ­these genres are predominantly single, named individuals, such as Memmius in DRN, Maecenas in the Georgics, a Caesar in Manilius, Lucilius in Seneca’s letters, and all of the twenty-­two addressees in Horace’s Epistles 1 and 2.72 In the few instances where t­ here is more than one addressee, one or more of the addressees remains nameless, and the multiple addressees are not integrated into a single and collectively named group in the manner of the 67. Epist. 1.19.23–25, 32–33, Carm. 3.30.13–14. Rudd 1989: 172 further compares indicta (AP 130) with Carm. 3.25.7–8, dicam insigne, recens, adhuc / indictum ore alio. 68. Omitting quidvis (23), properly a pronoun, the second ele­ment of which is a verb. 69. The statistics in the AP overall for second-­person singular verbs and pronouns are as follows: thirty-­five verbs in the second-­person singular (in the imperative, indicative, or subjunctive) with sixteen instances of the pronoun tu and two of the adjective tuus. 70. Cf., e.g., Bing 1993: 99 on Aratus: “the intimate second singular” corresponds “exactly to our reception of the text as solitary readers,” and, in combination with an anonymous addressee, serves “to allow, or even encourage us to identify ourselves with this role in the act of reading.” 71. This feature also makes the Ars Poetica unique among Horace’s own verses: nowhere in the Epodes, Odes, or Satires does he address a named, plural group. 72. On which, see further White 1993: 224–39.

pis on e s   117

Pisones.73 By introducing into the Ars Poetica a plural named addressee that consists, moreover, of separate individuals, Horace is able to issue advice in the second-­person singular that applies both to any one of the Pisones and to the external reader. The vacillation from singular to plural might muddy the w ­ aters, but far from creating confusion, it has the effect of making the distinction between the poem’s internal and external readers less heavy-­handed and their ­union quite organic. More follows in the same vein, nestled among Horace’s litany of negative prescriptions—­and in fact, the constant barrage of instructions about what not to do might itself be sufficient to dishearten even the most committed wouldbe writer. In its emphasis on the negative, it may be noted, the Ars Poetica resembles one of Horace’s satires (the essence of satire being to show the undesirable) far more than it does Aristotle’s Poetics or Philodemus’s On Poems, guides that more frequently tell the reader what to do. “Nor w ­ ill you,” Horace warns, “begin thus as the cyclic writer once did” (nec sic incipies ut scriptor cyclicus olim, 136) by saying “the fortune of Priam I ­shall sing and the noble war” (fortunam Priami cantabo et nobile bellum, 137). Why not, we may well won­der? While we might have expected the answer to be that this has been done already (Horace has in mind the epic cycle of archaic Greece)74 and that our concern ­ought to be with ­doing t­ hings in our own way (proprie, 128), the justification given is rather, “what worthwhile ­thing could this promiser bring with such a gaping maw? The mountains w ­ ill give birth, a ridicu­lous mouse w ­ ill be born” (quid dignum tanto feret hic promissor hiatu? / parturient montes, nascetur ridiculus mus, 138–39),75 where the verb nasci serves also to connect to the birth and rebirth of leaves and words described ­earlier (62, 70) and the birth of a poem

73. E.g., Hesiod’s Works and Days addresses unnamed kings alongside Perses, and Empedocles’s On Nature contains, in addition to the numerous addresses to Pausanias in the second-­ person singular, six in the second-­person plural: one to “the citizens of Acragas as the φίλοι of the poet,” a second to ὦ φίλοι, Obbink 1993: 58, 76–80. 74. Already Porphyrio and Pseudo-­Acro see Horace at 137 translating the opening of a specific poem of the epic cycle by one Antimachus (Brink 1971: 214 accepts their assertion), which would render even more pointed the idea that fortunam Priami cantabo et nobile bellum has already been done. On the Greek epic cycle, see further Davies 1989 and the essays in Fantuzzi and Tsagalis 2015, especially their introduction, 1–40, and Fantuzzi’s chapter, 405–29. 75. Cf. Brink 1971: 212 (“it is a fault to begin in the manner of a ‘cyclic writer’, that is, with a proem ill adapted to the rest of the work”), Rudd 1989: 173 (“presumably we should regard the line [sc. 137] as [i] pompous in itself, and [ii] the wrong sort of prelude to an epic” b­ ecause “the scope of the announcement is too wide”), Fantuzzi 2015: 422 (who sees Horace objecting to works that, while referring in their incipit “to the ­w hole myth,” narrate “only a portion”).

118  ch a p t e r 2

that ­will come at 377 (sic animis natum inventumque poema iuvandis).76 The overt point is that a work’s opening must be indicative of and in harmony with what is to come (just as Horace’s poem reveals its own harmony through repetition of words and ideas from its first five lines), but the tone is scathing: the agent noun promissor, first attested at 138,77 stands in sarcastic apposition to scriptor (136);78 the dignity (dignum) of poetry is undercut by the grotesque and ludicrous gaping maw of the poet (hiatu), an image that would captivate both ­later Roman verse Satirists;79 and, as if the tiny mouse that results w ­ ere not obvious enough in its absurdity, Horace (in a departure from the Greek proverb that he other­wise seems to be translating)80 further brands it ridiculus, continuing the poem’s habit of presenting laughter as the universal response to incongruity and artistic failure. Epic as a subject for poetry is revisited ten lines l­ ater, and Horace further connects 136–42 to 128–30 through the term rectius, revealing his ongoing concern with “right”: quanto rectius hic, qui nil molitur inepte! ‘dic mihi, Musa, virum, captae post tempora Troiae qui mores hominum multorum vidit et urbis.’ (140–42) The suggestion that the Pisones (or some other putative writer) may be in danger of undertaking a literary work “ineptly” is stinging. The counterexample that Horace provides—­for we are invited to read lines 141–42 as an example of something not done inepte—­evokes both Homer and Virgil’s Aeneid,81 76. Cf. also 82, et natum rebus agendis, and 122, iura neget sibi nata. 77. Brink 1971: 214 (“prob­ably a Horatian coinage”), Rudd 1989: 173. 78. If scriptor designates a real, official “doer” (see chapter 1, pp. 75–76), so does promissor, but in an ironic tone that accentuates the fruitlessness of the promises, i.e., Horace is speaking of a habitual promiser who cannot deliver. The negativity of promissor ­here, moreover, complicates in retrospect the phrase promissi carminis auctor (45), since promissor stands as a combination of the latter’s promissi and auctor. 79. Juv. 6.636, Pers. 5.3. 80. ὤδινεν ὄρος, εἶτα μῦν ἀπέτεκεν; Brink 1971: 215 traces its history briefly. As he and Rudd 1989: 173 point out, Horace may also have in mind the exiguus mus of Verg. G. 1.181–82. 81. Horace is assumed to have the Odyssey in mind (e.g., Kiessling and Heinze 1914: 315, Blakeney 1928: 75, Rostagni 1930: 44, Rudd 1989: 174, Antolín 2002: 197), though the lines are “a paraphrase in two verses rather than a close translation” (Brink 1971: 217), yet resonances of the Aeneid have received ­little mention (neither Brink 1971: 217 nor Rudd 1989: 174 acknowledge the possibility, though Laird 2007: 141 does). Notable in par­tic­u­lar are the placement of events ­after the Trojan war (captae post tempora Troiae), Aeneas’s encounters with numerous ­peoples (mores hominum multorum . . . ​et urbis), the echo in virum of Arma virumque, and the delayed relative pronoun qui which, coming directly a­ fter Troiae, exactly parallels that of Aen. 1.1, Troiae qui

pis on e s   119

indicating that it is foolish to undertake a well-­worn subject (the Trojan war) and one that has been attempted and executed with ­great success and fame, both in the past (Homer) and very recently (Virgil). Existing material and works can, however, serve as a jumping-­off point for subsequent efforts (the events captae post tempora Troiae). Only in this way w ­ ill an author succeed in writing a poem entirely his own, which ­will be able to live up to the promises contained in its opening lines: non fumum ex fulgore, sed ex fumo . . . ​lucem (“not smoke from a spark, but light from smoke,” 143).82 Yet another warning comes just a few lines l­ ater in 140–42, as Horace continues the insistent theme of the Homeric epics and the Trojan war introduced already at 120–22 with the mention of the implacably angry and bellicose Achilles familiar from the Iliad.83 ­There is no need to start from the very beginning (146–47); rather, the gifted author ­“snatches his listener” (auditorem rapit, 149, reiterating the sense of motion from 100, animum auditoris agunto) in medias res.84 Crucially, “­those ­things which he worries are not able to gleam through his ­handling of them, he leaves aside” (quae / desperat tractata nitescere posse relinquit, 149–50), as Horace once again, though perhaps more ­gently and obliquely this time since we have shifted to considering a hy­po­thet­i­cal third-­ person author, urges the would-be poet to leave untouched what he cannot do brilliantly (a brilliance Horace demonstrates by having nitescere pick up on the lux of 143). If we thought the Pisones ­were safe with the shift to the third primus ab oris. Moreover, Boileau’s rendering of AP 136–42 at 691–701 (“cet homme pieux / Qui, des bords phrygiens conduit dans l’Ausonie, / Le premier aborda les champs de Lavinie”) markedly describes Aeneas, suggesting that he perhaps saw Horace’s passage ­doing the same. 82. Brink 1971: 218 (cf. Otto 1890: 137), comparing the expression “where t­ here is smoke ­there is fire,” adduces parallels from Plautus (Curc. 53–54) and Livy (10.24.13), but ­these fail to satisfy entirely, especially since, as Brink 1971: 218 understands it, “the meta­phor is switched midway; the cyclic procedure ‘befogs’ the reader, is unclear.” Nevertheless, Horace’s point is ascertainable: lux (“light,” especially “daylight”), a lasting and sustainable form of brightness emerging from the smoke that the initial spark has created, is desirable, while fulgor (“brightness, brilliance; flame, flash”), though perhaps temporarily intense and impressive, is not, especially as it dies out quickly (Orelli 1844: 734 also considers the respective qualities of ­these forms of light; cf. Fedeli’s 1997: 1520–21 lengthy exposition). The two halves of Horace’s image are thus complementary, each required to grasp the other’s sense in full. Compare also dare lucem at 448: lux, that is, excellent art, can be made to emerge from darkness (parum claris). 83. Cf. Brink 1971: 200–201. 84. Proverbial for us but, as Brink 1971: 221 points out, memorable in this context, especially when not separated from what follows: in medias res / non secus ac notas, “into the ­middle of ­things not other­w ise than if they w ­ ere known.” We do well to remember with Reckford 1969: 11 (cf. Günther 2013: 1) that such precepts “wither like branches cut from trees” when they are removed from “their living contexts.”

120  ch a p t e r 2

person, however, Horace begins emphatically with tu again at 153,85 instructing his addressee to listen to what he and, along with him, the audience desires. Horace’s unusual placement of himself in opposition to, rather than allied with, his addressee is striking: ego et populus mecum comes as something of a surprise in a line that begins tu quid ego, given that elsewhere Horace makes a point of joining ego with, rather than against, tu/vos).86 The effect is to direct the spotlight back onto the poem’s addressees, perhaps even in such a way as to make them uneasy. Horace’s reminders that successful creative endeavor requires a robust awareness of one’s own limitations as an artist are so insistent that it becomes hard to see how the young scions of the prominent Piso ­family, the poem’s ostensible first readers, would not by this point be squirming in discomfort. Armstrong 1993: 211–12 reads the scene thus: “In fact we are in a setting more familiar from Victorian and ­later novels and memoirs than from Roman lit­er­a­ture, but easily enough imaginable in Rome. An aristocratic sprig, male or female, has announced a desire to be a professional artist. He/she is confronted by his/her ­family, deliberately, with The ­Great X (pianist, painter, writer of established standing), a f­ amily friend b­ ecause of their patronage of arts, and made to realize the limitations e­ ither of his talents or of the artistic life itself.”87 Where Armstrong sees gentle (if firm) redirection of the Piso youths’ ambitions, however, I see an adamantine gatekeeper of the literary and other arts who cannot and w ­ ill not be persuaded that any Pisones know enough to realize how ­little they know and to understand how inaccessible the Horatian life remains for them. Horace is in the habit, as should by now be clear, of cycling back to a topic one or more times, e­ ither through strategic deployment of individual words or discussion of larger concepts, in order to restate and clarify or refine it. The 85. Brink 1971: 227 notes the emphasis (though as he sees it “tu continues the instruction offered to ‘the poet’ or ‘poets’, scriptor 120 . . . ​and qui scribitis 38,” since “the reader has not yet been told that any especial interest attaches to the elder son and he is not told so h­ ere”). 86. Brink 1971: 228 also notes that AP 153, where Horace “represents himself as part of an audience,” contrasts with ­earlier uses of the first-­person pronoun (103, 105, 182, 188), in which Horace is predominantly letting “his opinion be known.” 87. As Armstrong 1993 explains, the figure of “The G ­ reat X” is first found in relation to the AP in Cristoph Martin Wieland’s 1782 translation (see Seiffert 1968: 591 with Kilpatrick 1990: 33, Frischer 1991: 95, Reinhardt 2013: 522–23). Reading the dynamics of the poem through the lens of Epicurean therapeutics (on which in Philodemus, see now further Tsouna 2007: 52–87), Armstrong sees Horace extending to the Pisones the possibilities of improvement, self-­ improvement, and redemption (Horace “takes a much more graceful tack” than some of the examples he gives from Victorian and ­later lit­er­a­ture, 215), though with Oliensis 1998: 198–223 (cf. Reinhardt 2013), I ultimately do not see Horace offering the Pisos any path to becoming poets.

pis on e s   121

instance of 263–74, which Brink 1971: 302 calls “one of the central passages of the poem,” is representative: non quivis videt immodulata poemata iudex et data Romanis venia est indigna poetis. idcircone vager scribamque licenter? an omnis visuros peccata putem mea, tutus et intra spem veniae cautus? vitavi denique culpam: non laudem merui. vos exemplaria Graeca nocturna versate manu, versate diurna. at vestri proavi Plautinos et numeros et laudavere sales, nimium patienter utrumque, ne dicam stulte, mirati, si modo ego et vos scimus inurbanum lepido seponere dicto legitimumque sonum digitis callemus et aure. (263–74) This section lies sandwiched between two o­ thers tenuously related to it in thematic terms, the first (251–62) dealing with the technicalities of the iambic meter and its frequency in Accius and Ennius, the second (275–84) outlining the history of drama, from Thespis’s invention of tragedy, through Aeschylus’s introduction of the mask and of the distinctive palla garment, to the birth of comedy. Brink is right to see meter (numeros, immodulata poemata) as a thread connecting 263–74 to 251–62,88 yet more remarkable is the verb versate, not merely the same verb as at 39 but now repeated twice in the same line (269), such that of the eight second-­person plural verbs in the entire poem, three are one and the same. Th ­ ere is, however, some variatio, as the first usage of versate makes use of its intellectual sense (“turn over [sc. in your mind],” that is, “ponder, consider”), the second two a more literal one (“turn over” with your hands, though the intellectual sense naturally provides the undertone).89 Also repeated in both places is the word culpa (cf. peccata, 266), as the exhortation to pore over Greek exempla day and night (versate) is presented as a solution to the prob­lem first introduced at 25–26 (brevis esse laboro, obscurus fio) and 31 (in vitium ducit culpae fuga, si caret arte) and ­here phrased (with some additional nuances) as vitavi denique culpam: / non laudem merui (267–68).90 This form of 88. Brink 1971: 302, who adds that 263 has been seen as parodic in that it is itself “a metrically incoherent hexameter” (no ­middle caesura); Rudd 1989: 194 concurs. 89. As Brink 1971: 123 points out, the usage at 39 would usually be accompanied by a word such as secum (“with oneself ”). 90. Brink 1971: 304–6 says that 263–69 has “given much trou­ble” on account of its “extraordinary sparseness of expression.” Horace is clearly presenting two opposing courses of action, in line with 25–26 and 31: a person may ­either write what­ever s/he pleases (idcircone vager

122  ch a p te r 2

reduplication, typical of Horace yet deployed far more noticeably in the Ars Poetica (perhaps ­because of its form as a single, long poem) than in any of his other works, serves on the one hand to stitch the pieces of the poem together into a more cohesive ­whole and on the other to communicate the relative importance of certain themes, w ­ hether through implicit mechanisms (without realizing, we feel the weight of a par­tic­ul­ ar, repeated point) or explicit ones (the repetition ­causes us to sit up and pay attention, making us consciously alert to subsequent appearances of the same point—in this instance, the need for relentless effort and study in order to avoid error). That the Pisones are pre­sent in versate is indicated by the mention of vestri proavi,91 and while invoking the illustrious ancestors of one’s illustrious addressees has the trappings of traditional respectfulness,92 Horace subverts the motif by proceeding to say that ­these ancestors behaved stulte, “stupidly,” a strong word not much softened by ne dicam which precedes it. That ne dicam stulte would be a “boorish way of referring to [the Pisos’] ancestors” is not reason enough to preclude them from being the referents of vestri proavi; nor is it convincing to say that “the last address to the Pisos (235) is no longer in view,” since if the entire poem is addressed to them it is reasonable to see them as on at least some level the intended recipients of any second-­person plural advice dispensed in it.93 Concluding the section are not one but two scribamque licenter?), expecting (it is implied) that some indulgence w ­ ill be given, or s/he may exercise excessive caution and (by implication) not write at all, out of fear that someone may perceive and point out some faults (peccata). Whereas at 25–26 and 31 the ideal is the mean, ­here the cowardly ­middle course between error and praise unexpectedly receives some criticism. 91. Brink 1971: 305 says “not the Pisos, as has often been thought” (for Rudd 1989: 195 they are “­people who expressed literary opinions about comic writers”), but Armstrong 1993: 208 for one reads the phrase as referring to and criticizing the Pisones’ ancestors. The referent of vos (268) has also been much discussed: Brink sides with Rostagni 1930: 76 in taking it to mean “not only the Pisos but the Romans,” while Pseudo-­Acro explains “vos. Scilicet, o Pisones, vel poetae.” It seems notable, however, that although vester occurs nowhere e­ lse in the poem, vos is other­wise found only at 291–92, vos, o Pompilius sanguis, suggesting that it is reserved as the marked form of address for the Pisos. 92. Rudd 1989: 199: “a ceremonious, but not ­here a solemn, address.” 93. Brink 1971: 307; cf. 308: Horace “would not say to the Pisos that their ancestors ­were stulti” (though, e.g., Fedeli 1997: 1560 considers the possibility), even though, as he admits, “ne dicam often emphasizes what it apparently apologizes for.” As Brink admits (my chapter 2, n91), vos at 268 includes the Pisos, a mere two lines before vestri proavi; Kilpatrick 1990: 45 describes how “in retrospect, the vos of [268] does appear intended to draw the three Pisos more personally into the prob­lem in anticipation of 291–94”). Zetzel 1982: 98–99 is right to take issue with Cairns’s 1972: 4 claim that “in ancient lit­er­a­ture it is impossible that a poem addressed to a

pis on e s   123

first-­person plural verbs: scimus (repeated from 11, where it was one in a group of three first-­persons plural) and callemus. First-­person plural verbs total only ten in the poem, and they have so far been used to envelop the Pisos qua poets within Horace’s embrace, as in “we know, and we ask for and we grant this indulgence by turns” (11), “we are deceived” (25), “we are owed [to death]” (63), or “we w ­ ill take our time over” (178).94 Strikingly, and evoking lines 25 (decipimur) and 63 (debemur morti nos nostraque), scimus ­here has the highly unusual grammatical subjects ego and vos, neither of which alone would merit a first-­ person plural verb form but which together, considered as a single, combined entity, may do so, particularly if Horace wants to emphasize the unity of ego (himself) and vos (the Pisos).95 Any feeling of intimate unity, however, in a pattern we are coming to recognize, is once again undermined by the rudeness of what Horace is saying: that the ancestral Pisos had poor taste;96 that the latter-­day Pisos may be able to redeem themselves by evincing proper artistic discernment (though si modo, 272, pre­sents this as a mere possibility, far short of a guarantee); and that only dedicated study of Greek models (which we may assume has not been practiced to Horace’s satisfaction thus far) is the remedy ­either to writing without any discipline or to refraining from writing anything at all (i.e., committing one vice while attempting to avoid its opposite). In addition, we won­der ­whether the category of Roman poets, said to be unworthy of indulgence or ­pardon (264), might include the Pisos and their ilk.

patron-­cum-­dedicatee should be uncomplimentary.” That Lucretius’s rudeness t­ oward Memmius, for example, has been recognized, however, while Horace’s t­ oward the Pisones has received less attention, must depend in large part on the high po­liti­cal and social standing that has been traditionally imputed to t­ hese figures (­whether Lucii or Gnaei). Armstrong 1993: 213, who reads Horace’s words as aimed squarely at the historical Lucii Pisones, suggests that Horace is able to “install a distance between the implied wider audience and his individual narratees without compromising their dignity in the manner of Perses and Memmius” (description from Konstan 1993: 18, who endorses Armstrong’s view). 94. The ten instances are: scimus, petimusque damusque (11); decipimur (25); debemur (63); morabimur (178); scimus, callemus (273–74); speremus (331); velimus (345). 95. Kilpatrick 1990: 44 sees this done “tactfully.” 96. Though “archaizing” (Brink 1971: 307) tastes are the object of attack elsewhere in Horace (Epist. 2.1.28–49), h­ ere the prob­lem would seem to be e­ ither that the proavi, evidently Plautus’s contemporaries, are admiring his plays ­because they are the current fashion (i.e., precisely not “archaizing”), or perhaps (if the force lies in the adjective Plautinos) that they are admiring works in the Plautine style, i.e., derivative ones. Horace seems to be suggesting, as at Sat. 1.10.67–71, that aesthetic preferences are determined in large part by the era in which one lives.

124  ch a p t e r 2

The eighth and final second-­person plural verb comes at 292, and Horace uses it alongside the fourth direct address to issue a strong exhortation to the Pisones, who are once again identified in terms that appear honorific:97 vos, o Pompilius sanguis, carmen reprehendite quod non multa dies et multa litura coercuit atque perfectum decies non castigavit ad unguem. (291–94) This turn to the addressees as a group, the final one in the poem, once again has implicit within it a critique of their poetic abilities, for the reminder that one must be self-­critical and hone one’s work into an acceptable state (to perfection, ideally) through a time-­consuming pro­cess of erasure and correction presupposes a recipient who is at best young and inexperienced and at worst a verbose hack poet, such as ­those we have met in the Satires.98 We have moved—­the shift imperceptible ­until it is complete—­from the history and evolution of drama that occupies 202–91, though it began as early as 179, to the debate over how the most outstanding poetry is produced: ­w hether through inspiration or effort. Horace has, however, been priming us all along with a few judiciously placed words and phrases to receive and absorb the view he begins to lay out more fully beginning at 291: sweat, we have already learned, is critical (sudet multum frustraque laboret / ausus idem, 241–42), as is the avoidance of hastiness (morabimur, 178; cf. also 131–32, si / non circa vilem patulumque moraberis orbem), both anticipating the famous limae ­labor et mora.99 If we are paying close attention, we may si­mul­ta­neously discern the importance of daring and boldness: ausus at 242, and the Latin poets who have dared (ausi, 287) to move beyond the traces left by the Greeks. Horace’s turn to the Pisones at 291–94 and the instructions he issues to them consequently strike the reader, who has been carefully prepared to accept them, as self-­evidently true. The discomfort and self-­critique that must go into a praiseworthy poem is apparent. The ­human physicality of the pro­cess is accentuated by the phrase perfectum unguem, now within the Latin expression of which it forms a part: Horace had ­earlier described a sculptor actually shaping his statue’s fingernails (32–35) and h­ ere refers to the same in meta­phorical terms to describe a work 97. Brink 1971: 322: “the fourth address by name . . . ​and easily the most solemn”; “archaic solemnity (over-­solemnity, it would seem).” 98. Crispinus (1.1.120, 1.4.13–16, 65–70), Cassius Etruscus (1.10.61–64), Furius (Bibaculus; 1.10.36–37, 2.5.39–41), ­etc.; see further Freudenburg 1993: 109. 99. Brink 1971: 321: “the πόνος and ἀγρυπνίη of the Callimacheans, the trou­ble and time demanded by ars–­τέχνη, a motif known from all parts of this poem.”

pis on e s   125

finished to perfection.100 Horace has, yet again, crafted emphasis and connections through pointed and painstaking repetition of individual words: unguem, in its second instance in the Ars Poetica at 294, is used again (for the third and last time) just three lines l­ ater (297), and this double repetition invites us to go back and reconsider its appearance at 32, in the vignette of the craftsman whose talents are confined to one area (sculpting hair and fingernails).101 Compelling parallels, once again, emerge: the craftsman, performing the proverb in a­ ctual fact, adds the final touches that make a work perfectly complete, and the care he exercises should be a model for the Pisones, and all readers, as they/we contemplate how to write. In addition, by giving multa dies and multa litura in the singular, Horace draws attention to each individual day and erasure and thus their collective multitude, stressing the importance of time in the creative pro­ cess, as he w ­ ill at 388–89, when he suggests that a work should be put aside for nine years before its eventual publication (though at the same time, he is being playfully disingenuous in suggesting that t­ here is a “magic number” that w ­ ill suffice to produce a worthy poem).102 The web of textual, and with them thematic, connections traced ­here between the first forty lines of the poem and the section from 267 to 304 is dense and fibrous, but naturally incomplete—­ the paradox that only a 1:1 scale map can accurately represent the totality of its material.103 Nevertheless, it should have become evident how ­these serve to fasten the poem’s parts into a ­whole. Key among ­these rivets are the rededications of the Ars Poetica to the Piso ­family in vari­ous and varying configurations that punctuate the length of the poem.104 In the fifth and final address, Horace departs from the previous four: rather than addressing the ­father and sons together, the Pisones in unison, he singles out the elder son (introduced conceptually already at 24) for individualized 100. See Otto 1890: 355–57 s.v. unguis 6 with Porphyrio ad Hor. Sat. 1.5.32 on this widespread idiom, used to describe a person or an object polished to perfection, from the practice of sculpting the fingernails last, as the culminating detail of a statue. On the textual prob­lem, pr(a)esectum (“cut”) versus perfectum (so Shackleton Bailey 2001), see Brink 1971: 323–25. 101. Moreover, castigavit (294) picks up on castigator (174), a key term for Horace’s self-­ representation in this poem, as discussed in chapter 1. 102. The numbers nine and ten are familiar from the evaluative pro­cess elsewhere as well: Maecenas is described as welcoming Horace into his fold in the (suggestively gestational) ninth month of their acquaintance (Sat. 1.6.61–62), and Catullus says Cinna’s Zmyrna was published ­after nine harvests and winters (95.1–2), while at AP 365 Horace describes a type of painting that may be viewed ten times with plea­sure. 103. Most familiar ­today from Borges’s “Del rigor en la ciencia” (“On Exactitude in Science,” at Hurley 1998: 325), though ­earlier at Lewis Carroll, Sylvie and Bruno Concluded (1893: 169). 104. Compare the mid-­book rededications such as in the sixth poem of Satires 1 (Fraenkel 1957: 101).

126  ch a p t e r 2

advice in a passage that may be, though has generally not been, read as breathtakingly rude: o maior iuvenum, quamvis et voce paterna fingeris ad rectum et per te sapis, hoc tibi dictum tolle memor, certis medium et tolerabile rebus recte concedi. consultus iuris et actor causarum mediocris abest virtute diserti Messallae nec scit quantum Cascellius Aulus, sed tamen in pretio est: mediocribus esse poetis non homines, non di, non concessere columnae. (366–73) The nuances are hard to miss, and that the dismissive tone of Horace’s address at 366–73 has gone largely unnoticed105 is perhaps itself proof that Horace succeeded in walking the tightrope between flattery and overly frank speech.106 As the son of an aristocrat, from a consular ­family—­true w ­ hether we think of Lucius Calpurnius Piso [Caesoninus or Pontifex] or Gnaeus Calpurnius Piso or generic Pisos/Rockefellers—­a ­career in law and politics would have been the obvious path for the elder Piso ­brother, even if his natu­ral talents in it w ­ ere few. In certain spheres of activity and professions, as Horace informs the elder Piso son, mediocrity (medium, mediocris, emphatically repeated) is acceptable: in the l­ egal profession, not every­one need be a Messalla107 or an Aulus Cascellius to nevertheless be valuable.108 On the other hand, poetry and mediocrity (the idea repeated with insistence for the third time, mediocribus), he announces with a decisive swipe, are incompatible in the eyes of all—­men, gods, and booksellers. Young Piso ­ought thus to abandon his poetic aspirations as being impossible (due to his mediocrity). The point is repeated through two similes at 374–78 and 379–81: first, a dinner party is spoiled by discordant ­music, thick perfume, and foods improperly seasoned, as none of ­these are necesary and thus constitute excess;109 and second, he who cannot play sports has the good sense not 105. Though not by all readers: Russell 2006: 337, for example, perceives that “the appeal to Piso (366) . . . ​remains somewhat discouraging.” 106. Philodemus’s On Frank Criticism (Περὶ παρρησίας), which warns against b­ itter frankness for all but the most obstinate students (Glad 1996: 42–43), lies in the background to this way of reading, as ­will be explored at greater length in chapter 3. 107. I.e., M. Valerius Messal(l)a Corvinus, the “­great orator,” Brink 1971: 375; Rudd 1989: 210 notes that he was praised by Tacitus (Dial. 18.2) and Quintilian (Inst. 10.1.113). 108. On the identity of this Aulus Cascellius, see further Elmore 1935. 109. Armstrong 1993: 212 sees Horace drawing on Philodemus’s On ­Music, where he explains that it is “a necessary plea­sure to eat but an unnecessary plea­sure to eat elegantly or eat to ­music.”

pis on e s   127

to participate at all. Likewise, “in poetry the only alternative to excellence is nullity” and “a poem is ­either good or void.”110 As ever, however, Horace goes on to soften his attacks, framing his “harangue”111 with such flattery as per te sapis (367) and id tibi iudicium est, ea mens (“you have that judgment, that [good] sense,” 386). And, in turn, the softness of this is somewhat scraped away by the implication that the goddess associated with both wisdom and art may very well not provide her necessary cooperation (“you w ­ ill say or do nothing with Minerva unwilling,” tu nihil invita dices faciesve Minerva, 385),112 especially pointed in that the line begins with the grammatically unnecessary pronoun tu,113 and recalling the admonition at lines 38–40.114 Nevertheless, the point remains, echoing in our ears: “the one who does not know how to nevertheless dares to craft verses” (qui nescit versus tamen audet fingere, 382) b­ ecause he is “­free and nobly born” (liber et ingenuus, 383), a member of the equestrian class as rated by his property holdings (praesertim census equestrem / summam nummorum, 383–84),115 and tainted by no vice (vitioque remotus ab omni)—­necessary qualifications, and ones certainly possessed by the elder Piso son, for becoming a ­lawyer, but of

110. Brink 1971: 376, 378; the sentiment is the same as in the “Materia” commentary: “si non fuerit summum, erit infimum” (Friis-­Jensen 1990: 375). 111. Brink 1971: 381; Rudd 1989: 212–13 takes a rather kindlier view of 386, understanding Horace, with “a dextrous touch,” to be saying to the young Piso that he “­will not write poetry merely b­ ecause it is the ­thing to do; he ­will be guided by his natu­ral, creative, impulse.” Massaro 1974, noting Horace’s frequent juxtaposition of sapere and derivatives with virtus (also AP 308–9, 366–70), argues for an essential connection between them, sapientia being “il primo obiettivo dell’uomo” (116) and virtus the quality of manliness itself. Imputing sapientia to the elder Piso son ­here would thus be a marked form of flattery tied to his ephebic, proto-­manly state (which Armstrong 1993 understands as key to the AP). 112. Hajdu 2014: 95 entertains the possibility: “it may well be causative: ‘you w ­ ill not [or even ‘should not’] write anything, ­because Minerva does not want you to.’ ” 113. The elder son’s singling out is emphasized at 367 in the same way: tibi, te. 114. Wickham 1891: 424. 115. The phrase quibus est equus et pater et res at 248 evokes youths of similar status: equus denotes the equestrian class, pater “the sort of ­father who could be mentioned in polite society” (Rudd 1989: 191), and res wealth, perhaps even the specific amount required for equestrian standing (so Rudd, Brink 1971: 293). Horace’s Pisones / any historical Pisos must have been members of at least the equestrian class and may well have had senatorial status (the ­matter is complicated by the fact that dif­fer­ent branches of a ­family could have dif­fer­ent aristocratic statuses and that one’s status could change during one’s lifetime, since the qualifications w ­ ere of both birth and wealth).

128  ch a p t e r 2

l­ittle direct value for becoming a poet.116 The irony, of course, is that Horace himself was praesertim census equestrem / summam nummorum.117 In a sense the Pisos themselves are not to blame for their artistic mediocrity; rather, Horace allows that the Roman educational system may be at fault. Where Greeks are given inborn talent (ingenium) and the ability to speak “with a rounded mouth” (ore rotundo, 323–24) by the Muse herself, Roman boys are drilled in mathe­matics (325–30), as they are trained to preserve their familial wealth.118 Greeks consequently develop the desire for praise alone, whereas Romans unreasonably expect to produce poems worth preserving (speremus carmina fingi / posse linenda cedro et levi servanda cupresso?, 331–32) despite the “rust-­ canker and concern for pocket money” (aerugo et cura peculi) that has tainted their souls (animos . . . ​imbuerit, 330–31). While Dilke 1958: 54 observes that “the arithmetic lesson beginning at line 325, a power­ful if humorous attack on the over-­commercial basis of Roman education, is particularly apt if Horace was addressing the ­family of a prominent Philhellene,” Horace’s juxtaposition remains cutting, for the f­ ather and sons have presumably, for all their philhellenism, nevertheless gone through the Roman educational system and are therefore condemned to produce nothing creative of note. This is the son addressed at 366–73, for whom Horace, employing the language of athletics that we are to imagine would be familiar to any young Roman,119 spells out the fact that 116. Oliensis 1998: 211 reads it the same way: “If the Pisones harbor the belief that their social qualifications automatically qualify them to write poetry, Horace politely sets them straight by assuming that they w ­ ill join him in mocking the absurdity of such a belief.” White 1978: 90 (cf. my chapter 2, n117) suggests that “the relationship between the craft of poetry and the equestrian census was so familiar that Horace had only to invert it in order to find a useful irony.” 117. True of, if not exactly the historical Horace, then of his poetic persona (Sat. 2.7.53; the posts of tribunus militum and scriba quaestorius that he held indicate the same, Sat. 1.6.48, 2.6.36, Suet. Vita Hor., with Fraenkel 1957: 14–15). It has been pointed out (e.g., White 1978: 88–90; see further Armstrong 2010) that a huge proportion of Republican poets ­were of equestrian standing, suggesting that this was the optimal position from which to write poetry, and perhaps even that a nonequestrian poet might accordingly seek to pre­sent himself as equestrian. 118. Hardie 2016: 11 speaks of the “ste­reo­typical view” of the Romans as “a hard-­headed, practical race, interested in material gain.” Horace may also have in mind Philodemus’s On Property Management, which warns against the pursuit of excessive wealth (e.g., col. 13–14; Tsouna 2007: 163–94, 2012: 34–41) and gives advice for maintaining one’s inheritance (e.g., col. 15–21; Tsouna 2012: 40–61). Philodemus specifically mocks the Roman penchant for bud­geting in col. 25 (Tsouna 2012: 70–71). 119. Armstrong 1993: 213 sees Horace implicitly praising the Piso boys’ athletic accomplishments, while g­ ently suggesting that their literary ones do not match t­ hese. Alternatively, and since ­there is nothing in the passage to support the idea that the boys addressed have actually attained athletic excellence, Horace may be read as explaining to Roman youths that in the arena

pis on e s   129

the risk one runs in attempting something that has a low probability of success is ridicule: “the one who does not know how to play sports abstains from arms on the Campus, and untaught in ball[-­games] or discus or hoop he rests, so that the assembled crowd may not raise a laugh with impunity” (ludere qui nescit, campestribus abstinet armis / indoctusque pilae discive trochive quiescit, / ne spissae risum tollant impune coronae, 379–81; qui nescit repeated for emphasis at 382). ­There is patently much in the Ars Poetica that should make us rethink the role of honored addressees that has traditionally been accorded to the Pisones. It can hardly be surprising, especially in light of recent trends in Horatian scholarship, that such figures as the Pisones, and even Maecenas, would by no means be u­ nder special protection from Horace’s protreptics for the need for clear-­eyed self-­ examination. Nor can it come as a revelation that Horace’s addressees, even when they bear the names of impor­tant contemporaries, are no less figures that he has crafted for a par­tic­ul­ ar use in his poetry. Rather than assuming that the Pisones are Horace’s addressees ­because they requested or commissioned this poem (the conventional and deeply unsatisfying answer), we should rather consider what having them as his addressees, and addressees that are partitioned up in vari­ous ways and also subjected to an ungentle h­ andling, allows Horace to do in what is, ­after all, a complete and internally coherent artistic work, likely the culminating one of his poetic c­ areer. The first of several answers that ­w ill be ventured ­here is that the Pisones, in being framed as “­father and young men worthy of the ­father” (pater et iuvenes patre digni, 24), with the elder son subsequently extracted from this group altogether (366), exist in this form in order to facilitate Horace’s interest in and discussion of ­fathers and sons. ­Fathers and sons, in turn, bring into and make seminal to the poem the allied themes of teachers and students, of youth and old age, of life cycles and generations. ­These “power­ful unifying themes” of “change, growth, and decay,”120 which both stand on their own and fuse in and out of one another, play a recurring role in the Ars Poetica, yet it is the Pisones who serve to bind them into the poem’s substance.

Ages and Aging, Young and Old Horace’s interest in the passage of time, perhaps especially evident throughout his Odes, where the reader is urged to carpe diem (1.11.8) and reminded that death comes to all (1.4.13–14) for “the time for living is always now. Tomorrow is of athletics (unlike that of writing) they have the sense and experience (from a lack of past successes) to understand their limitations. 120. Innes 2003: 22.

130  ch a p t e r 2

uncertain; death inevitable and inexorable,”121 forms one of several undercurrents to the Ars Poetica.122 At Epistles 2.1.28–49, with the illustrative aid of the sorites paradox, Horace had ridiculed the notion that poems, like wines, improve with age: he demands that his reader-­interlocutor justify why 100 years and not a moment less is the magic number that makes a work “old and honored” (vetus atque probus) and sighs in exasperation at the fact that death alone elevates art to being worthy of admiration. So in the Ars Poetica Horace won­ders “what indeed w ­ ill the Roman grant to Caecilius and Plautus that has been denied to Virgil and Varius?” (quid autem / Caecilio Plautoque dabit Romanus ademptum/ Vergilio Varioque?, 53–55), as he affects once again to be baffled and irritated by a putative con­temporary habit of praising literary works based largely or solely on their distance from the pre­sent day. Yet a critical difference from Epistles 2.1 and the Odes emerges in that throughout the Ars Poetica Horace evinces a par­ tic­u­lar interest in the effects of time on the scale of individual ­human lives, a theme enabled and facilitated by the presence of the Pisones, specifically, their personae as f­ ather and sons. Aging first appears at 60–72, where it is initially applied to words but quickly expanded to encompass also the deeds of men. “We are owed to death, we and our ­things” (debemur morti nos nostraque, 63), Horace startlingly declares directly ­after three lines on old leaves and old words and the new ones that emerge in their places, as he “widens mortality to all works of man.”123 The word order enhances our discomfort, for debemur morti is clear enough and not in need of the added emphasis created by nos (the position of which is precisely emphatic, rather than an afterthought), let alone the syntactical oddness of nostra (“our ­things”) as the (joint) subject of a first-­person plural verb.124 The abiding 121. Fish 1998: 99. This worldview overlaps significantly with the Epicurean one, as Fish (contra e­ arlier scholarship) rightly argues, where death is both wholly unavoidable and not to be feared or fretted about (see further Nussbaum 1989, Pearcy 2012). In On Death (see Tsouna 2007: 239–311, Henry 2009), Philodemus juxtaposes young and old, particularly in relation to ­dying and the proper ­human lifespan (12.34–14.2, 17.3–18.14, 19.1–19.37), and discusses the inevitability of ­dying and thus the foolishness of fearing it (e.g., 37.18–38.2, 39.6–25), as he does also in On Choices and Avoidances, col. XVI–­XX (Indelli and Tsouna-­McKirahan 1995: 34–39; see 61–70 on the authorship of this work, which is tentatively ascribed to Philodemus). 122. This theme’s prominence in the poem, though generally overlooked, has been noted also by Geue 2014 (“the dynamic between young and old, f­ ather and son is a refrain of the Ars that sneaks into the unlikeliest nooks,” 159), though he connects it to the Gnaei Pisones as addressees (“the dramatic situation of this poem is poised at a critical moment between past and ­future personified,” embodying “succession anxiety”). 123. Hardie 2005: 38 (cf. 2009: 61). 124. That the phrase is thought to derive from a sepulchral epigram ascribed to Simonides (79 Gow & Page = Anth. Pal. 10.105), θανάτῳ πάντες ὀφειλόμεθα, makes it all the more striking,

pis on e s   131

concern from ­here onward is not with age as in itself admirable and as engendering admiration in the absence of other qualities but rather with age as against youth, with life cycles, and with death and rebirth, all heralded by the way in which Horace has expanded “the simile of the leaves on a tree . . . ​into a statement about the ­human condition.”125 As if to mirror the cycle he describes, moreover, Horace reformulates debemur morti nos nostraque into mortalia facta peribunt (“mortal deeds ­will perish,” 68), the phrases bookending the intervening list of examples of t­ hings that do not last forever.126 Alongside this ring may be found ­those created by nata (62) and renascentur (70), and cadunt (61) and cecidere cadentque (70),127 line 70 itself being further chiastic.128 Anticipating cecidere cadentque is licuit semperque licebit (58), past and f­ uture tenses juxtaposed, as Hardie sees, to emphasize both “changeability” and “constancy.”129 Fi­nally, honos (69) and honore (71) form another join. We feel not merely the presence and power of Nature, but her relentless cyclicality130—­a mirror, perhaps, of Horace’s own recursivity in the Ars Poetica, for as Mitsis rightly notes of meta­phor in reference to Lucretius’s De Rerum Natura, passages such as 60–72 are “not just isolated bits of formal beauty or of self-­expression, but part of an intended communication.”131 The be­hav­iors Horace imputes at lines 60–72 to words—­living entities, they are said to be born, flourish like young men, die, and even be reborn—­parallel t­ hose of ­people not merely in since “nothing in the Greek epigram answers to nostraque” (Brink 1971: 150). Compare Horace’s stretching of the grammatical subjects of decipimur (25) and scimus (272–73). 125. Reinhardt 2013: 515; he adds that in d­ oing so, Horace “aligns events in nature with ­those in poetic practice and thereby grounds them.” Hardie 2005: 37 rightly sees the history of Horace’s Iliadic source for the image as “itself exemplif[ying] a genealogy of mutability.” 126. See also Brink 1971: 151 on the correspondence between 63 and 68. 127. The polyptoton cadunt . . . ​cecidere, cadentque is all the more remarkable for the fact that ­these represent three of only four instances in the AP of the verb cadere (the other at 53), with all four, moreover, having a word that means “words” as their subject (as do renascentur and nata). 128. Renascentur . . . ​cecidere, cadentque . . . ​sunt, Rudd 1989: 162. 129. Hardie 2005: 38–39, noting the “tense-­shifting polyptoton” of both phrases that creates a ring around the passage, compares Lucr. DRN 3.969 (cecidere cadentque in the same line-­final position). He adds that Horace views the licentia to make new words as “an under­lying law of nature, as it ­were.” 130. ­There is a kinship in outlook with Philodemus’s epigram (Anth. Pal. 9.412) on the deaths of Antigenes and Bacchius, which Gigante 1995: 59 says “centers on the notion of an all-­powerful death,” where rebirth is yet pre­sent in the arrival of spring. 131. Mitsis 1993: 114. Reckford 1969: 138–39 sees 60–72 signaling “the importance of the themes of time and right growth” to the AP, further noting that Horace “asks for new words what in the Carmen Saeculare and Odes IV he had asked for Rome: a chance for the ­future to be born and grow rightly in its turn.”

132  ch a p te r 2

general terms, but specifically as they are described behaving within the Ars Poetica (158–76). Iuvenum ritu (62) even evokes the Piso sons themselves, twice called iuvenes (24, 366). The overall effect of lines 60–72 is to remind the reader of his own mortality, yet in a way that is oddly reassuring, as if emphasizing how natu­ral and inevitable aging and death are—­the same outlook, in fact, that runs throughout Horace’s Odes.132 Such themes might seem out of place in an ars poetica, but the Pisones—­father and sons, old and young—­help to anchor and contextualize them. The idea that begins to emerge is of all ­things in existence—­whether ­human or made by ­humans or existing in nature, such as plants or ­water—as subject to the same forces, in par­tic­u­lar decay and death. Far from seeming extraneous to the poem, this theme is woven throughout its length and tacked firmly into place fives times by each of the five direct addresses to the Piso ­family. Horace’s notable interest is not merely in young men, but in young men in conjunction with their elders. The patterning is densely woven and consistent, from the poem’s opening lines to its closing ones. From its first appearance in line 24, the term iuvenes is intertwined with pater (maxima pars vatum, pater et iuvenes patre digni, / decipimur specie recti, 24–25), as Horace primes his reader to think of iuvenes throughout the poem both as interlinked with his addressees (first as Pisones in line 6) and in conjunction with their ­fathers. Next, as he turns his eye to words and leaves, Horace speaks not idly of how “the tongue of Cato and Ennius enriched the paternal conversation” (cum lingua Catonis et Enni / sermonem patrium ditaverit, 56–57), while figuring the words involved as behaving—­being born, flourishing, and thriving—­“in the manner of young men” (iuvenum ritu, 62). Young and old are most overtly juxtaposed and contrasted in lines 153–78, yet the life cycle outlined t­ here intersects with the abiding interest in f­ athers and sons that runs throughout the Ars Poetica more than in broad strokes, for key individual words are repeated. At 153–78 the young man (iuvenis) comes on the heels of a boy (puer) who has just mastered walking and talking but exists in thrall to his emotions and precedes a man who is reaching or has reached maturity (his attention is on public life), leading fi­nally to an old one (senex), caricatured as a curmudgeon. Connections are forged with iuvenes mentioned elsewhere in the poem through the presence of the Campus as the locus of athletics (162; cf. 379, campestribus), of ­horses (162; cf. 84, and also equus next to pater at 248; we may also think of the h­ orse’s neck from the poem’s opening, where the Pisones ­were pre­sent as well), of vice (vitium, 163; cf. 384, and also 31), and of the protection afforded

132. On this view in Lucretius and its Epicurean-­ness, see, e.g., Fish 1998, Reinhardt 2002, O’Keefe 2003, Taylor 2007.

pis on e s   133

by a parent or pedagogue (custode, 161, and monitoribus, 163; cf. 366–67).133 Beyond ­these, the spirit (animus) is able to move and be moved (animusque virilis / quaerit opes et amicitias, 166–67), as at 100 (animum auditoris agunto) and 111 (post effert animi motus interprete lingua), while within the passage ­there may be found repetition in aetas and anni, incommoda and commoda (169, 175), and puer (158, 174). In the course of delineating this life cycle, moreover, Horace continually reminds us that time is movable and ever moving on (for us, too, lest we forget), beginning with mobilibus (157), through mutatur (160), remoto (161), flecti (163), conversis (166), mutare (168), circumveniunt (169), and dilator, futuri, and temporis acti (172–73), all contributing to the feel of a system that is in constant and unstoppable flux. Th ­ ese crescendo to multa ferunt anni venientes commoda secum, multa recedentes adimunt (175–76), as Horace encourages us to think of aging not as a linear pro­cess that culminates in old age, but as a cycle that goes and comes around again (incommoda and commoda create an additional feeling of ring composition), just as he depicts it also at 61–62 (ita verborum vetus interit aetas et iuvenum ritu florent modo nata vigentque) and 70 (multa renascentur quae iam cecidere).134 As a passage richly textured both within itself and across the larger work, 158–78 succeeds in communicating Horace’s abiding interest in h­ uman development. In summing up the contents of lines 153–76 at 176–78, Horace pairs his four stages in an intriguing way that further reminds his reader about his quadripartite division of ­human life. As he ­will at 389–90, Horace counsels deliberate action and caution: ne forte seniles / mandentur iuveni partes pueroque viriles, / semper in adiunctis aevoque morabimur aptis. Having first presented the natu­ral progression from boy to young man to mature one to old one, Horace ends by contrasting old men with young (seniles . . . ​ iuveni) and boys with adult men (puero . . . ​viriles).135 The effect is to emphasize the father-­son, old-­young theme that runs throughout the Ars Poetica and remind us that it is a dichotomy, yet at the same time Horace manages to convey the seamless continuity from one stage to the next, rather than emphasizing any firm breaks between them. The final instance of iuvenes in the poem at 366 reiterates and expands upon the ideas pre­sent in and near its ­earlier occurrences. Although he stands on his 133. Armstrong 1993: 203 sees the description of the iuvenis at 161–65 as “aimed slyly at the younger Piso (and echoed ­later in words directly addressed to him).” 134. Brink 1971: 241 notes the contrast between recedentes and venientes and adduces Sat. 2.6.40, Epist. 2.1.147, and Carm. 4.11.19 as comparanda for “the meta­phor in H. of years rising to, or approaching, an acme, and then falling or receding.” 135. Cf. Brink 1971: 242: “The fourfold division is preserved . . . ​but the types are now interwoven”; moreover (242–43), “viriles rhymes with seniles” and “rhymes are formal means of bringing out features of the content.”

134  ch a p t e r 2

own, the “elder of the youths” (maior iuvenum) still cannot be separated from his ­father, who, with echoes of pater et iuvenes patre digni, appears in the very same line as a disembodied voice, voce paterna. Connections are forged with lines 24–25 not only through repetition of iuvenes and pater/paternus in close proximity to one another, but also by the reappearance of “right,” rectum (367; cf. recti, 25), further accentuated by recte (369), as well as of the notion of a brotherhood of poets (vatum, 24; poetis, 372), with the question implicit within it of ­whether the Pisones and the reader may claim membership or not. Tempted as one might be to seize the spot being offered among such rarefied com­pany, it should be noted that Horace’s vates are deceived by appearances (decipimur specie recti, 25) and that the poets of line 372 are “mediocre” (mediocribus . . . ​poetis). The superficial attractiveness of the titles poeta and vates, then, is sullied by the ways in which ­these figures are described, such that no reasonable person would or should aspire to be counted among e­ ither group.136 Next, although Horace flatters the elder Piso son by imputing to him good judgment (367, 385; though, in my reading, he in the same breath excoriates his lack of poetic talents), he reminds him at the same time that (what­ever his a­ ctual age) he is not yet a fully formed man since he remains ­under his ­father’s protection and tutelage, voce paterna fingeris ad rectum—­the same direction Horace, speaking to himself as if in the voice of his own ­father, exhorts for himself at Satires 1.4.134 (rectius hoc est). Horace reminds us of Piso se­nior again ­later in the section that had begun o maior iuvenum, when he describes some hy­po­thet­i­cal writings making a strange descent: si quid tamen olim scripseris, in Maeci descendat iudicis auris et patris et nostras nonumque prematur in annum membranis intus positis. delere licebit quod non edideris; nescit vox missa reverti. (386–90) While the verb scripseris, in a somewhat rare appearance,137 emphasizes writing, it is once again surrounded and overshadowed by words that emphasize 136. Newman 1967: 76 is, moreover, surely right that Horace uses poeta at 372 (mediocribus . . . ​ poetis) rather than vates to emphasize the mediocrity of ­these par­tic­u­lar poets. 137. The verb is, as Lowrie 1997: 56–57 notes, extremely rare in the Odes (see chapter 1, n100) in uncompounded form with the plain meaning “to write.” In the Satires, by contrast, Horace uses the verb considerably more often (e.g., 1.4.12, 1.4.13, 1.4.16, 1.4.23, 1.4.41, 1.4.57, 1.4.65, 1.4.75, ­etc.), as he does in the Epistles (1.3.7, 1.4.3, 1.6.43, 1.9.13, 1.15.25, 1.16.4, 1.19.3, 2.1.109, ­etc.). Its comparative rarity in the hexameters of the AP (a work that, it bears repeating, has writing among its explicit topics) is thus notable.

pis on e s   135

speaking: dices (385, with facies) and vox. In addition, it is rather at odds with the ­father and sons as described ­earlier: while in line 38 they are described as currently writing (qui scribitis), though notably not as “writers” (scriptores), ­here the elder son, at any rate, is presented as not yet having written anything at all (si quid tamen olim / scripseris),138 and Horace further sarcastically implies that his good judgment (id tibi iudicium est, ea mens) ­will prevent him from ­doing so. Saying and making (dices faciesve) are presented as ­future activities that require Minerva’s cooperation; writing, on the other hand, seems a distant possibility (si quid tamen olim). Above all, poetry, once released, has ­great longevity (as in Odes 3.30), ­whether for good or bad, so a writer must be certain that what he sends forth is not something he ­will ­later regret. Although line 387 is somewhat marred by textual prob­lems,139 it is clear enough that the writings are to be heard and judged by the named individual, by the Piso boys’ ­father, and by Horace himself, who accordingly becomes a father-­judge-­teacher figure, perhaps the only reliable one in the trio, since the boys’ ­father is unlikely to be appropriately critical of his ­children’s verses.140 Sandwiched between the opening (24–25) and closing (366–90) instances of iuvenes in conjunction with their ­fathers—­where the theme of didaxis is furthermore pre­sent—­are several further instances of the word iuvenes or a related form, and similar concerns are often evident. The effect of ­these isolated appearances of the key term iuvenis, and in a variety of eye-­catching variations, even neologisms, is to sustain the importance accorded to the Piso youths throughout the poem, even when they have receded from view for a moment. Thus at 83–85, Horace speaks of the “concerns of young men” (iuvenum curas) in defining lyric, but he associates the genre further also with epinician poetry and, thereby, with athletics (“and the victorious boxer and the ­horse first in the contest,” et pugilem victorem et equum certamine primum, 84), the purview of iuvenes as at 161–62 (imberbis iuvenis, tandem custode remoto, / gaudet equis canibusque et aprici gramine Campi). In 114–18 ­there may be found in compressed form a point similar to that made at 153–78 as Horace pre­sents a number of contrasting pairs, reiterating the need for each character to be recognizable as

138. Rudd 1989: 213: “The natu­ral inference is that the young man was not engaged in writing a satyr-­play or anything ­else.” 139. Variations including Maecii, Meci, Meti, and Mettii are found, and if the reading Maeci is correct, the individual in question is evidently Spurius Maecius Tarpa, appointed by Pompey in 55 BCE to select plays for inclusion in the festival to inaugurate his new theatre; cf. Brink 1971: 382–83, Rudd 1989: 213. 140. In addition, the reference to Tarpa as judge at Sat. 1.10.38 is “derogatory” (Brown 1993: 188), and Cicero had also critiqued him at Fam. 7.1.1 (cf. Gowers 2012: 324–25).

136  ch a p t e r 2

such: divinity and hero;141 old man and young one; married ­woman and nurse; wandering merchant and (stationary) small-­scale farmer; Colchian and Assyrian; Theban and Argive. Listed second, the senex and his counterpart are each further described in evocative terms: the old man is maturus and the young one depicted as “hot with still-­blooming youth” (an adhuc florente iuventa / fervidus), a formulation that allows Horace to avoid the much-­repeated term iuvenis itself. To be maturus is to have attained adulthood, to be experienced, and to be at the proper stage and ready for what is yet to come,142 and this adjective therefore somewhat softens the negative valences of senex, creating an image of a man who is older but not yet elderly—­a perfect advisor-­figure, in fact, rather than an old man at the end of his life. Florente, on the other hand, heightened by fervidus (which loosely anticipates Achilles, 120–22)143 and with the general idea reinforced by virentis144 directly below it two lines ­later, evokes iuvenum ritu florent (62), a correspondence strengthened by the repetition of iuventa-­ iuvenum as well and by the fact that the verb florere occurs only ­these two times in the poem.145 Into this contrasting pair, then, Horace manages to insert rather forcefully the idea that the young man ­w ill grow into an old one in an ever-­ moving cycle of birth, death, and rebirth.146 Fi­nally, Horace coins at 246 the verb iuvenor, apparently on the model of Greek νεωτεϱίζω,147 to issue the instruction that fauns not behave like young men, in a passage that has the overall effect of drawing our attention, once more, to Roman city youths and their 141. See Orelli 1844: 723–24, Brink 1971: 192–93, Rudd 1989: 168–69 on the readings divus and Davus, and the distinction Horace intends between divus and heros. 142. OLD 2, 4, 5. 143. For Brink 1971: 193 both epithets are “traditional,” though this somewhat undervalues the emphasis on youth and its bloom that repeats in vari­ous ways throughout the poem. 144. Certain of the manuscripts read vigentis, but virentis is the commoner and preferable alternative (so, too, Brink 1971: 193). On this color-­term as indicative of youth, see Clarke 2003: 144–51, 268–72. 145. The assonance within and between the two passages, which accentuates and is accentuated by the straightforward repetition of words, is also noteworthy: florente iuventa fervidus and iuvenum ritu florent. 146. This correspondence is, notably, absent from the other pairs listed: the hero w ­ ill not become a god, nor the nurse a matron, nor a person born in one city turn into one born in another (though in ­these latter pairs, nutrix, 116, and nutritus, 118, evoke birth). Moreover, as if to strengthen the internal echoes, Horace repeats nata at 122 (albeit in a dif­fer­ent context: Achilles denying that laws w ­ ere born for him, iura neget sibi nata), having first used it at 62, iuvenum ritu florent modo nata vigentque. 147. TLL s.v. iuvenor, though it should be noted that the Greek verb has the rather distinct sense of “make innovations.”

125-81019_FerrisHill_HoracesArsPoetica_5P.indd 136

8/8/19 8:40 AM

pis on e s   137

likeliest c­ areer paths (forenses: the Forum), as well as to the importance of their paternity and their interest in h­ orses (quibus est equus et pater et res, 248). Pater at 248 finds itself followed up by iura paterna (256), didicit patriae quid debeat et quid amicis (312), O maior iuvenum, quamvis et voce paterna / fingeris (366), and in Maeci descendat iudicis auris / et patris et nostras, 387–88, a crescendo of references to fatherhood that w ­ ill culminate in a jarring and disturbing act at 470–71. The final appearance of ­fathers and sons in the Ars Poetica problematizes the ­human life cycle with its intergenerational and familial relationships, all of which have so far been presented as natu­ral, ­wholesome, positive, and productive, and leaves our image of the Pisones as ­father and sons on a sour note. Amid the images of death, killing, madness, and animalistic rage that mark the poem’s final twenty lines ­there appears a son urinating onto his f­ ather’s grave. The context is the Horatian voice wondering what terrible act could have been committed by the person who is now afflicted by a madness that has rendered him a poet: nec satis apparet cur versus factitet, utrum minxerit in patrios cineres an triste bidental moverit incestus; certe furit. (470–72) Despite the pretext, this is an extremely odd and even insulting note on which to conclude the repre­sen­ta­tion of father-­son relations for a poem dedicated to a ­father and sons: beyond the obvious grotesqueness of the image, few crimes ­were considered worse in ancient Rome than t­ hose that broke the bonds of filial piety.148 Horace’s point would seem to be more than merely to make Piso se­ nior question the love and re­spect of his sons ­toward him. Rather, as ­these sons are in the predicament of wanting to be writers, and as being a writer requires being mad in the popu­lar imagining (though Horace somewhat critiques this view), Horace would seem to be offering the sons a sure route, albeit a terribly unpalatable one, to literary success: if they only urinate on their f­ ather’s ashes, they ­will be struck down with madness by the gods, and, presto, they ­will emerge as poets. We have seen young men grow into old ones, a natu­ral pro­cess evident also in leaves as in words; we witness father-­elders teaching their sons and son-­figures; yet we conclude with an image of sacrilege, degradation, and utter lack of filial re­spect. 148. The infamous punishment for parricide, a category that included patricide, required that the offender be tied up in a leather sack with a dog, rooster, snake, and monkey, and thrown into the ­water (OCD s.v. parricidium). Horace may have in mind at AP 471 also the type of tomb inscription that urged passersby not to urinate upon the bones of the deceased (Orelli 1844: 798, Brink 1971: 429).

138  ch a p te r 2

Horatius pater The passage at 366–73 proves critical in two ways. First, the dissection of the Pisones into ­father and sons allows each to be further molded according to Horace’s designs: f­ athers are teachers (paterna voce), and sons are s­ haped by their teachings (fingeris), two complementary roles that are at last made explicit at 366–69 (prior to which the Pisones as a unit have been the recipients of Horace’s advice). Second, as much as Horace talks to the elder Piso son about his ­father and describes the ideal be­hav­ior of each, we are given no indication of how the ­father and son interact with each other: they are separated only ­here and elsewhere treated as a unit. We are also left on our own to imagine what teachings exactly the ­father is dispensing to his son in an effort to mold him ­toward Right. The only father-­teacher figure we encounter, in fact, is Horace, who in the course of issuing such advice as hoc tibi dictum / tolle memor or si quid tamen scripseris or delere licebit quod non edideris seems to have a­ dopted the vox paterna for himself.149 Addressing his poem to Pisones thus enables Horace to pre­sent an explic­itly didactic, paternal persona—­one that reverses his posture from the Satires, where he appears as a son taught by his own ­father: insuevit pater optimus hoc me, ut fugerem exemplis vitiorum quaeque notando. cum me hortaretur parce, frugaliter atque viverem uti contentus eo quod mi ipse parasset, ‘nonne vides Albi ut male vivat filius utque Baius inops? magnum documentum ne patriam rem perdere quis velit.’ a turpi meretricis amore cum deterreret, ‘Scetani dissimilis sis.’ ne sequerer moechas, concessa cum Venere uti possem, ‘deprensi non bella est fama Treboni.’ (Sat. 1.4.105–14) My most excellent f­ ather trained me in this way, that I should flee from vices by noting examples of each. When he was encouraging me to live sparingly, frugally, and content with that which he himself had supplied me with [he would say], “Surely you see how poorly the son of Albius lives, how Baius lacks resources? It is a power­ful lesson that one should not wish to lose one’s inheritance.” When he was deterring me from the foul love of a whore [he would say]: “­Don’t be like Scetanus.” [He warned] that I should not pursue adulterous affairs, when I could enjoy lawful love [saying]: “The reputation of Trebonius, caught red-­handed, is not pretty.” 149. Geue 2014: 150 has also noted Horace’s “shameless act of . . . ​supplanting and surrogacy” as he becomes a “father-­teacher.”

pis on e s   139

As Leach has shown, with this portrait “so specific and well drawn . . . ​that it has had unquestioning ac­cep­tance as pure autobiography,” Horace casts his ­father as a kindlier, wiser version of the stern and overprotective ­father of Terentian comedy.150 Despite his impoverished early circumstances, noted in Satires 1.6,151 Horace se­nior, dubbed pater optimus, nevertheless nobly molds his son’s habits and character through their shared observation of still-­life scenes from the society around them: “My most excellent ­father trained me in this way, that I should flee from vices by noting examples of each.” As Horace and his ­father watch and learn from Albius’s son and Baius, Scetanus and Trebonius,152 so we do, too, ­these exempla being paraded before young Horace’s eyes as well as our own. At the same time, Horace has himself already throughout Satires 1.1–1.3 offered up moral counterexamples (positive ones are in short supply) in the form of dim interlocutors and offenders named and unnamed, requiring us to be the recipients of his teachings ­there already. Horace’s own ­father’s teachings take both positive and negative form (sive iubebat . . . ​sive vetabat, Satires 1.4.121, 124; “­whether he ordered [me to do something] or forbade [it]”), yet the superior didactic value of negative examples is left in no doubt: “Just as a neighbor’s funeral wakes up sick gluttons and compels each to treat himself better ­because of the fear of death, so other men’s shameful deeds often terrify tender souls away from vices” (avidos vicinum funus ut aegros / exanimat mortisque metu sibi parcere cogit, / sic teneros animos aliena opprobria saepe / absterrent vitiis, 126–29). Our poet fondly recalls, “thus he fashioned me, a boy, with his dictums” (sic me / formabat puerum dictis, Sat. 1.4.120–21), as Horace looks back on his youth from a ­later, more matured vantage point, where he is afflicted only by minor vices and ones of the pardonable sort from which he may yet be wholly f­ ree (ex hoc ego sanus ab illis / perniciem quaecumque ferunt, mediocribus et quis / ignoscas vitiis teneor. fortassis et istinc / largiter abstulerit longa aetas, liber amicus, / consilium proprium, 1.4.129–33). 150. Leach 1971: 618. 151. Horace repeatedly declares himself to be the son of a freedman, libertino patre natus (Sat. 1.6.6, 45, 46, 58, Epist. 1.20.20), a statement long taken at face value as autobiographical fact (and “confirmed” in Suetonius’s Vita Horatii, though on the prob­lems of ancient lives see Lefkowitz 2012, and further Graziosi 2009 on Horace’s own), with Williams 1995 even arguing for Horace’s ­father as one of the Venusian prisoners captured during the Social War. As noted by Freudenburg 1993: 5, 14, 205 and Gowers 2012: 222, however, the words are a virtual translation of Bion of Borysthenes’s ἐμοῦ ὁ πατὴρ μὲν ἦν ἀπελεύθερος (fr. 1), and t­ hese scholars accordingly espouse a more skeptical reading of Horace’s claim, aligned with generally heightened suspicions regarding Horace’s autobiography. 152. As Gowers 2012: 177 points out, ­these names, in that they suggest “ ‘Man A’, ‘Man B,’ ” reek of the poet’s invention.

140  ch a p t e r 2

Horace shows himself having absorbed his ­father’s lessons and putting them into practice just a few lines ­later, muttering to himself in imitation of the paternal voice he has long listened to, and concerned as ever with “Right”: “ ‘this is more right; ­doing this, I ­w ill live better; thus I may appear sweet to my friends; this indeed [was] not well [done]: surely I ­will not ever imprudently do the same t­ hing as that man?’ I keep g­ oing over t­ hese ­things to myself with pursed lips” (‘rectius hoc est; / hoc faciens vivam melius; sic dulcis amicis / occurram; hoc quidam non belle: numquid ego illi / imprudens olim faciam simile?’ haec ego mecum / compressis agito labris, 134–38). Leach 1971: 617 is right to point out that the young Horace was ­shaped by words and that he continues to use words to continue shaping himself, but the extension of this practice into the Ars Poetica is also notable, where sic me / formabat puerum dictis may be found remade as voce paterna fingeris.153 With his insistence on the Pisones as ­father and sons, and his explicit assigning of f­ athers and sons in general to their complementary roles, Horace succeeds in inserting into his Ars Poetica, tentatively at first but with increasing force, echoes of this other father-­and-­son pair, the one a teacher, the other his student, from a narratively (and, available evidence would suggest, chronologically) ­earlier point in his hexameter poetry. Leach 1971: 618 was tipped off by the discussion of comedy in Satires 1.4 to the comic traces within it, and the same form of signposting may be discerned in the Ars Poetica, where the dramatic genres loom large—­far larger than has been satisfactorily explained. She reads Satires 1.4 against Terence’s Adelphoe, with its per­for­mance of teaching by example (412–29). Demea, upon whom Leach sees Horace’s ­father modeled, may not be named in the Ars Poetica (neither is he in Satires 1.4), but Chremes does appear, raging angrily with distended mouth (iratusque Chremes tumido delitigat ore, 93–94)—­Chremes, the ­father in Terence’s Heautontimorumenos and Andria.154 If Horace remakes Terence’s Adelphoe for Satires 1.4, he does so again and in rather dif­fer­ent ways for his Ars Poetica. Horace ju­nior had been cast in Satires 1.4 in the role of Ctesipho, the country-­raised b­ rother for whom the f­ ather still harbored hopes that he might turn out well. His other son, Aeschinus, raised in the city by 153. 366–67: formare and fingere are interchangeable in sense, while both vox and dictum involve speaking aloud. Oliensis 1998: 206 compares Sat. 1.4.120–21 with AP 367–68. 154. Brink 1971: 177 is skeptical ­whether Horace’s Chremes is to be identified with the character in Heaut., but di Benedetto 1962: 42–43 thinks a specific allusion is likely. Though Plautus is named in the AP, Terence is absent, and conspicuously so, to the eye of di Benedetto 37–38 (it is also worth noting that Plautus is named as an unsuitable object of admiration, 270–72, which may be a hint that Terence was more to Horace’s taste; intriguingly, Horace describes Terence at Epist. 2.1.59 as supreme in ars, suggestive of the AP itself).

pis on e s   141

Demea’s ­brother Micio, the f­ ather considered already lost, an instantiation of the urban/rustic tension evident at vari­ous moments in the Ars Poetica as well. The very title of the play, in fact, “­Brothers,” should put us in mind of the Piso boys, and Armstrong indeed has argued that they resemble the respective characters of the Adelphoe: the elder one is “Terence’s Aeschinus (if t­ hings go wrong)” and the younger “a ­silent version of Ctesipho.”155 As Horace “placed the fable of his education within the traditions of new comedy”156 in Satires 1.4, it seems natu­ral to extend ­these traditions to that point where Horace himself becomes the educator, as he does in the Ars Poetica. We see the Horatian persona aging over the course of his hexameter writings, as Horace transforms himself from the lesson-­receiving son of the Satires into a lesson-­dispensing ­father figure in the Ars Poetica. This narrative arc, in turn, intersects with that of Ars Poetica 153–78: the mature and seasoned poet whom we encounter in the Ars Poetica is deeply concerned with dif­fer­ent generations and with the handing down of wisdom from one to the next.157 He takes up the mandate of the old man as described at line 174, castigator censorque minorum, a “chastiser and censor” of the young, where the coda se puero (“when he was a boy”) emphasizes not only the connection to the boy with which the life cycle began, but also Horace’s own boyhood as we have witnessed it in Satires 1.4.158 The Horatian senex of the Ars Poetica, moreover, is not the overprotective but basically kindly f­ ather of Satires 1.4, but rather the curmudgeonly and strict Demea of Adelphoe and the Chremes of Heautontimorumenos. So Horace becomes, not exactly his own ­father, but another version of the comedicized ­father character, a stock senex, as he portrayed him in Satires 1.4;159 and Horace 155. Armstrong 1993: 203; he also suggests Plato’s Lysis as a comparandum for teaching in the AP. Di Benedetto 1962 lists numerous further textual correspondences between the AP and plays of Terence. 156. Leach 1971: 620. 157. For Reckford 1969: 138, Horace’s “last writings,” among them the AP, “though full of good humor, are like a testament made in preparation for death.” Though Maresca 1966: 145 with n149 identifies Pope as the first major practitioner of the form that had already begun c. 1415 and 1450 with two works in Latin entitled Ars moriendi, Horace with his AP might well be thought of as the spiritual originator of the tradition. 158. Horace’s repetition of the term puer is insistently clustered at AP 158, 174 (where se puero creates a ring with the puer who set in motion the life cycle at 158), 177, 185 (this last one being of heightened import, since it is also the most disturbing—­Medea murders her boys); cf. 325–26 (pueri innocently in a schoolroom) and 340 (disturbing: devoured by Lamia). 159. Compare Horace’s self-­description at Epist. 1.20.25 as irasci celerem, tamen ut placabilis essem (Brink 1971: 234 compares this passage with AP 159, iram, the anger of the boy), which further suggests the poet embracing the role of grumpy old man. Armstrong 1993: 218 also briefly connects Sat. 1.4 with the AP: “In short Horace is behaving to the boy [sc. the elder Piso son]

142  ch a p te r 2

educates not his own son, but the Pisones, especially the elder son, and by extension all would-be writers, filial stand-­ins. Horace’s paternal and professorial persona, which surfaces ­here and ­there throughout his ­earlier hexameters, culminates in the Ars Poetica, where he embraces and proclaims it: respicere exemplar vitae morumque iubebo. Although the development of this persona in the Satires is not strictly linear (Horace has even been seen regressing from instructor to student as the collection goes on, and moreover one subject to lessons from increasingly “inept teachers,”160 even as he paradoxically places before our eyes ever more images that recall ­those his ­father placed before his own eyes), it finds itself completed by the time of the Ars Poetica, the e­ arlier vacillation perhaps allowing Horace to position himself increasingly as a father-­teacher while abjuring responsibility for such a role. This is not to say that Horace’s poetic corpus is straightforwardly autobiographical in any naïve sense; rather, that his Satires, long viewed as the rough experimentations of a young poet, appear so b­ ecause he pre­sents himself t­ here as relatively youthful and in need of teaching, and his Ars Poetica, long viewed as the mature and climactic achievement of a long poetic ­career, appears so ­because he pre­sents himself in it as a voice of authority, drawing on a lifetime of experience and dispensing his learning onward.161 Horace communicates his newfound paternal didactic authority by reinserting himself explic­itly as speaker into the Ars Poetica in and near passages where prescriptions are listed. He does so for the first time in adopting the characteristics of the senex described at 169–76 by delivering a litany of haranguing prescriptions, 179–201. Th ­ ese continue at 234–74, when Horace reminds us again that he is the speaker with the first-­person singular verbs amabo (235) and enitar (236) and the declaration made by Satyrorum scriptor of his status as a writer of Satires. ­These lines and t­ hose that follow are marked by a concentration of first-­person singular verbs, of which Horace’s use has been strategic and mea­ sured since the outset. He has used them to admit his own failings, in a form of captatio benevolentiae, as at 24–26 where decipimur, which implicates Horace like that good Epicurean therapist, his own ­father, who offered him attractive examples of virtue and success, and horrible examples of vice and failure, from among p­ eople to be seen in the streets.” 160. Barchiesi and Cucchiarelli 2005: 215; cf. Freudenburg 2013: 314–15. 161. This aligns with the medieval view that the vari­ous books in Horace’s poetic corpus seemed and therefore w ­ ere aimed at ­people in dif­fer­ent stages of life: “Horace clearly made four sorts of poems on account of the four ages: the Odes for boys, the Ars Poetica for young men, the Satires for men, the Epistles for old men and the fully-­formed” (“oracius fecit IIIIor diversitates carminum propter IIIIor .s. etates. odas pueris, poetriam iuvenibus sermones viris epistulas senibus et perfectis”; quoted [sic] at Hajdu 1993: 232, Friis-­Jensen 1997b: 179, 2007: 291).

pis on e s   143

and the Pisones among ­those who may be deceived by appearances, gives way to brevis esse laboro, obscurus fio—­Horace himself, we are told directly, is guilty of the very error he warns against at 31. At other times, the first-­person singular serves as a voice of authority: in rejecting as a model the craftsman who is competent or excels at only a single ­thing, Horace, in a way that ensures his reader cannot but agree with him, declares, “I myself would not wish for me to be this man any more than [I would wish] to live with a crooked nose, [though] admired for my black eyes and black hair” (hunc ego me . . . ​/ non magis esse velim quam naso vivere pravo, / spectandum nigris oculis nigroque capillo, 35–37; the added emphasis from ego me is striking). Elsewhere, too, a first-­person singular verb models expected or desirable be­hav­ior for the reader: male si mandata loqueris, / aut dormitabo aut ridebo (104–5); quodcumque ostendis mihi sic, incredulus odi (188); non ego paucis / offendar maculis (351–52). Similar is aut ego fallor (“­unless I am mistaken,” 42), which, as Brink sees, is “not a polite disclaimer” but rather a phrase in which the words mean the opposite of what they appear to, such that it paradoxically becomes “a way of insisting on the truth of what is said.”162 Horace is likewise the voice of authority—­while pretending merely to express a personal opinion or preference, often by way of a rhetorical question that serves to feign uncertainty—at lines 55–58 (ego cur, acquirere pauca / si possum, invideor . . . ?) and 86–88 (descriptas servare vices operumque colores / cur ego si nequeo ignoroque poeta salutor? / cur nescire pudens prave quam discere malo?). We may even catch him innocently in the midst of what appears to be an isolated, solitary action or expressing a personal thought (though the hint that he is modeling the one desirable reaction or be­hav­ior often remains unavoidably pre­sent): o ego laevus, / qui purgor bilem sub verni temporis horam! (301–2), ­Choerilus . . . ​/ quem bis terve bonum cum risu miror; et idem / indignor quandoque bonus dormitat Homerus (357–59), ego nec studium sine divite vena / nec rude quid prosit video ingenium (409–10), mirabor si sciet inter-­/ noscere mendacem verumque beatus amicum (424–25), or ‘qui scis an prudens huc se proiecerit atque / servari nolit?’ dicam, Siculique poetae / narrabo interitum (462–64). Beginning at 234 we can see Horace exploiting the effects of his own singular authorial persona established emphatically with amabo. “I ­shall trace my song crafted from something familiar” (ex noto fictum carmen sequar, 240), he says, building upon and clarifying his e­ arlier advice from 119 (aut famam sequere aut sibi con­ve­nientia finge). The correct course of action, we deduce, is the one demonstrated by Horace himself, but not just anyone can accomplish what he does: ut sibi quivis / speret idem, sudet multum frustraque laboret / ausus idem

162. Brink 1971: 130; both 42 and 272, he adds, “qualify affirmations in a quasi-­personal manner” (107). See also Reinhardt 2013: 512.

144  ch a p t e r 2

(240–42).163 Some ten lines l­ ater, he positions himself as judge, me iudice (244), again emphasizing his physical presence in the poem at this moment and appearing to shake his fin­ger at uncooperative fauns: “let them be careful” (caveant). At 263–68, with another careful placement of the word iudex (not explic­itly Horace himself this time, but we may feel the implication nonetheless),164 Horace again deploys first-­person singular verbs in the context of a rhetorical question: idcircone vager scribamque licenter? an omnis / visuros peccata putem mea, tutus et intra / spem veniae cautus? vitavi denique culpam: / non laudem merui (265–68).165 Ever instructing, ever advising, ever scolding, this is the voice of a f­ ather (pater) and of an old man (senex)—­two figures that frequently find themselves embodied as one in the senex iratus of Roman comedy who chastises his son. We see ­here both the senex of the Ars Poetica (172–74) and Horace’s own f­ ather, who, as he encourages certain be­hav­iors and discourages ­others, always insisting carefully on pre­ce­dent and example, shapes his boy with his words. At this late juncture in the Ars Poetica, having set it up by adopting the persona of the senex and repeatedly reintroducing himself as speaker into the poem as he dispenses advice, Horace at last makes explicit his didactic role: ergo fungar vice cotis, acutum reddere quae ferrum valet exsors ipsa secandi; munus et officium nil scribens ipse docebo, unde parentur opes, quid alat formetque poetam, quid deceat, quid non, quo virtus, quo ferat error. (304–8) With docebo, Horace declares his intention to teach.166 In the first place the subject ­will be munus et officium, “the gift and duty.” This is momentarily clarified to mean the poet’s activities (307–8), but Horace has made it clear, once again, that ingenium of a sort is required: if someone is fortunate enough to have been 163. Cf. Oberhelman and Armstrong 1995: 254: “the result may look ­simple, but a person would discover the limits of his own ars and ingenium who tried to equal it.” 164. “Not just any judge sees unrhythmical poems” (non quivis videt immodulata poemata iudex, 263), he says, implying that he himself is such a judge, and leaving it open to his reader (external and internal) to determine w ­ hether s/he merits inclusion in this group, too. 165. Brink 1971: 107 also notes the preponderance of first-­person singular verbs in the poem’s final third, e.g., 301–6 (purgor, fungar, docebo), 317 (iubebo), 351–60 (offendar), 388 (the adjective nostras), 409–10 (ego . . . ​video), and 463 (dicam), seeing ­these as reflective of Horace ­here speaking “of himself as a teacher (or critic).” 166. Hajdu 2014: 87–88 identifies t­ hese lines and 335–37 as the two places where teaching is discussed openly in the AP.

pis on e s   145

born with talent (conceived as a munus from the gods, perhaps), then it is incumbent upon that person to discharge the concomitant obligation (another sense of munus: “duty”) by producing poetry as a “gift” (munus again) for the ­people.167 The way in which anyone, even Horace, could teach a munus, however, remains opaque. Officium is similarly multivalent, denoting on the one hand one’s ­career (though it is unusual of poetry or art) and the accompanying social-­contractual tasks that dictated much of the upper-­class Roman male’s day-­ to-­day life,168 and on the other suggesting, through its far more common meaning, the po­liti­cal ­careers to which the Piso boys would be best suited. Horace seems fully pre­sent in the verb docebo, itself preceded by fungar, “I ­will perform/ discharge/ execute,” with which he declares in open and meta­ literary fashion that he is stepping into a role (in this instance, that of whetstone). As is the case in the Satires, the reader of the Ars Poetica is lulled into a similar sense of warm familiarity with the poet, even if Horace himself seems rather less pre­sent in the latter poem (certainly, he does not tell us anything about himself, as he does or purports to in the Satires).169 Any “real” Horace is, however, as ever elusive,170 since he allows us access only to a painstakingly curated simulacrum, not through to the true intimacy we desire and are perversely encouraged to believe we can attain if we only make a concerted enough effort. Yet Horace’s persona in the Ars Poetica is especially slippery (even if he spares us in this par­tic­u­lar work from having to ponder w ­ hether to believe each autobiographical detail he mentions): where exactly does he lurk among the trinity of senex / pater / Herr Professor Doktor, and how should we imagine him contained within or acting as an inanimate tool (cotis)? With the image of the whetstone, I see Horace alerting his reader to the fact of his larger, other per­ for­mance(s) within the poem—­performances further emphasized by terms such as partes (315) and persona (316). In this par­tic­u­lar instance Horace’s per­ for­mance as sharpening tool is one to which he is wholly committed: just as a 167. OLD 1–7. 168. Horace describes the same at Epist. 2.2.67–75, Sat. 1.9.43–60, and 2.6.23–39 (where he claims to be experiencing this mode of existence for the first time on account of his usefulness to ­others through his newly forged connection to Maecenas). Sedley 2014: 115 is rightly skeptical that munus et officium means “a lesson in how to write poetry,” as most commentators take it. 169. Wood 1978: 134–42 sees Pope and Boileau ­doing the same in their Arts, while in their satires both, like Horace, reveal more of themselves (or rather, of their personae as writers of satires). 170. Brink 1971: 107 is right to speak of Horace’s “astonishing practice of partly displaying and partly concealing his personality”; cf. 449: “The Ars, like most Horatian works, encourages the quest for the poetic personality that has given rise” to the poem.

146  ch a p t e r 2

whetstone is itself incapable of cutting but makes steel sharp (acutum / reddere quae ferrum valet exsors ipsa secandi, 304–5), so Horace, though “writing nothing” (nil scribens ipse), ­will shape ­others into poets, the meta­phor extended to its logical end.171 The context of lines 304–8, much excerpted, is vital: Horace declares his intention to teach and declares that he is “currently writing nothing” directly on the heels of a rejection of writing (nil tanti est, 304), if madness, detailed at some length in 295–303, is a requirement for writing. It is u­ nder t­ hese specific conditions, then, that Horace describes himself as nil scribens ipse, such that this phrase becomes not so much a statement that he is not currently engaged in composing poetry (­whether lyr­ics, or satire or other hexameter non-­poetry, or tragedy, or what­ever other way this has been construed),172 but rather announces that his writings (of what­ever kind) are not the product of the madness many mistakenly believe the poet must be afflicted by. Also notable is the verb scribere, which, as discussed above, is notably rare in the Ars Poetica as it was also in the Odes. In his disavowal of writing, we may thus perhaps see hints again of Horace’s repeatedly professed preference for speaking over writing and for living words that evolve organically over written ones ­under the tight control of a cabal of censorial authors. The verb scribere, especially by virtue of its rarity, knits 304–8 (with scribere in 306) to what follows (scribendi, 309), in which Horace, having declared that he ­will teach (docebo), proceeds to do exactly that: scribendi recte sapere est et principium et fons. rem tibi Socraticae poterunt ostendere chartae verbaque provisam rem non invita sequentur. qui didicit patriae quid debeat et quid amicis, quo sit amore parens, quo frater amandus et hospes, quod sit conscripti, quod iudicis officium, quae partes in bellum missi ducis, ille profecto reddere personae scit con­ve­nientia cuique. respicere exemplar vitae morumque iubebo doctum imitatorem et vivas hinc ducere voces. (309–18) 171. The image is found also at [Plut.] X orat. 838e, where it is attributed to Isocrates (active early fourth ­century BCE); cf. Orelli 1844: 769, Wickham 1891: 418, Brink 1971: 335, Rudd 1989: 202, Sedley 2014: 114. Sedley agrees that the meta­phor declares Horace’s intention to become a teacher (so Brink), rather than a critic (Rudd), though (as discussed in chapter 3 with n127) the two are not in opposition. 172. See introduction with nn85–86.

pis on e s   147

Lines 312–16 have rightly been seen as a response to 307–8,173 yet I would draw attention to two further details. First is the implicit presence of the Pisones in the passage above: they are t­ here in the social obligations and in the picture of public life, as well as in the insistent repetition of familial terminology—­patria, parens, frater. Second, the lines are marked by a blurring of writing and living: it is only through careful study of life that a writer can avoid dissonance and incongruity in his creations, for life, vita, is the source of the “living voices,” vivas voces, that must be the artist’s desideratum. At the heart of this entire enterprise lies wisdom, sapere, said to be “the starting point and wellspring of writing properly”—­“writing properly” (scribendi recte, Sat. 1.4.13) being the very ­thing Horace says his generic ancestor Lucilius, “chatty and lazy” (garrulus atque piger, 12), is not up to the task of ­doing on account of ­these shortcomings in his temperament.174 Thus is made apparent the multivalence of Horace’s prescriptions in the Ars Poetica, as life (vita, 317) and writing (309) are revealed both to have reason and wisdom (sapere) at their core (if done properly, recte, that is). The overall effect is to minimize the importance of writing, demoting it to but a portion, albeit an impor­tant one, of other life pursuits; and yet, it is hard to decide where we should place the ultimate emphasis. Is living of the utmost importance, or is Horace merely interested in real life in the ser­vice of art? That the Ars Poetica may not be primarily concerned with poetry, as I suggest especially in chapter 4, does not mean that it is not concerned with teaching of one sort or another. The teacherly role Horace embraces with docebo is nevertheless not fulfilled as completely as we might have hoped. While we may feel that the lessons begin immediately with that verb’s utterance, and while we can discern teachings in the lines that follow (309–18) emanating from the persona of censor castigatorque minorum beneath which Horace has clothed himself, the futurity of docebo continues to ring in our ears. Iubebo, as late as line 317, confirms our suspicions: perhaps the words we are reading right now are not the lessons and commands he means, and Horace ­will provide ­these only at some ­future, 173. Cf. Wickham 1891: 418, Norden 1905 with Brink 1971: 326, Wood 1978: 136 (“the instances of his role as critic and teacher are more noticeable” a­ fter 301–8). Brink says it was one of Norden’s “­great merits” to have “recognized that A.P. 306–8 contain a t­ able of contents”: 309–32 explain unde parentur opes and quid alat formetque poetam; 333–46, quid deceat, quid non; 347–452, quo virtus, quo ferat error, continued at 453–76, “de insano poeta” (Armstrong 1995: 386 calls Horace’s answer to quo ferat error, which occupies the poem’s final section, “a delightfully literal joke”). 174. Brink 1971: 338 and Rudd 1989: 202 also note the connection with Sat. 1.4.13, and Brink (423) rightly sees the dual valence of sapere at 456: “qui sapiunt puns: ­those who are in their right senses and ­those who have the savoir of ars poetica.”

148  ch a p te r 2

as-­yet-­undetermined date. Or perhaps the Ars Poetica is envisioned as a monument for ­future generations, as Horace with docebo and iubebo unhumbly (if accurately) prophesies his own poetic immortality as he had in Odes 3.30, foreseeing the status of the Ars Poetica as a teaching document.175 If this is the case, however, Horace manages both to provide education for the Pisones while at the same time abdicating responsibility for this task in f­ avor of a proj­ect that offers him greater potential. ­Either way, the promise of teaching contained and somewhat enacted within lines 304–18 degrades into the vignette of the Roman schoolroom, where inspiration is dimmed in ­little boys who are taught instead to count their pennies, and big ideas are crowded out by relatively trivial technical skills. Cicero’s Pro Caelio has been read as a comedy in disguise,176 and similar strains, paralleling ­those discovered by Leach in Satires 1.4, may also be discerned in the Ars Poetica. Horace’s hexameters may lack a meretrix, but the required iuvenis and senex/pater have prominent roles. Leach suggests that Horace, in the Terentian manner, wanted Satires 1.4 “judged . . . ​as an attempt to capture the temper of ordinary life,”177 and what could be more ordinary than the narrative scaffolding of the Ars Poetica: a ­father raising two sons. The Ars Poetica thus shows one of the most ubiquitous and mundane of all ­human activities, child-­ rearing, rendered as high (and low) art, as from b­ ehind his authorial mask, cloaked further in the mask (persona) of the senex, Horace instructs and helps to shape the Piso boys (and the general reader), like his own f­ ather had done to the poet’s youthful persona in Satires 1.4. This approach goes some way t­ oward explaining a long-­standing mystery of the Ars Poetica: why and how it is that Horace can purport to offer instruction in a genre, drama, in which he himself never wrote. In addition, terms from drama pervade the poem (persona, 126, 175. Horace describes this at Sat. 1.10.74–75 (“would you, mad, r­ eally prefer your poems to be recited in cheap schools?” an tua demens / vilibus in ludis dictari carmina malis; cf. Epist. 1.20.17–18 with Tarrant 2007: 64–65), clearly regarding it as undesirable, though it may also be taken as evidence that he knew his writings ­were already being put to this use in his lifetime. Griffin 1996: 205–6 is similarly amused by seeing in AP 304–6 how, “longer a­ fter his death than he could have ­imagined . . . ​Horace continued to teach the skill of writing,” and Sowerby 2012: 262 likewise says ad AP 304–5, “Horace accurately predicts his role in his own reception as critical guide and mentor.” 176. Geffcken 1973, Leigh 2004. 177. Leach 1971: 626. She notes further that Horace connects “his own diction with the sermo merus of new comedy” (619), which in turn is “almost indistinguishable from that of real life” (620)—­a focus on everyday, spoken, lived, natu­ral language that finds itself expressed also in the AP. Yona 2015: 229 likewise speaks of how “the exploration of intimate and familial affairs typical of Terentian drama is closely paralleled by the domestic origin of Horatian satire and its concern with privacy,” as he builds upon the work of Leach to reveal also the Epicurean ele­ment in Sat. 1.4.

pis on e s   149

192, 278, 316; actus, 129, 189, 194; scaena, 125, 179, 183, 260), as do ele­ments from Greek drama particularly (ephebic youths, satyr-­play, figures such as Aeschylus and Thespis, the evolution of tragedy and comedy). If the Ars Poetica is read as Horace’s own exploration, in the form of character studies, of the possibilities afforded by comedy, the poem becomes in a sense its own self-­justification: drama is pre­sent ­because Horace was making use of drama in the covert ways outlined ­here, all of which ultimately go back to the personae of the Pisones as the poem’s addressees. Above all, Horace is perhaps exploiting the opportunity afforded by comedy for “the damage done through youthful exuberance” to be “accommodated without any lasting harm to the ­family or to society at large”:178 Horace’s advice to the maior iuvenum, the representative of all young men who wish to become poets, ­will ideally result in any such individuals abandoning their literary whims in ­favor of law and public life, the lesson made more palatable through its being cloaked in the honey of comedy.179

Conclusions Although we encounter a g­ reat many poems in Horace’s corpus and elsewhere in Latin lit­er­a­ture (as well as in all lit­er­a­ture) where “at first glance the patron appears to be both the major catalyst for the poems and the sole person in whose praise t­ hese poems have been written,” t­ here exists in real­ity “a multiplicity of audiences and voices, including but not ­limited to patron and poet.” Although Gold 1992: 161 in making ­these comments has in mind Satires 1.1 and Odes 1.1, both prefatory poems to larger collections and both addressed to Maecenas (who, as Horace’s sole historically attested patron, is in a distinct category from other addressees), much the same dynamics are at play in the Ars Poetica, or in any stand-­alone poem, with the added complication that the business of openings and closings (not to mention ­middles, that is, the poem’s substance) must by necessity be handled in the course of a single work. Gold adds that “one of the main functions of t­ hese poems is to establish the authorial voice of the poet by delineating the vari­ous levels of audience to be addressed” (161–62). Although the Ars Poetica is not a programmatic poem along the lines of Satires 1.1 and Odes 1.1, Horace is nevertheless concerned with and succeeds in executing a stratification similar to that perpetrated by both of the latter among t­ hose who receive this work. While Pisones in line 6 performs one of the functions of Maecenas at both Odes 1.1.1 and Satires 1.1.1 by providing “the primary 178. Leigh 2004: 302, of Cicero’s Pro Caelio. 179. Ibid.: 301 speaks of Cicero’s aim to “make the jury study what he claims are the central issues in the case as if they ­were watching a comedy.” Lucretius famously speaks of concealing philosophy in the honeyed cup of poetry at DRN 1.936–47.

150  ch a p te r 2

audience, the dedicatee,” Horace manages in the course of the first thirteen lines of the Ars Poetica to expand the poem’s polyvocality beyond that of poet and dedicatee to include an “internal audience” (provided by the objection voiced at 9–10) and through the use of the vocative amici and the second-­ person plural verbs teneatis and credite, both an “authorial audience, the first-­ century B.C. Roman upper-­class writers and politicians” and an “­actual audience, the person who is reading or hearing a text at any given moment.”180 Horace unites and embraces all of ­these separate entities within his first-­person plural verbs as he sums up, “We know, and we ask for and we grant this indulgence by turns” (scimus, et hanc veniam petimusque damusque vicissim, 11).181 The Pisones thus serve—­through their nebulous relation to the amici and to the voice that speaks in lines 9–10—to help establish the typically Horatian polyphony that w ­ ill run the length of the poem.182 As Pomeroy 1980: 46 lucidly explains in relation to Odes 1.1, Horace as a poet accepts “that by himself he cannot create his own uniqueness”; rather, “it is up to his partner in the act of communication to reply and assert that Horace ­really is a poet,” and thus the situation requires “an outside observer (the reader) . . . ​to truly assess Horace’s worth.” The form that this delicate dance takes, and the ensuing validation (or not) of both poet and reader in their roles, is naturally dependent on the qualities and status of each. Regarding the Pisones themselves, with Gold 1992: 163 in mind we should ask “­whether [the addressee] is more impor­tant as one of the prominent themes which traditionally receive mention in any programmatic first line of poetry (e.g. arma virumque cano) or as an external referent who lends dignity and authority 180. As identified by Gold 1992: 162–63, who indicates (n4) her debt to Rabinowitz 1977 and 1986 (for him the possibilities include “real readers, implied readers, virtual readers, super-­readers, ideal readers, and narratees,” and he likewise discusses the concept of the “authorial audience,” i.e., “the audience whom the author had in mind when constructing the text,” 1986: 117–19, cf. 1977: 126). Oliensis 1998, on the other hand, drawing on Barchiesi 1993, sees as impor­tant the “overreader,” “an unnamed but other­wise specified other who may be i­ magined as reading over the addressee’s shoulder.” 181. By contrast, Carm. 1.1 ends with poet and patron still firmly separated (“but if you [singular] place me among the lyric poets, I s­ hall strike the stars with the top of my head” (quod si me lyricis vatibus inseres, sublimi feriam sidera vertice, 35–36), and in Sat. 1.1 Maecenas recedes into the background ­after the opening three-­line address to him (though “you” in line 14, ne te morer, audi / quo rem ducam, may denote him; cf. also the first-­person plural verbs at 27, quaeramus, and 119, queamus). See further Sharland 2009 on Horace’s voices in his Satires. 182. See Gold 1992: 185 on the same in Sat. 1.1 and Carm. 1.1: in t­ hese opening poems, Horace reveals “a special model of the technique he ­w ill use in his shifting changes of address and thereby introduces us to the vari­ous audiences through whose eyes he w ­ ill develop and be defined as a poet and an artist.”

pis on e s   151

to the poem by his name and presence. We should ask ­whether [the poem] enhances [the addressee’s] importance or [the addressee] lends to the [poem] a dif­fer­ent dimension and tone.” I see the Pisones allowing Horace to accomplish all of the above at once. They have traditionally and widely been read both as lending the poem a higher standing through their presence and as themselves being enhanced in importance through their inclusion in it—­a mutually beneficial system of endorsement.183 This is, of course, unavoidably tied up with the ­matter of their identity: if Lucii Pisones, they suggest literary patronage and a learned, philhellenic cultured milieu (one specifically encompassing Philodemus), along with success and standing in public life; if Gnaei Pisones, they bring the wealth, power, and influence of a distinguished and po­liti­cally prominent lineage, though no literary bona fides. Yet t­ hese figures may be unique among didactic addressees in being both illustrious (if Horace had not wanted the sheen of aristocracy upon his poem, he could and should have named a dif­ fer­ent ­family) and in the Hesiodic mold of νήπιοι. We, the external readers, have been placed by Horace in the considerable bind of wanting on the one hand to identify with the Pisones and on the other to distance ourselves from them184—­a bind further complicated by our inability to be entirely sure just which Pisones we are meant to alternately identify with or distance ourselves from. Are they one generation or another of the Lucii Pisones, interested, like the would-be writer reading the poem, in writing, but not talented enough to succeed, with the result that the would-be writer flatters himself as being (s/he imagines) better than they are? Or are they Gnaei Pisones, figures with apparently even less hope of success in writing, such that the would-be writer reading the poem feels sure (s/he thinks) of his/her own superiority? This form of flux is unusual, indeed, perhaps unpre­ce­dented, as far as didactic addressees are concerned, though it has resonances of the sorts of games Horace plays with the reader of his Satires, where it is evident that the receivership of a work is clearly stratified along intellectual lines, yet we remain always unsure of where to locate ourselves among ­these strata, and are forced continually to reevaluate any placement upon which we do ­settle (generally, and rather dispiritingly, in the downward direction). This dynamic is especially predicated upon the fact that no member of the Piso ­family (to our knowledge) ever acted as artistic patron to Horace. Despite 183. Though as White 2007: 199 points out, Horace “does not promise to immortalise friends in his poetry,” ­either in the AP or elsewhere. 184. Contrast Hesiod’s Works and Days, in which the reader’s sole inclination is “to dissociate ourselves from the addressees, derogated t­ here as ‘­idiots’ ” (Bing 1993: 99; cf. Mitsis 1993: 126 on the same in Lucretius); Bing in turn contrasts Aratus’s Phaenomena, in which we are given no reason “whatsoever to distance ourselves from the addressee.”

152  ch a p te r 2

their position of relative power, then, Horace was not beholden to them in the par­tic­u­lar ways he was to Maecenas, who, among other gifts, material and other­ wise, bestowed upon him a country villa. What particularly complicates the position of any Piso ­family in this regard and Horace’s attitude ­toward them, however, is that on the basis of their wealth and standing they could have been his patrons, as indeed two generations of Lucii Pisones ­were to Philodemus and Antipater of Thessalonica. Moreover, Horace may be seen alluding to the mechanics of patronage with such phrases as aere dato (“with money given,” 21, describing the circumstances u­ nder which a picture is painted) and promissi carminis auctor (“the author of a promised song-poem,” 45).185 Horace is thus in the Ars Poetica on the one hand addressing patrons and yet on the other remains ­free from certain of the constraints such a relationship might dictate in that ­these par­tic­u­lar patrons are not his own.186 Although the categories of “patron” and “addressee” overlap, and lit­er­a­ture on poetic patronage can be useful for thinking about the dynamics of power in the Ars Poetica,187 I nevertheless see this poem resembling Horace’s Epistles where “most often the occasion,” the stated reason for why Horace is writing at a par­tic­u­lar moment to a par­tic­ul­ ar addressee, “is but the thinnest pretext for a poem.”188 The Pisones are in the Ars Poetica, in short, ­because their presence—­and the par­tic­u­lar forms this takes—­ was useful for Horace’s larger poetic designs.189 As the Ars Poetica develops, and as we are made privy to additional details about the Pisones, we come to see that this name functions much like the declaration arma virumque cano: though it may take some readings to realize it, the Pisones, it transpires, herald the prominent role that w ­ ill be accorded within the 185. Though as White 1978: 86 explains, while we find poems described as “ ‘promised’ to the amicus,” we “never hear of arrangements which committed the amicus to pay for the poem,” and “a commission in the modern sense would have been out of the question.” He adds also that, judging from the evidence from Martial, such gifts as w ­ ere received w ­ ere too meager to provide the poet with any real sustenance. 186. Notably, patronage involves “the reciprocal exchange of goods and ser­vices” and a relationship that “must be a personal one of some duration” as well as being “asymmetrical” (Saller 1982: 1, emphasis his; see further Nauta 2002: 18–26 on Saller’s criteria). 187. Especially Gold 1987 and 1992, Saller 1982 and 1989, and Nauta 2002: 11–34. 188. White 2007: 199. 189. Mitsis 1993: 114 likewise sees the figure of Memmius, as described by Lucretius, serving to tie together the pieces of DRN, arguing that “by treating the reader as a child or népios, Lucretius’ poetic testament is strictly in keeping with a larger system of connected imagery in De Rerum Natura taken as a w ­ hole”; even if “Lucretius deliberately chose this par­tic­ul­ ar historical figure as an addressee ­because he displayed a very healthy range of intellectual and moral vices—or, I would also add, ­because he displayed all the typical shortcomings of the didactic népios—in his role as addressee Memmius never strays beyond the bounds of his allotted literary function” (125).

pis on e s   153

poem to ­fathers and sons and the relationships between them; to life cycles in nature; and to the tricky dynamics that underlie the intersection of friendship and criticism that form the focus of the next chapter. ­These are tied together, and feel securely established within the poem, in large part ­because of the presence of the Pisones; without them, the Ars Poetica might justifiably be criticized (as it unjustifiably has been) as disjointed, a series of “more or less random reflections” strung together into lines of hexameter.190 ­Because of their integral place within the poem itself, however, as argued ­here, the dispute over their precise identity (as over the poem’s genre, name, and date) is misplaced: not only are the historical realities of members of vari­ous Piso families ultimately of secondary importance to the reading of the Ars Poetica presented ­here, but ­there is, as I see it, no identity, no fixed and unaltering “real” Pisones, whom Horace had in mind and who require uncovering.191

190. Fairclough 1929: 442. 191. Compare the figure of Caesar in Manilius’s Astronomica: “The living Caesar, no ­matter who he is, is of value to Manilius only as a poetic device; his precise identity is of no importance, and he does not require reference in specific or individual terms” (Neuburg 1993: 256–57).

3 Amici, risum

essential to the artist is his audience—­there can naturally be no criticism (or appreciation) of a work without critics. So it is that as soon as Horace has drawn for us with his words a painting in the opening lines of the Ars Poetica, an audience is conjured up to view and react to it. This audience, however, is neither ideal nor idealized. Having completed his masterpiece, the artist invites his “friends” (amici) to admire it—an audience preselected to guarantee (or so we imagine the artist thinks) the desired outcome of praise—­yet ­these friends react instead with laughter (risum) that they strug­gle to restrain, the moment left narratively unresolved for now. Laughter w ­ ill prove germane to the ways in which artist and audience relate to each other, and from the first it is inseparable from friendship, itself intertwined with the persons of the ­Pisones. A reflex, per Aristotle distinctively ­human,1 laughter offers a way for the critical impulse, a manifestation of friendship, to be voiced. Both terms—­ amicus and risus—­run the length of the Ars Poetica,2 appearing sometimes singly and at other times in unison, such that the senses of each crystallize further, both individually and together, with e­ very appearance. The result is an artist-­audience relationship that never quite stabilizes, remaining subject to continual negotiation and renegotiation from the poem’s beginning through its end, at which point the reader may find that s/he needs to start again from the beginning.

Risum Laughter next appears in the Ars Poetica strikingly reduplicated and in a form distinct from that of the opening scene: ut ridentibus arrident, ita flentibus afflent / 1. Part. an. 673a. 2. Kilpatrick 1990: 38 with n41 notes also the numerous appearances of the term risus, seeing “ARS-­RISUS” as “the fundamental antithesis that drives the Ars and makes it work as a poem.”

154

a m ici , r isu m  155

humani vultus (101–2).3 Not an undesirable and reflexive reaction to something absurd, it is now a response explic­itly desired by the artist and producible only by laughter itself through a mechanism of mirroring sympathy, just as the sight of tears produces tears. The ability to affect the audience in this manner is the mark of a poem not simply “beautiful” (pulchrum) but also “sweet” (dulce). Yet the reciprocal laughter of lines 101–2 is unique in the Ars Poetica, for elsewhere it is presented only as the response to—­and as the only appropriate response to—­incongruity. Ridentibus arrident serves also to flag a further appearance of laughter a mere four lines l­ ater, where Horace says that if dramatic characters, illustrated by Telephus and Peleus, speak in a manner that is incompatible with their characteristics (male si mandata loqueris, 104; cf. 89–98), “I ­will ­either sleep or laugh” (aut dormitabo aut ridebo, 105), the reaction made all the more pointed for the fact that it is his own.4 The same point is restated less than ten lines l­ater: “If the words of the one speaking are out of tune with his fortunes, the Roman knights and foot-­soldiers ­will raise a cackle” (si dicentis erunt fortunis absona dicta, / Romani tollent equites peditesque cachinnum, 112–13). What began as already critical laughter in line 5 has thus intensified into an even more earsplitting and mean-­spirited response,5 as Horace further shows the universe of laughers expanded to include all ­people: second-­persons plural (5), the poet’s own first person (105), and now the third-­persons plural (tollent) of an all-­ encompassing Romani, itself further explicated with the “archaic formula”6 equites peditesque as consisting of the full spectrum of Roman society. The connection between laughter and incongruity at last made explicit, Horace provides ever more instances of the same, on the one hand refining his criteria each time, and on the other continually inviting his reader to think back to and reevaluate the poem’s opening scene. A portentous and lofty opening line, fortunam Priami cantabo et nobile bellum (137) should be followed by more in the same vein, but when mountains ­labor the resulting incongruous mouse is dubbed “ridicu­lous” (parturient montes, nascetur ridiculus mus, 139), the adjective corresponding to the verb ridere. Another form of artistic shortcoming may 3. Though Orelli 1844: 721 objects to the reading ridentibus arrident, finding the repetition “nimis aperta ac prope putida,” as he does also to flentibus afflent (see chapter 1, n37). 4. “As at 103, 153, 188,” Horace h­ ere identifies “himself with the audience in the theatre,” Brink 1971: 187. 5. Rudd 1989: 168 suggests “guffaw” and Pseudo-­Acro glosses it as “effuse ridebunt,” but the intensity of the same noun at Pers. 1.12 is recognized by the scholiast who explains, “cachinnus autem risus est lascivior cum voce” ( Jahn 1843: 250), Keane 2006: 123 adding that it “reconceptualizes satiric laughter as more intense and mocking than a risus.” 6. Brink 1971: 190.

156  ch a p t e r 3

be seen in the examples of the lyre player and of Homer (347–69), where Horace seems to have set his sights on the sin of repeated and habitual error, particularly stemming from the unwillingness to accept correction.7 On the one hand, the unfortunate lyre player, equated to the proverbial pessimus poeta Choerilus,8 who always makes a m ­ istake on the same chord, is laughed at (ridetur) for this form of error, even as Horace admits, “I marvel at [him] as good with a smile/laugh two or three times” (quem bis terve bonum cum risu miror, 358).9 On the other hand, finishing up the section, Horace professes himself offended (indignor) “whenever good Homer becomes drowsy” (quandoque bonus dormitat Homerus), even though “it is divinely permitted for sleep to creep over a long work” (operi longo fas est obrepere somnum, 360).10 The juxtaposition of sleep and laughter recalls line 105, though sleep is now not a response to lack of harmony or compositional failure but rather the artist’s own action, an error (even if he cannot help falling asleep) that leads to further errors perceptible in his work. In a ­later example, perhaps illustrating a more evolved stage of self-­awareness, a iuvenis11—­t he Pisones serving again to fuse a theme into the poem’s material—­avoids an activity in the understanding that, should it not go as well as hoped, the result could be ridicule: ludere qui nescit, campestribus abstinet armis indoctusque pilae discive trochive quiescit, ne spissae risum tollant impune coronae. (379–81) 7. Horace’s train of thought in 347–60 is difficult to tease out (the punctuation of 354–58 is debated, but Brink 1971: 365 insists, “­there is nothing wrong with the text”: “the sequence of thought is apt and lively, provided H.’s dialectical pro­cess is understood”). In Brink’s view, Horace is exploring the standard princi­ple that “occasional ­mistakes [are] venial, botching unforgivable.” 8. Brink 1971: 365–66; Choerilus was “one of the highly paid court poets of Alexander of Macedon.” 9. Brink 1971: 365 suggests that Horace experiences “amused surprise” when “the botcher . . . ​ gets one or two ­things right,” but in conjunction with bonum, risus may suggest a smile of real enjoyment, rather than laughter proper, and thus indicate a further expansion of the semantic range of ridere in the AP. 10. This resembles Philodemus’s advice at On Frank Criticism fr. 46.5–11, where he explains, according to Glad 1996: 39, that “the teacher should thus not hate the one who commits pardonable ­mistakes, ‘remembering that he himself is not perfect and that all men are accustomed to err’ ” (πῶϲ γὰρ μιϲεῖν τὸν ἁμαρτάνοντα μὴ ἀπογνώ[ϲ]ιμα μέλλει, γινώϲκω[ν] αὑτον οὐκ ὄντα τέλε[ι]�ν καὶ μιμνή[ϲκων, ὅτι πάντεϲ ἁμαρτάνειν εἰώ]θαϲιν;]; Konstan et al. 1998: 58). 11. He is not h­ ere named as such but need not be, since his interest in athletics on the Campus Martius (as at 161–62: imberbis iuvenis, tandem custode remoto, / gaudet equis canibusque et aprici gramine Campi) identifies him as one.

125-81019_FerrisHill_HoracesArsPoetica_5P.indd 156

8/8/19 8:40 AM

a m ici , r isu m  157

The phrase risum tollant, the idiom repeated from 113, tollent . . . ​cachinnum, suggests as before a sudden and uproarious outburst from ­those watching: laughter as again reactive and irrepressible and emanating, as in the poem’s opening, from a collective of respondents, perhaps to be ­imagined as being encouraged in their individual impulse to laugh by their participation in a crowd where ­others are behaving in the same manner.12 Horace’s interest in unintentional, reactive laughter is clear from its insistent reappearance in this marked form throughout the Ars Poetica: first in the poem’s opening; then as male si mandata loqueris, aut dormitabo aut ridebo and si dicentis erunt fortunis absona dicta, Romani tollent equites peditesque cachinnum; leading through parturient montes, nascetur ridiculus mus and the musician poets of 347–69 to risum tollant impune coronae, the laughers as always unpunished and unpunishable.

Amici Although the poem opens with friendship and laughter in close proximity and intimately connected, the two themes diverge and enjoy separate existences for over 400 lines. A ­ fter line 5, friendship next appears in the context of the Roman male’s life cycle (158–76), where in the most idealized of the four stages and the only one presented in wholly positive terms, the newly adult man (aetas animusque virilis), now sensible and concerned with honor, busies himself with “seeking resources and friendships” (quaerit opes et amicitias). The image, though compressed, gestures at Roman public life, assumed to be the goal of ­every aristocratic male, and the term amicitia is accordingly loaded. Used at Rome “for a wide variety of relationships, ­whether between equals or unequals, ­whether marked by deeply felt affection or by mere urbane politeness, w ­ hether founded on selfless devotion or on the interested exchange of goods and ser­vices,”13 fundamental to any definition of the term was that it designated a reciprocal bond involving mutual obligations.14 This ideal of reciprocity informs line 167 where, 12. Corona, ­here in the sense of “circulus hominum” (TLL s.v. corona III A), further hints perversely at the victor’s garland (the sense of corona at AP 250) that the athlete cannot hope to have earned. 13. Nauta 2002: 15; cf. also Brunt 1965, 1988, White 1978: 90–92 (who cata­logs such potential beneficia: “bequests, loans, gifts, sinecures, arranged marriages”), Saller 1982: 11–15, Konstan 1997: 122–48, and further my pp. 173–83, “Roman Friendship and Horace as Amicus Pisonum.” 14. It was also “formally undefined,” White 1993: 27–28: “In contrast to many other relationships into which a Roman could enter (as parent, spouse, master, soldier, or maker of a contract, for example), friendship carried no ­legal consequences,” but rather “operated according to an almost completely situational ethic” (with the downside that it “easily became an open-­ended commitment for the weaker partner”).

158  ch a p te r 3

overlapping with the spheres of opes and honor,15 amicitia denotes the young man’s incipient connections, the web of social contacts without which he could not become a successful actor in the all-­important po­liti­cal system. Friendship of the same sort, involving persons of similar social status, bound to one another by shared values, the same profession in public life, and perhaps even by ­family ties, is meant by amici also at 312: qui didicit patriae quid debeat et quid amicis, quo sit amore parens, quo frater amandus et hospes, quod sit conscripti, quod iudicis officium, quae partes in bellum missi ducis, ille profecto reddere personae scit con­ve­nientia cuique. (312–16) The individual described has learned, w ­ hether through personal experience or study of o­ thers, what amounts to the training of the ideal Roman male, and lines 312–16 elaborate considerably upon 166–68. Horace not only hits the highlights of acceptable senatorial c­ areers—[sc. pater] conscriptus, judge, general—­but also lists the classes of relationships that mattered most for his contemporaries: ­those with a friend, ­father or ­mother, ­brother, and guest-­friend, as well as fatherland. Not coincidentally, ­these are the familial relationships that Horace focuses on throughout the Ars Poetica, aided in his effort by the addressee framework he has painstakingly constructed. If it is plain to the Piso boys who their f­ ather (parens) and ­brother (frater) are, perhaps Horace is suggesting that he, the third party in the world of the poem, is their hospes—­a stranger in the pro­cess of becoming an amicus and one to be treated with cordiality and warmth. Above all, however, friends and friendship appear alongside stages of the ­human life cycle, intertwined with the Pisones, ­father and sons. Horace introduces a third grammatical form of friendship into his Ars Poetica alongside amicus and amicitia with the adverb amice. “Let that one cherish good men and advise them in a friendly manner” (ille bonis faveatque et consilietur amice, 196), Horace urges in a lengthy passage detailing the be­hav­ior and role of the ideal chorus. By organ­izing, directing, instructing, ordering, nudging, the chorus succeeds in propelling the actors and the play’s action forward in the best pos­si­ble way, but this is such a peculiar description of the role of the chorus that we would do well to examine closely the import of amice. The overarching concern is of the need for the chorus to provide helpful support combined with constructive advice—­the first of several moments suggesting that friendship and criticism are connected, just as friendship and laughter are. Amice occurs next at 411, in what seems at first perhaps to be an incidental usage, though 15. Brink 1971: 237: “opes ­here defines amicitiae as ‘useful connexions.’ ”

a m ici , r isu m  159

one that reaches back directly to 196, with the verbs the adverb accompanies (coniurat, consilietur) further joined through alliteration: natura fieret laudabile carmen an arte quaesitum est. ego nec studium sine divite vena nec rude quid prosit video ingenium; alterius sic altera poscit opem res et coniurat amice. (408–11) We are alert to this unusual adverb from its ­earlier appearance, and may quickly discern that it portends a clustering of terms for friendship, which total five at 408–52 (of only eight total instances in the Ars Poetica). Thus Horace, as he so often does with key ideas, concentrates his discussion of friendship into one section of the Ars Poetica, having expanded upon his thoughts on the m ­ atter with each subsequent instance of the term and now ­doing so to the fullest.

Laughter and Friendship At 408–52, plaiting together friendship and laughter once more, Horace takes up the prob­lem of amici as critics with which the poem opened: ut praeco, ad merces turbam qui cogit emendas, assentatores iubet ad lucrum ire poeta dives agris, dives positis in faenore nummis. si vero est unctum qui recte ponere possit et spondere levi pro paupere et eripere artis litibus implicitum, mirabor si sciet internoscere mendacem verumque beatus amicum. tu seu donaris seu quid donare voles cui, nolito ad versus tibi factos ducere plenum laetitiae: clamabit enim ‘pulchre! bene! recte!’ pallescet super his, etiam stillabit amicis ex oculis rorem, saliet, tundet pede terram. ut qui conducti plorant in funere dicunt et faciunt prope plura dolentibus ex animo, sic derisor vero plus laudatore movetur. reges dicuntur multis urgere culillis et torquere mero quem perspexisse laborent, an sit amicitia dignus: si carmina condes, numquam te fallant †animi sub vulpe†16 latentes. (419–37) 16. The text is variously printed with (Shackleton Bailey 2001) and without (Orelli 1844: 792, Rudd 1989: 73, Brink 1971) obelization (though Brink obelizes in his commentary, 1971: 411,

160  ch a p te r 3

This rapid sequence of scenes invites us to think back to the poem’s opening, not merely ­because they depict artistic evaluation in pro­gress, but especially ­because of the conjunction of derisor with amicum, amicis, and amicitia. Whereas Horace had e­ arlier narratively pressed pause, leaving unanswered the question spectatum admissi, risum teneatis, amici? and leaving us f­ ree to speculate about the thoughts ­going through the heads of the painters’ friends at the moment of near laughter (1–5), at 426–30 he allows the scene to play out: the poet’s verses cause his audience members to exclaim and weep. Surrounding this display of emotional excess are two further vignettes, and all three clarify and nuance the poem’s opening act of art criticism. First is the statement derisor vero plus laudatore movetur. The precise sense of derisor, and therefore the essence of Horace’s thought pro­cess, has caused puzzlement (some even seeking to emend the reading to adrisor),17 and the passage above has consequently not been viewed as terribly central to the poem. Rudd 1989: 220, evidently relying on the Oxford Latin Dictionary where derisor is defined as “a mocker, derider,”18 reveals the pervasive confusion when he goes on to muse that, nevertheless, “the man cannot be, in the ­simple sense, a mocker.” The reason given is that “he is the insincere counterpart of one who gives genuine praise (vero laudatore).”19 Brink 1971: 410 suggests that “the assentator of 420, mendax amicus of 425, turns into a derisor,” yet this established interpretation of the laugher as insincere fails to satisfy entirely since Horace’s schema would then show a fairly banal contrast between true praise (witnessed nowhere ­else in the Ars Poetica, it might be added) and false praise. If, however, we read Horace in line 433 as speaking not of a “true praiser,” vero laudatore,20 but rather as explaining, “no one to my knowledge has made satisfactory sense” of it), while ­others argue to emend (e.g., Peerlkamp 1845: 151, who suggests animi sub amica pelle latentes). If genuine, animi sub vulpe latentes is taken as referring to the Aesopic fable of “the crafty fox who persuaded the crow to sing and then made off with the cheese” (Rudd 221), and would in the context urge would-be poets to take care that they are not deceived (fallant) by “[sc. malicious] feelings (animi) lurking ­under the [flattering, tricky] fox” (to paraphrase Rudd). 17. Nisbet 1986: 229 (cf. Rudd 1989: 221); this despite the fact that adrisor is a hapax, its sole attested usage being a ­century l­ ater at Sen. Ep. 27.7 (TLL s.v. arrisor). 18. See also TLL; ­these definitions are, however, problematized by being based on a very few instances. 19. The prob­lem, then, lies in the need to accept that derisor means both “mocker” and “flatterer,” two distinct and even contradictory meanings; Rudd is accordingly forced to won­ der, “Could it be that the term came to be used in an attenuated sense as ‘a hired entertainer,’ who could behave in e­ ither way?” I see the idea of flattery in the scene residing rather in the laudator than the derisor. 20. In the “Materia” commentary it is already explained as “derisor, id est falsus laudator, movetur plus vero laudatore” (Friis-­Jensen 1990: 380), and subsequent commentators follow suit.

a m ici , r isu m  161

saying “a laugher is in fact/truly (vero) moved more than a praiser (laudatore),” a dif­fer­ent understanding of the scene overall can emerge. The separation of vero and laudatore (by plus)21 and the exclusively adverbial use of vero to mean “in fact” throughout the rest of the poem22 and ­here emphasizing the ele­ment of surprise support this interpretation. Horace’s key point is that laughter as criticism is the most spontaneous, heartfelt, and genuine of all emotional responses and is not capable of being feigned, a point heightened if we understand the phrase ex animo from the preceding line, where it describes the one whose grief is genuine, as extending to the derisor as well, thus making him one who laughs from the soul. That the outward emotional expression of the one who finds what he sees or reads absurd, and does not conceal or falsify his internal response, would be less effusive than that of the (false) praiser (who is, on the inside, amused even as he shouts pulchre! bene! recte!) is not paradoxical. Indeed, this is the point of the comparison set up in 431–32: the genuine mourner is less demonstrative than the hired mourner, and the degree of emotion shown by the truth-­telling laugher-­critic is accordingly outdone by that of the one who seeks to hide his emotional response by giving voice, and effusively, to its opposite. Scathing laughter, a response no artist desires, is nevertheless genuine;23 of praise, on the other hand, we can never be quite sure. The verus laudator is an impossibility in Horace’s critical universe, and only laughter, we have been made to realize bit by bit, can be a true and guileless expression of one’s internal life—­even though it may cause pain to its object. Rudd 1989: 220, commenting on lines 429–30, rightly complains that we see “physical manifestations” of emotions but not the a­ ctual emotions that underlie them (just as we ­were not privy to ­these in line 5). By contrast, with line 433 we get as close as we ever ­will in the Ars Poetica to seeing under­lying emotions in the context of laughter and to understanding what drives laughter and what purpose, in turn, it serves. Even grief, not to mention praise, plea­sure, and delight (427–30), can be feigned (431–32), to such a degree that hired mourners outperform the genuinely bereft in their displays (though presumably not their inner feelings) of sorrow.24 The exuberant delight of the single audience 21. Horace could have written derisor plus vero laudatore movetur had he wished to signal that vero was an adjective to be taken with laudatore. 22. At both 422 and 475 vero is the adverb; the adjective, in the form veris, is used only at 151 and 338, substantively and in the plural, to mean “true ­things.” 23. On the view that truth telling requires laughter elsewhere in Horace and in the Greco-­ Roman comedo-­satiric tradition more generally, see Ferriss-­Hill 2015. 24. Commentators have been puzzled as to why Horace speaks of masculine hired mourners (conducti) rather than the traditional feminine ones (as at, e.g., Lucil. fr. 959–60, which Horace

162  ch a p te r 3

member who is not quite a friend (only his eyes are dubbed such, 429–30) w ­ ill ­later be remade along with lines 208–19 by Persius (1.1–23) into what Freudenburg 2001 describes as “one of the most jarring and (meta­phor­ically) pornographic scenes in all of Roman satire.” The result is that Horace’s “calm, confident voice” becomes, as we are forced to read it again through the Persian lens, “unstrung, cynical and dismayed . . . ​feckless and passé.”25 Although Persius may cause distress as he wreaks destruction upon our “cherished notions of Horace,” he has managed to get right to the heart of the m ­ atter, for as Freudenburg sees, he “makes us think twice about all the writhing and pleas­ ur­able moaning we see and hear” in Persius’s remake and “remembering Horace, we ask: do they ­really mean it? Or is theirs, too, a purchased, but stunningly performed, fake orgasm?” The crux in Horace’s illustration of the ­matter lies in the fact that the person attempting to determine the reliability of a friend is wealthy (420–21), able to provide a ser­v ice of some type (lavish foods,26 sponsorship, l­ egal aid, 422–24).27 The second formulation at 419–37 with implications for the entire Ars Poetica is at 424–25: inter-­/ noscere mendacem verumque beatus amicum. I see this difficulty under­lying and informing the w ­ hole poem: how can one tell the difference between a lying (mendacem) friend and a true (verum) one? This, the paradox of the true friend, had been explored e­ arlier by Plato28 and Aristotle,29 as it would be ­later by Plutarch and Maximus of Tyre.30 Mere de­cades before is supposed to have had in mind; Cichorius 1964: 117–20, Brink 1971: 408–9), suggesting even that the text be emended to quae conductae (so, e.g., Rudd 1989: 220). Kiessling and Heinze 1914: 360, however, in seeing that Horace has in mind mendaces amici as the point of comparison (Brink 1971: 409 disagrees), come closest to the mark, though since Horace speaks of a singular mendacem . . . ​amicum at 425, I see conducti reaching back rather to the plural (and masculine) amici of the poem’s opening. 25. Freudenburg 2001: 161–68. Brink 1971: 407 also compares Pers. 1.48–49 with AP 428–29. 26. Brink 1971: 403: “unctum ‘a rich t­ able.’ ” 27. The specific forms of wealth enumerated and the benefits the imaginary figure can provide evoke persons of the Pisones’ standing, much as 369–72, 379–81, and 383–84 have also been read. 28. In the Lysis, Plato shows Socrates and his interlocutors working t­ oward a definition of friendship, a central premise being that friendship or love and goodness or beauty must be related; see especially 214b–222e with Shorey 1933: 113–18. 29. Especially in the Nicomachean Ethics, though on the familiarity of this work to Roman readers see Cic. Fin. 5.12 (who believed it to be by Aristotle’s son, Nicomachus) with Gill 2012: 34. As Brink 1971: 405 says, Cichorius 1964: 117–20 argued “plausibly that the true friend also has a Lucilian ancestry,” referring to the first satire of book 26, a view endorsed also by Fiske 1913: 1–2, 9, 1920: 446, 458–59. 30. See Konstan 1995: 333, 1996a: 7, and Glad 1996: 23. For Plutarch, as Konstan 1996a: 7 explains, παρρησία, “frankness,” is of the utmost importance ­because “candor is the sign of the true friend.”

a m ici , r isu m  163

Horace wrote the Ars Poetica, moreover, Cicero had devoted an entire dialogue to the topic of friendship, De Amicitia, published in 44 BCE. Cicero, like Plato, regards friendship as predicated upon goodness,31 an interdependence that leads him to discuss the dynamics also evident throughout the Ars Poetica. The double-­edged prob­lem, Cicero explains, is that “friends must often be warned and reprimanded, and ­these [warnings and reprimands] must be accepted in a friendly manner, ­under the circumstances when they are offered benevolently” (et monendi amici saepe sunt et obiurgandi, et haec accipienda amice, cum benevole fiunt, 88). The unwillingness or hesitancy32 to cause offense in this way, even as it is done without harshness or insult (monitio acerbitate, deinde ut obiurgatio contumelia careat, 89), leads to an even greater offense, in that it allows the friend to continue headlong in his destructive course of action. In short, as Cicero insists throughout, “both to warn and to be warned is characteristic of true friendship” (et monere et moneri proprium est verae amicitiae, 91).33 As for how one can detect a false friend, especially a clever and sneaky one (callidus ille et occultus, 99), he admits that this may be difficult and recommends merely “the application of diligence” (adhibita diligentia, 95).34 Also providing essential—­perhaps the most essential—­background to Horace’s treatment of friendship in the Ars Poetica and to lines 422–37 is the understanding of friendship in the Epicurean tradition and especially in the writings of Philodemus. The two most impor­tant works in this regard, as their titles aptly indicate, are On Flattery (περὶ κολακείας) and On Frank Criticism (περὶ παρρησίας).35 Sider 1995b: 44 makes two points that may prove critical for the pre­sent discussion: first, “it may well be significant that Philodemus addresses Vergil, along with Varius Rufus, Varus, and Plotius Tucca, in his work 31. When the good man (virum bonum) seeks o­ thers like himself, “stability of friendship” (stabilitas amicitiae) results, Amic. 82; cf. 83, 100. On Plato see chapter 3, n28. 32. Cicero’s term for this is obsequium, which he credits to Terence; it also denotes “compliance, deference, solicitude.” 33. He acknowledges, however, as Horace w ­ ill, too, the possibility that a person may be utterly unreceptive to the truth even from a friend (cuius autem aures clausae veritati sunt, ut ab amico verum audire nequeat), concluding merely that “his health/safety is a cause for despair” (huius salus desperanda est, 90). Konstan 1996a: 12 cites the retort attributed to Plato’s near-­ contemporary, Diogenes the Cynic: “Other dogs bite their enemies, I [bite] my friends—so that I may save them” (οἱ μὲν ἄλλοι κύνες τοὺς ἐχθροὺς δάκνουσιν, ἐγὼ δὲ τοὺς φίλους, ἵνα σώσω, Stob. Ecl. 3.13 [Peri parrhesias] 44, at Wachsmuth and Hense 1894: 462). 34. Cicero’s view of amicitia in the context of his own friendships is discussed by Konstan 1997: 122–37. 35. See Glad 1996: 31 on why On Frank Criticism is the most appropriate En­glish rendering of the Greek title, and Tsouna 2007: 91–118, 126–42 for useful overviews of both this work and On Flattery.

164  ch a p t e r 3

On Flattery,”36 and second, for Philodemus “the virtue opposed to κολακεία is παρρησία.”37 The first of Sider’s points raises the possibility that the Epicurean dynamics of friendship operating within the Philodemean set are to be seen lurking also u­ nder the surface of the Ars Poetica, as they have increasingly been discerned in the Epistles and Satires as well.38 The second should make us alert to the extent to which flattery and frankness are opposites in Horace, too. The viewpoint of Philodemus that I see expressed above all with consistency and clarity throughout the Ars Poetica is that of the necessity of complete openness among friends: “On Frank Criticism reflects an ideal of non-­concealment,” encouraging friends “not to hold back but to reveal their faults and innermost secrets to o­ thers,” for “the benefits of such an openness outweigh any conceivable setback.”39 Horace’s exploration of the dynamics of friendship in the Ars Poetica was thus in and of itself neither innovative in antiquity nor the last word on the ­matter, but the par­tic­u­lar and distinct uses to which he puts ­these in his epistolary, didactic, satirical poem are notably original. First, amicitia is located within the poet-­addressee framework (rather than that of teacher and student explic­itly, or friendship among equals, such as between two poets); and second, the topics of true as against false friends and of the form(s) that friendship should ideally take are deployed with a specific addressee group in mind, the Pisones. In place of a treatise dedicated to the theme, such as is found in Plato, Cicero, and Philodemus, then, friendship is in the Ars Poetica, like poetry, put

36. He adds, “That this par­tic­u­lar group offered friendly advice on each other’s poetry is strongly suggested by Horace, The Art of Poetry 438–52,” though on Horace as outsider to the group, see my introduction, n161. 37. Contra Gargiulo 1981: 104, who understands (in this work entitled On Vices and Their Opposing Virtues) the virtue corresponding to the vice of flattery to be φιλία, “friendship,” and while παρρησία seems the better counterpart (for one ­thing, like κολακεία, it involves speaking), Gargiulo’s argument shows how deeply dependent are friendship and frank/free speech. 38. As recently as 2010, Kemp was able to write that “Philodemus’ influence with regard to Horace’s ­handling of amicitia (unlike that of Aristotle, or Epicureanism generally) has as yet been somewhat underplayed” (75), though, e.g., Armstrong 2004 should be noted as an impor­tant example of its application to Epistles 1. See also now the work of Sergio Yona 2015, 2017, 2018a, 2018b (who likewise notes that the connection between Philodemus and Horace’s Satires “has not received the attention in modern scholarship that it truly deserves,” 2018a: 16, though it was explored already by Michels 1944 and despite the obvious connection between Philodemus’s On Frank Criticism and Horace’s satirical libertas, on which, see Konstan et al. 1998: 3–5). It should also be noted that Horace’s AP in relation to Philodemus received some attention in Rostagni 1930, on which see Porter 1995: 143–45, and in the 1995 PhD dissertation of Tsakiropoulou-­Summers. 39. Glad 1996: 48, 50.

a m ici , r isu m  165

to other and further uses, yet its limits and conditions are continually probed and interrogated no less than in explicit discussions. With this literary background in mind, lines 419–37 allow us to reconsider the poem’s opening, since in typically Horatian fashion, further iterations of a topic enrich and nuance e­ arlier ones. In both instances an artist has invited one or more friends to give their opinion. Although the second artist is a poet rather than a painter, ad versus . . . ​ducere is physical and kinetic in the manner of walking ­toward a painting to examine it and is an unexpected way to describe approaching a written work, strengthening the sense that Horace wants his reader to have viewing (as of a painting) in mind.40 On this latter occasion we are further informed that the poet is wealthy, and as a result his supposed amici approach his work with the expectation of a reward in return for their praise: they not only exclaim pulchre! bene! recte! but grow pale with emotion,41 leap about, and squeeze some tear-­drops “from their friendly eyes,” where the adjective amicis serves to draw our attention again to precisely the prob­lem at hand—­are ­these true friends or not?42 The artists of 419–37 are unpleasantly aware of all this (ignorance would indeed be bliss): as much as they may wish to think of their audience members as amici, the latter have been invited in the role of assentatores, “yes-­men,”43 and with full knowledge of their needy circumstances. The artist (as seems likely) of 422–25 has, ­after all, given or offered to give them food and financial or ­legal help, hoping that, despite their inherent unreliability (levi),44 they ­will turn out to be reliable. Consequently, he is himself responsible for the unenviable situation in which he must agonize over w ­ hether the praise he receives is genuine—­like a king, we are told, who plies ­people with wine in an effort to determine w ­ hether they are r­ eally worthy of friendship (amicitia). As we now think back to the opening vignette of the Ars Poetica, where Horace yelled “cut” before we could see w ­ hether the amici in fact succumbed to their 40. Brink 1971: 406 compares ad lucrum ire (420), strengthening the points of comparison between the poet and the praeco. 41. Pallesco is delightfully ambiguous, in that it may denote “the paleness of fear” (Brink 1971: 407) or that resulting from awe or amazement (OLD). 42. Brink 1971: 408 also notes the repetition of amicus (the adjective) at 429, picking up on the noun in 425. 43. Intriguing in this context is Cicero’s tarring at Pis. 29 of Piso’s Graeculus, i.e., Philodemus (see chapter 2, n8) with the label assentator/adsentator. ­There could hardly be a worse insult within the Epicurean-­Philodemean constructs of friendship, flattery, and f­ ree speech, as Cicero surely would have known, and for Horace to use this same rare term (see TLL s.v. adsentator) in the AP suggests 1) that he knew Cicero had applied it to Philodemus, and 2) that he employed it as a friendly (or not so friendly) dig at Philodemus. 44. Brink 1971: 404: “not only needy but worthless.”

166  ch a p te r 3

impulse to laugh, we won­der anew w ­ hether, u­ nder compulsion, they praised the artist’s efforts or ­whether—­true friends—­they ­were able to critique the work: the scene can play out ­either way. The two pos­si­ble courses of action, it is fi­nally made explicit at 419–37, are to express the laughter that is building up (which is, in a sense, irrepressible) and be a derisor, or to give voice, as enthusiastically as pos­si­ble and falsely, to its opposite, thus acting as a (sc. falsus) laudator, both agent nouns concretizing their referents as doers of t­ hese actions alone. Curiously absent from the Ars Poetica is the possibility of genuine praise—­a stance that is perhaps to be tied to the unsettling sense that the poem is, from beginning to end, a caution against writing by all not-­Horaces. If the genuineness of praise is unknowable, the genuineness of a friend is not: a friend who praises might seem a true friend, but if his praise is false (as it by necessity is, in the framework presented), so is he; on the other hand, a friend who laughs at someone’s faults may have caused that person pain, but the trustworthiness of his friendship is not in doubt.45 Having painstakingly and deliberately laid the groundwork on friendship and laughter throughout his first 437 lines, Horace provides near the end of the Ars Poetica an exemplum of how criticism ­ought to work. A teacher figure is envisioned correcting and improving the would-be writer’s efforts, and the scene stands in most direct contrast to the poetaster of 408–18 who feels entitled to write poems and pronounce them “amazing” (mira), all without putting in the training that is universally agreed upon as necessary in other areas of endeavor such as athletics and ­music: Quintilio si quid recitares, ‘corrige, sodes, hoc’ aiebat ‘et hoc.’ melius te posse negares, bis terque expertum frustra delere iubebat et male tornatos incudi reddere versus. si defendere delictum quam vertere malles, nullum ultra verbum aut operam insumebat inanem, quin sine rivali teque et tua solus amares. (438–44) Horace has in mind an ideal form of the creative pro­cess whereby an advisor figure is, like a personal trainer, able to push a student figure to achieve more than s/he thought pos­si­ble—­a keen-­sighted exposition of how the pro­cess of revision, so critical to writing, o­ ught to work at e­ very level (­after all, feedback is most useful, to students as to professional writers, when they can themselves 45. Brink 1971: 412 rightly notes the balance in the AP of the ideas that “the true critic is the true friend” and “the true friend is the true critic.” I have sought ­here, however, to elucidate how this balance of ideas relates to Horace and the Pisones, and thus how it informs the context and substance of the poem.

a m ici , r isu m  167

no longer see what changes are needed and regard their piece of work as finished). This dynamic, however, presupposes a student figure willing to make the effort,46 and we see at 442–44 that in the case of an utterly intractable student,47 the advisor figure, who in this instance suggests himself as none other than Quintilius Varus in the flesh,48 also becomes unwilling to continue trying. The one who w ­ ill not admit to even an obvious blot (delictum) is left with only one admirer—­himself. Yet even in what has been regarded as a wholly positive exemplum of the critical pro­cess,49 it is pos­si­ble to discern a sneaky critique of Quintilius in light of Philodemus’s thinking on the ­matter in On Frank Criticism. The first hint is contained in the fact that Quintilius makes a mere two or three attempts at correction—­neither a magic number like nine or ten that might have yielded results nor the actions of a dedicated and indefatigable mentor. Second, as Philodemus explains, it is the teacher who must tailor his

46. See Glad 1996: 44–54 and Konstan 1996a: 13 on the need for this. 47. The position of frustra admits of vari­ous readings, and it may be best to understand them all in unison working to communicate the futility of the entire enterprise: with expertum, it would indicate that the poet’s efforts have been “in vain” (despite his two or three attempts at correction); with delere, it suggests that any emendations made to the work are doomed; and with iubebat, it would mean that Quintilius’s advice is “in vain” since the would-be poet has no intention of taking it. 48. An identification made already by the scholiasts (Pseudo-­Acro: “hic Quintilius erat Varus poeta Cremonensis, amicus Vergilii”; Porphyrio adds, “eques Romanus”), though some confusion exists with a P. Alfenus Varus of Cremona (Rudd 1989: 222). According to the generally accepted identification (Nettleship 1883, Dilke 1958, Nisbet and Hubbard 1970: 279–80), this Quintilius Varus died in 24/23 BCE (per Jerome; ­whether he is to be understood as living or deceased at the moment of his portrayal in the AP has naturally been marshaled as evidence for the poem’s dating) and is likely the same Quintilius whose death Carm. 1.24 mourns (so already Porphyrio; some, e.g., Dilke 53, would have also the Varus of Carm. 1.18 be the same individual, though White 1993: 235 finds it “not likely,” yet regardless of the historical personage of Quintilius Varus, I see ­little reason to doubt that as a named person he is consistent across the Horatian corpus). The Quintilius of AP 438 would therefore be a member of the Herculaneum set alongside Virgil, Varius, and Plotius (see Armstrong 1993 and 2004, Tsakiropoulou-­Summers 1998, Cairns 2004, Gigante 1995: 49–61 and 2004 on the relations among t­ hese figures). Understanding Horace’s Quintilius as the Κοιντίλιε named in Philodemus’s On Slander (see introduction, nn159–61) would support the importance of Philodemus’s writings to the AP and raises the possibility that Horace, standing outside the group and looking jealously ­toward them in his hexameter persona, is casting aspersions on one or more of their members. 49. Rostagni 1930: 125 (“Esempio di un sincero amico a squisito giudice di poesia”), Brink 1971: 413 (“Quintilius is not an assentator, a false friend; he is the verus amicus ­because he is a true critic”), Armstrong 1993: 216 (who speaks of “Horace’s picture of the ideal Epicurean parrhesiasts and poetic therapists: the critic Quintilius”).

168  ch a p te r 3

teachings to each student, not the student who must bend to the teacher’s method.50 The failure of the student to learn, then, may lie not in him but rather in Quintilius, who has failed to arrive at the optimal way of reaching this specific student for, according to Philodemus, “­there are no incurable ones, no faults which cannot be redeemed.”51 What is illustrated at 438–44, then, much as readers have taken it to be complimentary ­toward Quintilius or even a fond reminiscence on Horace’s part of his time among the Philodemeans,52 is not true friendship. As the Quintilius vignette ends, Horace turns his attention to a vir bonus et prudens, the transition so swift as to suggest initially that the phrase is a description of Quintilius himself rather than a new, separate figure. Where Quintilius and the friendship dynamics of his situation fell short, however, the vir bonus et prudens succeeds. At first reading he appears to be an idealized self-­critic, one who has evolved beyond the need for an external teacher, being able on his own to provide himself with a genuine and useful critical voice: vir bonus et prudens versus reprehendet inertis, culpabit duros, incomptis allinet atrum traverso calamo signum, ambitiosa recidet ornamenta, parum claris lucem dare coget, arguet ambigue dictum, mutanda notabit, fiet Aristarchus. (445–50) No sooner does this scene, which appears to show a solitary figure fussing over his own verses (just as Horace exhorts Lucilius to do at Satires 1.10.67–71)53 end, however, than it becomes apparent that at least one additional figure must be pre­sent. Horace is speaking, it turns out, not of an ideal, self-­critical poet, but rather of an ideal critic-­friend, since this individual “­will not say ‘why should 50. See further Glad 1996: 35: “Dif­fer­ent types of students emerge, namely, ­those violently resisting frankness, ­those who are irascible, ­those who cannot tolerate frank criticism, and, fi­nally, ­those of a lesser intellectual ability. The capacity of the young of dif­fer­ent dispositions to bear the frank speech of the sage is a major concern of the handbook.” 51. Ibid.: 42 (though at the same time, Philodemus seems to hold that “poetry is not a rational science that can be articulated and taught,” Ferenczi 2014a: 12). Philodemus discusses how to deal with especially obstinate students at On Frank Criticism fr. 10 (Konstan et al. 1998: 32; cf. Glad 1996: 40–44). 52. On which in Horace’s poetry more generally, see Gigante 1995: 26–27 and 2004 with Armstrong 2004: 268. 53. West 1967: 59–60, who traces the “vitality” of meta­phors from arboriculture across Horace’s works, convincingly explains duros (446) as “hard woody growth,” though Brink 1971: 416–17 is skeptical.

a m ici , r isu m  169

I offend a friend over trifles?’ ­These trifles ­will lead to serious evils the one [who has been] laughed at once and poorly received” (nec dicet ‘cur ego amicum / offendam in nugis?’ hae nugae seria ducent / in mala derisum semel exceptumque sinistre, 450–52). This idealized figure, so enlightened that he appeared at first to be critiquing and improving his own verses, checking and cutting them back in a highly physical series of images,54 is even given a name: Aristarchus, for a famous Alexandrian critic.55 The fact that his name means “best official” (arist-­ from ἄριστος, “best” + archus from ἄρχω, “begin; lead; rule, govern, command”), however, has gone entirely unnoticed.56 Although Brink sees a transformation57 taking place, I rather see Horace presenting two separate and distinct figures, Quintilius and the vir bonus et prudens. ­These two are compared, moreover, with the result that the former falls short of the ideal, embodied by the latter who is given the title to prove it—­Aristarchus. Quintilius gives up when faced with recalcitrance (443), whereas the vir bonus et prudens does not, never wavering in his efforts to improve his friend’s writings even at the risk of offending him. This, and only this, w ­ ill lead to positive improvements to a work, though notably never to outright, genuine praise.58 The question Aristarchus is made to ask in line 450–51 (‘cur ego amicum / offendam in nugis?’) is rhetorical, since one should precisely not refrain from causing offense over even the minor artistic blemishes of a friend—­only through 54. Rudd 1989: 223 notes how “from 445 on, we have a series of words linking the authority of the critic with the authority of a l­egal figure,” w ­ hether prosecutor, censor, or judge, and “intertwined with this we have another, more imaginative, series, in which the critic is seen as a grower of fruit trees or vines.” 55. Used “to typify the complete critic” also at Cic. Att. 1.14.3 (OCD s.v. Aristarchus [2]). Aristarchus lived c. 216–144 BCE and produced critical recensions of key Greek texts (Homer’s Iliad and Odyssey, lyric poets, e­ tc.) in addition to original works on “grammatical, etymological, orthographical, literary, and textual criticism” (OCD). Frischer 1991: 58, working to connect Horace and the AP with Lucius Calpurnius Piso Caesoninus, notes that he is named also at Pis. 73. 56. Comparable is Horace’s play at Epist. 1.19.23 on Archilochus (Ἀρχί-­λοχος, “Rule-­swarm”) as dux reget examen (“the leader w ­ ill rule the swarm”), discerned by Katz 2008. 57. Brink 1971: 416: “Quintilius the man, already a somewhat exemplary figure, changes wholly into a type—­the honest critic, ‘an Aristarchus.’ ” 58. The idea of the vir bonus et prudens is itself perhaps somewhat problematized by its use in Epist. 1.16 to describe one Quinctius (whose name has resonances of Quintilius). As Armstrong 2004: 288–89 points out, the praise that ­ought to be reserved for the truly wise is wrongly accorded to him by the populus, since he is nothing of the sort. In this poem, then, which Armstrong regards as the central piece of Epistles 1, while describing the sort of interaction “expected between fellow students and progress-­makers in specifically philosophical studies,” Horace “rakes a young nobleman Quinctius over the coals as an aspirant in philosophy.”

170  ch a p t e r 3

painful critique w ­ ill he prove himself a true friend. The consequences of a failure to do so are spelled out (hae nugae seria ducent / in mala derisum semel exceptumque sinistre, 451–52) as Horace contrasts (as he had at 353–60, where laughter also features) occasional and habitual error.59 In a sense, we have been expecting derisum, or at least the sound of it, since it was preceded by risus (358, 381) and derisor. If the derisor was the (true) friend who laughed at someone’s efforts, derisum is its counterpart:60 the one laughed at.61 The notion of “having been laughed at [derisum] once and poorly received” has been much puzzled over, just as derisor itself perplexed at 433. “We are meant, of course, to think of a recital,” says Rudd 1989: 224, while Brink 1971: 420 insists that the poet described in this line “is not the rich dilettante of the previous section,” though he is right to say that the individual “recites his verse without first submitting it to an Aristarchus and consequently disgraces himself.” I see derisum as executing a connective and signaling function, reaching back to derisor and thus allowing both terms to stress the seminal role of laughter to the creative pro­ cess as Horace describes it throughout the Ars Poetica. If derisor describes the critic-­friend whose laughter is a reflex, difficult to stifle, yet laden with good intentions, derisum denotes the recipient of this corrective laughter, which is not mean spirited but rather, again, the irrepressible critical reflex as which laughter has functioned—in conjunction with friendship—­since the beginning of the poem. Any artist must be open to criticism: having once made an error and received corrective laughter in response to his poor per­for­mance (exceptum sinistre), it becomes incumbent upon him to prune, refashion, and improve his verses. The poet-­performer who fails to do so (­whether ­because unable or unwilling) is the one ultimately criticized by Horace: he is the poetaster attacked off and on throughout the Ars Poetica, the one who fancies himself destined to write ­because he is f­ ree born and a member of the equestrian class. Again, we 59. Although sēria, “serious,” is not etymologically related to serō2, “link together, entwine, interlace; join in a series, string together” (on which see chapter 1, with nn247–51), sēria mala “serious errors” nevertheless suggests *seria mala, the repetition over and over again of an error rendering it serious. Sēria ­earlier appeared at 226, vertere seria ludo, to describe the activities of satyrs (and ­there perhaps similarly evoking sermo < serō2). 60. Previous commentators, however, having understood derisor as “mocker/flatterer,” cannot make the connection with derisum, e.g., Brink 1971: 420 (“a reference to the assentator and derisor . . . ​would be misplaced”; “semel would convey nothing if derisum referred to the derisor of 433, and meant ‘fooled’ or ‘flattered’ ”; he is himself arguing against Pseudo-­Acro and ­later followers who in his view “misunderstand” line 452). 61. Nugae is used ­here in an unusually negative sense to mean “trifling faults” (Brink 1971: 420) rather than “poetic trifles,” a self-­effacing but ultimately positive usage far more common among Horace and the Augustan poets; see Ferriss-­Hill 2015: 53 with Catull. 1.4, Hor. Sat. 1.9.2.

a m ici , r isu m  171

may well won­der how the Pisones received this rather strong apotreptic of their ostensible literary aspirations. Aside from the ideas that a good friend criticizes his friend’s literary errors even at the risk of causing short-­term pain for the sake of avoiding worse pain in the long term, and that the best critic is a good friend (well placed to offer suggestions, and most likely to have ­these suggestions taken seriously), what is Horace’s point? Why is Quintilius named and his efforts perhaps critiqued as falling short? And what role do lines 408–76, the poem’s concluding section, play in the poem as a w ­ hole? The section began with the proposition, in real­ity a summary of the preceding lines (though a summary that may well recast what we thought the preceding lines w ­ ere about): “­whether a praiseworthy song-poem is made by nature or by art was the ­matter ­under investigation” (natura fieret laudabile carmen an arte / quaesitum est, 408–9).62 We would as ever do well to consider this theme not in abstract terms, but as a part of the Ars Poetica, which constitutes a coherent narrative ­whole, and I once again view the Pisones as key: they may not be addressed by name in this section, though they are pre­sent in the multitude of second-­person singular verbs employed. Horace, too, reinserts himself formally into the poem with an emphatic ego,63 as he muses that neither talent alone nor training in the absence of talent ­will suffice (409–10), and he has by this point said enough ­things directly to his addressees on the topic of talent that we are invited to think of them at this moment as well. The Pisones, accordingly, might be seen as the reges (434) who, tortured by uncertainty and doubting o­ thers’ intentions, in turn torture a friend in an attempt to discern ­whether he is true or not; and Horace himself might be seen as the socioeco­ nom­ically inferior friend whose truthfulness (especially if he speaks praise) remains always slightly in doubt. Rather than being a Quintilius who throws in the towel ­after a few tries (a reference all the more pointed if it denotes a real member of the circle of Lucius Calpurnius Piso Caesoninus, Philodemus’s patron), Horace seems to promise that he w ­ ill be for the Pisones an Aristarchus—­ and in ­doing so assures them with a wry wink that any pain he ­causes them is a sign his friendship is true. Historical Pisos aside, Philodemus’s On Frank Criticism is of the utmost importance for the dynamics of friendship on view throughout the Ars Poetica. Philodemus recommends kindness and patience, combined with per­ sis­tence: a student ­w ill “be treated” (θ]ε[ρ]απε)θῆναι) successfully “not by ­doing this [sc. censuring him] continuously, nor on all ­things, nor ­every 62. Brink 1971: 395 notes the technical language: quaesitum est makes use of education’s quaestio (“investigation” or “subject of discussion”). 63. Cf. Brink 1971: 395: “ego: the critic as at 306 and frequently thereafter.”

172  ch a p t er 3

­ istake including chance ones, nor with ­those being pre­sent who do not need m to be, nor through ridicule, but by taking up his ­mistakes in a sympathetic way and without mockery and insult upon [them]” (μηδὲ +)νεχῶϲ αὐτὸ ποιεῖν, μηδὲ κ4τὰ πάντων, * μηδὲ πᾶ5 ἁμάρτημα καὶ τὸ τυχόν, μηδ’ ὧν οὐ χρὴ παρόντων, μηδὲ μετὰ διαχύϲεω+, ἀλλὰ ϲυνFαθῶ[ϲ] τ[ὰϲ ἁμαρ]78αϲ ὑπολαμβάνειν καὶ μὴ] καθυβρίζειν μηδὲ λοιδορεῖ]5 ἐπὶ, On Frank Criticism fr. 79).64 Moreover, the teacher should guide his students ­toward “­doing ­things in accordance with the good [qualities] that they possess” (τἀκόλουθα τοῖϲ ἀγαθοῖϲ πράττειν, οἷϲ ἔχουϲι, fr. 68.6–7). Armstrong reads Horace tailoring his “value-­relative” Epicurean therapeutics65 precisely to a Piso son of ephebic age, appealing to his addressee “on the basis of his pre­sent beliefs, in this case as an aristocratic young Roman . . . ​not to abstract, ideal standards.”66 While also reading the Pisones as integral to the Ars Poetica and as integrated into it through the many references to what are well read as their own interests (youthful athletic pursuits on the Campus Martius) and prescribed path in life (lawyering, vel sim.),67 I ultimately do not see Horace offering his reader or the Pisones any path to a successful outcome as far as literary production is concerned. Horace does not concretely help his addressees, in all their multiplicity, become better poets or become poets at all ­because he cannot; in fact, no one can. Instead, Horace tells the Pisones, to whom he is speaking in the first place, to become ­lawyers or politicians, a course of action slightly distinct from what Philodemus apparently had in mind.68 If an under­lying message of the Ars Poetica might thus be that what Philodemus prescribes is not in e­ very case pos­si­ble, the poem becomes something of an indictment of On Frank Criticism. That neither it nor the Satires and Epistles, to which Philodemus has been more widely applied, have been read as such is further testament to the influence of Philodemus: Horace, though critical of his literary protégés of the moment, is never so overtly harsh ­toward them as to make his resolute opposition to their literary prospects evident. 64. Greek text as at Konstan et al. 1998: 82–84; as they point out, 85, n117, the “necessity of avoiding ridicule when criticizing faults” is treated also in fr. 37, 38, and 60. 65. On the model of Nussbaum 1986b. 66. Armstrong 1993: 193; cf. 194: “attentive to the par­tic­u­lar case”; “ ‘individual-­relative’ rather than based on community values.” 67. Though ­these are not specific to the Pisones themselves, being rather the interests and occupations common to any aristocratic Roman male over a period of many centuries. 68. Konstan et al. 1998: 74, cf. Glad 1996: 38; Philodemus does not appear to recommend that a teacher steer his students ­toward wholly dif­fer­ent activities, as this would involve giving up on teaching the activity in question, which is nowhere endorsed.

125-81019_FerrisHill_HoracesArsPoetica_5P.indd 172

8/8/19 8:40 AM

a m ici , r isu m  173

Roman Friendship and Horace as Amicus Pisonum Apparently drawing on their (near-)con­temporary En­glish model, early scholars of amicitia at Rome defined it as ­little more than a useful connection or “party relationship.”69 As Brunt 1965 argues,70 however, this overlooks the ample evidence for the rich tapestry of forms that friendship could take,71 encompassing bare acquaintances out of politeness; “the network of relationships between leading senators”; friendships that could withstand po­liti­cal disagreements and even deep divides; and ­those instances in which a friend was loved as a sort of alter ego—­after all, amicitia derives from amo. The extent to which the language and conception of Roman friendship intersected with Roman power structures begins to emerge in the work of Peter White, whose 1978 article analyzes amicitia in the context of the poet-­patron relationship. While the latter, certainly at first glance, had control over the former, White eschews the terminology of patronage altogether as neither useful nor appropriate and, in fact, ahistorical,72 arguing that the “familiar code of amicitia fully explains the treatment of poets in Roman society.”73 Even where the Romans could have used (and frequently did use) the terms patronus and cliens, moreover, t­ hese w ­ ere felt to be rather too blunt, even gauche, and therefore to be avoided, replaced instead with the language of friendship.74 What emerges as critically impor­tant from this brief survey is that two individuals designated amici may be quite 69. On this form of reading and its drawbacks, see Brunt 1965: 1, Konstan 1996a: 16 (who criticizes the tendency t­ oward reducing amicitia to a relationship of “quid pro quo reciprocity and all but evacuating it of sentimental or emotional content”; cf. 1997: 122–24). 70. Reprinted with revisions as Brunt 1988: 351–81. 71. Brunt 1965: 20: “the range of amicitia is vast. From the constant intimacy and goodwill of virtuous or at least of like-­minded men to the courtesy that etiquette normally enjoined on gentlemen, it covers ­every degree of genuinely or overtly amicable relation”; cf. also Brunt 1965: 5–12, Saller 1982: 7, Konstan 1995, Nauta 2002: 15. 72. White 1978: 78; cf. 2007. 73. White 1978: 74 (cf. 1993: 29); he establishes that “­there did not exist a Roman code of literary patronage,” and that rather, “it could only have been in the elaborate code of amicitia that a gentleman would find the tacit understandings and the standards of behaviour which would govern his relationship with a poet.” This approach is further nuanced in the work of Saller 1982: 8–22 (who discusses in detail the terminology involved), Wiseman 1982, Zetzel 1982, White 1993 and 2007, Konstan 1995 and 1997 (who speaks of the “tense dialectic” between amicitia and patronage, 1995: 341), Nauta 2002 (who is useful especially on ­later periods), and Griffin 2003. 74. As Cicero states at, e.g., Off. 2.69, on which see further Griffin 2003. On the patronus-­cliens terminology and attendant relationship(s), see further Gold 1987, Saller 1982 and 1989, White 1993: 29–32, Konstan 1995: 328, Nauta 2002: 14–15.

174  ch a p te r 3

disparate in social standing, influence, and wealth, or in artistic talents and the access they can offer to an intellectual milieu (though any disparity is concealed, and intentionally so, by the use of the same term to denote both parties),75 yet upon such a necessary asymmetry is predicated a balanced friendship.76 Horace experiences dif­fer­ent degrees of asymmetry depending upon whom he is addressing at any given moment of the Ars Poetica. With the maior iuvenum, Horace’s relationship is that of a teacher ­toward a student, and Horace, even as a social inferior, occupies the position of authority and superiority.77 This is not the case e­ arlier in the poem, however, where Horace addresses the Pisones as a group and with apparent deference describes them as pater et iuvenes patre digni (24), even o Pompilius sanguis (291–92). This may go some way ­toward explaining Horace’s five addresses to this ­family and the forms they take: by dividing them up into several configurations, Horace succeeds in introducing continual flux into the patterns and dynamics of power in the poem. If we posit a friendship between Horace and the Pisones in the Epicurean/Philodemean mold specifically, the criticism bordering on rudeness that I discern Horace evincing ­toward his addressees becomes all but required. Paradoxically, and yet wholly in keeping with ancient thinking on friendship, as evidenced especially in Cicero’s De Amicitia, it emerges that the only way for Horace to “prove” that he is a true amicus, worthy of their amicitia, is by being utterly frank with the Pisones, which requires that he be unflaggingly critical of his addressees’ literary prospects. The Pisones, in turn, painful as this may be for them, must respond with good cheer, gratefulness, and concrete action to e­ very detail of the criticism they receive. This uneasy back-­and-­forth, idealized as “true friendship,” is communicated above all in lines 411–50, with its curious concentration of amic-­terms. The insistent repetition of amice, amicum, amicis, amicitia, amicum may seem heavy-­handed once it is noticed, yet each reappearance adds a further nuance to the overall picture. Encapsulated in the phrase coniurat amice (411),78 which overtly describes the way in which Horace envisions the “hard work” of studium 75. White 1978: 80: “­W hether a man is superior, equal, or inferior in standing to another, both are called amici, and the relationship itself is amicitia,” though distinctions could be and ­were made by adding a qualifying adjective such as mediocris/modicus, as against magnus/beatus (see also Saller 1989: 57, Konstan 1997: 140–42). 76. See chapter  2, with n186 for Saller’s 1982: 1 criteria for friendship, among them “asymmetry.” 77. See Armstrong 1993: 194 on this par­tic­u­lar type of “asymmetrical” relationship, which Nussbaum 1986b sees exemplified in that of patient and therapist. Konstan 1996a: 13 notes that col. IVb, Xa, and XIVa of On Frank Criticism give “practical advice on how to treat students of high station.” 78. Brink 1971: 397 explains the phrase as “a friendly pact.”

a m ici , r isu m  175

and native ingenium cooperating to produce worthwhile verses (laudabile carmen, 408), each “demanding assistance” (poscit opem, 411) from the other, is the work of friendly and collaborative critique by which excellent art is created. This adverbial usage of amice was already anticipated by its e­ arlier appearance at 196 to describe not conspiracy but cooperation and advising, where its application to the chorus’s proper role is so unexpected as to intimate that Horace is talking about something ­else entirely, namely, the way in which he, amicus Pisonum, organizes, directs, instructs, ­orders, nudges his addressees throughout the Ars Poetica “in a friendly manner” or, better, “in the manner of a friend” (amice). In the progressing vignettes of Quintilius, the vir bonus et prudens, and Aristarchus, t­ here may be seen an idealizing vision of enlightened friendship, of self-­awareness, and, as a result, of superior artistic production, unattainable by any other means or pro­cesses. Beneath this, however, the relative positions of Horace and the Pisones are ­under continual negotiation: is Horace Quintilius, attempting to teach and yet failing and giving up in disgust ­after issuing a few half-­hearted instructions? Or ­will the Pisones, through their own be­hav­ior, allow him to act as an Aristarchus for them? While the idea that “ingenium, without which art cannot be, but which, left to itself, would destroy art and its prac­ti­ tion­ers” (Brink 1971: 421–22) has long been recognized as the current that powers the final third of the Ars Poetica, it is the deployment of unusual terms such as amice in conjunction with the ars/studium versus ingenium debate that serves, through intersecting with the friendly interactions of Horace and the Pisones, both to ground the theme in the poem’s fabric and to enhance the appeal of Horace’s ­handling of a traditional point of literary theory. Horace’s essential message, which he strives to communicate through judicious yet insistent repetition of amicus and related terms, especially near or in conjunction with laughter, is that the poetics of criticism and the poetics of friendship are intertwined, and I accordingly regard 411–50 as critical to understanding the ­whole poem. Horace employs the overlapping structures of friendship, in its many forms, making use of them to craft a poem in which the portrayals of friendship remain inscrutable and unstable. Regardless of whom he is addressing, however, Horace would have been the inferior party in a friendship with anyone who bore the rarefied name Piso, rendering his positioning of himself as “friend” (amicus) rather presumptuous, even if such presumptuousness was allowed and encouraged within the framework of amicitia. In addition, although the relationship between Horace and the Pisones does not qualify as a patronus-­cliens one in any meaningful sense, this institution necessarily overlapped with that of amicitia in its asymmetrical manifestations. The result is that Horace is able to make use of the notion of the Pisones as literary patrons—as indeed they ­were, just not for him—­even as it does not strictly apply in the Ars Poetica. I would add to the debate on the dynamics of friendship and patronage the observation that if the

176  ch a p te r 3

relationship between Horace and his Pisones is viewed as one of unequals, both social and literary, and consequently as a form of poet-­patron interaction, it would be especially forward of Horace to be the first to designate it an amicitia: that title should rightly be bestowed upon it as a sign of affection or mere good manners by the superior party.79 Horace’s status as amicus Pisonum, and consequently theirs as amici Horatii, is also susceptible to serious doubt ­because ­these figures are mentioned nowhere e­ lse in his writings. This is a highly unusual situation: Maecenas, naturally, appears frequently: he is addressed in Epodes 1 (as amice, no less), 3, 9, and 14, in Satires 1.1, 1.3, 1.6, in numerous Odes (1.1, ­etc.), and in Epistles 1.1, 1.7, and 1.19, and he is named several times in Satires 1.5, 2.6, and 2.8, as well as in 1.9, 1.10, 2.3, and 2.7, and in Odes 4.11—­a regular and punctuating presence. The two addressees of the literary epistles likewise appear elsewhere: Florus (Epistles 2.2) is the recipient also of Epistles 1.3, while Augustus, to whom Epistles 2.1 is dedicated as Caesar (Caesar Augustus in line 48), is addressed as such also at Odes 1.2.52 and 4.15.4, and as Auguste at 4.14.3.80 Several addressees from Epistles 1 likewise reappear in the Odes,81 and within the books of Odes, in addition to Maecenas, we find Virgil as a repeated addressee (1.3, 1.24), while C. Asinius Pollio (Odes 2.1) and C. Valgius Rufus (2.9) are listed among the poet’s friends at Satires 1.10.81–86. Th ­ ere is, in sum, a reassuringly constant universe of p­ eople named by Horace throughout his poetry books, both as addressees of individual poems and as figures mentioned elsewhere in passing. Yet in the Ars Poetica, Horace asks his reader to accept without questioning or hesitation that t­ hese Pisones, of whom we have seen neither hide nor hair in his writings yet and never ­will again, are his friends, and he theirs.82 All this suggests that amicitia in the 79. Notably, as White 1993: 32 points out, the language of friendship had evidently changed by the third c­ entury CE, such that Porphyrio found Horace’s salutation of Maecenas as “friend” at Epod. 1.2–4 (amice . . . ​Maecenas) inappropriate, adding that he should have called himself “client” (“clientem”) out of “deference” (“verecundiae”). Gold 1987: 116–17 notes further that in Sat. 1.6, in which Horace recalls the origins of their relationship, “Horace himself never calls Maecenas his amicus . . . ​but rather puts the statements in the mouths of o­ thers,” suggesting that he had himself not yet attained that degree of warm familiarity with his patron. 80. He is also referred to in the third person with ­either Caesar or Augustus or both numerous times throughout the corpus. 81. Nisbet and Hubbard 1978: 167: “For friends commemorated in epistles as well as odes one may compare Maecenas (Epist. 1.1, e­ tc.), Albius Tibullus (1.4), Manlius Torquatus (1.5), Septimius (1.9), Aristius Fuscus (1.10), Iccius and Grosphus (1.12),” as well as Quinctius (Epist 1.16 and Carm. 2.11). 82. While it is pos­si­ble that Horace was working to cultivate a new set of patrons (as White 1993: 28–29 discusses, several Augustan poets had multiple such friendships, even si­mul­ta­ neously), such a theory would want additional confirmation.

a m ici , r isu m   177

Ars Poetica is acting in a way other than to describe historical real­ity. Rather, as I have tried to show, it is wrapped up in the matrix of themes that the Pisones as addressees fix into the poem, and it allows Horace both to introduce a popu­ lar theme from his Satires and Epistles—­friendship—­into this other exemplum of hexameter poetry, and to explore and exploit that theme in the playful and cheeky manner characteristic of his sermo. Comparison with Philodemus may once again help to illuminate further ­these strands of friendship and patronage and of praise (with its attendant anxiety) and criticism (with attendant hurt), that I see as so fundamental to the Ars Poetica. While confronting explic­itly the prob­lems of friendship and truth-­ speaking in his works On Flattery and On Frank Criticism, Philodemus also put ­these dynamics to work covertly in one of his epigrams, where they coexist alongside the prob­lem of who in the poet-­patron relationship truly holds the power: αὔριον εἰϲ λιτήν ϲε καλιάδα, φίλτατε Πείϲων, ἐξ ἐνάτης ἕλκει μουϲοφιλὴς ἕταροϲ εἰκάδα δειπνίζων ἐνιαύϲιον· εἰ δ᾽ ἀπολείψειϲ οὔθατα καὶ Βρομίου Χιογενῆ πρόποϲιν, ἀλλ᾽ ἑτάρουϲ ὄψει παναληθέαϲ, ἀλλ᾽ ἐπακούϲῃ Φαιήκων γαίηϲ πουλὺ μελιχρότερα, ἢν δέ ποτε ϲτρέψῃϲ καὶ ἐϲ ἡμέαϲ ὄμματα, Πείϲων, ἄξομεν ἐκ λιτῆϲ εἰκάδα πιοτέρην. (Palatine Anthology 11.44 = 22 Gow-­Page)83 Tomorrow, most beloved Piso, your muse-­loving comrade drags you to his ­simple hut a­ fter the ninth hour, entertaining you at an anniversary dinner on the twentieth day of the month.84 If you ­will miss the dish of udders and the Chios-­born draught in honor of Bromios, you w ­ ill nevertheless see your all-­true comrades, you ­will nevertheless hear ­things sweeter than the land of the Phaeacians, and if you ever turn also your eyes to us, Piso, we ­will spend a fatter twentieth.

83. Text as at Sider 1995b: 46. 84. I follow Sider’s 1995b: 48 interpretation of the text, whereby ἐνιαύϲιον modifies not εἰκάδα (to produce the untenable idea of an “annual twentieth of the month”) but ϲε, such that Philodemus “is inviting Piso to come for his, Piso’s, annual—or perhaps, merely infrequent—­ visit to one of the monthly Twentieths.” As Sider further explains, 47–50, the twentieth of ­every month would have been immediately recognizable to Philodemus’s readers as the day on which Epicurus’s “followers would meet to remember him.” On Epicurean friendship, see further Konstan 1996a, 1996b, Cairns 2004.

125-81019_FerrisHill_HoracesArsPoetica_5P.indd 177

8/8/19 8:40 AM

178  ch a p t e r 3

The adjectives φίλτατε and μουϲοφιλὴς, both alone and through their repetition, may be “designed to evoke the central Epicurean idea of friendship”; at the same time, “Piso, the addressee of this poem . . . ​would know that the friendship alluded to in this poem could also evoke the patron-­poet relationship.”85 Sider, having noted the marked Epicurean background to the epigram, further reads it in light of the tradition of “begging songs” where a poor man visits a rich man’s or king’s ­house.86 Conceding that “at first glance, the situation would appear to be the complete opposite to that of the invitation to Piso,” since in Philodemus’s poem Piso “is the rich man coming to the home of a poor, or at least frugal, man,” Sider nevertheless suggests that “in a certain sense he, like the beggar, comes hat in hand hoping to take away with him something that can be found only at Philodemus’ h­ ouse. If the surface of this invitation poem puts Philodemus in the position of beggar, the subtext suggests that it is Piso himself in need.”87 Operative is the truism that “it is not the poets who are the clients, but the patrons”:88 the power appears to lie with the latter, but in real­ity “poets have their own kind of power: they command, at least potentially, the scarce resource of immortal fame.”89 As a result, “what by Roman standards is an inherently unequal relationship is made to approximate, however obliquely, the equality demanded by Epicurean friendship.”90 Horace, in exploring ­these same themes in his Ars Poetica, may be deliberately evoking, especially through his image of kings and their friendships that are ever subject to doubts (434–36), Philodemus’s epigram and thus casting a wide-­enough net to encompass the Lucii Pisones among his Pisones. If the relationship between a poet and his addressee, ­whether patron or not, is a form of gift exchange, where the poem itself (and potentially the fame that ­will accrue to its addressee) is the gift, then what sort of gift to the Pisones is the Ars Poetica? On the surface and by the traditional reading, rather a good one: Horace, an established public literary figure by this point (if we date the Ars 85. Sider 1995b: 47. 86. The connection with the Eiresione, one of the best known of t­ hese begging songs, comes philologically through the word ἐνιαύϲιον: in the song, a group begins, “Let us turn ­toward the ­great h­ ouse of a power­ful man” (δῶμα προϲετραπόμεϲθ’ ἀνδρὸϲ μέγα δυναμένοιο), “for much wealth is pre­sent inside” (Πλοῦτοϲ γὰρ ἔϲειϲι πολλόϲ), explaining, “I come yearly” (νεῦμαι ἐνιαύϲιοϲ). 87. Sider 1995b: 49. This notion that “the beggar, if offered something, has the power to bestow benefit on the ­house” was common to begging songs. 88. Zetzel 1982: 101. 89. Nauta 2002: 27, who rightly makes the distinction between asymmetry of social status (where higher status should lie with the patron) and asymmetry of power (where even a poet of markedly lower status can command the higher degree of power). 90. Sider 1995b: 50.

125-81019_FerrisHill_HoracesArsPoetica_5P.indd 178

8/8/19 8:40 AM

a m ici , r isu m  179

Poetica ­after the Carmen Saeculare),91 has honored the Pisones by writing this poem for them (and in response to their request, as was commonly believed). The gift is rendered more special by the fact that it is given without the slightest hint that the Pisones ­will provide Horace with any f­ avor in the f­ uture, as they ­were not his patrons, and perhaps not even his amici.92 (At the same time, of course, this could be how a true amicitia begins—­with an opening volley such as a poem, the Ars Poetica, dedicated to a ­family already known to be patrons of the arts, for the Romans viewed “beneficence as creative of relationships”93). If the Pisones or the Rockefeller-­reader in general detect any criticism in the poem, however, they would instinctively (even if they managed to hide it well) be a ­little reluctant to accept it on the terms on which Horace offers it: even if “frankness or παρρησία” ­toward a social superior from a true friend was an indicator of “the force of affection or φιλία,” as Konstan explains,94 criticism would nevertheless sting (and therein lies its effectiveness and the thorniness of the entire m ­ atter). Nevertheless, refusing a poem-­gift, even one that c­ auses the recipient pain, was not, so far as we know, an option. If the Pisones further saw Horace exploiting this fact, their reluctance could have turned to outright resentment, and yet they would still have had to accept it u­ nder the social norms dictated by the conventions of amicitia. To refuse or to cry foul at the joke Horace may be playing—­for neither they nor we can be sure w ­ hether the amicitia at hand is genuine or w ­ hether Horace is abusing the conventions of true 91. Augustus had appointed Horace to write a poem/hymn, the Carmen Saeculare, to be performed by a chorus of boys and girls on the occasion of the ludi saeculares in 17 BCE, an ancient festival that he had reconfigured as a way of inaugurating his new age (see further Putnam 2000). Horace indicates in Odes 4 (published 13 BCE) that this unique honor conferred upon him subsequently brought a new-­found fame: he describes being recognized on the street (4.3.22–23) and envisions a girl who had sung in the chorus reminiscing about the event l­ater as a married ­woman (4.6.41–44). 92. Contrast Philodemus’s epigram, where he “quite clearly, if not shamelessly, is angling for reciprocal benefits from Piso in the ­future,” Sider 1995b: 47. He explains further, 47–50: “Piso, that is, ­will provide patronage, while Philodemus ­will provide both Epicurean ambience and poetic delights: Piso and Philodemus ­will both be amici—­i.e. patron and poet—­and φίλοι, i.e. two members of an Epicurean fellowship,” the balance achieved by the fact that each is “in his own way in need; and each has the power to aid the other.” 93. Griffin 2003: 98 (she adds that for Seneca, “amicitia is more often viewed as the result than as the cause of an exchange of benefits”). Zetzel 1982: 92–93 is right, however, to note the curiosity that, although we might expect Horace to have been most concerned with patronage at the outset of his ­career, “the only time that any major Augustan poet seems seriously to be influenced in his choice of subject by the wishes of a patron is at the end of Horace’s life: Odes 4, Epistles 2, and in a dif­fer­ent way the Carmen Saeculare.” 94. Konstan 1995: 336.

180  ch a p te r 3

friendship in order to guard his poetic domain from outsiders who would encroach upon it—­would have exposed them as unable to laugh at themselves, as unable to take precisely the sort of criticism that Horace is patiently explaining is essential. However the poem is received by the Pisones, on the other hand, Horace still wins in that he w ­ ill be recognized as an amicus Pisonum—­this being the primary gift that he, a poet, could expect from this form of exchange.95 In short, the Ars Poetica is a very poor sort of gift, the sort that an addressee would not want at all. It has begun to resemble a satire in that re­spect, of which no one sensible would wish to be the subject (and thus object). If having an addressee is a predicament for a writer, then it is one rivaled only by being an addressee.96 That we may encounter amicitia in a problematized and even satirically exploited form in the Ars Poetica in the manner outlined ­here should hardly be surprising, given the widespread presence of many dif­fer­ent forms of friendship throughout the Satires and Epistles. Brink says that “the preoccupation with amicitia of H.’s hexameter poems is striking,” 97 while Kilpatrick’s 1986 study of Epistles 1 is tellingly named The Poetry of Friendship, and several scholars have further argued for its particularly Epicurean nature.98 Foremost among the friendship poems are Epistles 1.17 and 1.18, which have even been dubbed “Horace’s own version of a De Amicitia.”99 The former, written to one Scaeva, explains “in what way, in short, one o­ ught to associate with greater men” (quo tandem pacto deceat maioribus uti, 2),100 and Horace’s advice centers on the importance of not overplaying one’s hand: “­those who are s­ ilent about their poverty before a king ­will carry off more than the one who demands” (coram rege sua de paupertate tacentes / plus poscente ferent, 43–44), while of the one who loudly makes demands, as in the case of the boy who cried wolf, Horace says, “Soon ­there ­will be no trust in true misfortunes and pains” (mox / nulla 95. Cf. White 1978: 87–88. 96. Cf. Konstan 1993: 21: “It is never an unadulterated plea­sure to be the addressee of a didactic work.” 97 . Brink 1971: 401 (cf. also Kemp 2010: 65); he points especially to Sat. 1.3 (with Van Rooy 1968: 56–71) and 1.4, Epist. 1.18, and of course the entire premise of the Epistles, where one amicus writes to another; see also Mayer’s 1989 essay, “Friendship in the Satirists.” 98. See Armstrong 2004: 269–70, 284 and now also Yona 2015, 2017, 2018a, 2018b. 99. White 2007: 195. 100. I print tandem, the reading of the codices, in place of Horkel’s emendation tenuem, ­adopted by Shackleton Bailey 2001: 281. Horace’s words to Scaeva (Quamvis, Scaeva, satis per te tibi consulis et scis [“although, Scaeva, you give advice to yourself by yourself well enough and you have knowledge”]) recall ­those directed to the elder Piso son at 367 (per te sapis) and 386 (id tibi iudicium est, ea mens). On the identity of Scaeva, perhaps P. Paquius Scaeva, see Mayer 1994: 231; on the poem, see further Kilpatrick 1986: 43–48.

a m ici , r isu m  181

fides damnis verisque doloribus adsit, 56–57). Epistles 1.18 reminds one Lollius Maximus,101 addressed as liberrime (1), which Konstan renders as “of an exceptionally ­free or in­de­pen­dent temper,”102 to tread the mean between unctuous servility and rudeness. Kemp 2010 notes Horace’s wish in the Satires “to emphasize the genuine nature of the friendship” between him and Maecenas “in part due to o­ thers’ suspicions of Horace’s flattery,”103 and yet at the same time Zetzel 1982: 89 is right to see that “Virgil’s attitude to Maecenas and Augustus is more straightforward and respectful than that of Horace.” As a counterpart to the striving t­ oward true friendship (if not exactly among equals, then among men who are each able to offer the other comparable benefits), t­ here may be found in certain poems of the two books of Satires excruciating portraits of scurrae (“professional buffoons,” in this period) and umbrae (“ghosts,” that is, hangers-on) and the men who enjoy their attentions. Satires 2.5 places this within the framework of legacy hunting (captatio),104 Satires 2.8 within that of a dinner-­party,105 but in both cases the motive of the false friend is “profit.”106 Another perverted form of friendship may be seen in Satires 1.4, as Horace explains that far from enjoying the pain he ­causes ­others (laedere gaudes, 78), he ­causes such pain in order that ­people may improve themselves. Other men, however, and ­these are the ones you ­really ­ought to watch out for (hunc . . . ​caveto, 85), do the same for their own amusement, gossiping about friends ­behind their backs or failing to defend them against ­others’ attacks or to keep their secrets (81–85).107 Alongside t­ hese stands Satires 1.5, with its portrait of friendship among Maecenas, Horace, Varius, Plotius, and Virgil, as well as Cocceius, Fonteius Capito, and even the scurra Messius Cicirrus—­the Herculaneum set of Virgil-­Varius-­Plotius-­Quintilius pre­sent as ever in some configuration in Horace’s poetry.108 Amicitia, it is clear, is a multifaceted and multivalent 101. See Hunter 1985, Kilpatrick 1986: 49–55, Mayer 1994: 8–9, 111 (“identity . . . ​unclear”). 102. The point, as Konstan 1995: 338–39 notes, is that “libertas ­here corresponds to the Greek παρρησία or frank speech”; see chapter 3, nn37–38. 103. Kemp 2010: 65, discussing Sat. 1.6.45–48 and 1.9.43–60 (though as he notes, Horace’s insistence that his friendship with Maecenas was sincere itself constitutes flattery t­oward Maecenas). 104. Kemp 2010 and Yona 2018a: 201–32 read this poem in light of Philodemus’s On Flattery. 105. See Baker 1988. 106. Kemp 2010: 67. 107. Cf. Leach 1971: 628–29. 108. Armstrong 1993: 197 observes that “Horace speaks from beginning to end of his ­career as a fifth member of the same Epicurean sodalitas or contubernium addressed in the now Philodemus text,” regularly listing the names of its members (e.g., Sat. 1.5.40, 1.10.44–45, 81) in the same order in which they appear in the dedications to vari­ous books of On Vices and Their Opposing Virtues (see introduction, nn160–61).

182  ch a p t e r 3

institution, not merely in general but especially in Horace’s own poetry, and this fact should alert us to its potential complexity in the Ars Poetica where, as has been observed of it elsewhere in his verses, “friendship is not merely a focus in itself but also a framework in terms of which Horace pre­sents other subjects.”109 Epistles 1.18, one of the pair of poems that deal most explic­itly with friendship, may further inform the Ars Poetica in an intriguing way. Horace explains at line 9 that “virtue is the m ­ iddle between vices, [the m ­ iddle] led away from both sides” (virtus est medium vitiorum et utrimque reductum). The context is how one o­ ught to conduct oneself in relation to a social superior within the construct of amicitia (professus amicum, 2): neither being overly obsequious (1–4, 10–36) nor aiming to be so ­free in speech as to become deliberately hurtful (5–8) is the proper way.110 Rather, one should be deferential, attentive, and respectful (37–40)—­be­hav­iors that, if exercised in just slightly the wrong manner, easily cross over into toadiness.111 The superior friend, likewise, seems ever on the verge of being able to take advantage of his lesser companion.112 Although Horace succeeds in putting the ­matter as benignly and blandly as pos­si­ble, his advice to Lollius remains, uncomfortably, to “yield to the gentle ­orders of your power­ful friend” (tu cede potentis amici / lenibus imperiis, 44–45). The fact that ­these ­orders are “gentle” fails to transform them from coercive command into friendly suggestion that the lesser party may choose not to act upon; nor, if he does act upon them, w ­ ill he avoid treading uncomfortably close to the dreaded vice of flattery. In ­these ways the importance of the Aristotelian ­middle path is emphasized in Epistles 1.18,113 though it is also shown to be potentially difficult and distasteful for the lesser party to navigate. The m ­ iddle path that avoids extremes is a constant in the Ars Poetica as well, where it is likewise framed in the language of virtue and vice: the sentiment of virtus est medium vitiorum et utrimque reductum from Epistles 1.18 is enshrined as in vitium ducit culpae fuga, 109. White 2007: 206. 110. On Epist. 1.18 in the context of Philodemus’s On Flattery, see further Konstan 1995: 338, Kemp 2010: 66–69. 111. Noted also by Konstan 1995: 339–40, who ultimately reads Epist. 1.18 in rather negative terms (as does Hunter 1985: 484): “Association with the power­ful tends to undermine the very basis for friendship, which resides in honesty, self-­lessness, and reciprocity.” 112. See especially lines 40, 45–48, 65–66. Konstan 1995 accordingly adds a fourth criterion to Saller’s 1982: 1 three, namely, that the “relationship is at least potentially exploitative.” 113. Cf. Mayer 1994: 243 ad Epist. 1.18.9, Kemp 2010: 66 (who explains Horace’s message at Epist. 1.18.9 as being “the Aristotelian princi­ple that virtue is a mean to how to conduct oneself with a superior. The extremes to be avoided on ­either side are, explic­itly, flattery and [by implication] rudeness”).

125-81019_FerrisHill_HoracesArsPoetica_5P.indd 182

8/8/19 8:40 AM

a m ici , r isu m  183

si caret arte (31), its language also echoed by phrases such as primo ne medium, medio ne discrepet imum (152), quo virtus, quo ferat error (308), and in the “mediocre” ­lawyer who lacks the “virtue” of certain famed figures in the field (consultus iuris et actor / causarum mediocris abest virtute, 369–70). If we read Epistles 1.18.9 into the Ars Poetica, it would seem that Horace’s continual advice to hew to the m ­ iddle path between extremes of be­hav­ior applies not merely to the production of art, but also to his relations with the Pisones. Accordingly, and in keeping with his advice to Lollius in Epistles 1.18, Horace has sought to avoid both flattery and rudeness (or rather, rudeness so overt that it is too easily recognized as such). Alongside this set of verbal correspondences may be discerned at least one more: semel emissum volat irrevocabile verbum (“a word once let out flies away, not able to be called back,” Epist. 1.18.71), to describe the spilling of a power­ful friend’s secrets, a betrayal that cannot be undone, becomes nescit vox missa reverti (“a voice sent [out] does not know how to be turned back,” AP 390), describing the need to delay publication by storing a work away for nine years. Fi­nally, the caution that Horace issues at Epistles 1.18.86–87 may also be seen as operative in the Ars Poetica: “The cultivation of a power­ful friend is sweet to t­ hose who lack experience; the experienced one ­w ill fear it” (dulcis inexpertis cultura potentis amici: / expertus metuet, 86–87). Horace, with the voices of senex, pater, and doctor that he exercises in the poem, surely qualifies as expertus; and the weariness he evinces ­toward the Pisones as would-be poets is, to the greatest degree pos­si­ble, muted, in large part through the trappings of amicitia, so as to avoid provoking their open ire. The Ars Poetica thus admits of a reading whereby it challenges the Pisones, or the reader more generally, to determine where Horace falls along the spectrum from flatterer to true friend and ­free speaker.114

Horatius iudex Under­lying the vari­ous exempla of true and false friends in the Ars Poetica and of artistic criticism that may be beneficial or absent is Horace’s own role as arbiter of quality—­the very “best official” (Arist-­archus), in fact. The idea of a single judge as a voice of authority is introduced at 77–78, as Horace casually remarks, “Which author, however, [first] sent forth slender elegies, the grammarians dispute and the case is still ­under [the consideration of] a judge” (quis tamen exiguos elegos emiserit auctor, / grammatici certant et adhuc sub iudice lis est). Judges (or rather, a single preeminent one) are thus presented as impartial decision makers, ones who take their time to reach a verdict, and the sense of 114. Glad 1996 outlines finding this proper place as a key prob­lem in Epicureanism, especially as practiced by Philodemus.

184  ch a p te r 3

iudex as a l­ egal figure is already cemented in its first appearance by the presence of lis, “lawsuit, dispute.”115 The next time we see the term iudex, Horace has appropriated it and positioned himself as the sole and final judge of ­whether certain characters (in this case, fauns) behave in a manner consonant with their nature and surroundings: silvis deducti caveant, me iudice, Fauni / ne velut innati triviis ac paene forenses /aut nimium teneris iuvenentur versibus umquam / aut immunda crepent ignominiosaque dicta (244–47). ­These words are, let us recall, directed at the Pisones, named at 235, as Horace subtly yet unmistakably informs them that their literary efforts, what­ever they may be, must pass his l­egal, judicial muster (me iudice). Language of law and judging—­iudex, lex, ius—­thus permeates the Ars Poetica almost from the start, building bit by bit to the appearance of Aristarchus, who acts in certain re­spects as an extension of Horace himself. “The instances accumulate,” and although at first they “do not cohere to bear a par­tic­ul­ar semantic burden,” the “wealth of more casual references to the law and politics sensitizes us to their intensification at several points.”116 The first appearance of lex is not in its plainer sense of a piece of legislation, however, but rather to denote the “law of genre,” operis lex (135),117 which serves (alongside “shame,” pudor) to deter the writer from embarking upon an unsuitable topic.118 This notion of literary law is operative throughout the Satires, as Horace speaks of transgressing the bound­aries of his genre: “­There are ­those to whom I seem too fierce in my satire, and to strain my work beyond the law” (sunt quibus in satura videar nimis acer et ultra / legem tendere opus).119 This usage of lex as genre policeman is supported by the several poems that deal with the law and laws in a more conventional sense: ­these institutions provide the narrative framework for Satires 1.7 (a courtroom argument) and 1.9 (where an acquaintance or wouldbe friend attempts to inveigle Horace into appearing at a trial),120 while at the end of Satires 2.1 mention is made of the regulations prohibiting slanderous utterances (where mala carmina, the “spells” or “slander” forbidden by the Twelve ­Tables, puns also on “bad poems,” 80–86).121 Law also appears in the Ars 115. Although as Brink 1971: 167 points out, the phrase adhuc sub iudice lis est “only sounds l­ egal.” 116. Lowrie 2014: 131–32, of Horace’s method in the Odes. 117. Orelli 1844: 727 suggests that communia (128) may also have a l­ egal sense. 118. Pedem, as often in Latin poetry (esp. Ov. Am. 1.1–4), refers to the metrical foot and thus the writer’s verses themselves; cf. AP 80, 252. 119. On Horace’s interest in and use of the law in his Satires, see further LaFleur 1981, Cloud 1989, McGinn 2001, and the numerous studies on individual poems, especially 1.7, 1.9, 2.1, and 2.5. 120. On this poem, see further Ferriss-­Hill 2011; on Sat. 1.7 and 1.9, also Gowers 2012 ad loc. 121. The poem’s final lines are dense with terminology denoting the apparatus of ancient Roman law and with puns on literary rules and regulations; see further Muecke 1993: 113–14.

a m ici , r isu m   185

Poetica as a force from which freedom may be desired: the mythologized first spectator at an ancient tragic per­for­mance, who is “done with his sacred rites and drunk” (functusque sacris et potus, 224), is described as exlex, “exempt from the law” (more literally: “de-­lawed”). The line comes, however, between the first and second appearance of Satyri in the poem, and if the idea of Horace’s Satires lurks in ­these (especially the fourth, Satyrorum scriptor, 235), then a further hint at this genre’s self-­professed generic discomfort and claims of l­ egal transgressiveness may also be discerned. Th ­ ese appearances of the term lex culminate in lines 281–84, where Horace describes, with “emphatic”122 polyptoton, the introduction of a law to curb Old Comedy: successit vetus his comoedia, non sine multa / laude; sed in vitium libertas excidit et vim / dignam lege regi. lex est accepta chorusque / turpiter obticuit sublato iure nocendi. The final appearance of lex in the poem is thus as a controlling agent imposed and enforced on lit­er­a­ture. Just as lex is intertwined with lit­er­a­ture throughout the Ars Poetica, so too is ius, albeit with dif­fer­ent emphases. Ius is one of the three ingredients alongside arbitrium (also legalistic) and norma loquendi that make up usus, the nebulous force that controls the birth and death of words and thus the flow of language (70–72), notably that as spoken by ­people rather than that controlled by poets. Ius is what Achilles denies pertains to him (iura neget sibi nata, 122), and it appears again a mere nine lines ­later (and four lines prior to an appearance of lex, 135) to denote a person’s ability to make common themes such as ­those from my­thol­ogy his own: “public material ­will be of private owner­ship” (publica materies privati iuris erit, 131).123 ­There are also the “paternal rights” (iura paterna, 256) that pertain to meter;124 the l­ awyer (consultus iuris et actor / causarum, 370–71) who offers a vision of young Piso’s f­ uture; and the marriage laws laid down by a mythical poet-­as-­civilizer (fuit haec sapientia quondam, / publica privatis secernere, sacra profanis, / concubitu prohibere vago, dare iura maritis, / oppida moliri, leges incidere ligno, 396–99). Th ­ ese instances culminate in the plea, close to the end of the poem, to allow the poet who wishes to commit suicide, represented by Empedocles, to do so: sit ius liceatque perire poetis (466). Horace thus speaks of judges and judging from early on in the Ars Poetica in such a way that we may not realize he has in mind himself as paradigmatic 122. Brink 1971: 316, who notes also the “­legal connotation” of iure nocendi (284) and the repetition of ius at 466. 123. Pseudo-­Acro: “materia iam dicta, iam ab alio edita”; Rudd 1989: 172: “publica materies is the body of mythological themes already handled by previous writers.” Brink 1971: 209 adds that “H. amusingly combines the language of Roman law with that of literary theory” (cf. Rudd), and notes (ad 131ff.) a pre­ce­dent for the sentiment from Philodemus. 124. Brink 1971: 299: “the spondee is being ‘­adopted’ into the iambic ­family, and can now enjoy the rights, iura paterna, of the iambus’ patrimonium, as though he ­were a natu­ral son.”

125-81019_FerrisHill_HoracesArsPoetica_5P.indd 185

8/8/19 8:40 AM

186  ch a p te r 3

judge ­until this self-­appointment has already been finalized. The figure of judge enters the poem as a ­legal authority and at the same time a literary one at 77– 78, its technical force underlined by the term lis that immediately follows.125 At line 244, Horace appoints himself an arbiter of generic appropriateness (me iudice) as he builds up to 263: “not just any judge sees unrhythmical poems” (non quivis videt immodulata poemata iudex). This bifurcation of the critical community into ­those who do and do not possess discernment is reinforced by the next line, in which Horace criticizes the excessive indulgence that has been shown to Roman poets (264): w ­ hether this fault was confined to past generations or extends to the con­temporary one Horace does not specify, but the suggestion lingers as potentially valid up ­until the pre­sent moment, Pisones again included. Horace’s point speaks to the specialization and professionalization of the field of writing that he is trying to impress upon his readers / the Pisones (though, as ever, the criticism implied by the fact that Horace alone is iudex is softened as he expands ­those who possess discernment in ­matters of taste at 272–73 with si modo ego et vos / scimus).126 The judge (iudicis) appears in line 314 as a ­simple example of a public ­career, but its significance is heightened by the verb iubebo, with its odd futurity, at the end of 317, which not merely aurally but also etymologically connects to and strengthens it. In this manner, and promising with iubebo that this ­will be ongoing into the ­future, Horace proclaims himself a judge in the Ars Poetica, just as he had promised a few lines before with docebo that his teachings would continue. Horace’s power is absolute: “writing nothing himself ” (nil scribens ipse), he ­will teach and evaluate the writings of ­others—­teaching and criticism having been shown to be dif­fer­ent names for essentially the same activity in this Epicurean milieu.127 The act of passing judgment on creative works, already represented in interrupted form in the poem’s opening lines, may thus be felt to gather speed as the poem progresses, and it culminates in two connected passages (361–65 and 385–90). The first shows the pro­cess of evaluating art in detailed, extended form: ut pictura poesis: erit quae, si propius stes, te capiat magis, et quaedam, si longius abstes. haec amat obscurum, volet haec sub luce videri, iudicis argutum quae non formidat acumen; haec placuit semel, haec decies repetita placebit. (361–65) 125. Striking also is the verb delitigat in line 94, a compound of lis that appears to have been coined by Horace (see chapter 1, n244). 126. Cf. Wickham 1891: 415 (Horace “compliments the Pisos by taking it for granted that their taste is as his”), Rostagni 1930: 78. 127. See chapter 2, n171 on how previous commentators have argued for one or the other.

a m ici , r isu m  187

Where the poem’s beginning showed us the artist’s completed work and the moment of its unveiling to his friends, this moment shows ­every possibility that comes next. The artist is nowhere to be found: only a solitary judge, iudex, is pre­sent, and we are given no sense of how he came to be the arbiter of quality and taste selected for this occasion. This evaluator, moreover, is the reader (stes . . . ​te . . . ​abstes),128 just as we ­were placed into the persons of the amici at the poem’s start. It seems also that he is an individual possessed of good judgment (argutum . . . ​acumen) and able to inspire dread in the sub-­par.129 Before this judge are ­imagined ­every type of painting, the variety of possibilities heightened by the deliberately lopsided ways in which Horace refers to them: he begins with mismatched quae and quaedam,130 then employs the aty­pi­cal phrase haec . . . ​haec twice.131 It has proven difficult to discern what Horace intends at 361–65, and in par­ tic­u­lar ­whether he means for one in each of the three pairs of artworks to be superior to the other.132 It seems likeliest, however, that he is presenting six separate ways in which poetry and painting may please: some give the best effect when scrutinized up close, ­others (as in Impressionism) are best enjoyed from further away; dif­fer­ent ones appear best ­under dif­fer­ent degrees of light;133 and some works give a unique plea­sure upon the first reading/viewing that cannot 128. Rudd 1989: 209 sees this usage of tu as the “ideal or generalising second person singular.” 129. That one work “does not dread” the judge’s “keen incisiveness” suggests that o­ thers do or should. 130. Rather than hic and ille, the standard way of expressing “this one . . . ​that one.” 131. While hic . . . ​hic (or haec . . . ​haec) can be used in place of hic . . . ​ille (Gildersleeve and Lodge 1895: 194, section 307 note 3), Horace’s choice of the more unusual construction nevertheless forces us to think about w ­ hether he is discussing one and the same work at dif­fer­ent moments in time or contrasting two dif­fer­ent works. 132. Porphyrio sees Horace’s equation of poetry and painting and his pairs of artworks as absent of value judgments, and Pope’s rendition of the lines (An Essay on Criticism 171–74; see further Golden 2010: 400–401) suggests that he read Horace the same way. Rudd 1989: 209, on the other hand, asserts that “in the third pair (365) the second picture is clearly meant to be superior,” but as he himself admits this pattern does not hold for the other two pairs, and Brink 1971: 370, evidently disagreeing with ­those who read Horace as not pronouncing one work better than another, says “some have commended H.’s broadmindedness.” Pseudo-­Acro allows for both possibilities—­that one artwork in each pair is meant to be better than the other or that each is pleasing and successful in its own way. Brink and Rudd 1989: 209 puzzle over at length what Horace means at 361–65 to ­little satisfactory resolution. 133. Amat obscurum is other­wise assumed to be a criticism: Rudd 1989: 209 says it “does not sound like praise,” while Orelli 1844: 780 explains it as a work that does not lack virtues but has shortcomings that may remain unnoticed ­unless thoroughly scrutinized.

188  ch a p t e r 3

be recovered or replicated l­ ater, while o­ thers give plea­sure (­whether the same plea­sure, or dif­fer­ent forms of it) with e­ very rereading.134 Perhaps we are to deduce that none of t­ hese types alone is sufficient; rather, the best work of art would be one that combines all six qualities, perhaps as the Ars Poetica itself has for some readers: the poem repays consideration on both the “atomic” and the “macroscropic” scale, as Philodemus prescribed;135 it may be enjoyed through a cursory or passive form of reading, as well as an active, deeply engaged one; and the effect it produces on the first-­time reader is irreplicable, even as it yields new and dif­fer­ent effects in subsequent readings. Lines 361–65 lead directly to the address o maior iuvenum, which itself introduces the second culminating treatment of the evaluative pro­cess where iudicium and iudex appear in consecutive lines: tu nihil invita dices faciesve Minerva; id tibi iudicium est, ea mens. si quid tamen olim scripseris, in Maeci descendat iudicis auris et patris et nostras nonumque prematur in annum membranis intus positis. delere licebit quod non edideris; nescit vox missa reverti. (385–90) Through dubbing himself a judge (as at 244, me iudice) and mentioning judges in connection with the Pisones (314; cf. 78, 263), the sense that Horace intends to pass judgment on his addressees’ literary efforts has become more per­sis­tent, and the ­matter is made explicit ­here, where he appears alongside Maecius and Piso pater, all three “honest critics,”136 or the first two perhaps not,137 to hear what­ever the elder of the youths might perhaps one day write. The image of written words entering the ears as if spoken fuses the passage to ­earlier moments in the poem—­segnius irritant animos demissa per aurem (180), tardior ut paulo graviorque veniret ad auris (255), legitimumque sonum digitis callemus et aure (274)—­all of which deal with the appreciation and evaluation of verse and ­music. 134. Contra Brink 1971: 370: few paint­ers “would agree that their pictures should ‘please only once.’ ” For Meerwaldt 1936: 155–56, who discusses in detail the ancient lit­er­a­ture on aesthetic appreciation, the distinction is between art that has an immediate and ineffable effect upon the viewer (“quae insigni quodam effectu audientium animos statim et subito capiant”) and art that is appreciated for the ­great care that has been taken over it (“quae diligentissima omnium partium cura otiose singula consideranti se commendent”), and he connects the passage to other moments in the AP. Houghton and Wyke 2009: 4, however, would seem in their paraphrase to understand ut pictura poesis as I do. 135. Cf. Armstrong’s 1993: 187 comment, cited in my chapter 1. 136. Kilpatrick 1990: 51. 137. See p. 135.

a m ici , r isu m  189

Lines 385–90 also contain Horace’s ardent wish, which has been expressed and continues to be expressed in fragmentary fashion, that his readership (including but not l­ imited to the Pisones) ­will grow to become their own best judges and worst critics. Only then ­will they succeed in writing well. While the elder of the Piso boys is perhaps flattered for an instant by Horace’s imputing of good judgment to him (id tibi iudicium est, ea mens), the certainty that he ­will do nothing without Minerva’s cooperation may prove disappointing: if she refuses to cooperate, he cannot write, yet it remains unclear w ­ hether he is aware of the status of Minerva’s attitude t­ oward him (a skeptical reader might suppose that he remains ignorantly confident of her support). And ­were they to study closely the vignette of Quintilius leading into the vir bonus et prudens and culminating in the evocatively named Aristarchus, the Pisones would see displayed ­there a failed form of student-­teacher relationship, in which they might or might not discern themselves, followed by the mutually beneficial interdependence of friendship and criticism in ideal form—­with Horace as omnipresent, omnipotent arbiter.138

From Laughter to Madness The Ars Poetica began with talk of laughter near which already lurked sick men’s dreams. The latter topic is unexpected in an Ars Poetica’s opening and when combined with laughter can have the effect of making the reader feel especially uneasy. As laughter and madness recur throughout the poem, separately or near each other, we may begin to sense, however, that far from being trivial they form an integral part of the poem’s fabric, even if we would rather they did not and even if Horace’s exact aims remain for a time opaque. The risum (5) and aegri somnia, vanae . . . ​species (7–8)139 of the beginning find themselves mirrored and reduplicated first in the vignette of Democritus, whom Horace has rendered a warden of sorts, concerned with keeping out “sane poets” from Helicon, that is, allowing in only the insane: ingenium misera quia fortunatius arte credit et excludit sanos Helicone poetas Democritus, bona pars non unguis ponere curat, 138. It has not to my knowledge been suggested that the preponderance of ­legal terminology in the AP, which I see as both an extension of Horace’s interest in the law in his Satires and as serving in the AP to establish Horace as the ultimate artistic authority, is perhaps further to be connected to the fact that any historical Pisones would likely have been members of a po­liti­cally and juridically active f­ amily: numerous leges Calpurniae are attested (149, 70, 67–63, 61, 55, and 52 BCE; Brill’s Neue-­Pauly s.v. ambitus, lex, leges, and repentundarum crimen). 139. Cf. also the species recti (25) that deceives.

190  ch a p te r 3

non barbam, secreta petit loca, balnea vitat. nanciscetur enim pretium nomenque poeta si tribus Anticyris caput insanabile numquam tonsori Licino commiserit. o ego laevus, qui purgor bilem sub verni temporis horam! non alius faceret meliora poemata. verum nil tanti est. ergo fungar vice cotis . . . (295–304) Democritus’s entry into the poem may come as a surprise, but his reputation as “the laughing phi­los­o­pher” (though Horace betrays no overt sense of this in the description above) serves to tie him into the Ars Poetica through the prominence accorded to this theme throughout. He is also a recurring character in the works of Philodemus, as Horace perhaps points once again t­ oward On Flattery and On Frank Criticism.140 The guise in which Democritus appears ­here, however, is the one familiar from Cicero (“indeed, Democritus says that no poet can be g­ reat without madness,” negat enim sine furore Democritus quemquam poetam magnum esse posse, Div. 1.80),141 but other aspects of his personage also lurk in Horace’s passage, where they serve further to nuance our understanding of risus in the Ars Poetica, and send the poem hurtling t­ oward its conclusion. The ­mental health of the individuals described neglecting their physical appearance and social niceties142 in the pursuit of art is not altogether clear, nor perhaps is it meant to be: are they genuinely not sane, relieved that at last they have been given permission to stop having to pretend to be fully functioning members of society? Or are they fakers who have ­adopted the trappings of madness out of a misguided belief that this ­w ill enable them to produce

140. The first portion of On Poems 4 deals with him at some length ( Janko 2011: 208–15, 245– 51), and Philodemus names him also in On Death (e.g., 29.28, 39.14–15; see Henry 2009: 68–69). 141. Cf. also De orat. 2.194 with Janko 2011: 214. ­W hether Democritus wholly rejected ars, however, is debated, though Horace certainly makes it seem so in his facetious portrait; see further Rudd 1989: 199–201, Ford 2002: 161–87. 142. Nails and beard stress the physical, secreta loca petit describes social interactions (where secreta reveals upper-­class Romans’ suspicion of privacy), and balnea bridges the divide, being both a locus for socializing and for cleaning one’s body. In addition, unguis at 297 reaches out to the sculptor near the Aemilian school who “presses out fingernails” (unguis / exprimet, 32–33) and to the proverb ad unguem at 292–94. By inverting the sense—­the nails of line 297 are unkempt—­Horace highlights the paradox of ­whether someone who cannot properly care for the fingernails on his own body can add the finishing touches (the proverbial ungues) to an artwork.

a m ici , r isu m  191

profound and lasting art?143 Rudd 1989: 200 says that Horace’s argument at Ars Poetica 299–301 “seems to run like this”: Democritus says true poets are mad. As a result, a lot of ­people pretend to be mad by letting their hair grow. . . . ​I [sc. Horace], however . . . ​who could be genuinely mad, am ­doing the exact opposite and taking a cure. The individual whom Horace has in mind at 297–98 would thus be “actually an un-­inspired, would-­be poet; t­ here is nothing wrong with his head.” Yet Rudd’s explanation rests, as he says, upon the words tribus Anticyris insanabile, since “if H. had seriously meant that the man’s furor poeticus was incurable he could hardly have said in the next line that he himself was taking a cure—­a cure which would change him from an insanus poeta to a sanus criticus.” This explanation, however, glosses over the import of tribus Anticyris, for Horace is not saying that the madness is wholly “incurable,” but rather that a specific and even very large144 dose of medicine w ­ ill not cure it (though, it is implied, an even larger dose might; of course, as hellebore is poisonous, it could also kill the patient). The man whose madness is tribus Anticyris insanabile is thus both suffering from a genuine affliction and from one that would require a gargantuan quantity of medicine to dispel. That Horace describes himself taking such a cure forces us to confront the uncomfortable fact that he wishes himself (or his poetic persona) to be understood as afflicted with (genuine) madness prior to this treatment, but any such discomfort is not reason enough to wish such madness away from him entirely.145 Rather, in saying that he has chosen to rid himself of it, Horace is able to boast implicitly that unlike o­ thers (for example, the Pisones) he was once divinely chosen and gifted with the trait necessary for poetic excellence. 143. Rudd 1989: 200 adduces Epist. 1.19.1–20, where “the proof of Bacchic inspiration is the ceaseless consumption of wine,” and Wickham 1891: 417 summarizes that both passages illustrate “foolish attempts to simulate inspiration by adopting peculiarities.” For Orelli 1844: 768, AP 295–97 is a perversion of Democritean thinking: “sed permulti . . . ​Democriti auctoritate abusi, ingenium atque ἐνθουσιασμόν, furorem poëticum, quem ipsi sibi tribuunt, longe anteponere solent arti.” 144. Tribus Anticyris is generally taken to mean “three times the output of Anticyra” (Rudd 1989: 201), though it could also be understood as “three doses of the stuff from Anticyra.” 145. Brink 1971: 335, for one, is concerned with preserving Horace’s sanity intact: “ego draws an unexpected and humorous conclusion: since he w ­ ill not give up his sanity and therefore cannot write (true) poetry, he proposes to teach poets rather than be one. This is crazy logic, but it allows H. to be sane and yet write (though not ‘poetry’) and it puts Democritus in his place, for his doctrine abhors teaching.” Williams 1980: 267, on the other hand, allows Horace a customary self-­mockery: “The idea that if he w ­ ere prepared to be insane he would be a better poet than any” is “unmistakable.”

192  ch a p te r 3

Indeed, this is precisely in keeping with his self-­re-­presentation in the Odes: “an infant full of spirit thanks to the gods” (non sine dis animosus infans, 3.4.20), he was covered with wreaths by doves as he slept (9–13) in an “amazing” (mirum, 13) event, and he continues to enjoy divine protection from dangers (28–36).146 As a poet touched by the gods, then, Horace would, prior to his self-­medicating, have been among ­those whom Democritus let in to Helicon, on the view that poets are born, not made.147 What emerges is “a seriously grim and au­then­tic vision of the artist’s life,” “written from inside the experience which it professes to ridicule.”148 Through his emphasis on flawed and failed artists over successful and perfect ones, sustained from the Ars Poetica’s opening through its close, Horace succeeds in bringing to the fore the painful essence of artistic endeavor; yet while pain and effort are explic­itly required for the creation of worthwhile art, Horace prefers to provide us over and over with examples of misplaced suffering that yields nothing of value. Nevertheless, ambiguity remains: are ­those who recognize the man of lines 297–98 as a poet right to do so? Or is he fooling us by feigning madness (the signal for poetic ability) where none exists? Comparison with the form of madness seen in the poem’s opening in this case brings no clarity: is the creature of the opening image proof that the poem’s first artist was truly mad (­whether with the fever dreams of one who is temporarily sick or the hallucinations of a permanently diseased mind)149 or does it rather show that, having misunderstood Democritus’s meaning, he was feigning madness out of the desire for better art? In fact, it turns out that Horace is the one who has committed something of a sleight of hand by wholly conflating ingenium with madness such that if we accept that the poet is mad, we must also accept this as proof of his ability.150 Yet the two are certainly not the same, though they may be thought of as overlapping somewhat,151 and to represent native talent entirely as madness gives some indication of Horace’s position on the longstanding debate in 146. See further Graziosi 2009: 147–56 on the literary tropes of Carm. 3.4. 147. Cf. Pseudo-­Acro: “poetam putat non arte fieri sed natura nasci”; Porphyrio: “ait enim Democritus, poeticen natura magis quam arte constare.” 148. Golden 2010: 411, Brink 1971: 516, respectively. Brink adds, “This cautionary story could never have been written without a generous mea­sure of the quality so caricatured.” Golden 412–13, following Oliensis 1998: 219, elaborates further that in the AP’s final lines is described “the psychic pain that stalks the artistic temperament forced into confrontation with troubling issues of life and death in a vocation that permits no evasions or escapes.” 149. E.g., Kiessling and Heinze 1914: 290, Brink 1971: 90 understands aegri somnia to be the former. 150. Cf. Rudd 1989: 200: “H. mischievously confuses furor poeticus with insanity.” 151. Rudd 2009: 249 puts it well: “In the Ars Poetica inspiration is associated not only with madness; it also belongs to ingenium.”

a m ici , r isu m  193

ancient thought between ingenium and ars,152 which we should not be surprised to see h­ ere ­under its official terminology since it has been gestured at already through other means. Foreshadowing its appearance at 295, the term ars has occurred repeatedly in the Ars Poetica, for example at 31, 214, 262 (especially intriguing if Horace is responsible for the title, since ars means also “handbook”), in senses slightly distinct from though overlapping with ­those that make it the complement to ingenium.153 Ingenium appears nowhere prior to 295 (though several times afterward), yet the idea of something “in-­born” and of birth itself have been pre­sent throughout the first two thirds of the poem. As often, however, it is only through further reduplication of the theme that we ­w ill learn where Horace stands. The short phrases of lines 301–4 read as spoken words, but are t­ hese Horace’s own words and, if so, what is their tone?154 The cry o ego laevus, / qui purgor bilem sub verni temporis horam!, where ego appears emphatically and suddenly, well encapsulates Horace’s dilemma: he purges himself of (black) bile, that is, removes from his body the substance and its accompanying state (μελαγχολία, melancholy) held responsible for artistic genius.155 While this is better for his (­mental) health, it leaves him unable to make the best poetry, and thus laevus, “unlucky, unfortunate,” but also “deluded, perverse.”156 Heightening his

152. Already evident in the poetry of Pindar (cf. Orelli 1844: 768) and in Old Comedy (see O’­Sullivan 1992, Ford 2002: 165–72); see also Hardison and Golden 1995: 31, Fedeli 1997: 1595– 96, Antolín 2002: 225. Many have sought to pinpoint Horace’s own position in the debate: Newman 1967: 57 traces what he sees as a change over the course of his writings (from an initial insistence on the importance of ingenium over ars, Sat. 1.4.43, to a l­ ater emphasis on ars and an “ironical” treatment of ingenium, Sat. 1.10.63, AP 295–97); Armstrong 1993: 209 concludes, “clearly he puts a much higher value on ars than on ingenium”; while Feeney 1993: 45 says, “Horace wanted, as the dichotomy of the day had it, both ingenium and ars (naturally inspired genius and perfection of technique),” as, too, Tsakiropoulou-­Summers 1995: 197–210. 153. The TLL entry on ars runs to twenty pages, which gives a sense of the word’s scope. 154. Lines 299–301 might even be punctuated with quotation marks to make them the dim-­ witted interjection of another voice, as in lines 9–10: to this interlocutor’s statement that “indeed, he ­will obtain the prize and the name ‘poet’ if he never entrusts his head, incurable with three doses of hellebore, to the barber Licinus,” Horace would then throw up his hands in despair and exclaim, “O, I am unlucky, I who purge myself of my bile just before the season of spring time!” 155. In Greco-­Roman medical understanding, bodily and m ­ ental health depended on the coexistence of the four humors—­blood, phlegm, yellow bile, and black bile—in the correct proportions (Brink 1971: 332–34, Rudd 1989: 201). 156. See Brink 1971: 333.

194  ch a p te r 3

frustration, he does so right before the optimal time for artistic creation.157 Although Horace elsewhere and ­later in the poem prescribes sweat and pain as essential to the creation of art, ­here his discomfort and distress are all in vain. Wistful about what could have been (non alius faceret meliora poemata),158 Horace nevertheless unequivocally rejects madness: verum / nil tanti est.159 In rejecting writing (nil scribens ipse), Horace embraces instead the role of one who molds other artists, rather than crafting his own objets d’art.160 This leaves rather a prob­lem, for as much as he disavows ­here and ­there that his writings qualify as poems, Horace manifestly is a writer (we are reading his lines and his poem, ­after all),161 so in what ways can lines 301–4 be considered valid? The phrase, I suggest, refers back to the Odes, not in the s­ imple sense that hexameters do not count as writing while lyr­ics do,162 but rather to allow a nostalgically retrospective reading: if the younger Horatian persona of the Odes was divinely inspired, the older persona of the Ars Poetica has grown to find all that rather exhausting and, having now renounced his former (persona’s) madness, is forced also to acknowledge that he no longer writes that way. 157. Rudd 1989: 201 explains that “according to the physician Celsus . . . ​the best time for the treatment” with hellebore was as spring approached (so, too, Porphyrio, Wickham 1891: 418; Orelli 1844: 769 rightly explains verni temporis horam as meaning the season of spring, where horam denotes “anni tempus” as at Carm. 3.13.9), but if springtime is thought of also as the time for rebirth and thus for resuming creative activities (compare Carm. 4.7, Diffugere nives), Horace’s frustration may be heightened by his abjuration of poetry just before this optimal moment. 158. The force of the imperfect subjunctive faceret as contrafactual is worth noting, i.e., it refers to an action in the past that could have taken place but did not (Gildersleeve and Lodge 1895: 171, section 258); see Brink 1971: 334, Rudd 1989: 201. 159. Wickham 1891: 418 explains the sense of nil tanti est rather differently as “it is nothing of importance” and Brink 1971: 335 concurs (he views it as “not a probable Horatian sentiment” to understand Horace as saying that being an inspired poet “is not by any means worth it,” with nil “a strong negative”). This seems rather weak in the context, however, and Cic. Att. 2.13 (iuratus tibi possum dicere nihil esse tanti), the comparandum adduced by Wickham and Brink, admits of being understood as both “it is nothing of importance” and “nothing is of such g­ reat importance”; Hajdu 2014: 88 also f­ avors the latter with his “nothing is worth it,” as do Landinus (at Bugada 2012: 124), Rostagni 1930: 88 (“non ne vale affatto la pena”), and Antolín 2002: 213 (“no vale la pena”). 160. Cf. Hajdu 2014: 88, who glosses Horace’s ironic stance as, “if as a not-­mad person I cannot create poetry I ­will teach ­others to do it.” 161. Oberhelman and Armstrong 1995: 240 persuasively argue that, for all of Horace’s faux-­ humility, “he is a poet, he is writing poetry,” and Rutherford 2007: 261, n32 likewise considers “the disclaimer . . . ​to be clearly disingenuous,” pointing out the paradox of Horace disavowing his status as a poet within a poem (so, too, Hajdu 2014: 87). 162. See introduction, nn85–86 for the long-­standing interpretation of nil scribens ipse in precisely ­these terms.

a m ici , r isu m  195

Horace appears to leave aside the ­matter of ars versus ingenium with his declaration ergo fungar vice cotis, but it lingers nonetheless: only someone who sees the value of ars, “skill,” would propose becoming a teacher, on the implicit understanding that inborn talent alone w ­ ill not suffice; training, too, is needed. This is mirrored by the emphasis on learning, didicit, that follows (312–16) as essential to the success of the dramatic author. A play or story (fa­bula) that lacks charm (Veneris, 320; as at 42) but is attractive in spots (speciosa locis, 319) and has proper morals (morataque recte, 319)163 and heft (pondere inerti, 320)164 ­will be preferred by the ­people over “verses poor in resources” (versus inopes rerum, 322) and “tuneful trifles” (nugaeque canorae). It is not obvious why Horace should be right on this count—it would be quite easy to see how “tuneful trifles” give more plea­sure than a turgid, moralizing work—­but we have l­ittle choice except to take his word for it.165 When Horace next begins with Grais ingenium, Grais dedit ore rotundo / Musa loqui (323–24), then, the reappearance of ingenium is to have been expected, as the intervening lines since its last mention have continued to deal covertly with the proper balance of ingenium and ars. In contrast to the inspired Greeks, Roman boys study: the former are given (dedit) ingenium, the latter learn (discunt, 326), the verb describing what goes into ars just as its active counterpart, docebo, does at 306. The debate reappears formally at 408,166 ­after several discussions of the importance of training and practice (354–56, 379–82), care and effort (358–60, 385–90). Having stated the prob­lem, natura fieret laudabile carmen an arte / 163. The perfect passive participle mōrāta, “having been mannered, having been given manners,” formed from the noun mōs, “custom, habit” (seen at 317: exemplar vitae morumque), is an uncommon word (see TLL s.v. mōrātus), and the appearance of the far more usual and similar-­sounding morātur, from the verb moror, “delay,” just two lines ­later serves to draw further attention to it. 164. Sed pondere inerti, Shackleton Bailey’s 1982: 103, 2001 emendation for the manuscripts’ sine pondere et arte, is perhaps too radical (especially as the previous line provides the redemptive qualities of the poem that justify valdius oblectat in the next), though it yields good sense. Sine pondere et arte would mean “lightweight” (Brink 1971: 345–46; cf. Pseudo-­Acro, Wickham 1891: 419), i.e., the opposite of Shackleton Bailey’s sense, though nevertheless equally “artless.” 165. Janko 2000: 190–91 (cf. Armstrong 1995: 218) sees Horace, like Philodemus, “persuaded that worthwhile content in poor verse was better than the versus inopes rerum nugaeque canorae of the euphonists,” who held that sound alone mattered. The difficulty at AP 322 lies in the fact that Horace does not usually impute good judgment to the masses, so if he was in agreement with Philodemus that content mattered over sound and appearance, it is not clear why the common mass of p­ eople should also have discerned this (Brink 1971: 346 suggests, “­Here the populus have a very partial view of the Horatian truth”). 166. Cf. Brink 1971: 394: “Before he comes to this section the reader has been exposed to varying answers to the apparently ­simple prob­lem, ‘natura an arte.’ ”

196  ch a p t e r 3

quaesitum est, Horace proffers his own understanding of the m ­ atter, ego nec studium sine divite vena / nec rude quid prosit video ingenium; alterius sic / altera poscit opem res et coniurat amice (409–11), before detailing what ars and studium look like: qui studet optatam cursu contingere metam167 multa tulit fecitque puer, sudavit et alsit, abstinuit Venere et vino; qui Pythia168 cantat tibicen, didicit prius extimuitque magistrum. (412–15) Brink 1971: 394 says of natura and ars, which, coming first and last in their line (408), embrace within them the “praiseworthy poem” (laudabile carmen), that “this duality underlies the w ­ hole poem.” ­These themes are pre­sent the length of the Ars Poetica and critical to it, where they are also intertwined with other key themes that they serve to elevate rather than exclude. At 408–15 they serve to bring friendship to the fore, and friendship in turn ensures their mutual cooperation: coniurat amice. Horace has at long last spelled out his view on the debate between ingenium and ars: neither on its own can do any good (ego nec studium sine divite vena / nec rude quid prosit video ingenium), and they not only require each other (alterius sic / altera poscit opem res) but must be on friendly terms, as suggested already at 295 by the way in which ingenium and arte frame the line.169 ­There is the slightest hint that t­ hese two ingredients, the vast multitudes each contains hinted at by the numerous terms Horace has used for them throughout the Ars Poetica,170 if combined in the proper portions (what­ever

167. See Brink 1971: 397–98 and Rudd 1989: 217 on w ­ hether meta denotes the turning-­or finishing-­post, as well as on the nature of the race in question. 168. Pythia has been understood to mean e­ ither “at the Pythian games” (Orelli 1844: 788: “certamina”; Brink 1971: 289: “a per­for­mance at one of the games”) or “the Pythian songs” (Pseudo-­Acro, Rostagni 1930: 119, Rudd 1989: 218, Antolín 2002: 225; Brink rejects this: “not Pythian songs”); the interpretation depends somewhat on ­whether cantat or certat is read (see preface). 169. Cf. Brink 1971: 330: “the carefully balanced order of words—­a symbol perhaps of ars rather than ingenium.” 170. Ingenium and ars at 295 on the quarrel’s first appearance (evidently to frame the issue in a familiar and clear way), with natura ­later taking the place of ingenium (408) and studium (standing for ars) juxtaposed with divite vena (representing ingenium, which appears momentarily as rude ingenium). The possibilities that emerge include the idea that ingenium, “inborn talent,” is granted by Nature and is also a component of h­ uman nature, though on its own it is coarse (rude), like a newly discovered mineral deposit (Orelli 1844: 787, Brink 1971: 395–96), while ars is training, effort, and the application of oneself to a task (studium)—­“grit” in current educational parlance.

a m ici , r isu m  197

t­ hose may be), ­will give rise to a final artistic product that is greater than merely the sum of its parts. More in­ter­est­ing than the Horatian persona’s position on this issue is why he tells the Pisones this and why he does so at this juncture. I see the presence of amice, the first of the four instances of the term clustered together in lines 411–36, as key for elucidating this. In speaking overtly about the friendly collaboration of ingenium and ars, Horace succeeds in inserting into the discussion the presence of himself and his relation to the Pisones. If they respond appropriately to his teaching and prompting and molding, it is implied, all of which is facilitated by the framework of amicitia within which he has located them and himself, they may succeed in achieving the ideal balance of ingenium and ars. On a positive reading, in which we impute ingenium to them, Horace is offering the Pisones the possibility of literary success. If we begin from the premise that they lack this essential ingredient, however, then no amount of training (ars—or reading of an Ars Poetica) ­will help. It is pos­si­ble, moreover, that they are not willing to do what is necessary u­ nder the dictates of ars, which requires sacrifice, suffering, and, intriguingly, fearing their teacher: multa tulit fecitque puer, sudavit et alsit, / abstinuit Venere et vino; qui Pythia cantat / tibicen, didicit prius extimuitque magistrum. They may even be like the arrogant speaker of lines 416–18, who with misplaced confidence declares, “I make amazing poems,” dismissing ­others in an infantile manner with occupet extremum scabies.

­Toward the End of the Poem Derisum (452), the final appearance of laughter in the poem, coming ­after the fourfold appearance of friendship (411–36), leads directly to madness. At the poem’s outset we witnessed the arrival of critic-­friends; at its close, a single audience member (certainly not a friend, yet the label critic does not fit ­either) is all that ­will be left ­after all ­others have been chased away (indoctum doctumque fugat recitator acerbus, 474), their flight reminiscent of other closural moments in Horace.171 While considering the Ars Poetica overall alongside Horace’s Satires and even his Odes, we should also be attuned to its internal reciprocity, 171. Sat. 2.8.93–95, for example. For Segura Ramos 1989: 115, the point of comparison is Verg. G. 4. Another comparandum may be found in Lucretius, where the “last action described in the De rerum natura is that of the mourners returning to their homes (redibant)” (P. Fowler 1997: 114, who notes the prevalence of this closural motif in Homer). The plague that concludes the DRN, moreover, which has caused readers discomfort and led some to suggest that the poem is unfinished, also mirrors the ending of the AP through their shared grotesque unpleasantness.

125-81019_FerrisHill_HoracesArsPoetica_5P.indd 197

8/8/19 8:40 AM

198  ch a p t e r 3

reading its ending against its beginning, not so much as an example of ring composition by themes, but b­ ecause it continues what was left incomplete. The poem’s opening vignette is of arbitrium interruptum: we see the completed artwork, we are told that the painter has invited his friends in for a viewing, and we are whisked away at the exact moment that they have laid eyes on the work and are about to react. Horace asks the paint­er’s friends (us, it turns out) w ­ hether we would be able to hold back our laughter, yet we are not permitted and privileged to find out what happens next. Th ­ ese strands of friendship and laughter wend their way in unison throughout the poem, overlapping consistently with the idea of criticism. As the Ars Poetica draws to its close, we are granted a final “­after” to complete the “before.” The artist, it turns out, is a vampirical creature: ravenous for attention, he consumes and destroys the one person and body he gets hold of (all o­ thers have fled: indoctum doctumque fugat, 474),172 and if the point is that Horace has given up poetry to become a critic and teacher, “the final image of the poem is an illustration of the futility of the critic’s work.”173 The irony, of course, is that in killing the single recipient of his art, the creator is left with no reader-­spectator at all. This leech glutted with blood may be satisfied for a moment, as s/he releases her/his victim, but we are left suspecting that the ­whole gruesome episode ­will be repeated in perpetuity. Although Horace has in lines 408–11 gotten as close as he ever w ­ ill to both articulating the prob­lem at hand regarding the tension between divine inspiration and effort in Greco-­Roman thought and venturing an answer, he ­will end not by engaging with this thorny m ­ atter head-on but rather by showing what divine inspiration taken ad extremum would look like.174 The poem’s ending forms a contrast with the rest of the Ars Poetica, where Horace repeatedly intimates that the Pisones, and all Piso-­like readers, should contemplate w ­ hether they might lack the inborn talent (ingenium) necessary for writing—­nowhere 172. The echo of Lucil. fr. 591–93 ( / nec doctissimis ; Manium / Persium haec legere nolo, Iunium Congum volo) suggests that the reciter has chased away precisely the sort of audience that Lucilius has said he does not desire. 173. So Hardison and Golden 1995: 82. 174. Golden 2010: 410, noting that the poem’s final section “has come to be known as ‘the caricature of the mad poet,’ ” identifies Brink as “the most authoritative proponent” of this interpretation, and Citroni 2016: 232–33 likewise explains that “for Horace, a sober, rigorous artist, the figure of the mad demoniac poet, devoid of any art, but who believes himself to be ­great and wants to impose his presumed genius on every­one, making himself unbearable and ridicu­lous, is the surest negative reference point, and the mad poet has the privilege of seeing the ­grand finale of the Ars poetica (453–76) dedicated to him.” I would caution, however, that it is not so obvious that the mad poet is, in fact, not a genius, and therein lies the prob­lem: ­every prospective artist who is not mad is left wondering ­whether he could improve his art through madness (­whether induced or natu­ral).

a m ici , r isu m  199

has it been suggested previously that a prospective poet might have an excess of genuine ingenium (again represented by madness) combined with no capacity at all for ars. In addition, Horace now and most unusually depicts inspiration as something awful inflicted by divinities upon p­ eople against their w ­ ill. In the world of the Ars Poetica, a positive form of divine inspiration—­the one widespread in Greek culture, where the poet acts as ­little more than a vessel to or through which the all-­remembering, reminding Muse speaks175—is an apparent impossibility: ­there is only ingenium/madness inflicted upon an unfortunate in retribution for a violation of divine law, from which he emerges a poet (though even then not a good one).176 It is only once we have read the ­whole of the Ars Poetica that we realize that the question posed in line 5, spectatum admissi risum teneatis, amici?, was extraordinarily complex. If we thought that Horace was asking about laughter (­whether we would or would not be able to stifle it when faced with a painter friend’s absurdly flawed “masterpiece”), we now understand the moment to have been a test of friendship: true friends, we should laugh, even at the risk of causing pain and when (false) praise would come to us more easily,177 and the painter should improve his work accordingly (this onus is of course upon him), and value his friends for having saved him from greater humiliation at a ­later date. ­These mechanisms of Epicurean friendship, explored also by Philodemus in his prose works as in his poetry, provide for the Pisones a guide for reading the Ars Poetica and for understanding Horace’s attitude ­toward them. A failure of friendship and laughter, as we ­will see in chapter 4, results in bad art, and with it not merely humiliation for the artist but even death for the audience. Yet amid this grimness and degradation, as I show in the next chapter, Horace locates the sublime, and so concludes his Ars Poetica by showcasing his own artistic superiority, which neither the Pisones nor any reader can hope to match.

175. On the conventional ancient association of Musa and meminisse/memorare, see O’Hara 1996 and Horsfall 2000: 424. 176. Brink 1971: 429 speaks of Horace’s satirical logic whereby “the mad poet’s verses are regarded as a consequence of divine punishment. The only question is why he is thus punished.” 177. The “Materia” commentary emphasizes the ease with which enemies ridicule and friends lie: “non loquor de inimicis qui etiam benefacta inimicorum facile derident, sed de amicis qui etiam malefacta amicorum facile laudant” (Friis-­Jensen 1990: 339).

4 The End of the Poem

if readers have found the beginning of the Ars Poetica frustrating and perplexing, they have said the same of its ending—­and with far greater vehemence. The closing sections give the appearance of being unfinished (as indeed they—­ and the ­whole poem—­have been accused of being),1 and the ­whole declines to come to a conclusion that gives at least the superficial impression of being satisfying. Readers have objected, for example, to the “jingle” of in mala, the first words of line 452, and ut mala, the first in 453, regarding it as accidental and an error, rather than as serving a “stapling” function, attaching the poem’s final section to what came immediately before.2 The adjective acerbus applied to the reciter (474), in fact, encompasses the ideas that the poet and his work are not ready for per­for­mance (“unripe,” “in an unfinished state”), that he is “pitiless, cruel, harsh” in the way he inflicts his work upon ­others, and that the work’s effect on his audience is “painful to the senses, grievous,” “distressing,” and “disagreeable.”3 The feeling engendered by reading the end of the Ars Poetica ­after exposure to its first 294 lines is not unlike reading Horace’s Epodes or Satires when one has grown accustomed to his Odes: readers long recoiled, 1. See Wilkins 1896: xvi–­xvii: “The view, which till recently has been the most generally accepted . . . ​regards the Ars Poetica as unfinished, and not published by Horace himself.” This may be found expressed as late as Branford and Pope 1953: 85: “He may perhaps have left it unfinished, or imperfectly revised, at his death.” 2. Brink 1971: 422 says, “The jingle . . . ​may or may not be intentional,” while Rudd 1989: 224 challenges “anyone who thinks the jingle intentional . . . ​to say what the intention may be.” It is, it should be noted, not a ­matter of mere repetition: mala at 452 is a neuter plural accusative substantivized adjective acting as a noun (“evils”) while at 453 it is a feminine singular nominative adjective. Similar “stapling” effects with variatio may be seen in tollere curet. / si curet (460–61), where the “changed metrical position” (Brink 1971: 426) ­causes the ictus to fall first on the first syllable (CU-­ret) and next on the second (cu-­RET), or in ordo . . . ​ordinis (41–42). The extent and effects of such “positional patterns” in Latin poetry generally are analyzed by ­Wills 1996. 3. OLD. Oliensis 1998: 215, moreover, understands the recitator acerbus to be Horace himself. 200

The E n d of t he P oe m   201

branding ­those Horatian poems that did not suit their taste the embarrassing experiments of an immature poet or dismissing them as incomplete. The manifest “turn” that has been seen taking place beginning with line 295, as Horace moves from the topic of ars to that of artifex,4 recalls ­those that characterize many of Horace’s Odes,5 though it is no more comforting for this faint resemblance. Yet, despite the jarring effect (and intentionally so) that the end of the Ars Poetica has upon the reader, it serves nevertheless as a surprisingly harmonious complement to its opening and indeed succeeds in bringing the poem as a ­whole to a conclusion that highlights its most salient qualities. The themes of ­human be­hav­ior, friendship and laughter, and the poem’s addressees are brought together once more in a way that solidifies their central roles in the poem, as Horace points also t­ oward his Satires and ­toward the idea that all forms of making and ­doing are in essence the same. The body of the Ars Poetica had emerged, Phoenix-­like, from an unpromising beginning, where a phantasmagorical image was likened to a sick man’s dreams, and it ends in parallel grotesquerie with the belching of verses and a mad poet falling down a well (457–63), the tale of Empedocles throwing himself into the caldera of Mount Etna (463–66),6 acts of profound sacrilege7 (including urinating on a f­ ather’s ashes, 470–72), and an escaped bear (simile for a poet) that hugs a man u­ ntil it is sated with his blood, becoming somehow a leech in the pro­cess (472–76). ­These grotesqueries, which rework ideas from ­earlier in the poem (the need for care and effort in art, the need for harmony and appropriateness, the need to avoid r­ unning to extremes and to avoid committing one error in attempting to avoid its opposite), invite us to see the poem’s opening vignette reflected in its final sections. The bear and the leech appear as a simile and meta­phor, respectively, for a person who wishes to inflict a recitation of his poetry on ­others, much as the painter of lines 1–5 had done to his amici. Crazed with the desire for even a single audience member, this 4. See especially Brink 1971: 325–29 for an overview, and my introduction, with nn142–43. 5. As Harrison 2004: 82 puts it, “Horace’s Odes sometimes employ a ‘turn’ in the centre of a poem, which serves to give additional structural variation to a monostrophic poem by providing a fulcrum in the m ­ iddle at which the poem in some sense diverts or turns away from its initial course.” 6. Empedocles’s suicide in the AP may be understood in terms of Epicurean views on death and the proper time for it, whereby suicide was not forbidden but discouraged (Armstrong 2004: 292); see further Philodemus’s On Death with Henry’s 2009 introduction. Empedocles’s sudden appearance in the AP is naturally to be connected with the didactic tradition: Obbink 1993: 52 dubs him “the avatar of didactic poets,” on the model of which Lucretius was honorifically titled Empedocles redivivus (Cic. Q Fr. 2.9.4 with Obbink 1993: 51 and n3). 7. Disturbing a bidental, a spot that had been struck by lightning and was therefore considered unable to be expiated (Pseudo-­Acro), was not permitted; see further Antolín 2002: 230–31.

202  ch a p te r 4

recitator acerbus is said to be “like a bear” (velut ursus, 472, a simile) that is strong enough to break the bars of his cage, and once he grabs hold of someone, he becomes a leech (meta­phor)8 who w ­ ill not let go of his victim ­until he is “full of blood” (plena cruoris, 476)9—­a quite evocative portrait of an artist in need of an audience that reveals the ­human ele­ment under­lying creativity in a way far clearer and more compelling than an image of perfection would permit. The unsettling effect of ­these final lines of the poem comes not merely from the two unexpected and disturbing animal images, but also from the way Horace has moved from one to the other, for they are in no way naturally compatible (and neither is their syntax, as the switch from simile to meta­phor jars, too). In fact, Horace seems to be picking ele­ments from the animal world at random, as did the painter in the poem’s opening vignette. In place of a single Franken-­creature constructed from a variety of mismatched parts, however, Horace ends by being unable to decide what life-­form his reciter is—­whether ­human, bear, or leech—­and the shifts among ­these possibilities are unsettling, all the more so for not being clearly defined. Individually disconcerting, the images become even more unappealing to contemplate when cast before us in jumbled succession, and we may find ourselves wondering w ­ hether we have become, once again, the amici of an artist. Addressed as such initially and told that laughter ­will be our likely reaction, we are now left on our own to identify that we are viewers both of the reciter’s scene and of Horace’s poem as a ­whole and to determine ­whether our reaction w ­ ill be horror or laughter. The ending of the Ars Poetica thus evokes its beginning through their shared futurity, as with risum teneatis, amici? (5) and missura (476) Horace leaves the outcome of both situations up to each reader’s imagination.10

8. Brink 1971: 431: “no longer a comparison—­velut ursus—­but an apposition”; he notes the “startling suddenness” of the image a­ fter the e­ arlier simile. On the often-­noted absence of any overt discussion of meta­phor in the AP, see Williams 1980: 280 (who objects to the characterization), Oliensis 1998: 221–22 (she notes that Horace nevertheless uses meta­phor to showcase his own poetic superiority), Innes 2003. 9. Cruor denotes “blood (fresh or clotted) from a wound,” often released through force, as opposed to sanguis, “blood (in its normal fluid state)” (OLD). Horace’s point, then, is not merely that the leech is full of blood, but that this blood is in a place where it does not belong, and that the pro­cess has involved vio­lence to the integrity of the h­ uman body. Cruor may perhaps also be seen ratcheting up Lucretius’s sanguine from the final lines of DRN (6.1285–86), where “bodies are left b­ ehind” (corpora desererentur) not unlike the way that Horace’s reciter leaves a sack of skin. 10. Also parallel is the delay of si in its own clause: at 1–2 it is the seventh word, at 474 the fourth (cf. Brink 1971: 429). Oliensis 1998: 216 is right to point out that whereas in the poem’s beginning our response to the artwork we are viewing is aided by that of the amici standing by, at the end

The E n d of t he P oe m   203

The illustrations of inconsistency-­become-­madness and of the creative pro­ cess gone awry are supported by a scaffolding of resonances on the levels of individual images and words as well. The poem’s opening is explic­itly colorful (varias, 2; atrum, 3), the ending implicitly so (blood, blackbirds, and a fiery volcano). Diana, described as “angry” (iracunda, 454),11 and the leech with its grammatically feminine gender echo the female head of the opening creature, revisited also in the interim with the mention of Diana in line 16 and the numerous other alarming female personages that surface throughout the poem, including Medea (twice), Scylla, and Lamia (­these coexist alongside more benign female entities such as Natura and Musa).12 Animals are pre­sent in both scenes, first as a single, poorly assembled fantastical being, and ­later as separate living t­ hings (man, bear, leech) that inexplicably blend into and become one another. In Horace’s hexameters animal imagery is often Aesopic, yet this aspect would seem to be wholly missing from the Ars Poetica.13 In the beginning parts w ­ ere forced together (albeit not with g­ reat success), but as the poem ends, the parts ­will not even hang together, and the result is a landscape of wreckage and carnage. Amid t­ hese fragments, which are distributed all about as if a­ fter an explosion, is the ooze of bodily pro­cesses and fluids: foreshadowed by Horace’s own purgor bilem (302) come a poet who burps (ructatur, 457),14 a son who urinates (minxerit, 471), and a leech that is full of blood (cruor, 476) and consequently left clutching a depleted sack of h­ uman skin (cutem) as she/ he (feminine leech cum male reciter) kills what she/he desires through consuming it wholly.15 Fi­nally, it is unlikely to be coincidental from a poet so attuned to sound, especially in this poem, that the first letters of the poem’s last two “we are left to make of this monstrous design what we can, without the assistance of the author, who drops off just when we may feel we need him most.” 11. The adjective is traditionally explained through Diana’s association with the moon and therefore with ‘lunacy’ (Pseudo-­Acro, Orelli 1844: 795, Rudd 1989: 225). Ira/iratus occurs a notable number of times throughout the AP (106, 109, 121, 159, 197), however, suggesting that Horace perhaps had in mind Philodemus’s On Anger (Indelli 2004: 106 sees the same in the Aeneid). 12. Lamia, with her ass’s feet (Antolín 2002: 217) and w ­ oman’s body (cf. Rostagni 1930: 98 “parte donna, parte bestia”), also echoes the opening creature’s hybridity as well as her femininity. 13. With the pos­si­ble exception of the corrupt phrase numquam te fallant †animi sub vulpe† latentes (437); see chapter 3, n16. 14. Although Brink 1971: 424 calls Wilkins’s 1896: 417 claim that the verb was no longer felt to be coarse “not implausible,” the context, i.e., its juxtaposition with other similarly unpleasant verbs, suggests other­wise. 15. The significance of cutem and of the image that we are invited to fill in (i.e., what the appearance of this skin is once the leech is full of blood) is foreshadowed by scabies (453; its own

204  ch a p te r 4

words, cruoris hirudo, chiastically mirror the first letters of the poem’s opening two words, Humano capiti, all four words furthermore balanced by their three syllables.16 This remarkable aural ring composition finds its counterpart in the imagery it is employed to describe, as the poem is bookended by, on the one hand, a human-­headed, horse-­necked, fish-­tailed beast with assorted limbs, and on the other a ­human reciter, figured as a bear who mauls and turns out to be a leech.17 The primary fault illustrated in the closing lines of the poem is, as in its opening, the failure to avoid extremes. This failure takes absurdly literal form: the mad poet, even as he burps (ructatur), is sublimis, “sublime” (457), yet the aesthetic sense of the term is complicated as its physical sense of elevation takes pre­ce­dence when the poet plummets into a well or pit (in puteum foveamve, 459). The mention of not one form of hole but two emphasizes the deepness and the drop, as does the way in which the fallen poet must be rescued: by letting down (demittere, 461) a rope. Perhaps he threw himself down ­there on purpose, Horace won­ders aloud, again drawing attention with proiecerit to the vertical distance travelled. A second anecdote retreads the same territory, though now the possibility of rescue is utterly absent—­rescue from a ditch or well is pos­si­ble,18 but rescue from the bottom of the volcano into which Empedocles has jumped (insiluit, 466) is not. A further form of contrast is introduced by the juxtaposition of the volcano’s heat with the poet’s coldness, the former’s adjective and noun embracing the latter: ardentem frigidus Aetnam (465).19 What­ever importance emphasized by its repetition from 417), which refers to vari­ous external skin afflictions (OLD, Brink 1971: 422). 16. Oliensis 1998: 216 says, “humano to hirudo, its mocking echo.” 17. Observed also by Brink 1971: 421 (“the two poetic pictures are related to each other and form as it w ­ ere a frame for the poem”), Oliensis 1998: 215 (“although the oppositions that structure the opening recur, the stakes have been raised, and the pressure increased”), Laird 2007: 137 (“subtly themed ring-­composition” involving “rapid conjunction of ideas and images involving sickness and hybridization of the ­human with the bestial”). 18. The poet’s cries for help make use of the Roman custom that required a citizen to help a fellow citizen in distress (Brink 1971: 423, 425). The cry of cives also conveys a wish to rejoin civil society (rejected also by the unwashed, reclusive poet of 297–301), though as Brink points out, “no one ­will help, the madman has put himself outside the community.” 19. Brink 1965: 138 (cf. 1971: 427) rightly observed that “to make poetic sense heat and coldness must be related to the situation,” perhaps in the form of a reference to Empedoclean theory (while no fragment directly supporting this can be identified, it is clear that “opposites ­were impor­tant in Empedocles’s system,” Rudd 1989: 226). Rejecting the usual explanations, Brink summarizes them as follows: “Empedocles felt chilly and went to get warm by jumping into Mt Etna,” perhaps taking “the plunge ‘cold with fear’ or, conversely, ‘in cold blood’ ”; “old ­people have a sluggish blood-­circulation and therefore a low temperature”; or “coldness of blood near the

The E n d of t he P oe m   205

e­ lse Horace intends to communicate (and as usual, it is a multitude of t­ hings at once), the dangers of ­running to extremes are illustrated to absurd and unpleasant lengths.20 The dizzying descents from high to low, moreover, may be discerned also in the opening image of the Ars Poetica, which similarly moves from high (caput, superne) to low (fish tail), such that this form of motion—­ from opening head, to closing well, volcanic abyss, and death—is made to run the length of the w ­ hole poem as well. An autumnal tone with death as its climax pervades the ending of the Ars Poetica—­a form of ultimate punctuation for a work, as it is for life. Armstrong 2004: 292 describes how Philodemus, in “the showiest” of usages, employs a “daring self-­referential closure” to his treatise On Death, speaking of death “as the ‘period’ and ‘paragraph’ of life in a concluding periodic sentence u­ nder which paragraphos itself is written in the papyrus as a close.” He concludes that Horace had “read this lecture and liked it,” as Epistles 1.16 closes with the words mors ultima linea rerum est (“death is the last finishing-­line of ­things”) in “a reference to the favorite Philodemean meta­phor of the paragraphē and corōnis of life.” I would suggest that Horace is ­doing the same in his Ars Poetica and raising Philodemus’s stakes by ending not merely with death but suicide and double murder (467, 475). If we reflect upon such passages as 60–72 and 158–76, however, the overwhelming presence of death should not come altogether as a surprise, for it has been a constant undercurrent of the poem, even if presented previously in reassuring and redemptive form as a natu­ral and immutable pro­ cess that contains within it the possibility of rebirth. Leaves fall and words die out, yet o­ thers are born and flourish (60–62); death is unavoidable (debemur morti nos nostraque, 63; mortalia facta peribunt, 68), yet “many t­ hings w ­ ill be reborn which have already fallen, and the words that are now [held] in honor ­will fall” (multa renascentur quae iam cecidere, cadentque / quae nunc sunt in honore vocabula, 70–71). Similarly relentless in its ever-­moving cyclicality is the course of a h­ uman life that sees a boy become a young man, then a mature man, and fi­nally an old one, the image ending with the ebb and flow of years: anni venientes . . . ​recedentes (175–76). We seem to be encouraged to fill in the gaps in the Ars Poetica’s ending from elsewhere: perhaps Horace, in this poem that seems so much the work of a poet’s last years, is declaring himself done with his ­human lifespan, with the unspoken understanding that, as he expresses the hope for elsewhere, his poetry w ­ ill live on. Perhaps this is the point of the story about heart was the cause of stupidity” (Orelli 1844: 797, cf. Wickham 1891: 430); the term may also have stylistic undertones (as at Ar. Thesm. 170). See also further Fedeli 1997: 1609–10. 20. For Newman 1967: 76, with n1, Empedocles “is chosen as the arch-­example of poetical humbug,” and called poeta ­because “the vates-­concept was too precious to waste like this” (the same would apply to 299, nanciscetur enim pretium nomenque poetae).

206  ch a p t e r 4

Empedocles: that his wish for immortality took not merely the wrong form but an un-­Horatian one. Rather than a “famous/storied death” (famosae mortis), he ­ought to have wished for his writings to live on, even as he himself as a h­ uman body and life ceased to be talked about.21 The ­dying that marks the end of the Ars Poetica is of an altogether dif­fer­ent quality from that seen elsewhere in the poem. It is strongly desired and deliberately sought out,22 no longer occurring as the result of the passage of time but able to be unnaturally preempted before its proper time. It is also absolute in its finality, since ­there follows no form of rebirth. Also new is the possibility of killing—­not only oneself, but another. The first of what ­will be two mentions of killing another person is the gentler one: Horace points out that someone who saves a man who wishes to die, and in d­ oing so forces him to live, has committed an act tantamount to killing (occidenti, 467). The second appearance of the verb occidere disturbs more (though Horace has succeed in inuring us somewhat to the word, as he does for scabies through its double appearance at 417 and 453), as we are witness to a frenzied reciter holding his victim fast and “killing him by reading” (occidit . . . ​legendo, 475).23 Horace has come very close to showing his reader the murder and metamorphoses he had e­ arlier declared forbidden (185–87),24 yet he manages to remain just within the bounds he himself spelled out of propriety and good taste by suggesting, rather than describing outright, the transformation of man to bear to leech (in a rather confusing and opaque manner),25 and by having an animal (the leech) rather than a ­human proper kill his single audience member, thus making it not quite murder. Horace seems concerned at this juncture primarily with the poet’s

21. Philodemus discusses in On Death the fear of not being remembered (e.g., 35.34–36.31), and Empedocles in the AP demonstrates a humorously idiosyncratic interpretation of this in wanting to be remembered not for his life but rather his death; cf. Orelli 1844: 797. 22. That the poet figure cries for help a­ fter he has, in a moment of distraction, fallen into a hole indicates that this fall was accidental, and Horace’s suggestion at 462–63, in the voice of an interlocutor not further identified, that perhaps he meant to throw himself down is therefore at odds with the story as presented (though Oliensis 1998: 218 suggests that perhaps the “accident” of the “absent-­minded poet” serves to “advertise his credentials”); nevertheless it sets up the second anecdote, that of Empedocles’s deliberate jump (Brink 1971: 426). 23. Rudd 1989: 229 compares Pope, Epistle to Dr Arbuthnot 32 (Pope 1735: 7, see with Maresca 1966: 73–116): “If Foes, they write, if Friends, they read me dead.” 24. Compare Aristophanes’s claim at Nub. 518–62 that his play contains none of the usual overused jokes about baldness, characters who rush in with torches shouting iou, or a grumpy old man who hits bystanders with his walking stick, despite the fact that each of ­these may be found in the play (1, 545, 1297–1302, 1476–1509). 25. Sharrock 1996 is useful on repre­sen­ta­tions of metamorphosis in lit­er­a­ture.

T he E n d of the P oe m   207

physical body: sit ius liceatque perire poetis (466), he pleads in doubled l­egal terms,26 between the example of Empedocles’s leap into Etna and the warning that saving someone (specifically a poet, as it transpires) who does not want to be saved is equivalent to murder. The audience, too, exists only as a body—­one that is turned into no body at all as its liquid contents are sucked out into another (hirudinous, invertebrate) body. Quodcumque ostendis mihi sic, incredulus odi (188)—­not an altogether inaccurate description of how the ending of the Ars Poetica has itself been received.27 Although not previously noticed, pre­sent in the Empedocles vignette and enacted in the grotesquerie that follows is also the idea of the aesthetic sublime, evoked overtly by the adjective sublimis,28 which Horace purports to use in all innocence to describe the distracted mad poet (vesanus poeta, 455) walking about with his head in the proverbial clouds.29 Porter has identified twenty-­ one traits as “logical and thematic markers of the sublime” in ancient thought,30 codified most famously in Pseudo-­Longinus’s On the Sublime. Remarkably almost ­every one of Porter’s criteria is pre­sent in the ending of the Ars Poetica ­after the appearance of the flag word, sublimis, in 457.31 An unexpected fall into 26. Alongside ius, licere (“be permitted”) also has a legalistic flavor; Brink 1971: 428 compares line 72. 27. For Brink 1971: 431 the transformation with which the poem ends is “startling” and for Rudd 1989: 229 “disconcerting” (cf. Laird 2007: 137). While most grant that the poem’s close is meant to be humorous, they do not give the impression of liking it very much and devote precious ­little space to discussing the final five to ten lines (cf., e.g., the scholia, Wickham 1891: 431, Kiessling and Heinze 1914: 365, Rudd 1989: 228–29, all of which simply fizzle out). 28. That Horace had a notion of the sublime and used sublimis in this sense elsewhere is clear from Porter’s 2016 discussion: “Sublimis is already circulating as a literary critical term by the time of Ovid, and it is commonly used in Horace and in l­ater Roman poets,” e.g., Carm. 1.1.35–36 (cf. Porter 354 with n215), 1.15.31, 3.1.45–46, 3.26.11–12, 4.2.5–8 (of Pindar; Porter 332), Epist. 1.12.15–19 (sublimia . . . ​concordia discors; Porter 501 with n349), 2.1.165. Porter suggests that the “kind of voguish appeal” it may have had in “late Augustan Rome” (1–4; cf. 32–33) is “in part a Lucretian inheritance” (28; cf. 463 and Hardie 2009: 197–202). Although the date of On the Sublime is, like its authorship, uncertain, the idea of the sublime was certainly current for Horace, and evidently predated this canonical discussion of it. 29. Brink 1971: 423–24 renders it “head high” at AP 457 (adding that it “connotes ‘high-­minded’ ”) and sees it corresponding to μετέωρος, Pl. Tht. 174a; so, too, Orelli 1844: 796. They, with Rudd 1989: 225, are right to reject the possibility that sublimis is an accusative plural modifying versus. 30. Porter 2016: 51–53; I refer to the items in his list by numbers that I have assigned to them for the sake of clarity. He cautions rightly that the list is not—­n or is it meant to be—­exhaustive. 31. Absent from lines 457–76 are perhaps only criteria 7 and 8, yet t­ hese can be discerned elsewhere in the poem: 7 in Horace’s discussions of Homer (AP 74, 359, 401), while the ideas of

208  ch a p t e r 4

a relatively shallow pit, though one where the unfortunate poet’s life is nevertheless in danger, l­ater reiterated and ratcheted up in horrifyingly magnified terms as Empedocles’s suicidal leap into Etna, makes use of “im­mense heights” and “profound depths” (Porter criterion 1); “sudden or extreme, often violent, motions or changes” including “falling” and “coming into or out of a view” (2); “gaps” such as “fissures, rips and tearing, hiatuses, abysses, vast distances” (3); “limits revealed in their transgression,” definable as every­thing “contained in the root prefix or preposition huper” (4); “unthinkably large masses and quantities and surfaces” including “mountains” (5); “lasting and everlasting qualities” including “undying fame” (9); “sharp collisions and contrasts (contrastive opposites)” including “large versus small” (10) or, in the Horatian Empedocles’s case, cold meeting hot32; “uncontainable forces” (12); “ephemerality, evanescence, (sudden) epiphanic appearances or disappearances” (15); “blinding moments of plea­sure or pain, fear and awe,” or “excessive noise or silence” (16)33; “moments of intense and vital danger, risk, and crisis (the kairos)” (18); “a sudden profound awareness of vitality, especially the vitality of life itself ” (19); and “natu­ral, mythical, divine, or literary phenomena embodying any of the above,” of which “Etna’s deadly volcano” (20) is listed as an example. The dizzying heights and vertiginous abysses of Ars Poetica 457–66, emphasized by such verbs and details as decidit, demittere funem, se proiecerit, insiluit, clearly instantiate the descent and movement from high to low and the ancillary anx­i­eties around mortality so fundamental to the sublime. Moreover, the sacrilege committed by urinating upon one’s ­father’s grave (471), or by moving a spot defined as sacred ­because it has been struck by lightning (471–72), one or more of which may cause a divinity to inflict madness upon the guilty person, further embody the criteria having to do with limits and their transgression (4), cosmic events including specifically “thunderbolts” (12), and overwhelming religious “awe” (16), as well as “nature and nature’s law” (21). Fi­nally, the transformation from poet to bear to murderous leech, with its “overwhelming focus on details that blot out all ­else” that indeed succeeds in eliciting from the reader “an utter, rapt absorption in their particularity and their features” (17), exploits the notions of “sudden or extreme, often violent, motions or changes (reversals, transformations)” (2), “sharp collisions and contrasts (contrastive opposites)” (10; large bear versus small leech, for example, where the leech in its “tininess,” 5, may also be considered sublime), “sudden excellence, perfection, and incomparable virtue (8) occur throughout (and perhaps even in its final twenty lines in, e.g., the figure of Empedocles). 32. The sublime as residing in contrastive opposites offers a new explanation for the juxtaposition of ardentem frigidus; see chapter 4, n19. 33. As in the poet’s shouts, which receive no reply.

The E n d of t he P oe m   209

intrusions of opposites into one another” (also 10, as ­human blood enters the leech, where it ­ought not to be), and “indefinability (elusiveness)” (14, as we strug­gle to follow Horace’s descriptions), while the jumping from scene to scene that takes place from 457–76 exemplifies “bold or sudden expansions or compressions” (6). The w ­ hole is a mass of “sharp antagonisms and tensions” (11) that gives life to “the vivid and terrifying collapse of form and order” (13) and yet reminds us of the “vitality of life” (19) and its “ephemeral” quality (15). Starting from the understanding of the sublime as “a sense of absolute structural impossibility and of total deadlock” that produces “profound ­mental or spiritual disruption,” Porter goes on to add the following: Standing for a shattering and dislocating excess, it appears to exceed the grasp of systematic closure. Unlike beauty, grace, charm, and other of the more domesticated aesthetic virtues, the sublime, which has a bit of the rogue and dysfunctional f­ amily member to it, seems to speak more directly to one’s experience: it betokens an overpowering immediacy and a bruising contact with some Real. (Porter 2016: 5–6) This description fits the ending of the Ars Poetica uncannily well. Although beauty, grace, and charm can (with some effort) be discerned in the poem’s close, this close precisely brings us into “bruising contact with some Real,” namely, death and madness, from which we would like to turn away but cannot, and which, through the inability of ­humans to truly grasp them,34 have a “shattering and dislocating” and disruptive effect upon the reader. This disruption, brought about by the unwillingness or inability to understand what we are seeing, may manifest itself, as it has throughout the poem, as laughter. The Ars Poetica’s potential as satire—in meter, content, and Weltanschauung—is the source of its rescue: taken at face value, much of the poem becomes unbearable, but if we read as we do when reading Horace’s Satires, its ending in par­­ tic­u­lar allows a vacillation where a hyperbolic false sublime overlays and allows glimpses at a true one.35 Where does this ending leave the Pisones and Horace himself? For ­those willing to read the ending in the terms suggested h­ ere, the Ars Poetica concludes with a display by the poet, an artistic tour de force, that reveals his grasp on the sublime: he parades this before us,36 daring us in the 34. Cf. Porter 2016: 6: “­There are further ways of describing the sublime, for instance by defining it as what­ever prompts or results from reflection on thought at the limits of the humanly conceivable.” 35. Consider Lowrie’s 1997: 352 closing words to her monograph: “We face a choice between irony and the sublime. Maybe the two are not incompatible.” 36. Compare Horace’s humorous recusatio that he cannot write epic verses “followed by superb epic verses!” (Oberhelman and Armstrong 1995: 234, n4) at Sat. 2.1.12–15 and Epist. 2.1.250–59.

210  ch a p te r 4

first place to recognize it for what it is (did he expect the Pisones to do so?), and, second, taunting us, if we do recognize it, with the unique and ideal balance of ars and ingenium that he possesses and that allows him to write poetry such as this.37 Just as the sublime is “not so much found in [the] sorts of c­ auses” detailed by Porter as it is “provoked by them,” so the lessons of the Ars Poetica, once again, are to be found not in what it appears to prescribe, but rather in what it is. Agamben is right to say that in general, “modern works of poetics and meter have not considered . . . ​the end of the poem insofar as it is the ultimate formal structure perceptible in a poetic text. Th ­ ere have been inquiries into the incipit of poetry (even if they remain insufficient). But studies of the end of the poem, by contrast, are almost entirely lacking.”38 Perhaps our dissatisfaction with the ending of the Ars Poetica, then, has much to do with a lack of consensus as to what the ending of a poem is supposed to accomplish. The prob­lem, as Agamben frames it, is that since any “poem tenaciously lingers and sustains itself in the tension and difference between sound and sense, between the metrical series and the syntactical series,” “what happens at the point at which the poem ends?” Having e­ arlier defined poetry as offering always “the possibility of enjambment,” he concludes that this must mean that the “last verse of a poem is not a verse. Does this mean that the last verse trespasses into prose?”39 This uncertainty as to where poetry ends and prose begins is especially thought-­ provoking in our pre­sent context of Horace’s disavowal that his hexameter writings qualify as poetry.40 Horace’s (faux) crisis of confidence in this regard in turn opens up the possibility that, like the poets Agamben has in mind, Horace, too, was “conscious of the fact that h­ ere ­there lies something like a decisive crisis for the poem, a genuine crise de vers in which the poem’s very identity is at stake” (113). If the Ars Poetica is intended in any meaningful way to teach the art of writing—­not through the rules it states but through its own nature and 37. Oliensis 1998: 219, too, says that “although Horace teaches what can be taught, although he supplies some of the rules, some of the fundamentals, with which a practicing poet must in fact be equipped, it is, in the end, the art that cannot be taught that holds the key to Helicon.” Similar is Hajdu 2014: 96: “The text can also be interpreted as a show of poetic excellence designed to intimidate pos­si­ble imitators. It does not want to teach you how to create ­great poetry, but to prove how difficult it is.” 38. Agamben 1999: 112. This despite Poe’s saying, as Borges cites him (at Mihailescu 2000: 50), that “a story should be written for the sake of the last sentence, and a poem for the sake of the last line.” Among the few studies of the topic, see Herrnstein Smith 1968, Kermode 2000, and (within the field of classics) Fowler 1989, and Roberts, Dunn, and Fowler 1997 (especially the essays by Peta Fowler and Don Fowler). 39. Agamben illustrates his point with the “trivial” yet illuminating observation that “­there can be no enjambment in the final verse of a poem.” 40. See chapter 3, n161.

The E n d of t he P oe m   211

attributes as artistic artefact—­then perhaps Horace is showing that the only way to end a poem at the “point in which sound is about to be ruined in the abyss of sense,” ­unless the writer employs a neat coda of some form,41 is in “the state of poetic emergency,” which in the pre­sent instance admits also of the sublime. Thus, perhaps the ending of the Ars Poetica is brightly and darkly cacophonous ­because Horace is illustrating for ­others in self-­referential fashion the prob­lems inherent in bringing a poem to its end—­especially a poem so long that we no longer have fresh in our minds its beginning or even its ­middle parts.

The Art of ­Doing Every­thing The Ars Poetica’s ending, bizarre and gruesome, matches its opening: both confront us with images we would rather not picture, and both explore the fraught relationship between the one who creates art and the ones who view or hear it. In addition, both beginning and end make notably greater use of color than might be expected from a poem that is supposed to be about poetry, not about visual art.42 Although readers who know the poem as the Ars Poetica come to it with the natu­ral expectation that it is to be primarily about poetry, the opening privileges painting (if of a rather poor sort) over writing: pictor (“painter”) appears prominently in the opening line; tabulae (“painting”) comes ahead of librum (“book”) in line 6; and Horace’s imaginary interlocutor objects, “paint­ ers and poets” (in that order) “have always had the same opportunity to dare what­ever they wish” (9–10).43 One of the poem’s most famous phrases, frequently excerpted, ut pictura poesis (361), likewise positions poetry second. ­Others have noted this unusual emphasis on painting, on “not-­poetry,” but it has been regarded more as a curiosity than as integral to the poem’s course and meaning. Laird is right to remind us, however, that “it should never be forgotten that the initial verses of the poem lead the first-­time reader to believe that they herald a didactic poem about painting,” numerous ­later writers having been 41. Herrnstein Smith 1968 and Agamben 1999: 112 provide examples. For Brink 1971: 421, “like many poems of H. the Ars is open-­ended” in that “no attempt is made to bring to a close the conceptual schema of his literary theory.” Compare the plays of Aristophanes, in which an initial premise for the plot is forgotten at its end, as the ­whole devolves into joyful and chaotic feasting. 42. Hardie 1993: 120 argues that Horace is “one of the least pictorial of Latin poets” (even as he admits that in the AP Horace “displays his ability to create word-­pictures as if to demonstrate that it is not through lack of ability that he other­w ise tends to avoid such t­ hings”). Although it remains true that “pictorialism, understood as the sustained and detailed evocation of a framed view, is alien to the lyric manner of Horace” (122), his interest in color and in compelling his reader to picture color is profound throughout the AP. 43. Brink 1971: 92: “The two arts brought together as 7 ff., below 21, 361.”

212  ch a p te r 4

inspired by ­these lines to do just that.44 I attempt to show ­here that Horace’s interest in painting and other creative arts, signaled by the degree to which color saturates the poem, is far more than a sideshow, a complement to the main event (Lit­er­a­ture);45 rather, it is an essential reflection of what the Ars Poetica may be read as ­really being and as being about. The first ten lines already contain hints at the vibrancy that is to come, as we are told that the creature’s feathers are “multi-­colored” (varias)46 and its fish tail “black” (atrum). It is furthermore hard to picture the image, as we are being instructed to do and even talked through the pro­cess of ­doing, without mentally assigning a shade to the h­ orse’s head (brown? white? black? we are left to choose for ourselves) and contemplating how its color and texture juxtapose with that of the ­woman’s skin. The next vignette blurs painting and writing, as did the opening one, through the verbs describitur (necessarily suggestive of writing), scis simulare (broad), and pingitur (unambiguously painting). This scene is again visual, yet it makes use of further arts as well: inceptis gravibus plerumque et magna professis purpureus, late qui splendeat, unus et alter assuitur pannus, cum lucus et ara Dianae et properantis aquae per amoenos ambitus agros, aut flumen Rhenum aut pluvius describitur arcus. sed nunc non erat his locus. et fortasse cupressum scis simulare: quid hoc, si fractis enatat exspes navibus aere dato qui pingitur? amphora coepit institui: currente rota cur urceus exit? denique sit quidvis, simplex dumtaxat et unum. (14–23) Horace has managed to compress an impressive array of dif­fer­ent creative activities into this short passage. While describitur, as a compound of scribere, may first suggest writing, its primary senses pertain rather to drawing: only secondary and ­later are found “to rec­ord in writing, write down.”47 This overall 44. Laird 2007: 138. See further my epilogue. 45. Ibid.: 138–39 likewise sees that “Horace’s interest in the visual field is . . . ​maintained beyond ­these initial verses,” and indeed the “Materia” commentary proper begins by stating that Horace begins as he does “quoniam inter picturam et scripturam, inter pictorem et scriptorem maxima est similitudo” (Friis-­Jensen 1990: 338). 46. The adjective suggests variatio, yet what is often presented as a desirable quality has plainly ­here been taken too far. Cf. Poet. 1450b1, where Aristotle cautions against the use of “beautiful colors” (τοῖς καλλίστοις φαρμάκοις) “indiscriminately” (χύδην). 47. OLD primary definition: “to represent by drawing, draw, mark out; to describe (geometrical figures or sim.); to trace out (on the ground or other surface); to mark (with a pictorial

T he E n d of t he P oe m   213

impression of painting (visual over literary purple patches) is confirmed by pingitur.48 Horace concludes explic­itly with the plastic arts, as the ceramic (and so richly clay-­colored) object that was supposed to turn out as an amphora comes off the wheel instead as a plain urceus, “jug,” an amphora apparently being beyond this par­tic­ul­ar artist’s abilities.49 In describing this, Horace shows off his own linguistic and poetic mastery, as Ahl discerned, creating an internal echo and symmetry with currente (“­running”), cur (“why”), and urceus (“jug”): CURrente rota CUR URCeus exit.50 We are a mere twenty-­three lines into this 476-­line poem, but already we should have developed some sense that prescriptions such as denique sit quidvis, simplex dumtaxat et unum apply far more widely than merely to the creation of poetry. If sit quidvis reminds us explic­itly that ­these rules should apply and be applied broadly, we also find versions of simplex dumtaxat et unum elsewhere in the poem, too, as several terms from line 31 (in vitium ducit culpae fuga, si caret arte) ­will turn out to be crucial (and may well already remind us of moments from e­ arlier in Horace’s writings): vitium, culpa, and, above all, ars. Fi­nally, the “purple patch” (another phrase we owe to Horace’s Ars Poetica) has occasioned much comment. It is clearly a form of “decoration”51 that Horace does not endorse,52 yet the fact that it has largely been read meta­phor­ically repre­sen­ta­tion).” Cf. Laird 1996: 92–93: “The Latin describere (more like the Greek graphein, ‘to write’ and ‘to depict’) means ‘to draw’, ‘to mark out’, ‘to write’, as well as ‘to describe.’ ” 48. The scholarly commentary on lines 19–21 concentrates on the custom among shipwrecked sailors of commissioning an artwork depicting their rescue, which they would then dedicate to a god (Brink 1971: 100–101). Th ­ ere is also discussion of the cypress tree as funereal symbol (appropriate to the almost-­drowned sailor; Brink), and of the Greek proverb about a painter who, able only to paint cypress trees, would offer to put them in even where least appropriate, including a scene at sea (the scholiasts, Brink, Rudd 1989: 152–53). 49. The point lies e­ ither generally in the artist’s lack of control over his artistic product (he cannot translate his vision of an amphora into real­ity) or, better, in his inability to make an amphora specifically, on the understanding that this form of vase is harder to make than an urceus (which is thus a misshapen amphora); see further Wickham 1891: 391, Rostagni 1930: 9 (who stresses the urceus’s small size), Brink 1971: 101, Rudd 1989: 153. 50. Ahl 1985: 37, who further points out that “poetic art imitates the potter’s” through the “linguistic metamorphosis” whereby “the Greek amphora, when put on the Roman potter’s wheel, ‘turns into’ a Latin urn.” 51. Rudd 1989: 152. 52. As Brink 1971: 96 points out, however, “the fame of the ‘purple patch’ must not make the reader forget that the expression is still unexplained.” Pseudo-­Acro, for example, states that the habit of embellishing a work in this m ­ atter is “pessimam” if it results in incongruity, yet if done properly and subtly, “pannus bonum additamentum.” Orelli 1844: 700–701 explains that what makes the purple patches undesirable is that they are not necessary (“supervacaneas”; cf. the

125-81019_FerrisHill_HoracesArsPoetica_5P.indd 213

8/8/19 8:40 AM

214  ch a p te r 4

obscures the fact that Horace clearly conceives of it as a physical object, able to be “sewn on” (assuitur) to something else—­yet another form of creative activity and one that, as Ferenczi points out, results in “an embarrassingly direct connection” between making a poem and “the very concrete activity of sewing.”53 Moreover, in addition to being purple or crimson (a deep, dense color with long-­standing royal associations),54 it gleams (splendeat),55 and thus has both a color shade and light, as aes, “bronze” (21) also does.56 In giving examples of such purple patches (16–18), Horace indulges in them, executing them well, even if ironically.57 And yet a­ fter a moment he disavows the rhetorical/literary stratagem, the words sed nunc non erat his locus serving to draw our attention, especially through the unexpected tense of erat that must refer to the lines immediately past, to the fact that he has done just this:58 lines 16–18 ­were, in fact, not the proper place for the purple patches Horace stitched onto his poem ­there. Striking among the images of 14–34, moreover, is that of “Materia” commentary’s explanation that while they may be lovely, “preciosus/boni,” they are out of place), while for Wickham 1891: 390 purpureus pannus indicates that the rest of the work is comparatively inferior (“the merits of the piece turn out to be two or three good descriptions”). Laird 2007: 139 describes its afterlife thus: “This apparently uncontroversial assertion that description is a form of detachable digression was quoted by Servius [ad Aen. 10.653] and endorsed by many other ancient authors.” 53. Ferenczi 2014b: 78. Brink 1971: 96 speaks of “ ‘adornments sewn on’ a piece of clothing” and Rudd 1989: 152 likewise of “a piece of fabric sewn onto the outside of the garment,” as described at, e.g., Ov. Ars am. 3.169, Juv. 2.124; both argue against reading it as a reference to the latus clavus (Orelli 1844: 700). Laird 1996: 93 adds that from the literary perspective, pannus is “a kind of appendage.” 54. OLD s.v. purpureus, Bradley 2009: 189–211, Goldman 2013: 40–52. 55. Rudd 1989: 152: “con­spic­uo­ us sheen.” 56. Aes occurs several more times in the AP: 33, 164, 345. Maxwell-­Stuart 1981: 1, Bradley 2013: 127 note the Romans’ attentiveness to light and dark, and Goldman 2013: 2 the effects of light on color. 57. As the echo in flumen Rhenum of Horace’s attack on Furius Bibaculus at Sat. 1.10.36–37 suggests (Orelli 1844: 701 notes the correspondence, as do Brink 1971: 98, who is skeptical, and Rudd 1989: 152, who adds that the scenes described ­were commonplace assignments in the rhetorical schools). 58. Erat has puzzled: Wickham 1891: 391 and Brink 1971: 99 work to explain it as equivalent to est (Wickham: “It is not though you thought it was”; Brink: “not a reference to a past time,” rather, “the pre­sent time is referred to”), while Rudd 1989: 152 rightly notes that “the impf. implies that something has happened which o­ ught not to have happened” and Laird is correct to connect Horace’s words to the passage at hand and see “the reflexivity of this ‘meta-­ekphrasis’ ” (1996: 92), whereby “Horace amusingly negates the cliché est locus with non erat . . . ​locus” (Laird 2007: 139). Comparable, again (see chapter 4, n36) are Sat. 2.1.12–15 and Epist. 2.1.250–59, and ­here the point would be, “­don’t sew on purple patches (as I just did).”

T he E n d of the P oe m   215

the “rainy bow” (pluvius arcus), that is, a rainbow. Although arcus was commonly used to mean “rainbow,”59 the phrase arcus pluvius is unique to Horace, unattested before or ­after,60 the adjective pluvius perhaps pre­sent to intensify the other forms of w ­ ater flowing through the passage (aquae, flumen, and the shipwrecked sailor). Horace’s rainbow also answers varias (“multi-­colored”), the adjective from sixteen lines ­earlier.61 Even when images and objects are not assigned a specific color, they can suggest one for themselves, as they did also in the poem’s opening lines: the deep green of the cypress tree, for example, and the marine colors of the vari­ous forms and bodies of ­water. Horace continues for some time yet on the theme of painting (appingit), which is still bound up in some unspecified way with poetry (vatum), addressing pater et iuvenes patre digni (24) on the dangers of “the appearance of right” (specie recti, 25). To illustrate the princi­ple of line 31 (in vitium ducit culpae fuga, si caret arte), Horace explains, “He who desires to vary a single theme unnaturally paints a dolphin in the woods, a boar on the waves” (qui variare cupit rem prodigialiter unam, / delphinum silvis appingit, fluctibus aprum, 29–30). As at 12–13 and 19–22, Horace juxtaposes ele­ments that do not belong together: first snakes and birds, tigers and lambs; then a cypress tree in a shipwreck scene; now a dolphin transposed to the woods, beside a boar in the waves. Although no color adjectives are assigned to t­ hese last items, we once again cannot picture what Horace is asking us to without assigning appropriate shades to woods and waves, especially as the latter two appear more than once (19, 30; 17–18, 20–21, 30). The verb is appingit, repeating e­ arlier pingitur, and the overall visuality of the passage is further accentuated by the noun specie, “appearance” (itself repeating species from line 8), as well as by the suggestive image of an overly timid creative figure crawling on the ground, serpent-­like, and cowering from a squall (serpit humi tutus nimium timidusque procellae), even as Horace now seems to have in mind speech or writing, professus grandia turget, the words picking up on the less overtly critical inceptis gravibus . . . ​et magna professis (14) through repetition of the verb profiteor and with gravibus and magna anticipating the sound and sense of grandia. The section ends, as did 14–23, with a maxim. If the point ­earlier was of an artist skilled in painting only one t­ hing and therefore naturally 59. E.g., Cic. Nat. D. 3.51, Lucr. DRN 6.526, Verg. G. 1.381, Aen. 5.609, 5.685, 9.15, Livy 41.21.12, Tib. 1.4.44 (imbrifer arcus), Prop. 3.5.32 ( purpureus pluvias cur bibit arcus aquas, intriguingly suggestive of both pluvius arcus and purpureus pannus). 60. TLL s.v. arcus and pluvius. The closest is the idea of the rainbow (arcus) as bringing or portending rains (pluvias), as at Prop. 3.5.32 (see chapter 4, n59) or Sen. Q Nat. 1.3.5 or Plin. HN 2.150; on the rainbow, see further Bradley 2009: 36–55. 61. See Goldman 2013: 135–60 on the numerous Latin words for multicolored and the significance of this.

216  ch a p t e r 4

wanting to include it everywhere, even nonsensically, now Horace has in mind variatio taken too far (prodigialiter),62 as in the grotesque image with which the Ars Poetica opened. As much as 24–31 overlap with 14–23 in imagery and, it would appear, overall objective, t­ here is a reason why Horace has included both sections, and the distinctions between them become most apparent in the words with which each concludes. First, it transpires, our attention was being directed to the importance of internal unity and the need to avoid excessive, distracting adornment; next (as if this w ­ ere a second evolutionary stage now that we have begun to self-­correct our efforts), we are warned not to go too far. The reader who perseveres ­will be rewarded at line 38 by some advice that at long last pertains specifically to writing, as Horace, turning to address “you who write” (qui scribitis), explains to the “author of a promised song-poem” (45–46) how to create words—­all markedly literary in focus.63 Yet the technicolor assault is not quite over, and one area of artistic endeavor—­sculpture—­ remains unmentioned before Horace, in his Ars Poetica, ­will turn his attentions in more focused fashion to writing: Aemilium circa Ludum faber unus et unguis exprimet et mollis imitabitur aere capillos, infelix operis summa, quia ponere totum nesciet. hunc ego me, si quid componere curem, non magis esse velim quam naso vivere pravo, spectandum nigris oculis nigroque capillo. (32–37) The sense (and text) of this passage has been much debated: are we to read unus or imus in line 3264 and on what grounds (which depends partly on which 62. This adverb, found only ­here and at Columella, Rust. 3.3.3 (Brink 1971: 114), resembles our “prodigiously” but its root meaning is “in the manner of a prodigium,” that is, a monstrosity or a marvel. As a result, “the meaning could be ‘monstrously’; but it could also be ‘marvellously’ ” (Brink) as Horace communicates that in aiming at dazzling variatio, the inept artist achieves only horror. 63. Many commentators have resisted Horace’s interest in creative activities other than lit­ er­a­ture: Pseudo-­Acro (cf. Porphyrio), Orelli 1844: 700, and Wickham 1891: 390, for example, insist that beginning at line 14 he is speaking exclusively about writing (contra Rudd 1989: 152 comments ad 12–13, “H. is still talking of unity in visual terms” and Brink 1971: 93, ad 14–23, that “­there is still no straightforward literary argument,” their “still” suggesting that Horace is ­doing something contrary to the reader’s expectations). 64. According to Brink 1971: 118, Bentley found unus in a late MS, and this reading is accepted and endorsed by Orelli 1844: 705, Brink, Shackleton Bailey 2001, Antolín 2002. The sense would be “one craftsman,” i.e., a person whose par­tic­u­lar identity is of no further importance. The predominant reading of the MSS, however, is faber imus, which, though favored by Wickham 1891: 392 and Rudd 1989: 154–55, does not yield a clear sense (Orelli, in rejecting it, explains that

The E n d of t he P oe m   217

reading we choose) is the craftsman being held up as an example of excellence and/or failure?65 Of greatest interest for the pre­sent purposes, however, are the emphasis once again on color, which has gone unnoticed, and the connections that Horace forges with what has come before. Horace describes the color of what would seem to be his own eyes and youthful hair (Epist. 1.7.26),66 drawing attention to the words through pointed repetition: nigris oculis nigroque capillo. Although the adjective is a dif­fer­ent one, niger evokes ater,67 the black fish tail of lines 3–4, while the fact that the sculptor works in bronze (aere) echoes the payment made (aere dato, 21) to the painter some lines ­earlier. It should be noted above all that in t­ hese first thirty-­seven lines Horace is not describing one and the same form of error across disciplines but dif­fer­ent types of error that are able to plague all the creative arts in equal mea­sure (it is implied):68 some men may fail as artists b­ ecause they attempt too much variation, resulting in a disjointed w ­ hole; ­others ­because they excel in one area but not in another and are therefore not able to produce a work of which e­ very part is up to the same standard69 (they may even put in the one ele­ment at which they do excel when it is least called for); and yet ­others ­because they cannot translate the image of their mind’s eye into real­ity.70 it is unlikely to mean the poorest sort of other­wise unknown craftsman, to be a proper name, or to refer to his location; Brink adds, “nor is his small stature in question”). 65. The point must be, however, that the craftsman has g­ reat skill in at least one area, and less or none in ­others (so Orelli 1844: 704), not that he is a bad craftsman in e­ very re­spect (Rudd 1989: 155) or that “the correct repre­sen­ta­tion of nails and hair was a trick of art easily caught” (Wickham 1891: 392). In fact, sculpting the fingernails and hair was regarded as difficult (Orelli 1844: 705, Brink 1971: 119), and so this craftsman is paradoxically envisioned as skilled in the hardest aspects, engendering admiration of his work, yet has failed in the essential ­earlier steps (though his success in the ­others perhaps serves to conceal this). For Brink (117) the term faber is also impor­tant: “the ability to set down totum (A.P. 34) distinguishes the true artist or poet from an artisan, faber (32).” 66. Cf. Mayer 1994: 162; ­there is also a striking resemblance to the description of Lycus at Carm. 1.32.11–12 (Lycum nigris oculis nigroque / crine decorum, see with Clarke 2003: 100, though the correspondence was noted already by Landinus [at Bugada 2012: 103]), perhaps indicating that in speaking of the sculptor, Horace has in mind the crafting of his own description with words in the Odes. 67. The two adjectives overlap almost wholly in their senses (OLD). 68. For many commentators, however, Horace is speaking at 21–22 entirely about poetry: Pseudo-­Acro explains, “currente rota. Id est, carmine procedente” and Orelli 1844: 702 that the amphora/urceus represents “poëmata deformia.” 69. Cf. Brink 1971: 106 ad 25: “excellence attempted but unattained.” 70. Brink 1971: 113, noting that “the doctrine of unity” is paramount throughout the AP (77–85), explains how Horace “juxtaposes a series of concrete cases which put the reader in

125-81019_FerrisHill_HoracesArsPoetica_5P.indd 217

8/8/19 8:40 AM

218  ch a p t e r 4

Color, having thus been introduced as an insistent theme in the poem’s opening few dozen lines, recurs throughout the rest of it, with the effects of extending the thematic texture of the poem’s opening sections throughout its body and continuing to suggest that Horace is interested in material beyond a narrowly conceived ars poetica. “Gold and purple” are, as often,71 mentioned together as the “royal” colors of garments worn by gods or heroes (regali conspectus in auro nuper et ostro, 228), and “black” or “dark” (ater) may be the meta­p horical shade of a lawsuit (422–24)72 and the literal (as well as figurative)73 shade of the ink used to mark deficient lines of verse (445–47). Purplish, too, are the wine lees used by Thespis’s itinerant actors to stain their ­faces for their shows (dicitur et plaustris vexisse poemata Thespis / qui canerent agerentque peruncti faecibus ora, 276–77), and imbuerit, which like peruncti at 277, means “drench, steep; imbue,” appears at 331, where it is used meta­phor­ically, though its subject, aerugo, is itself laden with color:74 “Can it r­ eally be that, when this rust-­canker and concern for pocket money has tinged the soul once and for all, we may hope that song-poems are able to be crafted that must be anointed with cedar-­oil and preserved in smooth cypress?” (an, haec animos aerugo et cura peculi / cum semel imbuerit, speremus carmina fingi / posse linenda cedro et levi servanda cupresso?, 330–32).75 Similar liquidity is pre­sent also in the verb manat as Horace says, “­Every superfluous ­thing drips out from a full breast” (omne supervacuum pleno de pectore manat, 337),76 as well as in the bodily fluids spilled at the Ars Poetica’s close. Even the word color itself appears alongside Horace’s liberal use of words that assign specific colors to objects: “Why, if I am incapable of and ignorant at preserving the described turns and colors of works, am I greeted as a poet?,” Horace won­ders (descriptas servare vices operumque colores / place of the artist, struggling five times over to find the right road and missing it”; he tabulates the types of errors at pp. 115–16. 71. Brink 1971: 281–82 adduces Verg. Aen. 4.134, 5.132–33, 11.72, 12.126, Lucr. DRN 5.1428, Livy 34.3.9. 72. Atris is the reading of most codices (Pseudo-­Acro explains it as “noxiis, [et] tristibus”), though artis, “narrow, constricting,” Bentley’s emendation, is attractive and endorsed by Brink 1971, Rudd 1989, Shackleton Bailey 2001 (Wickham 1891: 427 praises but does not print it). 73. Wickham 1891: 429 explains: “not only ­because the ink was black and the lines scored strongly, but also in the meta­phorical sense = ‘triste’ as the ‘nigrum theta’ of the judges’ mark of condemnation”; cf. Pseudo-­Acro, Orelli 1844: 794, Brink 1971: 417, Rudd 1989: 223. 74. Rudd 1989: 204 defines aerugo as “verdigris,” i.e., bluish-­green, while Wickham 1891: 420 says “the figure suggests that it is caught from the ­handling of rusty coin” and Brink 1971: 351 compares “robigo, rust.” 75. Cedro, “cedar oil” (Brink 1971: 351) is also liquid. 76. Cf. Pseudo-­Acro, Orelli 1844: 775, Rudd 1989: 204 (“the hearer’s mind is thought of as a vessel”).

125-81019_FerrisHill_HoracesArsPoetica_5P.indd 218

8/8/19 8:40 AM

T he E n d of the P oe m   219

cur ego si nequeo ignoroque poeta salutor?, 86–87). Although “the word is not on rec­ord before this passage and 236 tragico . . . ​colori” (Brink 1971: 173) to mean “the ‘tone’ or ‘style’ appropriate to dif­fer­ent genres of poetry,” commentators are nevertheless quite sure that this is what Horace means by color: “the dif­fer­ ent types and styles of poetic works . . . ​i.e. the vari­ous genres” (Rudd 1989: 165).77 I see Horace, however, being rather more literal (just as sublimis is both literal and aesthetic) and wondering, in a manner that pertains directly to the Ars Poetica, why he (or anyone) deserves to be called a poet if he cannot depict ­things properly, including their color.78 In this way, the phrase operumque colores becomes highly suggestive of the image and activity with which the Ars Poetica opened—­a description of the appearance, including its colors, of a painting (opus meaning “labor” but also the product of this labor, a “creation,” ­whether of engineering, lit­er­a­ture, or art). The Greco-­Roman vocabulary of color may well be qualitatively dif­fer­ent from our own,79 but Horace’s spectrum is nevertheless surprisingly narrow even as color is a prominent theme: red/purple/crimson, black, and yellow, along with vivid “multi-­colored,” dominate the poem, pre­sent at its opening and its close. Cruor, “blood, gore” (476), by any mea­sure a strong word and also evocative in terms of color, is defined variously as ­either red or black80—­the precise colors that Horace has primed us to see throughout the Ars Poetica, united in its final line into a single, horrifying w ­ hole. In addition, the shades of lines 470–76, where beyond bloodred ­there may be found the whitish-­yellow of lightning and urine, and the gloomy darkness of the bidental and the bear, are preceded by the same in lines 453–69: morbus regius, “jaundice,” ­causes the whites of the sufferer’s eyes to become yellow;81 “blackbirds” (merulis) are black or dusky; 77. Brink 1971: 173 says that in Horace’s usage ­there is “a slight technical implication and the meta­phor [is] almost faded.” The sense of color in the hands of Cicero, Seneca, Quintilian, Lucretius, and ­others is discussed by Quinn 1991 and Bradley 2009: 56–127; see also Clarke 2003 s.v. 78. Porphyrio’s scholium may suggest this more concrete reading, too: “ordinem positum per species si non possum [signorum] persequi, quare poeta [dicitur] dicor?” 79. The difference between modern and ancient terminology for color is commonly noted (e.g., Maxwell-­Stuart 1981: 1–11), in par­tic­u­lar, the need to use several modern color names to render a single ancient one or vice versa; see further Lyons 1999, MacLaury 1999, Bradley 2009: ix–xi, 1–35, 2013: 128–30, Goldman 2013: 2–24. 80. See TLL s.v. cruor, 4.0.1244.72–1245.46 (“epitheta”): ater at Verg. Aen. 4.687, 9.333, 11.646 and Hor. Epod. 17.31–32. Postdating the AP ­there may be found niger (Sen. Oed. 189–90), purpureus (Sedulius, Carm. pasch. 5.288), and rutilus (Ov. Met. 5.83). No colors other than t­ hese (i.e., blacks and reds/purples) are attested with cruor. 81. Pseudo-­Acro: “quibus oculi virides sunt” (reflecting, again, the conflation of green and yellow in ancient thought). The name morbus regius is explained (weakly, Brink 1971: 422) at Celsus, Med. 3.24 as stemming from the (regally) expensive entertainments that sufferers ­were

220  ch a p t e r 4

and burning Etna, red (ardentem . . . ​Aetnam). ­These ­causes and consequences of madness / furor poeticus would in fact seem to have ­little in common beyond the array of deep, vibrant hues they possess and conjure up. Even the Pisones, the poem’s scaffolding, have once again participated in this theme of the Ars Poetica: Horace’s address to them, o Pompilius sanguis (291–92), reads as vibrantly colored in retrospect and as prefiguring the dark red blood that flows as the poem concludes. It is perhaps especially notable that the colors permeating Horace’s poem overlap almost wholly with the ones Empedocles, named in the midst of them (465), identified as most fundamental: white (λευκός), black (μέλας), red (ἐρυθρός), and yellow (ὠχρός).82 The specific forms that color takes in the Ars Poetica may thus be seen as nodding to Empedocles and, in ­doing so, serving to prefigure his eventual appearance in the poem. The per­sis­tence of color and colors throughout the Ars Poetica and Horace’s manifest interest in arts beyond writing is perhaps stated most emphatically at 361: ut pictura poesis. Horace’s phrase and thought go back to the same preserved in Plutarch, who attributes it to Simonides: τὴν μὲν ζωιγραφίαν ποίησιν σιωπῶσαν προσαγορεύει, τὴν δὲ ποίησιν ζωιγραφίαν λαλοῦσαν (“Simonides calls painting ­silent poetry, and poetry painting that talks”; Mor. 346F / De gloria Atheniensium 3.346).83 Yet Horace is not likely to be straightforwardly regurgitating the Greek lyricist.84 First of all, the words come shortly ­after Horace’s arresting, quid ergo est?, “what’s the point of all this?” (to paraphrase 353), just as we might have indeed been wondering what was the point, not merely of the section at hand, but of the entire Ars Poetica, of which we have now labored through exactly three quarters. ­These words, I contend, are central to the poem, in which I see Horace putting forth the notion of all art—­whether visual, literary, or plastic—as something made. Painting is not like poetry in any vague thought to require. Wickham 1891: 429, Brink, Rudd 1989: 224 rightly point out that it is not clear ­whether the point at AP 453 is that Horace thought of it as contagious; for Lowrie 2014: 137–38, the key word is regius, which picks up on the reges of 434 and which she suggests evokes “the divinization of Julius Caesar and the emperor worship Augustus kept trying to keep at bay.” 82. See Rowe 1972: 341, Ierodiakonou 2004, 2005, Kalderon 2015. Empedocles’s four colors correspond to ­those that ­were most readily available and most widely used by ancient artists: white from Melos, Attic yellow, red from Sinope, and lamp-­black (see Gage 1981: 2 and Ierodiakonou 2005: 13 on Plin. HN 35.32, and Hardie 2014: 47 on the presence of Empedocles’s theory in the opening of the AP); Rowe and Gage discuss also the per­sis­tence of the four-­color theory among ­later Greek and Roman writers. 83. On t­ hese and further attestations of the idea, see Brink 1971: 369–71, Laird 1996: 84–85 with n20. 84. The “ramifications are endless” (Rudd 1989: 209); see, e.g., Johnson 1982: 13, Hardie 1993, Laird 1996, Sharrock 1996: 104 (“the exact nature of the parallel . . . ​is open to debate”), Ford 2002: 96–101.

125-81019_FerrisHill_HoracesArsPoetica_5P.indd 220

8/8/19 8:40 AM

T he E n d of t he P oe m   221

way; rather, they are akin in being forms of “fabrication” or “creation.” Poesis (ποίησις) is, ­after all, from the Greek verb ποιεῖν, with its two basic senses of “make” and “do” that encompass “make, produce; bring about, cause; bring into existence; invent; do, act; prepare,” that is, making and creating in a multitude of forms.

Ars Poetica, Ars Poiētikē (ποιητική) If we are to retain Ars Poetica as the title (and a meaningful description of the contents) of Horace’s poem, we would be well served by reconfiguring our understanding of it as an Ars Poiētikē in the widest sense—­not merely an Art of Poetry, but an Art of Producing, Making, Creating. The resonances of the verb ποιεῖν in the terms poema (ποίημα), poesis (ποίησις), and poeta (ποιητής) would have been more obvious to the ancient ear, as Ford 2002 has persuasively argued with regard to the Greek tradition, and as indeed each word’s primary definition reflects: ποίημα is “anything made or done,” ποίησις is “fabrication, creation, production,” and ποιητής the “maker” of both.85 We, on the other hand, are no longer attuned to recognizing poets as “makers” and a poem as a “made ­thing” (ποίημα), no dif­fer­ent in its crafted nature from, say, a chair. That Horace had in mind the Greek verb ποιεῖν and the notion of making and crafting in broad and far-­reaching ways as he wrote the Ars Poetica is suggested by the preponderance of verbs that mean “make, do, fashion, shape, craft, mold, create” throughout the poem and, furthermore, by the pointed juxtaposition of ­these with words derived from ποιεῖν itself.86 Horace’s holistic view of h­ uman activities has been recognized elsewhere: throughout the Satires, readers have seen that his “ethical princi­ples are fully consistent with his aesthetic princi­ples” and that “­every lesson he teaches on the proper style of life contains a second 85. LSJ gives further definitions of ποίημα as “work, invention, product, poem . . . ​deed, act,” of ποίησις as “(fabrication, creation, production) of poetry” and “poetic composition, poem,” and of ποιητής as “inventor, lawgiver, workman, composer of a poem, author, poet, composer (of ­music), author (of a speech, opp. deliverer of it).” See further Ford 2002: 93–157, who locates the development of this new conception of song as verbal craftsmanship and, thus, artifact in the early fifth ­century and in par­tic­u­lar in the hands of Simonides, Pindar, and Bacchylides. That Horace would seem to have a­ dopted this language to describe the making of poems in the AP is especially thought-­provoking given the prominence of Pindar throughout the Odes. 86. ­There is, it should be noted, no etymological connection between ποιεῖν and its derivatives and the terms Horace employs to translate them: the cognate of facio is τίθημι (Watkins 2000: 17); of fingere, θιγγάνειν (Watkins 18); and of pangere, πήγνυμι (Watkins 61); while formare may derive from Proto-­Indo-­European *merph-­(Watkins 55). Ποιεῖν and its derivatives may be traced back to PIE *kwei-­2, “to pile up, build, make,” which, while productive in the Greek branch, does not appear to have been so in the Latin.

125-81019_FerrisHill_HoracesArsPoetica_5P.indd 221

8/8/19 8:40 AM

222  ch a p te r 4

literary application on the proper style of poetry.”87 Yet this way of reading has not been applied to the Ars Poetica, perhaps b­ ecause it is only once we have worked our way through the entire poem, and many times over, that the essential sameness in Horace’s imagining of writing, painting, sculpting, pottery-­ making, and even living, begins to emerge. Throughout the Ars Poetica, perhaps even more than elsewhere in his poetry, Horace creates meaning by repeating specific terms, clustering them close together or layering them at intervals. The reader is compelled by t­ hese means to intuit that a par­tic­ul­ ar idea is or ­w ill prove impor­tant, but this is achieved subtly, often in such a way that on the second or third recurrence of a term we feel we have met it before but cannot say so for certain or recall where exactly that might have been. Horace toys throughout the Ars Poetica with what I see as its essential notion of the artist as maker in precisely this manner, deploying from his opening image the key verbs fingere, facere, and formare, all of which translate ποιεῖν, as well as ­others that evoke it aurally. Having traced how an artist (pictor) can “join” (iungere)88 and “attach” (inducere) ele­ments, Horace debuts fingere to describe the production of something that is not desirable: veluti aegri somnia, vanae fingentur / species, ut nec pes nec caput uni reddatur formae (7–8). Forma, the noun from which formare is derived, appears alongside fingere to denote the “single shape” (uni . . . ​formae) that is h­ ere not achieved, though its complementary adjective formosa has appeared to describe the (apparently lovely) female portion of the hybrid creature in line 4. Poetis is found in the same line as formae but without any specified relation to fingere or formare as yet. In ­these first nine lines of his poem, then, just as he has heralded or hinted at the many themes explored in chapters 1 (Humano), 2 (Pisones), and 3 (Amici, risum), Horace has begun the pro­cess of priming his reader to be alert to the verb fingere, the set of words forma-­formosa-­formare, and the term poeta and its relatives. Horace begins slowly a­ fter the casual association of forma and poeta, first employing verbs of making in conjunction with terms semantically related to poema, while still avoiding the term itself. So words, verba, as “­things made” are explored at 45–72: the “author of a promised song-­poem”89 (promissi carminis auctor) is exhorted to exercise caution and discernment in both his use of words 87. Freudenburg 1993: 186; see further Mette 1961, Bramble 1974, Cody 1976, Gowers 1993, Möller 2004, Rutherford 2007: 253–55. 88. As Sharrock 1996: 125 points out, “iungere (‘to join’) is an obvious signifier of hybrids.” 89. Carmen is formed from cano, “sing,” with both often etymologized (incorrectly) in antiquity as related to thoroughly Latinate Camena (“Muse”; cf. Pfaffel 1981: 109–10, 150–55, Coleman 2001: 70–72, Ernout and Meillet 2001: 90), and means in the first place “a solemn or ritual utterance,” e.g., “chant, spell, incantation, oracle, prophecy,” and subsequently “song, poem, play”

The E n d of t he P oe m   223

(in verbis etiam . . . ​/ dixeris egregie, notum si callida verbum / reddiderit iunctura novum) and his creation of them (fingere cinctutis non exaudita Cethegis / continget dabiturque licentia sumpta pudenter; / et nova fictaque nuper habebunt verba fidem), the physicality of the image accentuated by procudere nummum, 59, the image of striking a coin. Horace compares the life cycles of words and h­ umans, rounding out the passage with mortalia facta peribunt and thereby adding the verb facere to the growing nexus of terms having to do with making and creating. Terms including versus (e.g., 75, 89) and carmen (e.g., 91) occur as frequently as might be expected in a poem having to do with poetry, and they are increasingly interwoven with verbum (e.g., 97, 106) and poeta (87, as Horace won­ders why he is called a poeta if he cannot accurately depict t­ hings as they r­ eally are) and, at last, poema (first at line 99), the nexus expanding ever further. Horace is working up to figuring the writer as a maker/shaper/creator at 108–27. He begins format enim Natura prius non intus ad omnem / fortunarum habitum, presenting Nature herself as a creator that shapes a temperament for each of us individually, as argued in chapter 1 (and she comes on the heels of natu­ral features themselves refashioned by ­human agency, 63–68, though Nature is destined to refashion them back to their original states once sufficient time has passed). Following a list of dramatic characters, with the instruction that each must speak in the way proper to it, Horace turns to his addressee and commands him, aut famam sequere aut sibi con­ve­nientia finge. / scriptor.90 The association suggested between the writer (scriptor) and Nature as entities that create and shape o­ thers, suggested by the similarity in sound as well as sense of the verbs formare and fingere, is cemented as Horace reframes the guiding princi­ple issued at 119, sibi con­ve­nientia, as sibi constet (127), introducing the edict by saying, si quid inexpertum scaenae committis et audes / personam formare novam. The writer is addressed in the second-­person singular as at 119, but the verb used of the pro­cess by which he creates a new character (novam personam) is formare:91 not merely a variation on fingere used eight lines ­earlier but pointedly the verb used to describe the activities of Natura ten lines before that,92 and “poetry, song.” The contrast between the Latin-­ness of carmen and the Greekness of poema is evidenced at TLL s.v. poema 10.1.2494.31–32: “(carmen) [poem]a a Graecis dictum.” 90. Even if the punctuation of Shackleton Bailey 2001 is ­adopted, as ­here, with a period at the end of line 119, scriptor remains the person addressed by finge; Brink 1971: 59 accordingly preferred to print finge, / scriptor. 91. Persona, moreover, has valences far beyond “a character in a play,” as it is plainly prima facie used at 126: the source of our “person,” it meant already in Latin “the ­actual being of someone, individual personality/person” (OLD) in addition to “mask, character.” 92. Solmsen 1963: 474 notes that the earliest Greek cosmogonies (Hesiod’s Works and Days and Theogony) contain instances of “divine ‘fashioning’ or ‘making’ ” that employ the verbs ποιεῖν

125-81019_FerrisHill_HoracesArsPoetica_5P.indd 223

8/8/19 8:40 AM

224  ch a p t e r 4

and more suggestive of sculpting than writing. The passage is rounded off by carmen (129) and verbum (twice in 133). Through this deliberate and painstaking pro­cess of interleaving and overlap, Horace has begun to suggest both that the writer does something rather more profound and hallowed than drafting a rudimentary outline when inventing a new character and that writing a new character is a form of creation with analogues beyond the literary sphere. Pseudo-­Acro touches on this when he insightfully draws attention to the difference between retelling a story and inventing one: commenting on line 119, he says, “omnis poeta aut certam historiam dicit, aut fingit,” as he shows that some poets merely relate, ­others create. What so far might seem merely like a sprinkling around of possibly loaded terms grows more deliberate and insistent as the Ars Poetica progresses, and the seeding of the poem’s opening third with the marked language and ideas outlined ­here allows Horace to bring t­ hese and the connections among them to fruition ­after its midpoint and as he heads into its close. ­There is, in fact, a curious pause from lines 136–219, in which Horace manages to talk at length about writing in the context of mythological themes, about plays, their characters, and what the audience wants, and about certain apparently Aristotelian precepts for drama, all without employing the terms poeta or poema, carmen, versus, or fingere, facere, or formare. A ­ fter this, however, Horace begins g­ ently with a proto-­ tragic song-­poem (carmine qui tragico vilem certavit ob hircum, 220), then mentions verses and words once again (versus, 231; verba, 235), before fi­nally presenting a juxtaposed pair of words of exactly the sort we have been led to feel we might be given: fictum carmen (“a made song-­poem,” 240). ­There is, however, as yet no overt explanation of exactly how and why ­these two words might be connected: unlike poema, carmen does not at its root mean “a made ­thing,” nor does the verb of which fictum carmen is the direct object, sequar (“I s­ hall follow”),93 add anything etymologically or aurally. Terms derived from Greek ποιεῖν undergo an explosion in frequency beginning at 263. Having seen poema only once and poeta twice so far (in 262 lines of verse), we may well be surprised to be confronted with them a total of nine times in the space of forty-­five lines (from 263 to 301). “Not just any judge sees unrhythmical poems,” Horace begins, “and an unmerited indulgence has been given to Roman poets” (non quivis videt immodulata poemata iudex / et data and πλάττειν, and he discusses also similar conceptions in Plato and Aristotle (although the Poetics is generally thought of as most impor­tant for Horace’s AP, “de partibus animalium and de generatione animalium include a good number of instances in which nature acts like a modeller, a painter, a cook, a carpenter, a h­ ouse­builder, or a channelbuilder,” 489; see also Shorey 1933: 349 on similar imagery in Plato’s Timaeus). 93. ­Future, as also at Wickham 1891: 412, Rudd 1989: 190.

The E n d of t he P oe m   225

Romanis venia est indigna poetis, 263–64), deploying the rhe­toric of criticism and positioning himself as iudex supreme (see chapter 3). Poemata reappears at 276 as Horace recites the history of drama beginning with Thespis (whose poems are actually carted about, object-­like, on wagons). And as Horace turns to Rome he speaks of “our poets” (nostri . . . ​poetae, 285), somewhat jarringly using the Greek term for ­these earliest of Latin authors, who wrote comoediae praetextae and togatae (288). Poetae has further attention drawn to it by facta at 287 as the poets in question “dared to abandon Greek traces and celebrate domestic deeds” (vestigia Graeca / ausi deserere et celebrare domestica facta). ­These vari­ous instances are followed by poeta again at 291 (si non offenderet unum / quemque poetarum limae l­ abor et mora), 296 (excludit sanos Helicone poetas / Democritus), and 299 (nanciscetur enim pretium nomenque poeta), with carmen in the interval at 292 for good mea­sure (vos, o / Pompilius sanguis, carmen reprehendite).94 Horace is, we may now reasonably feel, building up ­toward some ­great revelation, as indeed happens momentarily: o ego laevus, qui purgor bilem sub verni temporis horam! non alius faceret meliora poemata. verum nil tanti est. ergo fungar vice cotis, acutum reddere quae ferrum valet exsors ipsa secandi; munus et officium nil scribens ipse docebo, unde parentur opes, quid alat formetque poetam, quid deceat, quid non, quo virtus, quo ferat error. (301–8) In the context of the ars-­ingenium debate, as he pre­sents madness as (facetiously) interchangeable with divine inspiration and inborn talent, Horace throws up his hands and declares that had he stayed mad/inspired no one ­else would or could have made better poems. H ­ ere at last, facere (“do, make”) is juxtaposed with poema (“poem,” but more critically, “a made ­thing”). This seminal idea of art as object and of the artist as its maker is given additional nuance in two ways. First, Horace emphasizes the sound and sense of faceret . . . ​poemata with the phrase formet . . . ​poetam, further borrowing a 94. Newman 1967: 76 is concerned primarily with distinguishing which uses of poeta are “contemptuous or playful” (he identifies the instances at 291, 420, 455, 466 as such), which “neutral” (9, 264, 285, 296, 307, 333, 463), and which must be explained in some other fashion (87, 299, 372), since he considers the term opposed to vates, yet as I argue ­here the term poeta does far more in the Ars Poetica than provide a foil for vates. I disagree also with Newman’s assertion that “in a work which ranged over Roman poetry generally and had many allusions to the Greeks it was inevitable that poeta should find wide employment” is a complete explanation for why Horace uses poeta fourteen times in the AP and vates only twice.

226  ch a p t e r 4

meta­phor from a concrete and elemental form of making—­a whetstone sharpening a blade. The image is used to express not how a written object can be brought into existence, but rather how the maker of this object can. Even as formet poetam reinforces faceret poemata, however, it also recasts it, and Horace goes on to state explic­itly that it is the making of the maker that is his aim. His persona, it transpires, is less interested in making written objects (nil scribens ipse) than in making the makers of t­ hose written objects. And yet, as much as poets must be physically molded (so that they may physically mold), Horace’s self-­re-­presentation as a whetstone prefigures the most explic­itly physical shaping and hammering out of verses that appears in the Ars Poetica: Quintilius, if he is given writing that needs further work, o­ rders the “poorly rounded verses” to be returned “to the anvil” (male tornatos incudi reddere versus, 441), as though they ­were metal items that had come out roughly when melted and poured and w ­ ere in need of fine tuning (this image is itself followed by arboricultural recidet, 447). The two phases are distinct yet complementary: metal is not worth sharpening u­ nless it has already been ­shaped into approximately the desired object (one would not apply a whetstone to a hunk of metal that has not been poured to at least resemble a sword); at the same time, while the written object requires smelting and hammering into shape, it is in Horace’s imagining the object’s maker (the poet) who requires honing, as though he ­were the blade, in order that he may then further shape his verse-­objects (through incisive cuts, to continue the meta­phor). Poeta/poema and verbs synonymous with ποιεῖν in Latin are associated rather more loosely with one another, as well as with synonyms for poema including versus, carmen, and verbum, over the 150 lines that remain, yet with certain notable results. Song-­poems (carmina) are molded (fingi) at 331, where the use of carmen rather than poema may be pointed: given the priorities of the con­temporary Roman educational system, it is unreasonable to hope for poemata of the sort that might be produced by the round-­mouthed Greeks with their Muse-­given ingenium (323–24), since even their Latin equivalents (carmina) are not worth preserving (linenda cedro et levi servanda cupresso, 332). Against this rather disheartening backdrop, Horace immediately turns to what poetae ­ought to strive for, that is, being useful and giving plea­sure (aut prodesse volunt aut delectare poetae, 333), commanding t­ hese makers, “Let t­ hings made for the sake of plea­sure be very close to true ones” (ficta voluptatis causa sint proxima veris, 338). With a jangling juxtaposition of Greek and Roman, Horace indicts Roman efforts as compared to Greek still one more time: “the lofty Ramnes-­tribe men pass over austere poems [or, ‘Greek made ­things’]” (celsi praetereunt austera poemata Ramnes, 342). With ut pictura poesis (361) Horace introduces a third ποιεῖν-­derived term, one unusual enough, both in general and in the Ars Poetica (where it appears

T he E n d of the P oe m   227

only this one time),95 that it should cause us to prick up our ears and consider not merely why Horace is employing poesis at this moment, but also why ­there is such a concentration of its relatives, poema and poeta, in this poem. Relying on Porphyrio’s claim that Horace co-­opted Neoptolemus of Parium’s “praecepta eminentissima” for his Ars Poetica, it has been supposed that this trio of nouns—­ poema, poesis, poeta—­reflects the fact that Horace is following Neoptolemus and adopting for his Ars Poetica Neoptolemus’s ostensible tripartite arrangement. Th ­ ere are, however, two clear prob­lems with this interpretation. First, understanding “praecepta eminentissima” to mean “a division of the material into three parts: poema, poesis, and poeta” would be a rather unintuitive reading of the phrase. Indeed, as Greenberg explains, despite the widespread supposition other­w ise, “Porphyrio never claimed that Neoptolemus supplied the orga­nizational princi­ple of Horace’s work,” and in fact Porphyrio’s next sentence reads, “The first precept is about sequence/coherence” (“primum praeceptum est περὶ τῆς ἀκολουθίας”), that is, nothing to do with poema, poesis, or poeta.96 Second, ascribing to Neoptolemus the under­lying structure of Horace’s Ars Poetica assumes that Horace both knew well and endorsed Neoptolemus’s material. Our knowledge of Neoptolemus’s work, however, comes from Philodemus, who moreover had before him only “the introductory statements of [Neoptolemus’s] treatise on poetry” and “did not know Neoptolemus’ work first-­hand, but was working with the handbook of a certain Philomelos.”97 Philodemus is plainly not well disposed t­ oward the ­earlier writer and his theories: Philodemus declares it “ridicu­lous” (καταγέ[λ]α[ϲ]τον) that Neoptolemus considered the poet to be a subdivision of poetry (ποιητική) and “astonishing” that he made “the subject alone belong to poesis since the poema and every­thing ­else in general are part of it also” ([τὸ] τῆ[ϲ] ποιήϲεω[ϲ] εἶναι τ[ὴ]ν ὑπόθεϲιν [μ]1νον, καὶ τοῦ ποήματο[ϲ καὶ] πάντων ὅλωϲ τῆϲ ποιήϲ[ε]ωϲ ὄντων).98 As Armstrong 1995: 218 explains, Philodemus is so committed to the unification of form and content that for him “any treatise simply dividing the subject into poetry, poem, 95. “Only ­here in H.” and “not the technical term of the literary critics” ­either (Brink 1971: 371). 96. Greenberg 1961: 284 (cf. 266: “­there is simply no evidence” that “Neoptolemus’ use of the terms poiema, poiesis and ‘poet’ ” refers “to the disposition or arrangement of his treatise on poetics”); Porphyrio at Hauthal 1866: 649. On the precepts for which Horace perhaps relied upon Neoptolemus, see also Nettleship 1885: 173–82 with Armstrong 1968: 7–8, Immisch 1932: 21, Greenberg 284–85, Laird 2007: 134–35. 97. Greenberg 1961: 266. 98. On Poems 5 cols. xi 23–­xii 26 = cols, xiv 23–xv 26 Mangoni. Text (papyrological notations and apparatus criticus removed) and translation from Oberhelman and Armstrong 1995: 249–50 (cf. Armstrong 1995: 219); cf. also Greenberg 1961: 276–84, Porter 1995.

125-81019_FerrisHill_HoracesArsPoetica_5P.indd 227

8/8/19 8:40 AM

228  ch a p t e r 4

and poet is wrongly conceived from the outset.”99 Moreover, “to Philodemus, the third division, the poet, is mostly an illusion: the poet and his training are only vis­i­ble to be criticized in the poetry and poem anyway.”100 Above all, however, “even where such a division is con­ve­nient, ­these subjects ­will continually ‘bleed’ into each other: for the w ­ hole poem is like a web or tapestry (οἷον ὕφη) of which the separate bits of poetry are the small parts or ‘works’ (ἔργα).”101 Philodemus also makes a fine distinction between poesis and poema in this passage, saying that “a poesis is also a poema, for example the Iliad; but the first thirty verses are just poema, but not poesis” (ἡ μὲν [γ]0ρ πόηϲιϲ καὶ π[όημά γ’ ἐϲτίν,] οἷον ἡ Ἰλι[άϲ,] οἱ δ[ὲ πρῶτοι] ϲτίχοι τρι[άκ]οντα τα[ύ]τηϲ πόημα μ[έ]ν, οὐ μέντοι ποί[η]ϲιϲ). Given Horace’s interest in and implicit approval of Philodemus on many other m ­ atters, a blind adherence on Horace’s part to a writer and work excoriated by Philodemus seems, on balance, less likely. In his usage of poesis at 361, Horace is e­ ither himself broadening the senses of the term considerably, from “poetic fabrication” to “the act of creation,” or (more likely) making use of an existing set of variable meanings.102 The single appearance of the term poesis in the Ars Poetica is especially infuriating as its precise sense ­there must be constructed from the context (whereupon it becomes a ­matter of interpretation) or supplied from external sources of vastly dif­fer­ent times and places. At the most basic level poema and poesis overlap considerably. Greenberg 1961: 269 explains: “As Posidonius said, a poiesis is also a poiema, i.e., it is a series of words (usually quite lengthy) which embodies 99. Cf. Oberhelman and Armstrong 1995: 249. 100. Armstrong 1995: 218; cf. Oberhelman and Armstrong 1995: 249. 101. Armstrong 1995: 218. 102. Although they had been impor­tant terms in Hellenistic literary theory, where “poiema characterized a literary production that was poetic in form, poiesis was a literary production that was poetic in form and content” (Greenberg 1961: 286), the meanings of poema and poesis in Latin w ­ ere no longer well established by Horace’s day (Greenberg 1961: 263, 267, 283; 266: he critiques the scholars who claim that Horace had gotten the poiema, poiesis, poietes scheme from Neoptolemus, yet lack any clear sense of what poiema and poiesis meant to ­either ancient author). “During about a ­century before the time of Horace (in Lucilius, in Varro, in Philodemus) poiēma had meant something like a short poem or a passage of poetry, poiēsis had meant a long poem,” Wimsatt 1970: 140 (in Brophy et al. 1970) explains, and he suggests that this is what Horace means by poesis and poemata in the AP: “some smaller, some larger,” while nevertheless noting “the shifty meaning” of t­ hese two terms. The concern with size is evident at Lucil. fr. 378–82 (e.g., “poema is a small part of poesis” [pars est parva ‘poema’ ] while “poesis is a ­whole work” such as Homer’s Iliad “and much larger than that which I ­earlier called a poema” [illa ‘poesis’ opus totum . . . ​/ et maius multo est quam quod dixi ante ‘poema’]) and Varro, Men. fr. 398, cf. Greenberg 1961: 268. Greenberg’s diachronic and detailed investigation of the terms remains valuable, and see TLL s.v. poesis 10.1.2514.60–2515.8 for further examples.

125-81019_FerrisHill_HoracesArsPoetica_5P.indd 228

8/8/19 8:40 AM

The E n d of t he P oe m   229

poetic features of a formal sort, but it must also express a content which is poetic.” More concretely (yet also so wide-­ranging as to be possibly even more unfathomable), “poiesis is the act of composing poetry, the product of such composing, and poetry itself in almost any vague and nebulous sense. Poiema is the product of poetic composition, again in a vague and general way.”103 If, in the first place, we suppose that Horace is likening poesis and pictura purely as objects (“a poem is like a painting,” as the phrase is so often understood), ut pictura poesis could further mean that made (written) objects and painted/ drawn objects are alike in that both are a form of mimesis: they take their vivas voces from life (318).104 The sense of neither word, however, is ­limited to the concrete (another aspect in which they are alike): poesis (unlike poema) may denote “the act of composing poetry, e­ tc.” and pictura itself may likewise be used “concerning the action/act or art of painting” (de actione vel arte pingendi).105 Although this is harder to grasp than the more straightforward notion that a written artistic object is the same as a drawn artistic object, it would satisfy the reader’s desire for poesis to be somehow distinct from poema: although both can and do mean “a t­ hing made,” surely Horace in line 361 had a reason for using the far less-­common term poesis (­unless he was seeking somehow to complete, for the sake of completing or signaling, Neoptolemus’s triad). In light of this, Horace’s engagement with Neoptolemus becomes rather more nuanced and multifaceted than Porphyrio’s straightforward scholium long led scholars to believe. The fact that the three classes that make up the poetic art in Neoptolemus’s view are easily discerned in the Ars Poetica need not mean that Horace is rigidly and unthinkingly relying upon them, as has often been supposed, or even that he endorses them, especially since the next item detailed has to do with “ ‘natu­ral succession’ or ‘consistency’ ” (ἀκολουθία), a form of thematic “associative connection” that does in fact describe the structure of the Ars Poetica.106 If Philodemus criticized Neoptolemus’s division (though apparently not his regard for ἀκολουθία), however, Horace’s position is harder to discern. We may infer that Neoptolemus was not an outdated or recondite 103. Greenberg 1961: 267. OLD defines poesis as “the production of a poet, poetry; a poetical composition,” i.e., not so much the act itself as the final product, which is thus not clearly distinguishable from poema (the same may be seen in the TLL entry: poesis is both “opus a poeta versibus factum” and “potius de arte vel actione,” 10.1.2515.30–36). 104. So, too, Russell 2006: 329 (as noted also at Reinhardt 2013: 509–10): “­Here, by relating poetry to painting, Horace makes a special point: both are forms of imitation, traditionally paired together.” 105. TLL s.v. pictura 2 (10.1.2082.57–2083.19); especially illustrative are Cic. Inv. 2.5, De or. 1.73, 3.26, Nat. D. 2.35. 106. Laird 2007: 137.

230  ch a p t e r 4

thinker for Horace: his theories ­were regarded as sufficiently current and impor­ tant that they w ­ ere subject to discussion and evaluation not only during Horace’s lifetime but could even be called upon and critically examined by Horace himself. Alternatively, it is pos­si­ble that since Neoptolemus’s theories w ­ ere critiqued by Philodemus, by versifying t­ hese theories Horace “intensified the fun” in his “mock didactic poem about poetics” dedicated to Philodemus’s patrons.107 While Horace certainly co-­opted recognizably Neoptoleman terms, then (poeta and poema would likely have raised no eyebrows, but the addition of the more unusual poesis, and only once at that, cements the connection), he puts them to use to construct a universe all his own, in which diverse creative activities are bound by the same rules. With the etymology and resonances of poema, poesis, and poeta in mind, we may now reconsider what more Horace means when he says, o maior iuvenum, quamvis et voce paterna / fingeris ad rectum et per te sapis (366–67). The character of the elder Piso son, who, we understand by now, harbors within the world of the Ars Poetica a wish to become a maker of poetry, is himself still in the pro­ cess of being made (fingeris). This might render him unsuitable for the task, yet it also allies him with the would-be poet mentioned ­earlier whom Horace, the pedagogic whetstone, aids in forming (formet . . . ​poetam). Mediocre poets (mediocribus . . . ​poetis), unlike mediocre ­lawyers (369–72), are presented as an adynaton ­because a poem (or any work of art) exists to please the spirit (sic animis natum inventumque poema iuvandis, 377). Despite all ­these warnings, however, he “who does not know how to nevertheless dares to craft verses” (qui nescit versus tamen audet fingere, 382), Horace exclaims in exasperation, employing the same verb—­fingere—­that he has already used to describe the way in which the older Piso ­brother is himself still a work-­in-­progress (fingeris) and the shaping of empty images (vanae / fingentur species, 7–8), of new words (fingere cinctutis non exaudita Cethegis, 50), of new stories (sibi con­ve­nientia finge, 119), and of poems that Romans cannot craft (carmina fingi, 331). The next words addressed directly to the elder boy become especially marked. Plainly taken, tu nihil invita dices faciesve Minerva; / id tibi iudicium est, ea mens (385–86) would mean “you ­will say or do nothing with Minerva unwilling; you have that judgment, that [good] sense,” but given how fingere, formare, and facere have been deployed throughout the poem, and given that the Pisones (or at least some of them) wish to try their hand at poetry, Horace’s words take on the air of a loaded warning, as we can now see in facies the vari­ous senses of “make,” “create,” “craft,” and even “shape,” all in the sense of creating art (poema). The sense, then, becomes “you ­w ill make nothing artistically” without Minerva’s cooperation, and yet Horace, while implying strongly that in this par­tic­ul­ ar case Minerva ­will 107. Cf. Frischer 1991: 65, Armstrong 1993: 192, 222–23.

The E n d of t he P oe m   231

not provide the assistance necessary to bring about the desired result, somewhat flatters his addressee among all this rudeness by imputing to him the good sense—­iudicium, mens—to realize the futility of his doomed efforts. ­There follow two more half-­formulations of the recurring idea of a poem as a made ­thing and of the poet as its maker—­half-­formulations in that they do not fully complete the ποιεῖν-­making equation yet participate in Horace’s program nevertheless. First, natura fieret laudabile carmen an arte (“­whether a praiseworthy song-­poem is made by nature or by art,” 408; which perhaps makes us think of Nature as a maker at 108–9); and second, nolito ad versus tibi factos ducere plenum / laetitiae (“do not lead a man full of gladness to verses made by you,” 427–28). Nestled between ­these is the remarkable phrase poemata pango, as Horace bemoans the fact that writers, unlike athletes and musicians, have not under­gone the arduous training pro­cess with its sweat, pain, and fear recognized as fundamental to other professions (412–15): an satis est dixisse ‘ego mira poemata pango. occupet extremum scabies; mihi turpe relinqui est et quod non didici sane nescire fateri?’ (416–18) Horace’s point—­that every­one thinks he can become a writer—is the same one expressed memorably at Epistles 2.2.117: “taught and untaught, we write poems indiscriminately” (scribimus indocti doctique poemata passim). The boastful tone of ego mira poemata pango is enhanced by the unusual verb pangere, which denotes in the first place highly physical actions such as inserting or driving a plant or a stake into the ground and, second, the setting of bound­aries, including the establishing of treaties. When used figuratively, it refers to composing of any kind (writing, speaking, singing, ­music), often with an archaic or solemn sense.108 Although this par­tic­ul­ ar writer (who ­ought not to boast since his writings are sub-­par) is also conceived of as a physical maker of verse-­objects, the nuances of pangere render his pro­cess rather more brutish than that described by ποιεῖν or facere—­still a form of physical making, Horace suggests, yet without 108. OLD and TLL; see also Maltby 1991: 440, 448. Among the ancient sources cited by Maltby, Paulus-­Festus explains that the verb is applied to verses b­ ecause they are fixed into wax (“versus pangi vel figi in cera dicuntur,” Lindsay 1913: 212; cf. TLL 10.1.207.69–70). Lucretius uses the term several times to describe nothing less than his own poetic enterprise ([versus] quos ego de rerum natura pangere conor, DRN 1.25; quod obscura de re tam lucida pango / carmina musaeo contingens cuncta lepore, 1.933–34 = 4.8–9), both showing how the verb may be applied to the poet’s activities and hinting that perhaps Horace is skewering Lucretius and his boastfulness at AP 416 (for Brink 1971: 399, who cites comparanda from Ennius and and Cicero, Lucretius’s usages are mock-­heroic and archaizing, and so Horace’s phrase “burlesques nonsensical claims made in the ­grand style,” though without being pointedly critical of Lucretius, as I suggest).

232  ch a p t e r 4

the necessary shaping and molding. Pangere may look and sound like a mash-up of fingo, pingo, facio, and ποιεῖν, but its connotations are not positive. Even as Horace strives to show the essential unity of vari­ous manifestations of the creative pro­cess, then, pango adds further nuance by being degrading to t­ hese par­ tic­u­lar verses as they are being created and to their creator. The phrase poemata pango is complemented by another that is equally evocative, alliterative, and potentially physical, though contributing slightly dif­fer­ent resonances to the overall system in the Ars Poetica of writing verses as a form of making: carmina condes (437). In the first place of founding cities, as in the Aeneid (tantae molis erat Romanam condere gentem, “such a burden was it to found the Roman race,” Aen. 1.33; cf. 1.5, 275–76, 522, e­ tc.), condere may be used also of putting a poem together, especially in conjunction with carmina as in Cicero (Tusc. 4.4), Lucretius (DRN 5.2), Virgil (Ecl. 10.50–51), Livy (27.37.7), and elsewhere in Horace (Sat. 2.1.82). Writing a poem, like founding a city, requires that a suitable topic (or location) be identified, a­ fter which begins the l­ abor of putting words (or bricks) into their proper places, one by one. Horace hints at this overlap in his description of Amphion as Thebanae conditor urbis (394): the mythological founder of Thebes in concrete terms, it was through song that he caused the city-­forming rocks to move, an achievement described in a cata­log of pre-­ historical and historical poets, found­ers of the art (conditores). The second and final notable clustering of poema/poeta comes in the last fourteen lines of the Ars Poetica, where Horace introduces yet another variation on a verb of making to describe the writing of poetry. ­After repeating at 453 the term scabies from 417—­unpleasant and unsettling enough when used once, its grotesqueness now heightened by the fact that its e­ arlier mitigating context, a ­children’s chant,109 has been stripped away—­Horace says that sensible p­ eople ­will avoid a poet: vesanum tetigisse timent fugiuntque poetam / qui sapiunt (455). First, the creation of verses by this madman is conveyed with the grotesque verb ructatur, rendered all the more striking through its juxtaposition with sublimis. Poeta appears again at 463 as Horace speaks of “the Sicilian poet,” Empedocles, and concludes the episode with sit ius liceatque perire poetis, “let ­there be the right and let it be permitted for poets to die” (466)—­three instances of poeta in twelve lines, and each time in line-­final position as though a Homeric oral formula. The closing notes of the Ars Poetica, madness and divine inspiration, culminate in speculation as to what the poet could have done to cause the madness that afflicts him: did he perhaps urinate on his ­father’s ashes (the theme of ­fathers and sons resurfacing for the final time) or disturb 109. On the origins of the phrase occupet extremum scabies (417) see Suetonius fr. 198 (Reifferscheid 1860: 346) with Brink 1971: 400, OLD; Armstrong 1993: 215 notes that the words are “specially delightful” if we understand the AP as addressed to a newly togate boy.

125-81019_FerrisHill_HoracesArsPoetica_5P.indd 232

8/8/19 8:40 AM

T he E n d of the P oe m   233

a spot made sacred by a lightning strike? What­ever the case, “he keeps making verses,” versus factitet, facere now appearing in its frequentative form, factitare, to suggest not merely an ongoing and repeated action but one that is performed compulsively, even against the subject’s w ­ ill. Fi­nally, echoed by facit, fecit, and fiet that surround it, as well as sonically by furit, frangere, and fugat—­a notable accretion of seven verbs beginning with the same letter in the course of eight lines—­factitet stresses, for the final time, poetry as a t­ hing made. This insistence on words and verses as physical artifacts becomes all the more remarkable in light of the fact that, as much as Horace talks about writers (scriptor), he never once uses the verb scribere itself in the Ars Poetica with versus, verbum, carmen, or poema as its direct object.110 Verses, words, and poems, then, are not t­ hings to be written; rather, they are ­things to be formed, crafted, molded, ­shaped, and brought into a physically real existence. The significance of the way in which Horace deploys Latin words in the Ars Poetica that derive from ποιεῖν to draw attention to their etymology and to broaden the understanding of what it means to be a poet is rendered more remarkable through comparison with other texts. In the context of Horace’s own writings, his interest in and continual play upon poema, poesis, and poeta in the Ars Poetica are especially striking since he does not do the same in his Satires, Epodes, or Odes.111 In fact, even his Epistles, the writings closest to the Ars Poetica both in time and in spirit, show ­little kinship with the Ars Poetica in the use of poema or poeta to reveal their etymologically derived senses: ­either term with a verb of making, fashioning, or forming is found on only one occasion, as Horace says, “But let the one who desires to have made a proper poem take up the spirit of an honorable censor together with his tablets” (at qui legitimum cupiet fecisse poema, / cum tabulis animum censoris sumet honesti, Epist. 2.2.110), reflecting the self-­critique and ­labor that must go into a lasting and genuinely poetic poem. Other­wise, poems are something that one, most mundanely and 110. Scribere is found in compounded form (describitur, 18; descriptas, 86; conscripti, 314) or used intransitively (qui scribitis, 38; quo scribi possent numero, 74; scribamque licenter?, 265; scribendi recte sapere est et principium et fons, 309). It is used transitively only at 306, nil scribens ipse, where Horace professes himself to be writing “nothing,” and 387, si quid tamen olim / scripseris, where the direct object is an anemic (ali)quid, “something, anything.” See chapter 1, n100, on Horace’s avoidance of scribere in his Odes. 111. In the Satires ­there are eight instances of poeta (with a ninth in the spurious eight-­line opening to 1.10, on which see Zetzel 2018: 59–61) and three of poema in book 1, along with one of poeta and one of poema in book 2—in short, plenty of opportunities for etymological playfulness that Horace does not take advantage of in the way he does in the AP. The Epodes, Odes 1–3, and Carmen Saeculare contain no instances of poeta, poema, or poesis; poeta occurs only at Carm. 4.2.33 and 4.6.30.

234  ch a p t e r 4

wholly unlike in the Ars Poetica, “writes” (Epist. 2.1.117, 2.2.65–66), though the evocative phrase poemata pango appears, again with derogatory undertones, at Epistles 1.18.40.112 The poet, likewise, may “fashion the tender mouth of a boy” (os tenerum pueri balbumque poeta figurat, Epist. 2.1.126) or “shape his chest with friendly teachings” (pectus praeceptis format amicis, 128), he may “do” many bad ­things to himself (multa quidem nobis facimus mala saepe poetae, Epist. 2.1), but he does not fashion, shape, or make poems in the Epistles with any regularity. The poet himself is similarly not s­ haped or molded, merely praised or admired (Epist. 2.1.61–65), terrified and chased away (Epist. 2.1.182), and even harassed by madness (qui minus argutos vexat furor iste poetas?, Epist. 2.2.90) or “spit out” by an age (veteresne poetas / an quos et praesens et postera respuat aetas?, Epist. 2.1.41–42). Against the backdrop of the other Augustan poets, poema, poesis, and poeta come into similarly sharp relief. Virgil uses poeta five times in the Eclogues and once in the Georgics,113 yet none of ­these instances exploits the notion of the poet as maker: he is divine (divine poeta, Ecl. 5.45, 10.17), the recipient of an ivy wreath upon his birth (Ecl. 7.25), and even a creation of the Muses (et me fecere poetam / Pierides, Ecl. 9.32–33; cf. Ecl. 10.70), though never himself an agent that actively creates.114 The same applies to its uses in Propertius,115 Tibullus,116 and the works of Ovid that predate Horace’s death in 8 BCE,117 where the closest we see to an expansion of the poet’s activities is that he “discovers” in Propertius (2.1.12, invenio causas mille poeta novas) or “puts together” in Ovid (Am. 2.1.10, conposuit casus iste poeta meos). The complete works of Virgil, Propertius, and Tibullus, moreover, along with the early works of Ovid listed above, contain among them not a single instance of e­ ither poema or poesis. Slightly e­ arlier, 112. Horace warns Lollius that he should not obliviously or selfishly work on his poems (nec . . . ​poemata panges) when his rich friend wishes to hunt. 113. Newman 1967: 51 provides a chronology and accounting of the uses of vates and poeta in the works of Virgil, Horace, Propertius, Tibullus, and Ovid. At Verg. G. 3.90, the reference is to Grai . . . ​poetae. 114. Notable in many of ­these instances, moreover, is the alliteration: pastores . . . ​poetam (Ecl. 7.25); poetam / Pierides (9.32–33); nec te paeniteat pecoris, divine poeta (10.17); poetam . . . ​Pierides (10.70–72). Cf. also Prop. 3.1.19, Pegasides . . . ​poetae; Tib. 1.4.61, Pieridas, pueri, doctos et amate poetas; 2.5.113–14, at tu, nam divum servat tutela poetas, / praemoneo, vati parce, puella, sacro; and 3.4.43–44, poetae / . . . ​Pieridesque. 115. 1.7.21, 3.1.19, 3.9.42, 3.17.20, 4.6.75. 116. 1.4.61, 2.5.113, 3.4.43. 117. Amores (published 16 BCE or l­ ater) has poeta at 2.1.2, 2.1.10, 3.1.16, 3.12.19, 3.15.13. The two other works that may have been or likely w ­ ere written before 8 BCE are the Heroides, which only has poeta at 19.137 and 21.110, and poetria, of Sappho, at 15.183, and Medicamina Faciei Femineae, which contains no instances of poeta. Ovid strongly f­ avors vates, using it liberally in the Amores, Metamorphoses, and Fasti over poeta (Newman 1967: 51).

125-81019_FerrisHill_HoracesArsPoetica_5P.indd 234

8/8/19 8:40 AM

T he E n d of the P oe m   235

Catullus also once juxtaposes poema with facere (hoc, iucunde, tibi poema feci, “I made this poem for you, dear friend,” 50.16) while at 22.15, a poem is conceived of as an object that can be touched (poemata attigit). Other­w ise, Catullus’s poetry contains ten instances of poema/poeta, many of which exhibit alliteration (14.5, 14.23, 16.5, 36.6, 49.5, 49.6; cf. n114), aligning his usages of the terms more with ­those of the other Augustan poets than with Horace. In fact, the only close con­temporary of Horace who seems interested in the etymology and etymologically based meanings of poeta, poema, and poesis is, perhaps not surprisingly, Varro. At De lingua Latina 6.77, Varro works to tease out the subtleties of three verbs that communicate ­doing and making: agere, facere, and gerere. Facere, he explains, describes what the poet does (poeta facit fabulam) and agere the actor’s part (contra actor agit et non facit); accordingly, “a play is made by a poet, not acted, [and] it is acted by the actor, not made” (sic a poeta fa­bula fit, non agitur, ab actore agitur, non fit). Gerere is reserved for the activities of generals. Although he does not explic­itly state that he considers poeta to be derived from ποιεῖν and thus to mean “maker,” Varro, in a manner fully in line with his explicatory practices throughout De lingua Latina,118 suggests it through juxtaposition. Isidore of Seville similarly credits to Suetonius (On Poets 2, early second ­century CE) an etymology that aligns poeta with ποιεῖν and further associates both with the (unrelated) term ποιότης: since the p­ eople wanted to worship the gods with due reverence they wrote poems, and “this kind of ­thing was given the name ‘poem’ (poema) ­because it is fashioned (efficitur) with a certain beauty known as ποιότης (i.e. ‘quality’), and its makers (fictores) ­were called ‘poets’ (poeta)” (“id genus quia forma quadam efficitur, quae ποιότης dicitur, poema vocitatum est, eiusque fictores poetae,” Etymologiae 8.7.1–2).119 Other ancient authors are more explicit.120 Commenting on the first three lines of Terence’s Andria-­ prologue, which themselves make use of the poeta-­facere connection,121 Donatus explains, “Poets are so called [from the act of making] from Greek ποιεῖν, ‘to make’ ” (“poetae [a faciendo] dicti sunt, ἀπὸ τοῦ ποιεῖν”).122 Fi­nally, Diomedes ponders the specialization of the term poeta to mean a maker of verses: “he who makes a verse is called, by the Greek word, a poet, but craftsmen, 118. See Ferriss-­Hill 2014. 119. Translation by Barney et al. 2006: 180, lightly edited. 120. See further Maltby 1991: 481, TLL s.v. poeta 10.1.2515.38–44. 121. Ter. An. prol. 1–3: poeta quom primum animum ad scribendum adpulit, / id sibi negoti credidit solum dari, / populo ut placerent quas fecisset fa­bulas, “when the poet first applied his mind to writing, he believed that this task alone was entrusted to him, (namely) that the stories which he had made please the p­ eople.” 122. Donat. ad Ter. An. prol. 3., Cioffi 2017: 10.

236  ch a p t e r 4

although they equally ‘do/make’ something, are not called poets” (“is qui versum facit dictus ποιητής, cum et artifices, cum aeque quid faciant, non dicantur poetae”).123 Varro, Suetonius (via Isidore), Donatus (and with him Terence), and Diomedes thus all provide compelling evidence for a widespread and well-­ established understanding of the Greek etymologies ­behind the Latin terms poema, poesis, and poeta, and for the senses of t­ hese as “the ­thing made,” “the creation/fabrication” (possibly also “the act/pro­cess of making”), and “the maker.”124 Moreover, the medieval commentary tradition offers some hints that other readers of the Ars Poetica have seen Horace playing in the poem upon ­these terms. The “Materia” commentary concludes its preface to the Ars Poetica by explaining the work’s title, ­w hether Poetria or De Arte Poetica (or Ars Poetica):125 “for poio, pois is ‘I make, you make.’ Whence poesis or poetria, i.e. a creation [fictio] or anything made [figmentum], and the poet is the one who makes [fictor]” (“Nam poio pois est fingo fingis. Inde poesis vel poetria, id est fictio vel figmentum, et poeta id est fictor,” Accessus 7.4–5).126 Similar is the beginning of the prologue to another anonymous twelfth-­century text of the poem: “­Here begins the ‘Book of Poetics,’ i.e. that deals with instructing poets to compose their poems well. Alternatively it is the ‘book of poesis,’ that is, fiction/making [fictionis]: for he teaches how to fabricate [fingere] fittingly—­the ancients said poio, pois, whence we have poesis and poeta, that is, the one who makes [­factor]” (“incipit liber poetriae id est qui agit de instructione poetarum 123. Ars Gramm. 3, Keil 1857: 1: 491.18–19. 124. Two further passages, one from Philodemus and one from Cicero, may also inform the pre­sent discussion. In On Poems 5, Philodemus uses the participle ποιῶν in addition to the noun ποιητής (e.g., col. 10), and although Greenberg 1961: 275 suggests that “the poion selects and arranges words so that they have poetic form, i.e., he produces poiema,” and that “just as ­every poiesis is a special sort of poiema, just so the poietes is a special sort of poion, i.e., he is the poion who produces a poiema which is also a poiesis,” ποιῶν serves also to draw attention, even more than ποιητής, to the nature of the poet as “the one making.” Second, Cicero said of Philodemus that “he makes a poem so lively, so clever, so elegant, that nothing can be made more clear-­voiced” (poema porro facit ita festivum, ita concinnum, ita elegans, ut nihil fieri possit argutius, Pis. 29; cf. Clay 1995: 13). The intriguing (if speculative) possibility arises that a) Philodemus, too, had played upon the idea of the poet as maker; b) Cicero, aware of this, reflected it in his attack on Lucius Calpurnius Piso Caesoninus; and c) Horace played upon poema facere in the AP knowing that Philodemus had done the same (and that some readers would recognize this). 125. See my introduction. Text from Friis-­Jensen 1990: 338, translation from Copeland and Sluiter 2009: 556. 126. Prior to this explication of poeta as fictor, the association is hinted at with “Utilitas huius operis est scientia poetandi, id est faciendi bona carmina” (“the utility of this work is the science of poetic composition, that is, making good verses”).

125-81019_FerrisHill_HoracesArsPoetica_5P.indd 236

8/8/19 8:40 AM

T he E n d of the P oe m   237

ad bene componenda sua poemata, sive liber poesys id est fictionis: docet enim con­ve­nienter fingere—­poio pois dicebant antique, unde et poesys et poeta id est ­factor dicitur”).127 Th ­ ese medieval commentators, then, seem both to have understood poeta more generally, as ancient writers also did, to have the sense of “maker” inherent within it, and further to have seen Horace’s Ars Poetica as concerned with the poet-­as-­maker. In light of this expanded definition of the terms poema and poeta, the patch sewn on that describes the work of Orpheus et al. (391–407) becomes integral to the poem in a new way. Not merely a testament to the civic128 functions of the poet modeled on Aristophanes’s Frogs, the ­whole becomes in Horace’s hands a paean to the poet as the true creator of any part of the oikoumene worth living in and to the fundamental unity of all forms of creation. The vast range of basically unpoetic activities with which the poet is credited—­Orpheus teaching wood-­dwelling men the proper way to eat or soothing wild animals with his song; Amphion moving rocks with his song; and men, now living in cities, receiving from poets laws, correct religious customs, and preparation for war—­ shows how all forms of making are interconnected in the Ars Poetica. This theme radiates outward into the rest of the poem from this vignette of the earliest poets and their activities, its significance for the Ars Poetica as a w ­ hole cemented by the fact that Horace is, as is evident from the discussion above, swimming against Augustan poetic currents to resurrect for poema and poeta the senses they ­were able to possess in Greek lyric poetry.

A Conclusion: Ars Poetica, Ars Vivendi The sermo-­like character of the Ars Poetica has often been noted. Scaliger already, for example, thought the Ars Poetica satire,129 and numerous scholars list correspondences between the Ars Poetica and Horace’s Satires,130 even using ­these 127. Text and translation (lightly edited) from Friis-­Jensen 1988: 137; reprinted at Copeland and Sluiter 2009: 557–58. Cf. also the eleventh-­century Scholia Vindobonensia (Zechmeister 1877: 1): “in hoc libro est intentia Horatii tractare de poetica arte, id est, arte fingendi et componendi. poesis enim graece, latine dicitur figmentum.” 128. This well-­established understanding of the passage is evident at, e.g., Rostagni 1930: 110–11, Brink 1971: 384–86, Antolín 2002: 222–23. 129. “Adeo sine ulla docet arte ut Satyrae proprius totum opus illud esse videatur,” cited at Armstrong 1995: 189; he notes that Frischer’s 1991 argument for the AP as “a parody with a detailed purpose: a hostile parody of Aristotelian criticism and critics by an author essentially Epicurean in his views” continues the Scaligerian tradition and “seems to advance the debate and improve its quality.” 130. E.g., Norden 1905.

125-81019_FerrisHill_HoracesArsPoetica_5P.indd 237

8/8/19 8:40 AM

238  ch a p te r 4

as evidence for the Ars Poetica’s closeness in date to the Satires;131 for the thesis that Horace’s Ars Poetica was based upon a satire of Lucilius;132 and, naturally, to argue that the Ars Poetica is itself a satire, ­whether b­ ecause it exhibits the genre’s characteristic parrhesia133 or adopts the persona of the speaker of the Satires.134 Such observations, however, are e­ ither made in passing, as though they w ­ ere not of much consequence; or they are marshaled in the ser­vice of other, ancillary arguments; or, fi­nally, they are put forth with the goal of determining once and for all ­whether the Ars Poetica is ­really, truly a satire, or an epistle. I espouse adopting an approach in line with that of ancient thinkers, who as noted in the introduction did not rigidly distinguish between the va­ri­e­ties of hexameter verse that we term Satires (or verse satire more generally), Epistles, and Ars Poetica. That Lucilius’s books contained a poem upon which Horace was thought to have based his Ars Poetica, or that Persius’s sixth satire takes the form of an epistle (as vari­ous of Lucilius’s did, too)135 also reveals the essential overlap among t­ hese forms. Any distinction among them, then, becomes one of focus: that of Horace’s Satires is ­human be­hav­ior, that of the Ars Poetica is lit­er­a­ture (the Epistles admit more readily of both), yet the dichotomy is far from rigid. In the ancient Greco-­Roman tradition, the conceit that the poet’s creations necessarily reflected his true self was a widespread one, memorably encapsulated in Seneca’s phrase, talis oratio qualis vita (Ep. 114; to paraphrase, “style is the man”).136 Horace plays with this notion throughout his Satires, and although Freudenburg was able as recently as 1993 to write that the “meta­phorical possibilities” of this collection “remain virtually unexplored” (186), recent de­cades have seen a surge of interest in this mode of reading Augustan poetry. So, for example, when Horace professes his oh-­so-­traditionally-­Roman rustic humility, saying, “I go home to a dish of leek and chickpea and oilcake. My dining needs are tended to by three slave-­boys, and a white stone slab supports two 131. So Elmore 1935. 132. Fiske 1913, 1920: 446–68. For Hardison and Golden 1995: 26–28, too, the AP is satire in the Lucilian mold. 133. Michels 1944. 134. Hardison and Golden 1995: 25 say that the AP is “in the same general form as the Epistles and is thus, like them, a modified form of Horatian satire,” especially as it “continues to use the device that Horace had developed in his Satires—­the satiric persona”; at 243 they speak of “Horace’s ‘conversation poem’ (sermo) technique in the Art.” Also advocating for recognizing the AP’s sermo–­like qualities are Reckford 1969: 141, Wimsatt 1970: 138 (in Brophy et al. 1970), Segura Ramos 1989: 112, Armstrong 1993: 199. 135. E.g., the first two poems of book 5 (see Krenkel 1970: 67). 136. See especially Möller 2004.

T he E n d of the P oe m   239

cups and a ladle; a cheap salt shell is ­there too, and an oil flask with a shallow bowl, plain Campanian ceramic ware” (domum me / ad porri et ciceris refero laganique catinum. / cena ministratur pueris tribus et lapis albus / pocula cum cyatho duo sustinet; astat echinus / vilis, cum patera guttus, Campana supellex, Sat. 1.6.114–18), he also in the same breath stakes his claim on Roman Satire, an earthy genre in its infancy and consequently much given to (at least half-­ironic) hand-­wringing about its unworthiness as poetry. Or when he notes the silliness of preferring to fill a cup or jug of w ­ ater from a huge river rather than from a l­ittle spring (Sat. 1.1.54–60), ostensibly a warning against greed, since one is likelier to drown in the former, it is hard not to see resonances of the rich tradition of flowing w ­ ater as a meta­phor for poetry, ranging from Pindar (cf. Hor. Carm. 4.2.5–8) to Callimachus’s muddy Assyrian river (Ap. 108–9) and Horace’s own marked preference for clear and slender trickles.137 If the Satires, which appear to be about living, may be read as (also) being about writing, the opposite may be discerned in the Ars Poetica, and indeed Horace gestures at the coherence of poetry and lifestyle, ethics and art, throughout the poem. Campbell, striving to identify a school with which Horace’s literary worldview might be most securely connected, commented, “The fact is, Horace’s views on lit­er­a­ture are derived primarily from himself,”138 the point being not that Horace is idiosyncratic in his approach to writing and to the analy­sis of writing, but rather that what we see in the Ars Poetica being applied to and recommended for lit­er­a­ture—­moderation, circumspection, deliberate care—­are the very same approaches recommended throughout the Satires as applicable to living and existing as well as pos­si­ble in society (from which literary readings have in turn been extrapolated). Moreover, it is only through careful study of life (exemplar vitae morumque, 317) that an artist (doctum imitatorem, 318, admits of more possibilities than merely writing) ­will be able to transport the “living voices” (vivas voces, 318) he discerns t­ here into his art. Poets, if they are to be successful, should provide both plea­sure and instruction (333–34), and what they write must be more than simply believable, it must be as close as pos­si­ble to real­ity: ficta voluptatis causa sint proxima veris (338). 137. See Freudenburg 1993: 187–90 and Ferriss-­Hill 2015: 173–74, 179. 138. Campbell 1924: 68. He goes on: “This statement may sound rather bald, but I believe it to be true to a much greater extent than has been yet realized, and if true, it has . . . ​an impor­tant bearing on the prob­lem of the sources, structure, and significance of the Ars Poetica. It makes no difference if he occasionally expressed t­ hose views in terms borrowed (to a greater or less degree) from Neoptolemus of Parium or ­others; . . . ​The position he takes up as a poet is fundamentally the same in Epodes, Satires, Odes, and Epistles.” Cf. Williams 1964: 191: “B[rink] examines the literary Satires and Epistles to show ‘the recurring features of thought and opinion in all of the critical poems.’ ”

240  ch a p te r 4

Laird 2007 is right to see Horace unifying h­ ere two distinct but interdependent ancient conceptions of imitatio: the “ancient rhetorical notion . . . ​as the adoption of the style of an ­earlier author” and “the semantic conception . . . ​as the linguistic repre­sen­ta­tion of objects.” So, “in telling the imitator to look at the exemplar of life and to draw living voices from t­ here, Horace is actually suggesting that the imitation of life in poetry comes down to the imitation of real-­life utterances. . . . ​Literary language may imitate life, but it must imitate everyday language in order to do so.”139 Horace is also, I have ventured, making a second suggestion with the words respicere exemplar vitae morumque iubebo / doctum imitatorem et vivas hinc ducere voces: exemplar vitae may be understood not only to refer to “the model of real life all around us,” but to Horace’s own Satires, where he has brought ­human habits to life.140 If exemplar vitae is taken to refer to a repre­sen­ta­tion of life, then, such as may be found in the Satires, Horace is recommending that his own Satires be studied for exempla of how to write convincing characters, a reading supported by the other­wise hard-­to-­construe prefix re-­ in respicere: “look back” to—­where else?—my ­earlier writings.141 The key question, however, remains how the manifest overlap between Horace’s Satires and his Ars Poetica is to be understood. An answer I propose ­here is to view the Ars Poetica and Satires 1 as the bookends—if not ­actual then narratological—to Horace’s ­career, in par­tic­u­lar, his hexameter ­career. Although material and ideas applied to lit­er­a­ture in the Ars Poetica also appear elsewhere in the Satires as well as in the Epistles and Odes,142 it is nevertheless with the diatribe satires, as they are often termed, the first three poems of book 1 of the Satires, and further with 1.4, Horace’s first programmatic satire, that the Ars Poetica most deeply intersects, as shown in chapter 1. This correspondence does 139. Laird 2007: 140–41. 140. Modern commentaries are generally s­ ilent on the ­matter, but Pseudo-­Acro explains Horace as follows: “Hoc dicit: Consuetudinem vitae respicere iubeo eum, qui vult recte imitari, ut, cum inducat adulescentem loquentem, debeat scire, quemadmodum iuvenes solent vivere. Quid est enim comoedia? [nisi] imago vitae cottidianae.” 141. As Brink 1971: 342 points out, respicere, which he glosses as “to scrutinize,” would other­ wise not have at AP 317 “the usual meaning of this or similar verbs”: used transitively, it ­ought to mean, “look around (back) at, look around and see, notice ­behind one; look around for (something or someone)” (OLD). 142. E.g., Sat. 2.2.54–64, where Horace contrasts gluttons and gourmands with one Aufidienus who is parsimonious to a fault (“in vain you avoid that vice if you turn yourself foully ­toward another” [ frustra vitium vitaveris illud / si te alio pravum detorseris; hac urget lupus, hac canis, aiunt], roughly our “stuck between a rock and a hard place”), or Epist. 2.2, where he speaks of the need to exercise moderation in spending money (190–94) and sums up that the ideal is to be “in second place to the leaders, [but] always ahead of the last” (extremi primorum, extremis usque priores, 204), which recall AP 25–26, 31.

The E n d of t he P oe m   241

not appear to have been noticed before, and it is unlikely to be accidental. Satires 1 was Horace’s poetic debut, published in 35/34 BCE, followed some five years ­later by Satires 2 and the Epodes. The Ars Poetica, on the other hand, belongs to the opposite end of Horace’s life, poetic and other­wise: its exact date is, as discussed in the introduction, disputed, but it is generally agreed to have been Horace’s last work before his death in 8 BCE. It would appear, then, that while Horace began his hexameter writings with his fin­ger pointed firmly at ­human foibles, and that he critiqued ­these in terms that have, as has been increasingly recognized, metaliterary resonances, he concluded his poetic oeuvre by turning his eye to poetry itself, all the while never losing sight of the ­human ele­ment informing it. The Ars Poetica, then, may be read as Horace’s earliest satires writ large(r) and transferred to the poetic arena. It alludes to his Satires through its form, content, and style while si­mul­ta­neously revealing, even betraying, its own nature as much the same. If “all of Horace’s poetry is metapoetry,”143 what is to be done with a poem that in fact is, or appears to be, overtly about poetry? If the advice Horace dispenses in his Satires about how to live reveals upon further probing to contain advice on how to write, then perhaps his explicit instructions on how to write contained in the Ars Poetica ­ought rather to be read as instructions on how to live. Just as the Satires, in the absence of this ambiguity, read as a tiresome litany of moral failings, so the Ars Poetica becomes enriched by being recognized as an Ars Vivendi, an approach that reveals also, once again and on a far larger scale, the interconnectedness of all h­ uman activities in Horace’s world view. The Satires are—as the genre’s etymology from satur, “full,” conveys—­stuffed and multiple, and they repay reading beyond. The Ars Poetica, equally dense, benefits from the same approach. Given the Ars Poetica’s many points of contact with the Satires and the fact that what passes for poetic advice in the Ars Poetica is often rather dubious, the poem clearly exists within the context of a larger Horatian schema, in which living and writing are to be governed by the same, interchangeable ethics. Just as literary meanings are generated for the Satires, so ones that reach beyond the narrow sphere of lit­er­a­ture emerge from the Ars Poetica. This is a not unexpected move from a writer of satire, who begins this, the final poem of his life, with the word humano and invites us in, his friends (we flatter ourselves in thinking), for a laugh at the incongruity caused by a failure in visual artistry. It is not so much a question, however, of w ­ hether Horace viewed (the teaching of) writing as more impor­tant than (the teaching of) living, or vice versa.144 Rather, his lessons and views on ­either make the best 143. Lowrie 1997: 97; a summary she says “may be true.” 144. Becker 1963: 64–112 (at Oliensis 1998: 207, Hardie 2014: 49), for example, describes how Horace transfers rules from recte vivere to recte scribere; cf. also Grimal 1964.

242  ch a p te r 4

sense when taken together and applied holistically, and as indivisible from one another.145 Making one’s way in the world, in what­ever fashion, is an art and a craft. One of the cruxes of Horace’s Satires, as of ancient satirical and comic writings more generally, is the degree to which they are serious. The bind in which the Ars Poetica has ended up is that in being framed as “more serious” (than, for example, the Satires), it has repelled ­until very recently the reading in it and the reading into it of any satirical ele­ments at all. Through the paradidactic premise that Horace acts as praeceptor artis poeticae in the poem, the bar of seriousness was raised to precisely the point where scholarly consensus would not admit of any unseriousness at its core. Yet we would do well to remind ourselves, with John Henderson 1989: 90, of an essential difference between didactic and satire: “Whereas, on the w ­ hole, ‘Student Textbooks’ d­ on’t usually invite . . . ​their readers to pick holes in their unargued prejudices—­Sorry: judgments—­, it could well be a standard feature of ‘Satire’ that you be expected to ask yourself continuously ­whether you agree, are supposed to agree, keep catching yourself agreeing, and so on (or ­whether you dis-­agree, ­etc.).”146 We can easily perform this exercise with the Ars Poetica. Even in the case of ­those pronouncements whose veracity we may not question, we repeatedly find ourselves asking why Horace is telling us a par­tic­u­lar ­thing—­again, not a question that readers of “Student Textbooks” typically need to ask themselves. This is not to say that the Ars Poetica is not didactic (or epistle, or satire); rather, much like Virgil’s Georgics, it is not didactic in the manner in which it purports to be.147 The ingenuity of the Ars Poetica lies in the fact that, once we have dispensed with the misconception that the poem’s instructional value lies in the vari­ous precepts contained within it and that Horace is acting as a straightforward literary guide, we may come to discern that it is in fact literarily didactic as an exemplar, as paradigmatic poetic artifact and object, and that Horace is, in this deeper and more covert fashion, a genuine praeceptor artis poeticae. It is the fashion, moreover, espoused by Philodemus, whose fingerprints can be discerned all over the Ars Poetica, from the passing interest evinced in anger, death, and property 145. Reckford 1969: 141 comes close to saying the same: “Horace does not say that to be a good poet one must first (in time) be a good man; no poet would say that. But he feels that the philosophical attitude somehow has primacy.” 146. Cited also at Gellar-­Goad 2012: 173, who uses it to show how Lucretius’s DRN engages with satire. 147. Cf. Thomas 1988: 7 (the Georgics are “didactic in appearance but without the intention of teaching its apparent subject”); Reinhardt 2013: 523 similarly concludes of the AP that “it would remain unnecessary to deny the narrative a genuine engagement with its primary theoretical subject ­matter.”

The E n d of t he P oe m   243

management to the abiding importance of friendship, teaching, and criticism. Philodemus, like Horace, did not believe that the poetic art could be taught; rather, it could only be demonstrated by t­ hose who had mastered it, and only discerned in such displays in turn by ­those both predisposed and motivated to master it themselves in turn. Accordingly, the following epilogue considers the Ars Poetica’s afterlife: not the survival of the poem itself or the uses to which it has been applied, examined already in the introduction, but rather the ways in which practicing poets have read Horace’s poem and crafted their own artes poeticae in the tradition of the archetype.

Epilogue Receiving the Ars Poetica

if the Ars Poetica’s poetic qualities have not always been clear to scholars of lit­er­a­ture, they seems to have been more evident to the practicing writers who, inspired by Horace’s poem, wrote artes poeticae of their own. My approach to the poem’s centuries of reception in this epilogue is therefore a par­tic­u­lar one: I focus on t­ hose artes poetriae, Arts of Poetry, Artes Poéticas, and Arts Poétiques that reveal their authors to have read the Ars Poetica as I read it ­here: not a handbook to be applied to the mechanical tasks that together constitute composition, but an artistic achievement in its own right; a hexameter poem by a poet whose other hexameter poems include Satires; and a work concerned not with what it may explic­itly state but rather with the implicit and interlaced themes introduced from its opening vignette: ­human life and be­hav­ior (chapter 1); ­fathers, sons, intergenerational relationships, and life cycles (chapter 2); friendship, criticism, and laughter (chapter 3); madness, death, and the sublime, and the overlap among all forms of creative endeavor, as well as between the didactic and satiric modes (chapter 4). In E ­ ngland in 1175, t­ here appeared the Ars Versificatoria by Matthew of Vendôme—­the first of the medieval artes poetriae written in Latin, and the first of what would be a ­great many original compositions on the model of the Ars Poetica over the next millennium in a variety of languages.1 Matthew’s work was swiftly followed by ­others written in Latin in prose, verse, or a combination of the two, including Geoffrey of Vinsauf’s Poetria Nova (­England c. 1202; partially revised 1208–13), Gervase of Melkley’s Ars Versificaria (­England, 1208–16), John of Garland’s Pa­ri­siana Poetria (c. 1220; revised 1231–35),2 and Eberhard of

1. Matthew’s work may, of course, merely be the earliest extant example, and some have argued that the anonymous Saint-­Omer Art of Poetry in the same style may be ­earlier (see Specht and Chesnutt 1987). 2. See further Lawler 1974, Copeland and Sluiter 2009: 640.

244

r e ce i ving the a r s p oet ic a  245

Béthune’s Laborintus,3 with the tradition culminating in the anonymous Tria sunt (­after 1256, before 1400).4 The traditional explanation for this sudden eruption of an entirely new form of reception for Horace’s poem has held that the Latin of the first ­century BCE was increasingly alien to student readers.5 On the one hand, the Ars Poetica remained central to the educational system at all levels; on the other, in order for it to remain so, a need was felt for a version that was more accessible to students than a mere Latin text with elucidating notes,6 even as the accompanying commentary tradition had grown robust, the “Materia” commentary being a prime example.7 Yet this narrative, which views the artes poetriae as “a systematized and updated version of the Horace-­ cum-­commentary package,”8 does not adequately explain why their authors did not simply produce more and better commentaries of the conventional type;9 indeed many ­others did continue to write precisely such works and 3. Dates from Woods 2010: 15; see also Specht and Chesnutt 1987: 15, Copeland 2016b: 28–29. Eberhard cannot be dated with certainty (Friis-­Jensen 1995b: 362). 4. Known by its incipit, this is one of “the most comprehensive of the medieval arts of poetry” (Copeland and Sluiter 2009: 670), bringing together material from the treatises by Matthew, Geoffrey, and Gervase, as well as from Horace’s AP and its twelfth ­c entury commentaries. 5. Copeland and Sluiter 2009: 545: “Students came to [Latin] from their own vernacular languages” and certain works “explic­itly recognized that students would be learning Latin as a second language.” 6. Copeland and Sluiter 2009: 548 speak of the “need in the grammatical curriculum for a focused and consolidated approach to composition” that the AP “could no longer fulfill by itself.” 7. Critical edition at Friis-­Jensen 1990; cf. 1991 and Copeland and Sluiter 2009: 551–56. Named for its opening word (“Materia huius auctoris in hoc opere est ars poetica”), it was likely composed in France between 1125 and 1175 (Friis-­Jensen 1990: 322). Friis-­Jensen 1990: 319 suggests that the “Materia” commentary be itself viewed as “in a sense also the first ‘new’ art of poetry,” seeing it as a “missing link” between the phases of early medieval commentary and artes poetriae (Friis-­Jensen 1990, 1995b, 2007: 300–301, Copeland and Sluiter 2009: 551–56). Friis-­Jensen 1995b: 364, e.g., describes Matthew of Vendôme’s impetus for writing as a “wish to systematize the exegetical material which generations of commentators had collected around Horace’s text.” Camargo 1988 has argued also for the artes dictaminis/dicandi—­manuals on prose writing developed in Italy in the eleventh and early twelfth centuries, that spread into France in the l­ater twelfth—as an impor­tant inspiration for the artes poetriae written by medieval Anglo-­French writers. 8. Friis-­Jensen 1995b: 393. 9. Such handbooks on rhe­toric and on grammar had been very popu­lar during the previous ­century; see Parr 1981: 2–3 and especially parts 1–3 of Copeland and Sluiter 2009.

246  E p ilo gue

recopy existing ones.10 Instead, t­ hese “manuals of a new sort” w ­ ere “self-­ consciously theoretical works on poetics,”11 suggesting that their writers w ­ ere motivated by something other than the desire for better pedagogical outcomes, even as ­these artes poetriae continued to profess their aims of teaching a classicizing form of verse and prose composition.12 So a poem by Horace appears to have become something dif­fer­ent altogether: not a document to facilitate the teaching of literacy, lit­er­a­ture, and Latin, but a guide that inspired the creation of new writings, while being itself the artistic work considered most worthy of imitation. The late twelfth and early thirteenth centuries are remarkable for being the first historical moment for which we can not merely conclude that the Ars Poetica (as opposed to works by Horace more generally) was being read but discern how ­people ­were responding to it and making use of it—­the first meaningful event in the Ars Poetica’s extensive, deep, and multiform reception. The four lengthy sections that make up the body of Matthew’s work are relentlessly instructional:13 118 paragraphs on the writing of descriptions (part I), forty-­six on the elegance of words (part II), fifty-­one on quality of expression (part III), and another fifty-­one on the execution of the material (part IV). The typical structure of a paragraph consists of a precept illustrated by a quotation from classical Latin lit­er­a­ture: Virgil, Ovid, Lucan, and Statius feature prominently among t­ hese authors, as does Horace and, especially, his Ars Poetica. So extensive is Matthew’s use of the Ars Poetica in his precepts, in fact, that some have suggested the Ars Versificatoria originated in lecture notes on Horace’s poem.14 Yet ­there are compelling points of contact beyond the obvious ones between Matthew’s poem and not only Horace’s Ars Poetica, but also other Horatian works. The Ars Versificatoria contains, for example, some overtly (and apparently gratuitous)15 sexual and scatological content (e.g., I 53, II 37–38, IV 48) more reminiscent of Horace’s Satires and Epodes. Matthew opens with a 10. See Friis-­Jensen 1995a: 232, 2007: 301–2. Friis-­Jensen 1995b: 362, Woods 2010, and Copeland 2016c: 59 rightly see the AP coexisting with the new artes poetriae. 11. Galyon 1980: 4. 12. See further Parr 1981: 3–4, Camargo 1988: 172. 13. Parr 1981: xii (cf. 14), echoing a critique not infrequently made of the AP itself, describes the work as “dull and repetitious.” See further Kelly 1966 and Copeland and Sluiter 2009: 559–72 for se­lections from the work and further explication (both are useful also on the differences among the artes poetriae and the place of each in the broader chronology and trends). 14. See Kelly 1966: 262 (who regards the work as aimed entirely at “beginners in the art of versification”), Galyon 1980: 17. 15. Cf. Galyon 1980: 16: “­There often seems to be no reason for his coarse sexual references except his own whims” and his schoolboy audience.

r e ce i ving the a r s p oet ic a  247

preemptive defense of his work to some critics (prologue 1–7), as Horace does in Satires 1.10 and 2.1, engaging in par­tic­u­lar with one Rufinus.16 And he even includes an especially sensitive portrait of Satire personified (II 6), describing the contrast between her timid and modest appearance (“fronte prodiga verecundiae, oculis indirectis”)17 and her slyness of mind (“mentis obliquitatem”), her excessive chatter (“labiis ex assidua garrulitate diffusis”), and her failure to blush even at nudity (“adeo suum pudorem praesumit dispensare, quod de corporis nuditate nequaquam erubescit”)—­a figure that markedly recalls Horace’s faux-­humble and mock-­insecure characterizations of his own Satires. While the body of the Ars Versificatoria reads as fairly mechanical, glimpses at Matthew’s authorial persona thus emerge especially from his prologue, and ­these suggest very much that he, like Horace, viewed didactic and satire as compatible in authorial and generic terms. This interrelation further suggests that Matthew appreciated the Ars Poetica as the product of the same mind that had produced the Satires, Epistles, and Epodes18 and regarded the satiric mode as intersecting with the genre of ars. If Matthew of Vendôme’s poem still resembles the Ars Poetica explicated, illustrated with examples, and somewhat reinvented, however, the Poetria Nova of Geoffrey of Vinsauf, which was to a far greater degree an original composition, marks a second phase of reception. This most popu­lar and widely read and used of the medieval artes, which differs notably from ­others in not quoting Horace directly but rather paraphrasing and reworking ideas from the Ars Poetica, quickly developed its own explicatory tradition and was disseminated early on accompanied by commentaries, the variety of which shows that the work “lent itself to teaching at all levels, from grammatical and literary instruction to professional training.”19 It also “announces in its title”—­New Art of Poetry—­ “the ambition to outdo” Horace’s Ars Poetica, which promptly came to be 16. Matthew’s attacks on this figure, identified as Arnulf of Orleans, have been taken at face value (e.g., Parr 1981: 8–13), but his pose may be seen as informed by the conventions of ancient Greek and Roman literary polemics, e.g., the “feuds” of the Old Comic poets (Rosen 1988, Biles 2011), Callimachus’s response to the Telchines, Terence’s prologues, or Hor. Sat. 1.10 and 2.1, on which, see further Ferriss-­Hill 2015. Comparable, too, are Boileau’s “utiles Ennemis” (Epître VII.58; see with Wood 1978: 86). 17. Latin text of the Ars Versificatoria from Faral 1958; on the textual prob­lems, see further Faral and Parr’s 1981 preface (xi–­xii). 18. Matthew does not quote from the Epodes, likely b­ ecause they are not in dactylic hexameters or pentameters, the two meters that make up the elegiac couplets of Matthew’s own verses as well as of all ­those he quotes from Classical Latin lit­er­a­ture (predominantly hexameter). 19. Copeland and Sluiter 2009: 594, cf. 560; see Camargo 1988: 167 (who notes the nearly 200 extant manuscripts containing the work from across Eu­rope), Woods 2010.

248  E p ilo gue

known as the Poetria Vetus.20 Many details in the instruction that Geoffrey provides are drawn directly from the Ars Poetica: the importance of harmony, where head and body go together and the w ­ hole is not marred by one defective area (60–70; revisited at 562–99); the ordering of one’s material (77–80);21 poetry as painting (747–48, 796–97); the combining of “seriousness and lightness” (“grave . . . ​levi,” 841), echoing Horace’s prodesse . . . ​delectare (333), utile dulci (343), and delectando . . . ​monendo (344)22; or the right to create new words (226–36) and revive old ones (756–64; cf. 1046–48, 1070–77, 1761–1841), for Geoffrey, like Horace, relishes language, delighting in it on the level of the individual word.23 Geoffrey’s personal and par­tic­u­lar reading of the Ars Poetica impresses beyond such moments of superficial overlap, however. The advice he dispenses is couched in the conversational tone familiar from all of Horace’s hexameter poetry. Wry and playful in the manner of the Horatian satirist and writing in the same hexameters,24 he begins by self-­consciously exploring, as Horace and Lucilius had both done, a metrical stumbling block (lines 1–8): the name of his dedicatee, Pope Innocent, does not scan (Īnnŏcēns), but calling him Pope 20. Hardison and Golden 1995: 92, 149; cf. Camargo 1988: 167, Copeland and Sluiter 2009: 595. The AP had previously been known by the title Poetria (see introduction with n35). The Latin text of the Poetria Nova (PN) printed ­here is that of Faral 1958, the translation that of Nims 2010 (lightly edited). 21. Geoffrey’s preference for the term ordo to express this idea surely owes something to Horace (AP 41–42; cf. 147), though it may be noted that dispositio, the “more technical rhetorical term” (Nims 2010: 85), is metrically problematic. 22. Other of the numerous points of contact include: Scylla (PN 157, 168; AP 145) and the creature in the opening vignette; the imagery of words entering through the ears (PN 85, 1070– 71) but the eye as “a surer arbiter than the ear” (“certior aure / arbiter est oculus,” 273–74) with AP 180–82; a mountain laboring and giving birth to a mouse (PN 449–50; AP 139; cf. Hardison and Golden 1995: 153). At 690–91 Geoffrey seems even to correct Horace and render his advice more useful: the latter’s brevis esse laboro, / obscurus fio (25–26) becomes in the former, “if you wish to be brief, first prune away ­those devices . . . ​which contribute to an elaborate style” (“si brevis esse velis, prius ista priora recide, / quae pompam faciunt”). 23. See especially his discussion of “conversion” (1588–1761; 1651: “regula quae supra docuit convertere verbum”), which Geoffrey offers as “a kind of machine for producing variations on standard expressions” (Hardison and Golden 1995: 155). While language also forms one of Matthew of Vendôme’s four areas of instruction, his discussion is not marked by the enthusiasm that permeates ­those of Horace and Geoffrey. 24. The first 990 lines of Eberhard’s Laborintus are likewise the author’s original composition in Latin verse, though in elegiac couplets rather than the Horatian hexameter (only at 775–816, 991–1005), and the Saint-­Omer Art of Poetry employs both elegiac couplets and the dactylic hexameter.

r e ce i ving the a r s p oet ic a  249

Nocent (“The Harming/Harmful Pope”) w ­ ill obviously not do.25 The description that follows of this pope as “a youth of ripe age” (“senex juvenis,” 23; 20–27) is unusual enough that we may won­der ­whether Geoffrey had in mind the ­fathers and sons and the focus on ages and aging that pervade Horace’s Ars Poetica.26 Geoffrey’s prospective writer is also, as he is by Horace, instructed to bear his limitations in mind: “Take stock of your mind and mea­sure your strength. If you are strong, dare g­ reat ­things; if you are weak, lay lighter burdens upon your shoulders” (“circumspice mentem; / et vires metire tuas. si fortis es, aude / grandia; si fragilis, humeris impone minora,” PN 294–96).27 Markedly lengthy and cutting rebukes are to be directed at “the man whose mind soars too high in prosperity” (“cui nimis in laetis mens surgit,” 276) and “the one puffed up by vaunting presumption” (“si quem jactatrix praesumptio durius inflet,” 292). While not incorporating Pisones of his own into the Poetria Nova, Geoffrey’s insistence on personal limitations and on the need to chide the arrogant (e.g., 455–60) take up Horace’s critiques of the Pisones, which I see as minimally redemptive and as offering no discernible path to becoming Horace. In addition, words are, as in Horace, reified (60–70)28 and their arrangement viewed in strikingly organic, natu­ral terms that evoke the life cycle of words from Ars Poetica 60–72 (PN 101–3, 134–41, 152, 256–63, 687–89). Even Nature herself, as throughout the Ars Poetica, appears personified in Geoffrey’s poem:29 Lucretian in heritage, she is credited with deciding what comes first and what last (112–18); she crafts and molds both Minos (159–66), whom she “adorns” (“pingit”), “refines” (“excoquit”), and “polishes” (“polit”), and a beautiful ­woman (563–97); and she is sprinkled throughout the poem as a mostly benign if sometimes intractable and capricious governing force (335, 393–408). Above all, the poet or writer is compared with a builder, a workman, a craftsman, a smith (43–59, 1615–16, 1683–84) as Geoffrey, like Horace, suggests that writing is merely one of many dif­fer­ent forms of “making,” ­every 25. See further Woods 2010: 2–5, Nims 2010: 19, 84 (even if “playing on metrically difficult proper names was common in medieval Latin poetry,” a resonance of Hor. Sat. 1.5.87 and Lucil. fr. 234–35 seems inescapable). 26. Such concerns may be seen also at PN 437–38, 674–86, 1842–52, while among the character sketches of The Saint-­Omer Art of Poetry are a youth (“iuvenis”) and an old man (“senex”); see Specht and Chesnutt 1987: 14, 19–20, 46–49. 27. Nims 2010: 86 rightly compares AP 38–40, noting that the meta­phor appears again at PN 1086, 1993. 28. Compare also PN 1065–66 “Words are instruments to unlock the closed mind; they are keys, as it ­were, of the mind” (“quae clausum reserent animum sunt verba reperta, / ut quaedam claves animi”), with AP 111 (post effert animi motus interprete lingua). 29. Matthew of Vendôme’s work likewise contains excursuses on the seasons (I 107–8) and on nature (I 111, II 3).

250  E p ilo gue

artifex being, in essence, one and the same.30 While the overarching themes of the Poetria Nova are the importance of unity and harmony, and while t­ hese are Horatian as well as Aristotelian and other­wise widespread notions governing the production and evaluation of art in the ancient world, Geoffrey of Vinsauf goes beyond ­these to reveal that he, too, perceived in the Ars Poetica its prominent currents of emotion, be­hav­ior, age, humanity, and nature. The advice and instruction provided in the Poetria Nova is far more easily put into practice than that found in the Ars Poetica. As one fourteenth-­century commentator on the Poetria Nova phrased it, Geoffrey explained “fully the art of poetry which Horace had taught confusedly and compendiously.”31 Not only are the majority of Geoffrey’s precepts given in positive as opposed to negative form, but they are capable of being concretely useful.32 Geoffrey even goes so far as to provide examples of the sorts of sections and themes about which he is giving instruction (e.g., 158–202, 276–689, 765–1060,33 1094–1576)—­poetic epideixis not through quotations from other works (as in Matthew of Vendôme’s Ars Versificatoria), but rather Geoffrey’s own verses.34 This distinguishes the Poetria Nova from all of the other extant artes poetriae,35 and reveals that Geoffrey understood the Ars Poetica as didactic ex arte rather than de arte: would-be 30. See Hardison and Golden 1995: 151–52, who label the poem’s first section “The Poet as Creator” and suggest that its “newest feature” is “its sustained comparison of artistic creation to divine creation as rationalized in neo-­Platonic interpretations of Genesis.” 31. Friis-­Jensen 1995b: 361 with n7, 2007: 301 with n39; the commentator is identified as Pace of Ferrara. 32. E.g., plan and outline before writing/building (PN 43–59); the function of an introduction and conclusion (71–76); use a proverb for your beginning (126–33), e­ tc. Negative prescriptions may nevertheless be found, e.g., 230–31. 33. Ad 844–52 Nims 2010: 88 says: “Geoffrey’s habit of combining example with precept is well illustrated ­here.” 34. The lack of quotations from other poets is all the more notable given that Geoffrey’s Documentum is structured like Matthew of Vendôme’s Ars Versificatoria or John of Garland’s Pa­ri­ siana Poetria, where the author’s explications in prose alternate with illustrative examples in verse drawn mostly from Classical Latin poets. Matthew of Vendôme also includes certain of his own writings among his exempla (albeit to a far lesser extent than Geoffrey), stating with rather Horatian disingenuousness that they are “merely for the sake of example” (“exemplorum gratia,” I 49; cf. 47): “let nothing in the following descriptions be understood as asserting my poetic ability” (“nihil in sequentibus descriptionibus esse dictum assertive . . . ​intellegatur,” I 49; cf. I 47; trans. Galyon 1980). Lawler 1974: xvii notes that John of Garland’s Pa­ri­siana Poetria also contains “many homemade examples.” 35. It also marks it as distinct from Geoffrey’s prose work, Documentum de Modo et Arte Dictandi et Versificandi, which quotes from the AP extensively; on the relation between this and the PN, see Hardison and Golden 1995: 91–92, 149–50.

r e ce i ving the a r s p oet ic a  251

writers ­will succeed in crafting excellent verses not by applying mechanically any rules enumerated but rather by developing a style and mastery akin to the poet’s own. Quite unlike Horace, however, who does not explic­itly offer up any verse or phrase of his Ars Poetica as an exemplum to be imitated, Geoffrey draws attention to the fact that he has done this. Where Geoffrey cries out to his reader, “Look at me,”36 Horace expects his reader to discern this strain in his Ars Poetica, yet what emerges from both poems is the same: only by becoming Geoffrey/Horace—an impossibility—­will you be able to write in the way, the best way, the only way that he is not so much telling you how to do as showing that he can do. While the commentary tradition on the Ars Poetica grew in size and scope during the f­ ourteenth and fifteenth centuries, aided (or handicapped, as the case may be) by the arrival of Aristotle’s Poetics and building as ever upon the efforts of late antique and medieval scholars, to our knowledge t­ here continued no parallel development in ­these centuries of artes poetriae, original works on the model of Horace’s poem. This strain of reception experienced a second and equally sudden flowering, however, beginning in the early sixteenth ­century in the form of neoclassical arts of poetry, which ­were written in a variety of meters and languages in Italy, France, E ­ ngland, and Spain.37 ­These seem to exist wholly apart from the new Aristotelian strain dominating literary criticism,38 but perhaps the climate that facilitated them was to a degree brought about by the interest around literary critical m ­ atters generated by the arrival of the Poetics. While ­those of Boileau and Pope are the best known ­today, Bogue identifies eigh­teen such works in total, ranging in date from 1527 (Marco Girolamo Vida’s De Arte Poetica, Italy) to 1829 (Don Manuel Norberto Pérez del Camino’s “Poética,” Spain).39 With most efforts being 1,000 to 3,000 lines in length, 36. E.g., “I word the statement more skillfully when, suppressing the w ­ hole, I imply that ­ hole from the parts in the way just exemplified,” “suppresso toto subtilius exprimo dictum / w quando modis dictis a partibus innuo totum,” 1028–29; cf. 1528–66. 37. Hardison and Golden 1995: 165 describe a “split” in approaches to the AP in this period: on the one hand, ­there is “the analy­sis of prob­lems inherent in the Art itself,” and on the other, “the use of the Art as a guide for writing poetry” as “­those who want to preserve it as a living influence on the creation of lit­er­a­ture write Horatian poems about poetry, not commentaries.” See Sowerby 2012 on “En­glish Horatianism” in the late-­seventeenth and eigh­teenth centuries, who especially notes “imitatio Horatii is a major feature of Horace’s reception in the period.” 38. Cf. Weinberg 1961: 111: although a tradition developed that combined the Horatian and the Aristotelian beginning in the sixteenth ­century, t­ here continued to be “writers who persist in the ­earlier tradition,” i.e., that of the AP alone. 39. Bogue 1975: 274–75. Bogue in fact lists twenty-­two works (274–78) that he considers to be “in the Art of Poetry tradition,” explaining (6 with n12) why he excludes four of t­ hese to reach the number eigh­teen. Other sources, e.g., Hardison and Golden 1995: 159–60, identify one poem

252  E p ilo gue

this body of work was “substantial”40 and the titles alone show the array of ways in which the Ars Poetica was being remade: into poems on satire (Walter Harte, An Essay on Satire, 1730),41 epic (William Hayley, An Essay on Epic Poetry, 1782), drama (Manuel José Quintana, “Reglas del drama,” 1791), and par­tic­u­lar aspects of literary composition (Went­worth Dillon, Earl of Roscommon, An Essay on Translated Verse, 1684; George Granville, Baron Lansdowne, An Essay upon Unnatural Flights in Poetry, 1701), and into works on the poetic tradition and heritage of individual nations ( Jean Vauquelin de la Fresnaie, L’Art poétique François, 1605; John Byrom, “An Epistle to a Friend on the Art of En­glish Poetry,” 1773). Although Bogue excludes Don Frey Lope de Vega’s Arte nuevo de hacer comedias en este tiempo (1609) as being “only marginally related” to Horace’s Ars Poetica, this work serves in the pre­sent context to demonstrate again how broadly inspirational Horace’s poem could be and indeed was.42 Undoubtedly reading Horace’s attitude ­toward the Pisones as critical and absent the potential for redemption, Boileau begins his work with a strong apotreptic against the writing of poetry by the untalented: C’est envain qu’au Parnasse un temeraire Auteur Pense de l’Art des Vers atteindre la hauteur. S’il ne sent point du Ciel l’influence secrete, Si son Astre en naissant ne l’a formé Poëte. Rash Author, tis a vain presumptuous Crime To undertake the Sacred Art of Rhyme; If at thy Birth the Stars that rul’d thy Sence Shone not with a Poetic influence. (1–4)43 that precedes that of Vida (though they, too, class Vida’s as the first in the new tradition of versified arts of poetry): the Nutricia of Angelo Poliziano (1486), “essentially a 780-­line enlargement” of AP 391–407 on “the civilizing power of poetry.” They also (243–44; cf. 172, and Golden 2010: 396) discuss Thomas Gray’s The Pro­gress of Poesy (1757) and James Beattie’s The Minstrel; or, The Pro­gress of Genius (1771). On Vida’s foundational poem, especially as it relates to the overall tradition, see further Russell 2006: 340–42, Sowerby 2006: 7–61. 40. Bogue 1975: 6. 41. Wood 1978: 19–22 is useful on the “literary satire” that spans the same centuries as the neoclassical arts of poetry and was in many cases written by the same authors, again revealing the inclination of writers of satire to write also arts of poetry. 42. The title’s Arte nuevo, moreover, like Geoffrey’s Poetria Nova, suggests a desire to trump the Ars Poetica (Poetria Vetus). A list of arts of poetry focused on the theatre may be found at Pavis 1998: 346–47. 43. The En­glish translation of Boileau used ­here is the Soames-­Dryden (at Hardison and Golden 1995, whose line numbering I also adopt; French text from Boudhors 1952), since this translation serves itself to illustrate the complex reception history of Horace’s poem. Though

r e ce i ving the a r s p oet ic a  253

Although he goes on to advise, as Horace does, that one should know one’s limits (“consultez long-­temps vostre esprit et vos forces,” 12),44 for Boileau it is pointless and an impossibility to attempt poetry without having been born a poet—­a view that Horace, though he certainly shows that he shares it, refrains from stating quite so boldly, let alone at the outset of his poem. Although this is rather obscured by the Soames-­Dryden translation, Boileau also begins his poem with a stirring evocation of the sublime, accentuating the height of Parnassus, at which the would-be poet gazes longingly from below, with the words “Ciel” and “Astre.” Having translated Pseudo-­Longinus’s On the Sublime, Boileau may well have discerned Horace’s application of such imagery at the close of the Ars Poetica, and he displays h­ ere prominently, at the opening of his own poem, his grasp of the same. Horace’s own vivid and illustrative opening image, on the other hand, is muted into general statements about the evils of unnatural, overly glittering verse (39–62, esp. 49–50). As in Horace, self-­knowledge and the importance of plain hard work (171–74, 184, 578) coexist in L’Art poétique as key themes alongside friendship: if the work is bad, a faithful friend ­will criticize it, while a faithless one (“le Flatteur,” 189) w ­ ill offer praise (186–230).45 The importance of staying the ­middle course and avoiding extremes is also a prominent theme (e.g., 47, 64–68). Perhaps above all Boileau discerned in Horace the Philodemean necessity of marrying sound (“Rhyme,” “la mesure,” “la Rime”) and sense (“Reason/Sence,” “le [bon] sens,” 28–29, 102, 154, 327–28), for he, too, impresses upon his reader how judicious placement can give rise to flawless unity: Il faut que chaque chose y soit mise en son lieu; Que le début, la fin, répondent au milieu; Que d’un art délicat les pieces assorties N’y forment qu’un seul tout de diverses parties. Each Object must be fix’d in the due place, And diff ’ring parts have Corresponding Grace: Till, by a curious Art dispos’d, we find One perfect ­whole, of all the pieces join’d. (177–80) Boileau also has fun with the figure of Horace himself, for despite the work’s being an art of poetry, Horace first appears in it as a writer of Roman understood to be that of Sir William Soames as updated by John Dryden, it is generally not pos­ si­ble to say which ele­ments each author is responsible for since Soames’s translation does not appear to survive. 44. Wood 1978: 1 also notes this as a point of contact between Horace (AP 38–40), Boileau (11–14), and Pope (52–53, 60–61). 45. Golden 2010: 396 identifies this as a major theme of Boileau’s poem taken from Horace.

254  E p ilo gue

satire, wittily censuring ­those who deserve such treatment (377–80), as Boileau emphasizes the cutting playfulness that is the hallmark of Lucilian and Horatian satire (373–76). Yet Soames-­Dryden, rendering the point that a person with faults would be unfortunate to become one of Horace’s targets (“Et malheur à tout nom, qui propre à la censure, / Put entrer dans un vers, sans rompre la mesure”), depart rather drastically from the French: “Unhappy was that Wretch, whose name might be / Squar’d to the Rules of their Sharp Poetry” (379–80). The novel words “Rules of . . . ​Sharp Poetry” serve inescapably to evoke the Ars Poetica, which is curiously absent from Boileau’s first portrait of Horace, who does not appear in L’Art poétique in his expected guise of author of the Ars Poetica ­until his last mention, in the poem’s conclusion (1081–85).46 Subtler Horatian details also run the length of L’Art poétique: the ear, hearing, and sound (105–10, 147, 155–58); excursuses on the history and development of the literary genres (111–50, 487–571); a variety of interlocutors (211–15); the figure of Nature (13, 286, 296, 534, 558, 788, 798, 840, 991); the need for characters to speak and behave appropriately (818–19); a ­father chiding his son (845–49); and hints, such as the presence of numerous literary genres47 or the example of a bad doctor who proved to be a good architect,48 demonstrate that Boileau, too, saw Horace’s Ars Poetica as concerned with writing (and living) beyond the genre of drama alone. Just as the opening of Boileau’s poem does not faithfully hew to that of the Ars Poetica, neither does its close, yet in both places t­ here may be found prominent Horatian themes. L’Art poétique opened with a strong and strident warning against writing by t­ hose who are not writers, and it closes with the importance of criticism and friendship: 46. In his second appearance Horace is envisioned as a kept poet whose lack of material worries lets him write (1041–43), and while Boileau names him in a list of poets beginning at 1050, Soames-­Dryden erase this reference, replacing him and most of the other literary ­giants Boileau mentions with En­glish equivalents (­these substitutions are attributed to Dryden rather than Soames, since Dryden describes his desire to “apply the Poem to En­glish Writers, [rather] than keep to the French Names,” Works II: 368, at Engetsu 2004: 189; cf. Hardison and Golden 1995: 173). 47. Boileau structured his poem into four cantos, with the second and third devoted to specific genres: two to pastoral, elegy, ode, epigram, “satyr” (satire), and three to tragedy, epic. 48. The mention of the architect (858–93), which leads to a Mason (883), evokes Geoffrey of Vinsauf ’s discussion of the same, which in turn goes back to Horace’s treatment of the artifex. Hardison and Golden 1995: 317–19 discuss the theme, noting its presence also in Keats’s Ode on a Grecian Urn.

r e ce i ving the a r s p oet ic a  255

Mais aussi pardonnez si, plein de ce beau zèle, De tous vos pas fameux, observateur fidele, Quelquefois du bon or je separe le faux, Et des Auteurs grossiers j’attaque les defaux; Censeur un peu fâcheux, mais souvent necessaire, Plus enclin à blâmer, que sçavant à bien faire. But ­pardon too, if, Zealous for the Right, A strict observer of each Noble flight, From the fine Gold I separate th’ Allay, And show how hasty Writers sometimes stray Apter to blame, than knowing how to mend; A sharp, but yet a necessary Friend. (1088–93) Perhaps feeling that Boileau was being rather too subtle, Soames and Dryden, having translated lines 1088–91 quite faithfully, reverse the order of the two closing lines in their version, reengineering their content such that their rendition of Boileau’s poem ends even more strongly on the note of friendship.49 This friend is left without a name, and the poet’s addressee, too, is anonymous,50 as though Boileau also saw Horace’s Pisones as generalizing rather than as specific, addressed individuals. Through a resemblance of spirit rather than of strict substance, then, Boileau’s final words evoke the close of the Ars Poetica: authors strug­gle to move beyond their craving for ­little more than a sentient audience and to appreciate one that offers criticism, even though this would lead to an improved artistic product. Pope strikes a similar note as he ends An Essay on Criticism: “The learn’d reflect on what before they knew: / Careless of censure, nor too fond of fame; / Still pleased to praise, yet not afraid to blame; / Averse alike to flatter, or offend; / Not ­free from faults, nor yet too vain to mend” (740–44). Like Boileau, Pope recognized the interwovenness of friendship and criticism in Horace’s Ars Poetica and the central role they play together in the poem. Also like Boileau and Horace, Pope was a writer of satires, suggesting not so much some essential link between an ars poetica/ars poetriae/art of poetry and the genre of satire, but rather that ­these l­ater writers ­were uniquely placed to see in Horace’s Ars Poetica the traces of his sermo51 (which had been vis­i­ble even 49. The b­ itter reciter with whom Horace’s poem ends is not altogether gone from Boileau: he may be found demoted to an ­earlier, less prominent location, 909–13 (cf. 221–24). 50. Cf. Wood 1978: 137–39. 51. See further Wood 1978 (154–84 and passim) on t­ hese three authors as writers of both satires and arts, Sowerby 1993 on Horace’s sermo as the model for An Essay on Criticism, and Moul 2010: 54–93 on Horatian libertas in Jonson. Indeed, Boileau says of himself/his poetic persona, “Pour moi, qui, jusqu’ici nourri dans la satire” (“For me, whose Verse in Satyr has been bred,” 1080).

256  E p ilo gue

to Geoffrey of Vinsauf, not himself a satirist). Both Boileau and Pope depart in a significant re­spect from Horace, however, in presenting a form of friendship that is not acknowledged as a possibility in the Ars Poetica—­that of the false criticizer. For Horace, e­ very criticizer is by definition speaking the truth, while ­every praiser should be suspected of being deceptive, but Boileau warns against ­those who, out of jealousy or ignorance, find fault with beautiful verses (916–19),52 while Pope goes so far as to open his poem with a statement about bad critics, though couched in a conversational, engaging, easy tone: “ ’Tis hard to say, if greater want of skill / Appear in writing or in judging ill” (1–2). The answer provided is that the latter is worse (3–4), for “in poets as true genius is but rare, / True taste as seldom is the critic’s share” (11–12). Yet curiously, while a “perfect judge” is pos­si­ble (233), a perfect poem is not: “Whoever thinks a faultless piece to see, / Thinks what ne’er was, nor is, nor e’er ­shall be” (253–54). In the view of Hardison and Golden 1995: 213, the “most striking feature” of Pope’s An Essay on Criticism is “its decision to turn the focus of the Horatian ‘poem about poetry’ from rules for writing verse to standards of criticism,” yet the turn need not be so surprising. Pope saw criticism to be a dominant concern of the Ars Poetica, and one inextricable from teaching, as I have also argued ­here, and chose to elevate it to an even more prominent place in his own work. Where Horace exhorts the would-be poet to know his limitations and strengths, for example, Pope instructs the critic to do the same (46–51).53 The En­glish poet even offers a solution obliquely to the prob­lem of what Horace means in describing himself as nil scribens ipse (306) before professing his intention to become a teacher, for Pope views teaching (complete with its critical bent) as the appropriate culmination to a ­career of successful writing, not as a substitution or poorer alternative for it: “Let such teach o­ thers who themselves excel, / And censure freely, who have written well” (15–16). Like Horace (and evidently unlike Philodemus), Pope views some individuals as unteachable, but whereas Horace does not suggest that giving up may be an acceptable and indeed sometimes the best course of action, Pope says precisely this: “ ’Tis best sometimes your censure to restrain, / And charitably let the dull be vain” (556–57; cf. 152–53). As Bogue 1975: 7 also discerns, in all of the neoclassical arts of poetry “the persona is an older poet giving advice and instruction to younger men interested in becoming poets,” indicating that t­ hese authors, too, saw this form of cross-­ generational didaxis to be a fundamental aspect of Horace’s Ars Poetica.

52. Similar are 920–26; at 927–40, in contrast, the ideal critic-­friend is described. 53. See Golden 2010: 398–99.

r e ce i ving the a r s p oet ic a  257

As notably Horatian as his elevation of the critic to the position of his poem’s protagonist is the central role that Pope preserves for Nature,54 even increasing it from the Ars Poetica to run almost the entire extent of the poem. Nature “affords at least a glimm’ring light” such that “most have the seeds of judgment in their mind” (20–21). She is the creating and determining force of all ­things (26–27, 52–68, 70–73, 159–60, 243–46, 486–89). And even the rules of lit­er­a­ture “discovered, not devised, / Are nature still, but nature methodised” (88–89). Horace’s advice to draw one’s characters from a close study of life and of the surrounding world is expressed as Virgil’s realization that “Nature and Homer w ­ ere . . . ​the same” (135), leading to the conclusion, “Learn hence for ancient rules a just esteem; / To copy nature is to copy them” (139–40), for “Nature’s chief master-­piece is writing well” (724). The per­sis­tence of this theme across so many l­ ater arts of poetry is remarkable, given that it has not been traditionally recognized as a key theme of Horace’s Ars Poetica, for which the focus was long on literary critical ­matters such as appropriateness and unity.55 Moreover, just as Horace views all creative activity, including writing, as essentially the same in quality, so, too, does Pope: “­Music resembles poetry,” he says, for “in each / Are nameless graces which no methods teach, / And which a master hand alone can reach” (143–45). In all cases the greatest pos­si­ble effort must be expended: “A ­little learning is a dang’rous ­thing; / Drink deep, or taste not the Pierian spring” (215–16). Even without the vivid and detailed pictures of the artist that we find in Horace, an impor­tant aspect remains, recalling especially the ending of the Ars Poetica: ­people are driven by “sacred lust of praise” (521), as it is not the critic but the artist himself who can prove to be a bad friend (519). Among all this plainly Horatian material, large and small, of An Essay on Criticism, one moment leaps out as revealing that Pope read the Ars Poetica just as generations of insightful readers had before him: Horace’s “precepts teach but what his works inspire” (660)—­ex arte over de arte, indeed. Beginning during and certainly a­ fter this second age of arts of poetry, the reception of the Ars Poetica becomes markedly more diffuse and multiform. In the place of individual works, quite similar to one another and marching across the literary-­historical landscape in discernible sequence, as had been the case with the artes poetriae and the first arts of poetry from the ­earlier sixteenth ­century, the ways in which and the loci into which the Ars Poetica is taken up 54. Sowerby 1993 is useful on nature in An Essay on Criticism. The theme is prominent also in James Beattie’s The Minstrel; or, The Pro­gress of Genius (1771), Byron’s Childe Harold’s Pilgrimage, and Goethe’s Natur und Kunst; see further Hardison and Golden 1995: 42–43, 243–48. 55. As Brink 1963: 229 notes, appropriateness/decorum has been widely accepted as the poem’s “chief subject”; the concern with the poem’s themes is especially evident in Immisch 1932.

258  E p ilo gue

become increasingly disparate and unconnected.56 Alongside the new, neoclassical arts of poetry and the “numerous annotated editions of the Art” that naturally continued to be published,57 ­there arose a group of arts of painting that evince some competitiveness with the literary branch. The possibility of reading the Ars Poetica as an art of painting may be traced back to at least Cennino Cennini, who in the early fifteenth c­ entury employed Horace’s opening plea for artistic license (for paint­ers and poets) in his handbook on painting,58 and the ­whole enterprise is often viewed as erected on the back of the dictum ut pictura poesis (AP 361).59 ­Later that ­century Leonardo da Vinci would make the same comparison (with Simonides somehow in mind)60 yet go further to assert the primacy of painting over poetry: “If you claim that painting is mute poetry, then the painter could say that poetry is blind painting. . . . ​If the poet, like the painter, is ­free in his inventions, the poet’s fictions are not as satisfying to men as paintings are.”61 A central work in this parallel tradition was Charles Dufresnoy’s De Arte Graphica of 1667, pointedly written in Latin and in hexameters at a time when arts w ­ ere beginning to be composed in vernacular languages.62 ­Those writing on painting, however, did not consider theirs to be a secondary tradition; rather, they continually equated painting and poetry as “­sister arts,” with Ludovico Dolce even declaring in the ­middle of the sixteenth ­century that “not only poets, but all writers, are paint­ers; that poetry, history, and in short, e­ very 56. Cf. Hardison and Golden 1995: 159: “The history of The Art of Poetry in the Re­nais­sance is rich and diverse, varying not only from period to period but also from country to country.” 57. Ibid.: 160. They cite the following figures: forty-­four printed editions of Horace’s Opera in Italy before 1500, ten more in Germany, and ten in France. 58. See Friis-­Jensen 1995a: 228, who is, however, skeptical as to ­whether Cennini knew the AP directly (239). 59. Hardie 1993: 120: “The long history in Western civilization of the relationship between the ‘­sister arts’ of poetry and painting has a­ dopted as its motto the three famous words ut pictura poesis”; cf. Laird 1996: 75. 60. See chapter 4, and also Friis-­Jensen 1995a: 228 with n2, 234 with n25, 239 (unlike Cennino Cennini, Da Vinci is thought likely to have read the AP). 61. “Se tu dimanderai la pittura mutta poesia, anchora il pittore potra dire la poesia orba pittura. . . . ​Se ’l poeta è libero, come ’l pittore, nelle inventioni, le sue finctioni non sonno di tanta satisfatione alli homini quanto le pitture”; translation and text from Farago 1992: 208–9 (cf. Friis-­Jensen 1995a: 228). 62. E.g., Girolamo Muzio’s “Dell’Arte poetica” (1551), Juan de la Cueva’s Ejemplar Poético (1606), and the works in French by Fresnaie and (just a few years ­later) Boileau. Dufresnoy’s poem begins with Horace’s tag, “Ut pictura poesis” (supplemented by the verb “erit,” and next reversed: “similisque Poesi / Sit Pictura”), calling them s­ isters, and also evoking Simonides with “muta Poesis / Dicitur haec, Pictura loquens solet illa vocari,” 3–4; see Allen, Haskell, and Muecke 2005 for text, translation, and commentary.

r e ce i ving the a r s p oet ic a  259

composition of learned men . . . ​is painting.”63 Just as poets could be termed paint­ers, so painting itself could be considered a form of poetry in this Re­nais­ sance milieu64—­all creative endeavor as, once more, interwoven, and in a manner that reflects the ethos of the omnicompetent statesman-­artist-­writer of this era. The reception of Horace’s Ars Poetica into drama-­criticism in the hands of Lope de Vega likewise shows the expansion of the tradition, while further revealing that some readers discerned the prominent role of drama, especially comedy, within the poem. Though Bogue dismisses the work as showing “­little Horatian influence,” Lope de Vega’s deep familiarity with Horace is not to be doubted, as the original ten-­line Latin poem in elegiac couplets near his work’s conclusion shows: beginning with the marked word “Humanae,” Lope de Vega goes on to describe comedy as a mirror of life (“Humanae cur sit speculum comoedia vitae,” 377), and he alludes further to the Ars Poetica through the juxtaposition of young men with old (“juveni . . . ​seni,” 378);65 to Roman comedy, so pervasive in Horace’s poem, through a continuing parade of stock characters (deceitful slaves, “fallaces servi,” 383; an unreliable w ­ oman, 383–84; a hapless lover, “ineptus amator,” 385); and to Horace’s Satires through his intermingling of seriousness and play.66 The seventeenth, eigh­teenth, and nineteenth centuries see translations of the Ars Poetica into a variety of languages, some for the first time;67 the commentary tradition continues, ever broadening in scope and approaches; and the poem continues to be imitated, w ­ hether directly and deliberately or in vaguer, more allusive, even second­hand ways. Th ­ ese strands of reception also begin to re­unite and overlap with one another. By way of example, Soames translated Boileau’s L’Art poétique, itself modeled on Horace’s Ars Poetica, into En­glish in 1680, a translation that was soon revised by John Dryden (1682).68 Dryden replaced Boileau’s (and Soames’s) Molière as the exemplary writer of 63. Lee 1940: 197, who calls Dolce’s view “rather more inclusive than the average.” 64. Interested readers should see further Lee 1940. 65. The line “quaeve ferat juveni commoda, quaeve seni” specifically echoes AP 169–75. 66. The phrases “lepidosque sales” (379), “gravia in mediis occurrant lusibus” (381), and “jucundis passim seria mixta jocis” (382) recall Lucilius’s “salting” of the city of Rome with his wit (Hor. Sat. 1.10.3–4) and Horace’s mention of Bion’s “conversations/sermones and black salt” (sermonibus et sale nigro) at Epist. 2.2.60, along with his effort to reconcile seriousness and play (Sat. 1.1.23–27; cf. 1.10.11 and further Ferriss-­Hill 2015: 70, 106). 67. The earliest, sixteenth-­century arts of poetry notably coincide with “surprisingly few translations” of the work (Hardison and Golden 1995: 163–64), while “no Spanish translations of the Art appeared ­until” 1591/1592 (shortly predating de la Cueva’s Ejemplar Poético, the first of the arts in Spanish), and none in Germany ­until the seventeenth. 68. Hardison and Golden 1995: 173; cf. my epilogue, n43.

260  E p ilo gue

comedies with Ben Jonson (822), and a few de­cades e­ arlier (it was published posthumously in 1640), Ben Jonson had translated the Ars Poetica of Horace (for whom he evidently felt a special affinity)69 from Latin into En­glish using the edition of Heinsius, which, following that of Riccoboni, proposed to rearrange the poem’s lines to produce a treatise of more acceptable structure.70 Dryden himself, moreover, who had also translated the Ars Poetica and had criticized Jonson’s version of the same,71 is named several times in Pope’s An Essay on Criticism (383, 458, 483, 617), as are Boileau (714) and Roscommon (725), both fellow writers of arts of poetry.72 Similarly, in his Dialogues of the Dead, Lord George Lyttelton (1760) ­imagined the deceased Pope and Boileau occupied in mild bickering amid the Elysian Fields: “Mr Pope, you have done me ­great Honour. I am told, that you made me your Model in Poetry, and walked on Parnassus in the same Paths which I had trod,” Boileau begins, to which Pope retorts, “We both followed Horace.”73 Byron, born twenty-­eight years a­ fter the publication of this “Dialogue of the Dead,” certainly had the Ars Poetica in mind as he wrote, and then rewrote, En­glish Bards and Scotch Reviewers,74 but it is his Hints from Horace, published posthumously, that “bears a very direct relationship to Horace’s AP,”75 amounting at many points to a translation or paraphrase of it.76 Many of ­these ­later works, in short, assume 69. See McGann 2007: 313, as well as Money 2007: 318, Rudd 2009: 239–40, and Moul 2009 and 2010. Jonson went on to revise his translation of the AP at least once (Moul 2010: 174–93), and the work is “notoriously ‘literal’ ” as Jonson “attempts to follow Horace’s text minutely, and to preserve where pos­si­ble even the details of Latin word order, sometimes to the extent of obscuring the meaning or movement of the En­glish line” (Moul 178; on the difficulties of preserving Horace’s “verbal corrugation” in translation, see further Tomlinson 1993: 240, Sisson 1993). 70. Jonson’s translation at Blakeney 1928. On Jonson’s use of Heinsius’s edition, see Hardison and Golden 1995: 168–69, Moul 2010: 179, n20; on Jonson and other readers of Horace’s AP in ­England, Sühnel 1996. 71. At Griffin 1996: 190 with n17; on how Dryden h­ andles and remakes Horace, see Hammond 1993. 72. Roscommon had also published a translation of the AP in 1680, and his Essay on Translated Verse (1684) begins by praising the 1682 Essay on Poetry by John Sheffield, Earl of Mulgrave ­because it “repairs so well our old Horatian way” (Sowerby 1993: 161; cf. 2012). 73. At Barnard 1973: 435–38. 74. See Hardison and Golden 1995: 257–90, especially their introduction to the poem. 75. Golden 2010: 404; see also Hardison and Golden 1995: 257–62, 291–312. Byron’s send-­off to Horace in Childe Harold’s Pilgrimage (Canto 4, 685–88) is also well known (Tennyson, for one, approved; cf. Griffin 1996: 199): “Then farewell, Horace; whom I hated so, / Not for thy faults, but mine; it is a curse / To understand, not feel thy lyric flow. / To comprehend, but never love thy verse.” 76. E.g., 1–20 with AP 1–13; 31–33 with AP 19–22; 42 with AP 25–26; 59–63 with AP 38–41; e­ tc.

r e ce i ving the a r s p oet ic a  261

not only a deep familiarity with Horace’s Ars Poetica but also knowledge of con­temporary debates about lit­er­a­ture, as well as of much of the intervening tradition.77 So the reception of Horace’s Ars Poetica becomes ever denser, not proceeding in any linear fashion but rather branching and folding back upon itself, yet always proliferating by virtue of accretions upon its accretions. As ever, each rendition of the Ars Poetica naturally continued to reflect the concerns of the age in which it was composed. The artes poetriae already reveal a Christian worldview, one that continues into the ­later arts of poetry. The genre representatives (Homer, Aeschylus, ­etc.) from ancient Greek and Roman literary history who dot Horace’s poem may find themselves replaced by more recent analogues, often natives of the new poet’s country of origin, or the two sets of figures may coexist, as each l­ ater poem deposits ever more material into the universe of arts of poetry. Boileau’s L’Art poétique would come to be “the definitive explanation of French neoclassicism,” while Pope “combines reminiscences of Horace with a sense of the awesomely power­ful image of nature revealed by Sir Isaac Newton and other pioneers of science”78—­one that goes back, w ­ hether t­ hese l­ ater writers ­were aware of it or not, to Lucretius, and before him to Empedocles and his On Nature. The “quarrel between the ancients and the moderns” is seen as providing the impetus for much of the literary production in the classical vein especially during the seventeenth c­ entury.79 In E ­ ngland, Horace was felt to be a kindred spirit, having already seeped into the collective consciousness of the young boys educated on his writings80—as boys had been for many centuries already—­and the first half of the eigh­teenth ­century saw a “Horatian outpouring”: “well over 100 imitations of individual Horatian poems, quite apart from complete editions,” including “10 on Ars Poetica,”81 among which was The Art of Cookery by William King (1708)82—­ perhaps not so strange if we bear in mind the place of Archestratus of Gela in the pre-­Horatian mock-­didactic tradition. In France, too, Horace was held in affection and regard, as both “poet-­cum-­critic” and “teacher-­cum-­thinker”—­the

77. Cf. Bogue 1975: 11, Russell 2006: 344 (who speaks of how the neoclassical arts “presuppose the close familiarity with the Ars that its use in education so long assured”). 78. Hardison and Golden 1995: 172, 174. 79. See further ibid.: 170–71, Goulbourne 2009. 80. Griffin 1996: 199–203 gives examples of poems from the late-­seventeenth and eigh­teenth centuries that show “the ease of the well bred man with his Horace,” and Harrison 2007d: 334–35 and 2009: 290, 304 notes how demonstrating familiarity with “this prestige poet” could reinforce “social capital and gentlemanly status.” See further Hopkins and Martindale 2012. 81. Money 2007: 321, giving statistics from the cata­logue of Foxon 1975: I 356–59. 82. Discussed briefly by Russell 2006: 344–45.

262  E p ilo gue

complementary personae of his Ars Poetica.83 As Goulbourne concludes, “Ultimately, studying the reception of Horace in eighteenth-­century France shows not just how eighteenth-­century French writers viewed Horace, but also how they viewed themselves.”84 In Hints from Horace, similarly, Byron uses “postclassical examples to illustrate themes common to it and Horace’s AP,”85 as Achilles becomes Drawncasir, Medea “Macbeth’s fierce dame,” and Ino, Ixion, Io, and Orestes, “Constance, King Richard, Hamlet, and the Dev­il!” (AP 119–27; Hints 173–82). Byron uses Hints from Horace, “an updated, satiric Ars Poetica,” “to criticise his romantic contemporaries from a more classical point of view.”86 As Hardison and Golden 1995: 236 put it, while the interest in Horace and in his Ars Poetica continued, “­there had been a fundamental change in the Eu­ro­pean literary climate around the m ­ iddle of the eigh­teenth c­ entury” whereby “the harsher doctrines of neoclassicism ­were already softening,” resulting in “a new relationship between The Art of Poetry and the literary world.” The simultaneous expansion and dilution of the tradition results from and reflects the changing socioeconomic and cultural environment of Eu­rope: the “combination of small, cheap books and literate citizens” and the new “popu­lar theatres,” especially in ­England and Spain, resulted in and went hand in hand with “changes in the way lit­er­a­ture was produced and the nature of the reading public.”87 The production and enjoyment of lit­er­a­ture was ceasing to be the province of the aristocratic amateur and becoming both less wedded to the upper classes and si­mul­ta­neously more professionalized. Karl Shapiro’s Essay on Rime provides, for the con­temporary reader, perhaps the most salient example of a document that reveals continually and everywhere the traces of the era in which it was created while managing to remain essentially Horatian. Published in 1945, the poem names such immediately familiar and recent figures as Ralph Waldo Emerson, Walt Whitman, and W. H. Auden, who are found alongside ones from Greece and Rome as well as the intervening tradition. Phrases such as “the wide style of the dry Americana” (759) or “the plainer scent of pure Chicago sweat” (1142), with “So went a long pro­ cession to the war / In thirty-­six” (1754–55), serve further to mark it as a

83. Goulbourne 2009: 263 quotes to this effect from Jean-­François Marmontel’s Poétique française (1763). 84. Goulbourne 2009: 270. 85. Golden 2010: 404; cf. 405–6 where he shows how Empedocles’s leap into Mount Etna (AP 463–66) becomes Budgell’s leap into the Thames (Hints 823–26). 86. Vance 1993: 200 (cf. 2015: 48–49) and Griffin 1996: 199, respectively. On Horace as read by Keats, Byron, ­etc., see Campbell 1924: 1–25, Brink 1981: 9, Hardison and Golden 1995: 240–53. 87. Hardison and Golden 1995: 248.

r e ce i ving the a r s p oet ic a  263

product of the 1940s United States.88 Yet Essay on Rime begins with a foreword that contains striking resemblances to Ars Poetica 1–4, as Shapiro works his way sidelong from “the zoo / Of rarities” (13–14) through “monsters” (15), “chimeras,” and “unicorns” (20) to “But look, / What is that perpendicular snake, that ­woman / With hot eyes gleaming in her viscera?” (23–25), a bodily reconfiguration of Horace’s hybrid, female-­headed being. Shapiro goes on to contemplate how inventive writers may be in creating new words (1238–1401; cf. AP 46–62), engaging also with the Lucretian atomological approach to language alluded to at AP 46–48, though revealing his skepticism: “But morally considered, is it not a danger / To atomize the language, construe its forms / As ions, and in the pro­cess to beget / Brain-­poems of such a nature?” (547–50; cf. also 43–45). He pre­sents portraits of both a suicidal poet (1402–52; cf. AP 463–69) and a “frenzied poet” (1852).89 This is all accomplished in an easy, conversational tone facilitated by the considered metrical choice of En­glish decasyllabic blank verse, which Shapiro himself explains in his note and acknowl­edgment as approximating “the cadence of prose speech” and “the rough flux and reflux of conversation” (2057–62), such that the work “feels spontaneous,”90 as does Horace’s hexameter sermo. Above all, Shapiro is concerned with working to articulate what ­really distinguishes poetry from prose—as in his explanation that “in the mathematical sense, rime is a power, / Prose raised to the numerical exponent / Of three or six or even n” (791–93)—­and, if something is poetry, with what makes it good (“instinct,” 773, for one)—­prob­lems treated by Horace throughout his hexameter writings. Had Shapiro read the Ars Poetica? While he names Horace91 alongside Pope, calling them “this pair . . . ​who argued in the voice of rime / And argued well” (47), Shapiro wrote Essay on Rime in a three-­month period while stationed in the South Pacific when, according to the established understanding, he had access to only two books.92 Yet certain marked correspondences such as ­those outlined above suggest that Shapiro knew the Ars Poetica intimately, and that he valued it for its sermo-­like qualities 88. See further David Leh­man’s foreword and Robert Phillips’s afterword in Shapiro 2003. 89. This figure is the protagonist of another work by Shapiro, Trial of a Poet, which features a mad poet thought to have been based on Ezra Pound; see further Phillips at Shapiro 2003: 111–15. 90. Phillips, at Shapiro 2003: 106. 91. At 1637–38 he also describes the twentieth-­century Irish poet (Louis) MacNeice as “a man of more Horatian tastes / Than any modern we know”; cf. Harrison 2007d: 341–43. 92. Phillips (Shapiro 2003: 105) identifies t­ hese as “Baudelaire’s Flowers of Evil and the Yeats edition of the Oxford Book of Modern Verse”; cf. also Golden 2010: 408–13. Lehman accordingly describes Essay on Rime as having “the virtues of an autodidact’s eccentricity,” and Shapiro himself describes how he wrote the poem “on schedule, about thirty lines a day” (Lehman at Shapiro 2003: vii).

264  E p ilo gue

and, more strikingly, for its portraits of damaged creators of verses such as are seen especially at the close of the Ars Poetica. Wallace Stevens’s Notes ­Toward a Supreme Fiction, written in the spring of 1942, is similarly a product of war­time Amer­ic­ a. Hardison and Golden note that Stevens’s authorial persona is “strikingly like the speaker of Horace’s Art” and that the poem “demonstrates the continuing vitality of the most basic Horatian tradition, the idea that the best way to write about poetry is in verse.”93 The correspondences go far beyond this, however, even as the structure of Stevens’s poem owes more to Dante’s Divine Comedy.94 The choice of an “ephebe” as addressee, an invocation repeated four times, indicates not merely that the poet’s persona “speaks like a teacher”95 but suggests that Stevens perceived in Horace’s Ars Poetica the prominent role of the iuvenis, especially as this youth is revealed at the conclusion of Notes to have become not a poet but a soldier, one of the occupations suggested for the Pisones (315). The young man’s growing up is communicated also through the poem’s insistence on the onward movement of time on the scale of both days and years and their seasons, beginning from Adam and Eve and Earth’s creation (I.4), which acts also to locate the poem within the Christian stream of artes poetriae. Like Horace, and putting into practice the pleas expressed in the Ars Poetica for poets to be allowed to bring back words that have fallen out of ­favor and to create new ones, Stevens uses unusual and archaic words, such as “closelier” and “spredden,” and coins compounds, including “ruby-­water-­worn” and “lapis-­haunted.” The phrase “satyr in Saturn” (II.1.14), moreover, suggests Horace’s own Satyrarum scriptor with its attendant wordplay on satire/satura, the genre known also to play on Saturnalia. In place of an opening chimera-­like figure, birds, bees, lions, elephants, and bears prowl the vibrantly colorful landscape of Notes (red, green, blue, violet, silver), and the poem ends with a mention of “blood” in what is the term’s fifth appearance (­after “blood,” “blooded,” “bloody,” “bloodless”), as Stevens creates for his poem a sonic and thematic texture highly reminiscent of Horace’s own. Golden identifies two further poems “of some significance” published during the twentieth ­century, both “bearing the title Ars Poetica,” that participate in the tradition: Archibald MacLeish’s 1925 “Ars Poetica” and Czeslaw Milosz’s “Ars Poetica?” of 1961.96 The latter, titled with a provocative question mark, shares with Horace’s Satires a worry evident also in Essay on Rime: that the 93. Hardison and Golden 1995: 325–26. 94. Ibid.: 327, though they note that the poem’s organ­ization becomes, like that of Horace’s AP, more “perplexing” the more it is examined. 95. Hardison and Golden 1995: 330. 96. Golden 2010: 408; see also Lehman at Shapiro 2003: viii.

r e ce i ving the a r s p oet ic a  265

pre­sent effort, in which this very ­matter is being discussed, somehow does not quite qualify as verse. “I have always aspired to a more spacious form / that would be ­free from the claims of poetry or prose,” Milosz opens, beginning his final stanza with, “What I’m saying ­here is not, I agree, poetry.” For MacLeish, on the other hand, whose “Ars Poetica” preceded Shapiro’s poem yet is more akin to the l­ ater twentieth-­century efforts discussed below in being (like Milosz’s) exceedingly compressed, the essential point had been that “A poem should not mean / But be,” encapsulating the Hortian tension between de arte and ex arte. Allied to this, MacLeish regards poems as artifacts, objects that can be touched and evaluated according to their physical properties: “A poem should be palpable and mute / As a globed fruit, / Dumb / As old medallions to the thumb.” That a poem’s words exist separate from its nature as poem-­object is clarified in the lines that follow: “­Silent as the sleeve-­worn stone / Of casement ledges where the moss has grown—­/ A poem should be wordless / As the flight of birds.” And, as ­others in the tradition also expressed, MacLeish values the atemporality of poetry: “A poem should be motionless in time.” The concentrated brews into which Horace’s Ars Poetica was distilled in the ­later twentieth ­century are by necessity quite distinct from one another, each forced by its length to emphasize only one or two concerns of the original, though often managing also to suggest ­others. The concerns elucidated in this study, moreover, go beyond t­ hose traditionally discerned in the Ars Poetica. Jorge Luis Borges’s “Arte poética” begins with three and a half opening stanzas that range over watching the flow of a river (consisting of w ­ ater and time), contemplating sleep and death, and understanding that single days and years are the units of men’s lifespans—­a reflection upon the inexorability of time that runs the length of the Ars Poetica. Borges sums up, “tal es la poesía / Que es inmortal y pobre,” suggesting, like MacLeish, poetry’s freedom from the constraints of time. For Borges, poetry ebbs and flows like the natu­ral pro­ cesses of sunrise and sunset (“La poesía / Vuelve como la aurora y el ocaso”), as he reflects the presence of Nature and its recurrent cyclicality in Horace’s Ars Poetica, as well as, again, the idea that an individual poem may possess longevity or eternity and that, beyond this, poetry as a w ­ hole exhibits a continued and continuing existence. The idea is restated through the simile of Odysseus weeping at the sight of his green and h­ umble home island of Ithaca (“Cuenta que Ulises, harto de prodigios, / Lloró de amor al divisar su Ítaca / Verde y humilde”): “El arte es esa Ítaca / De verde eternidad.” Yet “pobre,” reemphasized by “humilde,” suggests that poetry need not be richly monumental; rather, it can be small and s­ imple, even low and impoverished, that is, not-­quite-­poetry. Borges reflected upon ­these ­matters and more in his Norton lectures. Delivered at Harvard University in 1967–68 and now known by the title This Craft

266  E p ilo gue

of Verse (Arte poética),97 the six talks ­were “The Riddle of Poetry,” “The Meta­ phor,” “The Telling of the Tale,” “Word-­Music and Translation,” “Thought and Poetry,” and “A Poet’s Creed”—­vari­ous of the topics that comprise the Ars Poetica as well. Also like Horace, Borges “does not offer easy-­to-­munch-on teachings.”98 The Ars Poetica peeks in during the first essay as Borges quotes loosely from Plato’s Phaedrus, saying, “What is a book? A book seems, like a picture, to be a living being”99 (almost Horace’s ut pictura poesis), or when he casually glosses “poet” as “maker.”100 The discussion of the purple patch in the sixth piece not only engages directly with Horace’s purpureus pannus (AP 15–16), but would even appear to explicate it, as Borges (a mere three sentences ­after professing that his Latin is “very slight”)101 explains that he regards “purple patches” as “a ­mistake ­because they are a sign of vanity, and the reader thinks of them as being signs of vanity. If the reader thinks that you have moral defect, t­ here is no reason what­ever why he should admire you or put up with you.”102 Above all, for Borges “meaning is not impor­tant—­what is impor­tant is a certain m ­ usic, a certain way of saying t­ hings,” a point he illustrates by quoting a poem in Spanish and saying “the fact that many of you may have no Spanish ­will make it a finer sonnet.”103 Even as the tradition of the Ars Poetica has dissipated over several centuries, its participants ever more distant from one another and from their source, the vari­ous forms into which it can be recomposed remain strikingly in sympathy, even aside from any mechanical 97. The lectures survived as a video recording and ­were fi­nally transcribed and published more than thirty years ­after they ­were delivered and a­ fter Borges himself had died in 1986; see Mihailescu 2000: 143–50. Although he ordinarily wrote in Spanish, Borges spoke in En­glish on this occasion. It is not pos­si­ble to determine ­whether This Craft of Verse (alternatively, Of This and That Versatile Craft), not to mention the Spanish version, Arte poética, is the original title as bestowed by Borges. 98. Mihailescu 2000: 147. 99. At ibid.: 7, 125. 100. At ibid.: 43. 101. Borges similarly quotes a line from Virgil (Aen. 6.268) before (mock?) apologizing, “I won­der if I am scanning this as I should—my Latin is very rusty” (at Mihailescu 2000: 115), and he speaks e­ arlier in similarly self-­deprecating fashion of “a lesser poet—­a poet whose works I never read, but a poet whose works I have to write” (at Mihailescu 95). Mihailescu (147) marvels at Borges’s “soft-­spoken” manner in the lectures, which contrasts with “the brusque and idiosyncratic tone that characterizes most of his Spanish interviews and public lectures.” 102. At Mihailescu 2000: 111; on the Greco-­Roman tradition of a work of art as a reflection of its creator and on the creator as necessarily revealing his true self in his work, see my chapter 4 with n136. 103. Mihailescu 2000: 121, cf. also 77.

r e ce i ving the a r s p oet ic a  267

resemblance to the ur-­Ars, as each contributes “new shadings,” “fine shadings” to the story.104 Roughly contemporaneous with Borges’s poem are several ­others by Spanish-­language poets entitled “Arte poética,” named as though to extend the receptive strand of neoclassical arts of poetry written in Spain in the late eigh­teenth and early nineteenth centuries into the twentieth c­ entury and into Latin Amer­i­ca. The earliest,105 hardly much ­later than Perez del Camino’s “Poética” of 1829, which Bogue identifies as marking the end of the neoclassical group, was published in 1916 by the Chilean poet Vicente Huidobro. For Huidobro, as for MacLeish (and Horace), a poem (or verse, specifically) ­ought to be a physical object: “Que el verso sea como una llave” (1), where “llave” (“key”) is furthermore functional in a way that “medallion” is not. Huidobro’s poem is also marked by the unsettling presence of death: “El adjetivo, cuando no da vida, mata” (7).106 Most strikingly, Huidobro concludes his poem with a statement of the credo of Creacionismo: “El Poeta es un pequeño Dios” (18). This literary movement, of which Huidobro was the founder and principal exponent, understood the poet to be a creator in a physical and real sense and, consequently, his poem to be a physical and real creation,107 in a way resonant of Horace’s exploration of the basic senses of poeta and poema. Huidobro also makes the fine Horatian distinction between telling/describing and showing/doing: “Por qué cantáis la rosa,” he laments,

104. So Borges describes the retelling of existing stories (as opposed to the invention of new ones); at Mihailescu 2000: 48. 105. As identified by Vital 2011, who regards it as “un poema de transición entre la función ‘preceptiva’ y la función ‘arte poética’ ” (163); he sees Huidobro’s poem as coming on the heels of José Asunción Silva’s “Ars,” the last of the nineteenth-­century preceptive poems. The publication in this interim of the De­cadent poet Paul Verlaine’s Art poétique (1874; text and translation at Shapiro 2008: 126–27 and see further Hardison and Golden 1995: 319–21) should also be noted, especially since this French poem resembles t­ hose of the twentieth ­century writers discussed ­here in being comparatively short (thirty-­six lines). 106. “The adjective” represents all grammatical categories (Vital 2011: 167). Vital sees the interest in death as a hallmark of the twentieth-­century Artes poéticas, in that t­ hese are focused on an individual’s experiences and turmoil—as, indeed, Horace’s AP may also be read. 107. Huidobro provided a manifesto of this movement in his essay, “El Creacionismo” (Huidobro 1976: 733): “Os diré qué entiendo por poema creado. Es un poema en el que cada parte constitutiva, y todo el conjunto, muestra un hecho nuevo, independiente del mundo externo. . . . ​Dicho poema es algo que no puede existir sino en la cabeza del poeta. . . . ​Nada se le parece en el mundo externo; hace real lo que no existe, es decir, se hace realidad a sí mismo.” Compare also Stevens’s “in­ven­ted world” (I.1.2).

268  E p ilo gue

commanding that poets should rather “Hacedla florecer en el poema” (14–15).108 Vital 2011 identifies a group of six Artes poéticas that he regards as constituting their own subgenre from 1916–1949.109 Especially provocative is his argument that the Artes poéticas of the early twentieth ­century ­were filling the void left by the preceptive poetics of the nineteenth110 and offering a new way of describing what poetry was to be and what it could do. That an Ars Poetica could be antiprescriptive might seem counterintuitive, yet this view is wholly faithful to the spirit of the Horatian archetype. Although Vital regards Borges’s poem, published in 1960 as the last of ­these ­after a notable delay, as forming the epitaph of the movement,111 Artes poéticas continued to be written for two more de­cades, suggesting that Borges should rather be viewed as a nodal point. Only seven lines long, the “Arte poética” of Ramón Irigoyen (still living, Spanish, and with a degree in classical philology) reduces the tradition to “Un poema si no es una pedrada . . . ​es un fiambre de palabras muertas,” nodding to the Horatian ideas both of the poem as artifact and of the life cycles of words, as well as continuing the focus on death evident in Huidobro’s poem. At a mere three lines, the “Arte poética” (1974) of the Salvadoran Roque Dalton combines ele­ ments discernible in the poems of MacLeish and Irigoyen, as he notes simply that poetry is more than words: “Poesía / Perdóname por haberte ayudado a comprender / que no estás hecha sólo de palabras.” Marita Troiano, born in Peru in 1953, writes in her “Arte poética” about what poetry has meant and done for 108. Traces of Huidobro’s ­rose that the poet ­ought to make bloom may be seen in another Arte poética, that of Alfonso Reyes: “Asustadiza gracia del poema: / flor temerosa, recatada en llema. / Y se cierra, como la sensitiva, / si la llega a tocar la mano viva.” Though the opening words “Asustadiza gracia” suggest the sublime, attractive and terrifying in equal mea­sure, and Orpheus is named in this eight-­line poem, it does not other­wise have compelling resonances of Horace’s Ars Poetica. The declaration “no la creo” (6), moreover, reads as “una irónica y programática negación del creacionismo” (Vital 2011: 172), forging a strong connection (by way of rejection) to Huidobro’s poem instead. 109. Vital 2011 argues for the poems of Reyes, Huidobro, Borges, Manuel Bandeira (his poem, written in Portuguese, is entitled “Poética,” but Vital points out that in another poem, “Os sapos,” Bandeira rejects the notion of Artes poéticas: “Não há mais poesia, / Mas há artes poéticas”), Pablo Neruda (Chilean, 1904–1973), and Jaime Torres Bodet (1902–1974, Mexican) as representative of “un subgénero,” which he identifies as possessing five characteristics (enumerated at 186–87, ­these include a strong poetic “I”). Further poets who wrote poems entitled “Arte poética” include the aptly named Horacio Castillo (Argentine, 1934–2010), Galvarino Plaza (Chilean, 1930–85), Domingo Alfonso (b. 1935, Cuban; see Cohen 1967: 166), and Alfredo Lavergne (b. 1951 in Chile). 110. Vital 2011: 162; cf. my epilogue, n105. 111. Vital 2011: 182.

r e ce i ving the a r s p oet ic a  269

her: “La poesía me hizo suya,” she begins, concluding “La poesía fue mis alas hacia la libertad.” Yet the mentions of blood and skin (“Y fue mi sangre / Fue mi piel”) in this nine-­line work evoke the same from the conclusion of Stevens’s Notes, suggesting that both poets ­were captivated by the oft-­ignored ending of Horace’s Ars Poetica (cutem, cruoris, 476), valuing its saturation of color and its grotesque sublimity. More impor­tant and informative than their contents, however, even as ­these reveal numerous and unique points of contact with Horace’s poem, are the titles of t­ hese poems. As Levinson 1985: 29 points out, while Shakespeare may have been quite justified in saying that “a r­ ose by any other name would smell as sweet,” what­ever natu­ral objects are called “has l­ ittle bearing on our experience of them. What a work of art is titled, on the other hand, has a significant effect on the aesthetic face it pre­sents and on the qualities we correctly perceive in it.” In being named Arte poética, the poems that make up this mini-­genre of twentieth-­century Spanish-­language poetry do not so much describe their contents (­those could easily belong to another poem)112 as stake out their place in an august, millennia-­long line of writers who have viewed writing (and living) in similar ways.113 The tradition continues into the twenty-­first ­century and in English-­language poetry in the work of, for example, Elizabeth Alexander (nineteen poems titled “Ars Poetica” distinguished by the numbers appended to the title and sometimes by a further subtitle),114 Carol Ann Davis (“Ars Poetica inside an Evans Photo­graph,” 2003), Tom Disch (“Ars Poetica,” 2002), Dorothea Lasky (“Ars Poetica,” 2010), Amy MacLennan (“Ars Poetica Schmetica”),115 Ruth Irupé Sanabria (“Ars Poetica,” 2017), Eleanor Wilner (“Ars Poetica,” 2016), most of whom provide a further twist on the tradition by virtue of being, like Marita Troiano, w ­ omen amid what has 112. Consider, for example, the fact that Torres Bodet (epilogue, n109) also wrote a poem entitled “Poesía”; or Howard Nemerov’s “The Makers,” which, while it views the poet as a creator (e.g., “They ­were the ones that in what­ever tongue / Worded the world”) and names Orpheus and evokes Amphion (poets “sang the towers / Of the city into the astonished sky”), nevertheless does not and cannot participate in the same sustained conversation that arts of poetry do and can by their very names. Vital 2011: 162 rightly takes it for granted that each poet knew himself to be nodding to Horace’s Ars Poetica in entitling his poem “Arte poética”; for the idea, compare also Erica Jong’s “Arse Poetica” (1997), a humorously sexualized set of instructions for how to write a poem. 113. “No cualquier poeta escribe un ‘Arte poética,’ ” Vital 2011: 184 states, ­going on to say that ­those who do are motivated by the desire to play a prominent and mediating role on the literary scene. 114. Alexander 2005: 33–57. The closing words of #100, “are we not of interest to each other?,” perhaps hints at the idea of all artes poeticae communicating with one another in a long line. 115. At Banerjee and Szokolyai 2018: 40.

270  E p ilo gue

manifestly been the traditionally male preserve of artes poeticae.116 Equally, writing a poetic credo and not entitling it “Ars Poetica” is a decision imbued with meaning.117 It seems that the unusual popularity the Ars Poetica has enjoyed, then, is due to the generations of readers, many of them practicing poets, who saw it as far more than a mere handbook,118 rather than to the long-­supposed sincere use of it as an ars or technē (a narrative frequently accompanied by exclamations of bewilderment at the poem’s firm grasp on the title that in many re­spects fits it so inadequately). This survey of the reception of Horace’s Ars Poetica shows not merely that it has been read continuously since its composition, as outlined in the introduction, but that in e­ very time and place, its readers have remade it—­ whether through commentaries, translations, or adaptations—­into the forms necessary and valuable to them. In d­ oing so they have often elevated details of the Latin poem that the pre­sent study has also sought to bring to appropriate prominence: Nature, living (and death), the pro­gress of time, color, and the quality of poetry as craft and craftsmanship. This remaking is, accordingly, continual and inexhaustible, as Goethe and many o­ thers have seen: Passage 1983: 5 and Günther 2013: 61 say, respectively, that Horace is “im­mensely likeable, trustworthy, level-­headed, compatible with almost any society, and congenial with almost any mood” and that “a poet as inexhaustibly rich as he ­will speak to every­one in his own voice.”119 In wanting each to be “Horace for their own time and country,”120 Geoffrey of Vinsauf, Boileau, Pope, Jonson, and many ­others have not so much updated the rules contained within the Ars Poetica for con­temporary readership of what­ever type (beginning students or more 116. In the late twentieth ­century, see also the poems entitled “Ars Poetica” of Anne Nicodemus Carpenter (1982) and Rita Dove (1987), along with t­ hose of Stephen Corey (1987) and Gary Soto (“Ars Poetica, or Mazatlan, on a Day When Bodies Wash to the Shore”). Among the very few instances not in French, En­glish, or Spanish is the 1998 German “Ars Poetica 62” of Paul Celan. 117. Consider, for example, the variety of titles evident in Banerjee and Szokolyai’s 2018 collection of manifestos, itself entitled Credo. Some poems, in being doubly titled, e.g., “Manifesto, or Ars Poetica #2” by Krista Franklin (2017), are able to occupy more than one tradition si­mul­ta­neously. 118. Hardison and Golden 1995: 240 conversely point out, however, that none of the four major twentieth-­century commentaries on the work (they identify Rostagni 1930, Immisch 1932, Brink 1963 and 1971, and Rudd 1989) “is any more successful than Brink in suggesting why the Art might be of interest to a modern poet.” 119. See also Martindale 1993: 1 on how “many writers . . . ​have had a friend—in Quintus Horatius Flaccus.” 120. Wood 1978: 4 (similar is Bogue 1975: 12–13).

r e ce i ving the a r s p oet ic a  271

advanced students), as Geoffrey in par­tic­u­lar has been accused of d­ oing, but rather have produced works of poetry, literary artifacts, poem-­objects, that while they may say a variety of ­t hings about writing, are above all poems in themselves121—­living exemplars, just as the Ars Poetica is, of what the art ­ought ideally to look like and be like.

121. So, too, Sowerby 2006: 8: Vida’s De Arte Poetica, like Horace’s AP and Boileau’s L’Art poétique, “encodes in its own practice the creed which it advocates.”

bi bl io g r a p h y

Adams, ­Hazard. 1987. “Titles, Titling, and Entitlement to.” JAAC 46.1: 7–21. Adler, Ada. 1928–1938. Suidae Lexicon. 5 vols. Stuttgart: Teubner. Agamben, Giorgio. Trans. Daniel Heller-­Roazen. 1999. The End of the Poem: Studies in Poetics. Stanford, CA: Stanford University Press. Ahl, Frederick. 1985. Metaformations: Soundplay and Wordplay in Ovid and Other Classical Poets. Ithaca, NY: Cornell University Press. Alexander, Elizabeth. 2005. American Sublime: Poems. St. Paul, MN: Graywolf. Allen, Christopher, Yasmin Haskell, and Frances Muecke. 2005. Charles-­Alphonse Dufresnoy, De arte graphica (Paris, 1668): Edition, translation, and commentary. Geneva: Librairie Droz S. A. Antolín, Fernando Navarro. 2002. Quinto Horacio Flaco: Epístolas. Arte Poetica. Madrid: Consejo superior de investigaciones científicas. Armstrong, David. 2012. “Juvenalis Eques: A Dissident Voice from the Lower Tier of the Roman Elite.” In Susanna Braund and Josiah Osgood (eds.), A Companion to Persius and Juvenal, 59– 78. Malden, MA: Wiley-­Blackwell. —­—­—. 2010. “The Biographical and Social Foundations of Horace’s Poetic Voice.” In Gregson Davis (ed.), A Companion to Horace, 7–33. Malden, MA: Wiley-­Blackwell. —­—­—. 2004. “Horace’s Epistles 1 and Philodemus.” In David Armstrong et al. (eds.), Vergil, Philodemus, and the Augustans, 267–98. Austin: University of Texas Press. —­—­—. 1995. “The Impossibility of Metathesis: Philodemus and Lucretius on Form and Content in Poetry.” In Dirk Obbink (ed.), Philodemus and Poetry: Poetic Theory and Practice in Lucretius, Philodemus, and Horace, 210–32. Oxford: Oxford University Press. —­—­—. 1993. “The Addressees of the Ars poetica: Herculaneum, the Pisones and Epicurean Protreptic.” MD 31 (“Mega nepios: Il destinatario nell’epos didascalico”): 185–230. —­—­—. 1968. “­Toward a Theory of Structure in Horace.” PhD diss., University of Texas at Austin. Armstrong, David, Jeffrey Fish, Patricia A. Johnston, and Marilyn B. Skinner. 2004. Vergil, Philodemus, and the Augustans. Austin: University of Texas Press. Asmis, Elizabeth. 2008. “Lucretius’ New World Order: Making a Pact with Nature.” CQ 58.1: 141–57. —­—­—. 1995a. “Epicurean Poetics.” In Dirk Obbink (ed.), Philodemus and Poetry: Poetic Theory and Practice in Lucretius, Philodemus, and Horace, 15–34. Oxford: Oxford University Press. —­—­—. 1995b. “Philodemus on Censorship, Moral Utility, and Formalism in Poetry.” In Dirk Obbink (ed.), Philodemus and Poetry: Poetic Theory and Practice in Lucretius, Philodemus, and Horace, 148–77. Oxford: Oxford University Press. 273

274  biblio g r a p hy Austin, R. G. 1977. P. Vergili Maronis Aeneidos liber sextus. Oxford: Clarendon Press. Baker, Robert J. 1988. “Maecenas and Horace Satires II.8.” CJ 83.3: 212–32. Banerjee, Rita, and Diana Norma Szokolyai. 2018. Credo: An Anthology of Manifestos and Sourcebook for Creative Writing. [United States]: C&R Press. Barchiesi, Alessandro. 1993. “Insegnare ad Augusto: Orazio, Epistole 2, 1 e Ovidio, Tristia II.” MD 31 (“Mega nepios: Il destinatario nell’epos didascalico”): 149–84. Barchiesi, Alessandro, and Andrea Cucchiarelli. 2005. “Satire and the Poet: The Body as Self-­ Referential Symbol.” In Kirk Freudenburg (ed.), The Cambridge Companion to Roman Satire, 207–23. Cambridge: Cambridge University Press. Barnard, John. 1973. Pope: The Critical Heritage. Abingdon, UK: Taylor & Francis. Barney, Stephen A., W. J. Lewis, J. A. Beach, and Oliver Berghof. 2006. The Etymologies of Isidore of Seville. Cambridge: Cambridge University Press. Bate, Walter Jackson. 1952 (reprinted 2002, Miami: Wolf Den Books). Criticism: The Major Texts. New York: Harcourt, Brace. Beare, W. 1948. “The Roman Origin of the Five-­Act Law.” Hermathena 72: 44–70. —­—­—. 1946. “Horace, Donatus and the Five-­Act Law.” Hermathena 67: 52–59. Becker, Carl. 1963. Das Spätwerk des Horaz. Göttingen: Vandenhoeck & Ruprecht. Biles, Zachary P. 2011. Aristophanes and the Poetics of Competition. Cambridge: Cambridge University Press. Bing, Peter. 1993. “Aratus and His Audiences.” MD 31 (“Mega nepios: Il destinatario nell’epos didascalico”): 99–109. Blakeney, Edward Henry. 1928. Horace on the Art of Poetry: Latin Text, En­glish Prose Translation, Introduction and Notes, together with Ben Jonson’s En­glish Verse Rendering. Freeport, NY: Books for Libraries Press. Blank, David. 2014. “Philodemus.” In Edward N. Zalta (ed.), The Stanford Encyclopedia of Philosophy. https://­plato​.­stanford​.­edu​/­archives​/­fall2014​/­entries​/­philodemus. Bogue, Ronald Lynn. 1975. “The Art of the Art of Poetry: Graceful Negligence and Structure in Horace’s Ars poetica, Boileau’s L’art poétique, and Pope’s Essay on Criticism.” PhD diss., University of Oregon. Bolgar, R. R. 1954. The Classical Heritage and Its Beneficiaries: From the Carolingian Age to the End of the Re­nais­sance. New York: Harper & Row. Boudhors, Charles-­H. 1952. Épîtres, Art poétique, Lutrin. Paris: Société les belles lettres. Bowersock, G. W. 1971. “A Date in the Eighth Eclogue.” HSCP 75: 73–80. Bradley, Mark. 2013. “Colour as Synaesthetic Experience in Antiquity.” In Shane Butler and Alex Purves (eds.), Synaesthesia and the Ancient Senses, 127–40. Durham, UK: Acumen. —­—­—. 2009. Colour and Meaning in Ancient Rome. Cambridge: Cambridge University Press. Bragantini, Irene, and Mariette de Vos. 1982. Museo Nazionale Romano: Le Pitture II,1. Le decorazioni della villa romana della Farnesina. Rome: De Luca Editore. Bramble, J. C. 1974. Persius and the Programmatic Satire: A Study in Form and Imagery. Cambridge: Cambridge University Press. Branford, W.R.G., and M.W.M. Pope. 1953. “The Ars Poetica of Horace Rendered into En­glish Verse.” Theoria 5: 85–96. Braund, S. 1989. “City and Country in Roman Satire.” In S. Braund (ed.), Satire and Society in Ancient Rome, 23–47. Exeter: University of Exeter Press.

biblio g r a p h y  

275

Brink, C. O. 1995. “Second Thoughts on Three Horatian Puzzles.” In S. J. Harrison (ed.), Homage to Horace: A Bimillenary Cele­bration, 267–78. Oxford: Clarendon Press. —­—­—. 1981. “Horatian Poetry: Thoughts on the Development of Textual Criticism and Interpretation.” In Walter Killy (ed.), Geschichte des Textverständnisses am Beispiel von Pindar und Horaz [Vorträge gehalten anlässlich des 6. Wolfenbütteler Symposiums vom 18.–22. September 1978 in der Herzog August Bibliothek], 7–17. Munich: Kraus International Publications. —­—­—. (1971) 1985. Horace on Poetry: The “Ars Poetica”. Reprint, Cambridge: Cambridge University Press. —­—­—. 1965. “Horace and Empedocles’ Temperature: A Rejected Fragment of Empedocles.” Phoenix 23.1 (“Studies Presented to G.M.A. Grube on the Occasion of His Seventieth Birthday”): 138–42. —­—­—. 1963. Horace on Poetry. Cambridge: Cambridge University Press. Brophy, Brigid, Henri Peyre, Lionel Trilling, Northrop Frye, and W. K. Wimsatt. 1970. “Questionnaire on Horace.” Arion 9.2/3 (“Horace Issue”): 127–43. Brown, P. Michael. 1993. Horace: Satires I. Warminster: Aris & Phillips. Brunt, P. A. 1988. The Fall of the Roman Republic and Related Essays. Oxford: Clarendon Press. —­—­—. 1965. “Amicitia in the Late Roman Republic.” PCPS 11: 1–20. Bugada, Gabriele. 2012. Cristoforo Landino: In Quinti Horatii Flacci Artem Poeticam ad Pisones ­Interpretationes. Florence: Sismel—­Edizioni del galluzzo. Caballero, Marta Elena. 2007. “Simplex et unum: Formulación y ejecución.” Circe 11: 73–90. Cairns, Francis. 2004. “Varius and Vergil: Two Pupils of Philodemus in Propertius 2.34?” In David Armstrong et al. (eds.), Vergil, Philodemus, and the Augustans, 299–321. Austin: University of Texas Press. —­—­—. 1972. Generic Composition in Greek and Roman Poetry. Edinburgh: Edinburgh University Press. Camargo, Martin. 1988. “­Toward a Comprehensive Art of Written Discourse: Geoffrey of Vinsauf and the Ars Dictaminis.” Rhetorica 6.2: 167–94. Campbell, Archibald Y. 1924. Horace: A New Interpretation. London: Methuen & Co. Campbell, Oscar James. 1938. Comicall Satyre and Shakespeare’s Troilus and Cressida. San Marino, CA: Adcraft Press. Carettoni, Gianfilippo. 1983. “La decorazione pittorica della Casa di Augusto sul Palatino.” MDAI(R) 90: 373–419. Carpenter, Anne Nicodemus. 1982. “Ars Poetica.” Poetry 140.2: 96. Carroll, Lewis. 1893. Sylvie and Bruno Concluded. London: MacMillan. Casaubon, Isaac. (1605) 1973. De satyrica Graecorum poesi & Romanorum satira. A facsimile reproduction with an introduction by Peter E. Medine. Delmar, NY: Scholars’ Facsimiles & Reprints. Celan, Paul. Trans. Rita Dove and Fred Viebahn. 1998. “Ars Poetica 62.” Poetry (Special Issue: Con­ temporary German Poetry) 173.1: 8–9. Cichorius, Conrad. 1964. Untersuchungen zu Lucilius. Zürich: Weidmannsche Verlagsbuch­ han­dlung. —­—­—. 1922. Römische studien, historisches, epigraphisches, literargeschichtliches aus vier jahrhunderten Roms. Leipzig: Teubner. Cioffi, Carmela. 2017. Aeli Donati quod fertur commentum ad Andriam Terenti. Berlin: Walter de Gruyter.

276  biblio g r a p hy Citroni, Mario. 2016. “The Value of Self-­Deception: Horace, Aristippus, Heraclides Ponticus, and the Pleasures of the Fool (and of the Poet).” In Philip Hardie (ed.), Augustan Poetry and the Irrational, 221–39. Oxford: Oxford University Press. —­—­—. 2009. “The Ars Poetica and the Marvellous.” In Philip Hardie (ed.), Paradox and the Marvellous in Augustan Lit­er­a­ture and Culture, 19–40. Oxford: Oxford University Press. Clarke, Jacqueline. 2003. Imagery of Colour and Shining in Catullus, Propertius, and Horace. New York: Peter Lang. Clay, Diskin. 1995. “Framing the Margins of Philodemus and Poetry.” In Dirk Obbink (ed.), Philodemus and Poetry: Poetic Theory and Practice in Lucretius, Philodemus, and Horace, 3–14. Oxford: Oxford University Press. Clayman, Dee, Gregory Crane, and Donald Guthrie. 1993. “Response to Paul Keyser 1992, ‘Review: Shifting Paradigms: New Approaches to Horace’s Ars Poetica by Bernard Frischer.’ ” BMCR 4.1: 75–81. Cloud, J. D. 1989. “Satirists and the Law.” In S. Braund (ed.), Satire and Society in Ancient Rome, 49–67, 132–36. Exeter: University of Exeter Press. Cody, J. V. 1976. Horace and Callimachean Aesthetics. Brussels: Latomus, Revue d’études latines. Cohen, J. M. 1967. Writers in the New Cuba: An Anthology. London: Penguin. Coleman, Robert. 2001. “Varro as an Etymologist.” Papers on Grammar 3: 61–96. Collinge, N. E. 1961. The Structure of Horace’s Odes. Oxford: Oxford University Press. Commager, Steele. 1962. The Odes of Horace: A Critical Study. New Haven: Yale University Press. Copeland, Rita. 2016a. “Introduction.” In Rita Copeland (ed.), The Oxford History of Classical Reception in En­glish Lit­er­a­ture, Vol. 1: 800–1558, 1–20. Oxford: Oxford University Press. —­—­—. 2016b. “The Curricular Classics in the ­Middle Ages.” In Rita Copeland (ed.), The Oxford History of Classical Reception in En­glish Lit­er­a­ture, Vol. 1 800–1558, 21–33. Oxford: Oxford University Press. —­—­—. 2016c. “The Trivium and the Classics.” In Rita Copeland (ed.), The Oxford History of Classical Reception in En­glish Lit­er­a­ture, Vol. 1 800–1558, 53–76. Oxford: Oxford University Press. Copeland, Rita, and Ineke Sluiter. 2009. Medieval Grammar and Rhe­toric: Language Arts and Literary Theory, AD 300–1475. Oxford: Oxford University Press. Corey, Stephen. 1987. “Ars Poetica.” Poetry (Special Issue: The Seventy-­Fifth Anniversary Issue) 151.1/2: 35. Courtney, E. 1993. The Fragmentary Latin Poets. Oxford: Clarendon Press. Daly, Lloyd W. 1943. “The Entitulature of Pre-­Ciceronian Writings.” In Classical Studies in Honor of William Abbott Oldfather. Presented by a Committee of His Former Students and Colleagues: 20–38. Urbana: University of Illinois Press. D’Anna, Giovanni. 1983. “La cronologia dell’epistola di Orazio ad Augusto.” Vichiana 12: 121–35. Davies, Malcolm. 1989. The Greek Epic Cycle. London: Bristol Classical Press. Davis, Carol Ann. 2003. “Ars Poetica inside an Evans Photo­graph.” Poetry 182.5: 256–57. Davis, Gregson. 1991. Polyhymnia: The Rhe­toric of Horatian Lyric Discourse. Berkeley: University of California Press. Delattre, Daniel. 2004. “Vergil and ­Music, in Diogenes of Babylon and Philodemus.” In David Armstrong et al. (eds.), Vergil, Philodemus, and the Augustans, 245–63. Austin: University of Texas Press.

biblio g r a p h y  

277

Dessen, Cynthia S. 1968. Iunctura callidus acri: A Study of Persius’ Satires. Urbana: University of Illinois Press. Di Benedetto, A. 1962. “Echi terenziani in Orazio,” Rendiconti dell’ Accademia di Archeologia, Lettere e Belle Arti di Napoli 37: 35–57. Dilke, O.A.W. 1958. “When Was the Ars Poetica Written?” BICS 5: 49–57. Disch, Tom. 2002. “Ars Poetica.” Poetry (Special Issue: Ninetieth Anniversary Issue) 181.1: 20. Dove, Rita. 1987. “Ars Poetica.” Poetry (Special Issue: The Seventy-­Fifth Anniversary Issue) 151.1/2: 45. Duckworth, G. E. 1965. “Horace’s Hexameters and the Date of the Ars Poetica.” TAPA 96: 73–95. Earl, D. C. 1960. “Calpurnii Pisones in the Second ­Century B. C.” Athenaeum 38: 283–98. Ehrhardt, Wolfgang. 1987. Stilgeschichtliche Untersuchungen an römischen Wandmalereien von der späten Republik bis zur Zeit Neros. Mainz am Rhein: Verlag Philipp von Zabern. Elmore, Jefferson. 1935. “A New Dating of Horace’s de Arte Poetica.” CP 30.1: 1–9. Else, Gerald F. 1957. Aristotle’s Poetics: The Argument. Cambridge, MA: Harvard University Press. Elsner, Jaś. 1995. Art and the Roman Viewer: The Transformation of Art from the Pagan World to Chris­tian­ity. Cambridge: Cambridge University Press. Enenkel, Karl, and Henk Nellen (eds.). 2013. Neo-­Latin Commentaries and the Management of Knowledge in the Late ­Middle Ages and the Early Modern Period (1400–1700) (Supplementa Humanistica Lovaniensia 33). Leuven: Leuven University Press. Engetsu, Katsuhiro. 2004. “Dryden and the Modes of Restoration Sociability.” In Steven N. Zwicker (ed.), The Cambridge Companion to John Dryden, 181–96. Cambridge: Cambridge University Press. Erasmus, D., and J. Rivius. 1535. P. Terentii Afri comoediae ex D. Erasmi et Io. Rivii attendoriensis castigationibus. Cologne: Ioannes Gymnicus. Ernout, Alfred, and Antoine Meillet. 2001. Dictionnaire étymologique de la langue latine. Paris: Klincksieck. Fairclough, H. Rushton. 1929. Horace: Satires, Epistles, and Ars Poetica. London: William Heinemann. Fanti, Giulia. 2017. “The Failure of Memmius in Lucretius’ DRN.” Latomus 76: 58–79. Fantuzzi, Marco. 2015. “The Aesthetics of Sequentiality and Its Discontents.” In Marco Fantuzzi and Christos Tsagalis (ed.), The Greek Epic Cycle and Its Ancient Reception: A Companion, 405–29. Cambridge: Cambridge University Press. Fantuzzi, Marco, and Christos Tsagalis (ed.). 2015. The Greek Epic Cycle and Its Ancient Reception: A Companion. Cambridge: Cambridge University Press. Farago, Claire J. 1992. Leonardo Da Vinci’s Paragone: A Critical Interpretation with a New Edition of the Text in the Codex Urbinas. Leiden: Brill. Faral, Edmond. 1958. Les arts poétiques du XIIe et du XIIIe siècle. Recherches et documents sur la technique littéraire du moyen age. Paris: Librairie Honoré Champion. Fedeli, Paolo. 1997. Q. Orazio Flacco: Le opere. II, tomo quarto: Le Epistole, L´Arte Poetica. Rome: Istituto Poligrafico e Zecca dello Stato. Feeney, Denis. 2009. “Becoming an Authority: Horace on His Own Reception.” In L.B.T. Houghton and Maria Wyke (eds.), Perceptions of Horace: A Roman Poet and His Readers, 16–38. Cambridge: Cambridge University Press.

278  biblio g r a p hy —­—­—. 2002. “Una cum scriptore meo: Poetry, Principate and the Traditions of Literary History in the Epistle to Augustus.” In T. Woodman and D. Feeney (eds.), Traditions and Contexts in the Poetry of Horace, 172–87. Cambridge: Cambridge University Press. —­—­—. 1993. “Horace and the Greek Lyric Poets.” In Niall Rudd (ed.), Horace 2000: A Cele­ bration. Essays for the Bimillennium, 41–61. Ann Arbor: University of Michigan Press. Ferenczi, Attila. 2014a. “Introduction: The New Horace and the Ars poetica.” MD 72 (“New Approaches to Horace’s Ars Poetica”): 11–17. —­—­—. 2014b. “Following the Rules. Three Meta­phors in the Ars poetica.” MD 72 (“New Approaches to Horace’s Ars Poetica”): 71–83. Ferriss-­Hill, J. L. 2018. “Vivas Hinc Ducere Voces: Hearing Horace’s Satires in His Ars Poetica.” In Michael Fontaine, Charles J. McNamara, and William Michael Short (eds.), Quasi ­Labor Intus: Ambiguity in Latin Lit­er­a­ture. Papers in Honor of Reginald Thomas Foster, OCD, 143–64. [United States]: Paideia Institute for Humanistic Study. —­—­—. 2015. Roman Satire and the Old Comic Tradition. Cambridge: Cambridge University Press. —­—­—. 2014. “Varro’s Intuition of Cognate Relationships,” ICS 39: 81–108. —­—­—. 2011. “A Stroll with Lucilius: Horace, Satires 1.9 Reconsidered.” AJP 132.3: 429–55. Fish, Jeffrey. 1998. “Is Death Nothing to Horace? A Brief Comparison with Philodemus and ­Lucretius.” CErc 28: 99–104. Fisher, John. 1984. “Entitling.” Critical Inquiry 11.2: 286–98. Fiske, George Converse. (1920) 1966. Lucilius and Horace: A Study in the Classical Theory of Imitation. Madison: University of Wisconsin. Reprint, Hildesheim: Georg Olms Verlagsbuchhandlung. —­—­—. 1913. “Lucilius, the Ars Poetica of Horace, and Persius.” HSCP 24: 1–36. Fitzgerald, John T. (ed.). 1996. Friendship, Flattery, and Frankness of Speech: Studies on Friendship in the New Testament World. Leiden: E. J. Brill. Flobert, Pierre. 2003. Varron: La langue latine. Vol. 2, Book 6. Paris: Les Belles Lettres. Ford, Andrew. 2002. The Origins of Criticism: Literary Culture and Poetic Theory in Classical Greece. Prince­ton, NJ: Prince­ton University Press. Forsythe, Gary. 1994. The Historian L. Calpurnius Piso Frugi and the Annalistic Tradition. Lanham, MD: University Press of Amer­i­ca. —­—­—. 1990. “The Tribal Membership of the Calpurnii Pisones.” ZPE 83: 293–98. Fowler, Don. 1997. “Second Thoughts on Closure.” In Deborah H. Roberts, Francis M. Dunn, and Don Fowler (eds.), Classical Closure: Reading the End in Greek and Latin Lit­er­a­ture, 3–22. Prince­ton, NJ: Prince­ton University Press. —­—­—. 1989. “First Thoughts on Closure: Prob­lems and Prospects.” MD 22: 75–122. Fowler, Peta. 1997. “Lucretian Conclusions.” In Deborah H. Roberts, Francis M. Dunn, and Don Fowler (eds.), Classical Closure: Reading the End in Greek and Latin Lit­er­a­ture, 112–38. Prince­ ton, NJ: Prince­ton University Press. Foxon, D. F. 1975. En­glish Verse 1701–1750, Vol. 1: Cata­logue, Vol. 2: Indexes. Cambridge: Cambridge University Press. Fraenkel, Eduard. 1957. Horace. Oxford: Clarendon Press. Franklin, Krista. 2015. “Manifesto, or Ars Poetica #2.” Poetry 206.1: 6–7. Fredborg, Karin M. 1988. The Latin Rhetorical Commentaries by Thierry of Chartres. Toronto: Pontifical Institute of Mediaeval Studies.

biblio g r a p h y  

279

Freudenburg, Kirk. 2013. “The Afterlife of Varro in Horace’s Sermones: Generic Issues in Roman Satire.” In T. D. Papanghelis, S. J. Harrison, and S. Frangoulidis (eds.), Generic Interfaces in Latin Lit­er­a­ture: Encounters, Interactions, and Transformations, 297–336. Berlin: De Gruyter. —­—­—. 2002. “Writing to/through Florus: Criticism and the Addressee in Horace, Epistles 2.2.” Memoirs of the American Acad­emy in Rome 47: 33–55. —­—­—. 2001. Satires of Rome: Threatening Poses from Lucilius to Juvenal. Cambridge: Cambridge University Press. —­—­—. 1999. “Review: Horace’s Narrative Odes by Michele Lowrie.” CP 94.2: 234–38. —­—­—. 1993. The Walking Muse: Horace on the Theory of Satire. Prince­ton, NJ: Prince­ton University Press. Friis-­Jensen, Karsten. 2008. “Humanist Use of Medieval Commentaries on Horace’s Art of Poetry.” StudUmanistPiceni 28: 239–48. —­—­—. 2007. “The Reception of Horace in the ­Middle Ages.” In Stephen Harrison (ed.), The Cambridge Companion to Horace, 291–304. Cambridge: Cambridge University Press. —­—­—. 1997a. “Medieval Commentaries on Horace.” In Nicholas Mann and Birger Munk Olsen (eds.), Medieval and Re­nais­sance Scholarship: Proceedings of the Second Eu­ro­pean Science Foundation Workshop on the Classical Tradition in the ­Middle Ages and the Re­nais­sance (London, Warburg Institute, 27–28 November, 1992): 51–73. Leiden: E. J. Brill. —­—­—. 1997b. “Petrarch and the Medieval Horace.” In Marianne Pade, Hannemarie Ragn Jensen, and Lene Waage Petersen (eds.), Avignon and Naples: Italy in France—­France in Italy in the ­Fourteenth ­Century, 83–98. Rome: L’Erma di Bretschneider. —­—­—. 1995a. “Commentaries on Horace’s Art of Poetry in the Incunable Period.” Re­nais­sance Studies 9.2 (“Incunabula: Books, Texts and ­Owners”): 228–39. —­—­—. 1995b. “Horace and the Early Writers of Arts of Poetry.” In Sten Ebbesen (ed.), Sprachtheorien in Spätantike und Mittelalter, 360–401. Tübingen: G. Narr. —­—­—. 1993. “The Medieval Horace and His Lyr­ics.” In Walter Ludwig (ed.), Horace. L’oeuvre et les imitations. Un siècle d’interprétation, 257–303. Vandoeuvres-­Genève: Fondation Hardt. —­—­—. 1991. “Addenda et Corrigenda to CIMAGL 60 1990 319–88: Karsten Friis-­Jensen, The Ars Poetica in Twelfth-­Century France.” CIMAGL 61: 184–85. —­—­—. 1990. “The Ars Poetica in Twelfth-­Century France: The Horace of Matthew of Vendôme, Geoffrey of Vinsauf, and John of Garland.” CIMAGL 60: 319–88. —­—­—. 1988. “Horace liricus et ethicus. Two Twelfth-­Century School Texts on Horace’s Poems.” CIMAGL 57: 81–147. Frischer, Bernard. 1996. “Rezeptionsgeschichte and Interpretation: The Quarrel of Antonio Riccoboni and Nicolò Cologno about the Structure of Horace’s Ars Poetica.” In Helmut Krasser and Ernst A. Schmidt (eds.), Zeitgenosse Horaz: Der Dichter und seiner Leser seit zwei Jahrtausenden, 68–116. Tübingen: Gunter Narr Verlag. —­—­—. 1991. Shifting Paradigms: New Approaches to Horace’s Ars Poetica. Atlanta, GA: Scholars Press. Gage, John. 1981. “A Locus Classicus of Colour Theory: The Fortunes of Apelles.” JWI 44: 1–26. Gale, Monica R. 2000. Virgil on the Nature of ­Things: The Georgics, Lucretius, and the Didactic Tradition. Cambridge: Cambridge University Press. Galyon, Aubrey E. 1980. Matthew of Vendôme: The Art of Versification. Ames: Iowa State University Press.

280  biblio g r a p h y Gantar, Kajetan. 1964. “Die Anfangsverse und die Komposition der horazischen Epistel über die Dichtkunst.” SO 39: 89–98. —­—­—. 1954. “De compositione Horatii Epistulae ad Pisones.” Živa antika 4: 277. García Berrio, Antonio. 1978. “La contaminacion de autores, tecnica del comentario humanistico (Examen de la penetración horaciana en los seis grandes comentarios italianos del Siglo XVI a la ‘Poética’ de Aristóteles).” AMal 1: 225–53. Gargiulo, Tristano. 1981. “PHerc. 222: Filodemo sull’adulazione.” CErc 11: 103–27. Geffcken, Katherine A. 1973. A Comedy in the Pro Caelio. Leiden: Brill. Gellar-­Goad, T.H.M. 2012. “Lucretius’ De Rerum Natura and Satire.” PhD diss., University of North Carolina at Chapel Hill. Genette, Gérard. Trans. Jane E. Lewin. 1997. Paratexts: Thresholds of Interpretation. Cambridge: Cambridge University Press. Geue, Tom. 2014. “Editing the Opposition: Horace’s Ars Politica.” MD 72 (“New Approaches to Horace’s Ars Poetica”): 143–72. Gigante, Marcello. 2004. “Vergil in the Shadow of Vesuvius.” In David Armstrong et al. (eds.), Vergil, Philodemus, and the Augustans, 85–99. Austin: University of Texas Press. —­—­—. Trans. Dirk Obbink. 1995. Philodemus in Italy. Ann Arbor: University of Michigan Press. —­—­—. 1990. Filodemo in Italia. Firenze: Le Monnier. —­—­—. 1979. Cata­logo dei papiri ercolanesi. Naples: Bibliopolis. Gigante, Marcello, and Mario Capasso. 1989. “Il ritorno di Virgilio a Ercolano.” SIFC 7: 3–6. Gildersleeve, B. L., and G. Lodge. 1895. Gildersleeve’s Latin Grammar. London: Macmillan. Gill, Christopher. 2012. “The Transformation of Aristotle’s Ethics in Roman Philosophy.” In Jon Miller (ed.), The Reception of Aristotle’s Ethics, 31–52. Cambridge: Cambridge University Press. Glad, Clarence E. 1996. “Frank Speech, Flattery, and Friendship in Philodemus.” In John T. Fitzgerald (ed.), Friendship, Flattery, and Frankness of Speech: Studies on Friendship in the New Testament World, 21–59. Leiden: E. J. Brill. Gneuss, Helmut, and Michael Lapidge. 2014. Anglo-­Saxon Manuscripts: A Bibliographical Handlist of Manuscripts and Manuscript Fragments Written or Owned in E ­ ngland up to 1100. Toronto: University of Toronto Press. Gold, Barbara K. 1992. “Openings in Horace’s Satires and Odes: Poet, Patron, and Audience.” YCS 29: 161–85. —­—­—. 1987. Literary Patronage in Greece and Rome. Chapel Hill: University of North Carolina Press. Golden, Leon. 2010. “Reception of Horace’s Ars Poetica.” In Gregson Davis (ed.), A Companion to Horace, 391–413. Malden, MA: Wiley-­Blackwell. Goldman, Rachael B. 2013. Color-­Terms in Social and Cultural Context in Ancient Rome. Piscataway, NJ: Gorgias Press. Goulbourne, Russell. 2009. “Appropriating Horace in Eighteenth-­Century France.” In L.B.T. Houghton and Maria Wyke (eds.), Perceptions of Horace: A Roman Poet and His Readers, 256–70. Cambridge: Cambridge University Press. Gowers, Emily. 2012. Horace, Satires: Book I. Cambridge: Cambridge University Press. —­—­—. 2003. “Fragments of Autobiography in Horace Satires 1.” ClAnt 22.1: 55–91. —­—­—. 1993. The Loaded T ­ able: Repre­sen­ta­tions of Food in Roman Lit­er­a­ture. Oxford: Clarendon Press.

biblio g r a p h y  

281

Graziosi, Barbara. 2009. “Horace, Suetonius, and the Lives of the Greek Poets.” In L.B.T. Houghton and Maria Wyke (eds.), Perceptions of Horace: A Roman Poet and His Readers, 140–60. Cambridge: Cambridge University Press. Greenberg, Nathan A. 1970. “Metrical Expectations in the Ars Poetica.” Revue—­Organisation internationale pour l’étude des langues anciennes par ordinateur 3: 111–29. —­—­—. 1961. “The Use of Poiema and Poiesis.” HSCP 65: 263–89. —­—­—. 1955. “The Poetic Theory of Philodemus.” PhD diss., Harvard University. Published 1990, New York: Garland. Griffin, Jasper. 1996. “Horace in ­England.” In Helmut Krasser and Ernst A. Schmidt (eds.), Zeitgenosse Horaz: Der Dichter und seiner Leser seit zwei Jahrtausenden, 182–206. Tübingen: Gunter Narr Verlag. Griffin, Miriam. 2003. “De Beneficiis and Roman Society.” JRS 93: 92–113. Grimal, Pierre. 1964. “Horace: De l’art de vivre à l’Art poétique.” BAGB 23: 436–47. Grube, G.M.A. 1965. The Greek and Roman Critics. London: Methuen. Günther, Hans-­Christian. 2013. “Horace’s Life and Work.” In Hans-­Christian Günther (ed.), Brill’s Companion to Horace, 1–62. Leiden: Brill. Hajdu, Péter. 2014. “Horace’s Ars poetica as Pure Poetry.” MD 72 (“New Approaches to Horace’s Ars Poetica”): 85–96. Hajdú, István. 1993. “Ein Zürcher Kommentar aus dem 12. Jahrhundert zur Ars poetica des Horaz.” CIMAGL 63: 231–93. Halliwell, Stephen. 1986. Aristotle’s Poetics. Chapel Hill: University of North Carolina Press. Hammond, Paul. 1993. “Figures of Horace in Dryden’s Literary Criticism.” In Charles Martindale and David Hopkins (eds.), Horace Made New, 127–47. Cambridge: Cambridge University Press. Hardie, Philip. 2016. Augustan Poetry and the Irrational. Oxford: Oxford University Press. —­—­—. 2014. “The Ars poetica and the Poetics of Didactic.” MD 72 (“New Approaches to Horace’s Ars Poetica”): 43–54. —­—­—. 2009. Lucretian Receptions: History, the Sublime, Knowledge. Cambridge: Cambridge University Press. —­—­—. 2005. “Time in Lucretius and the Augustan Poets: Freedom and Innovation.” In Jürgen Paul Schwindt (ed.), La représentation du temps dans la poésie augustéenne / Zur Poetik der Zeit in augusteischer Dichtung, 19–42. Heidelberg: Winter. —­—­—. 1996. “Virgil: A Paradoxical Poet?” PLLS 9: 103–21. —­—­—. 1993. “Ut pictura poesis? Horace and the Visual Arts.” In Niall Rudd (ed.), Horace 2000: A Cele­bration. Essays for the Bimillennium, 120–39. Ann Arbor: University of Michigan Press. Hardison, O. B., Jr., and Leon Golden. 1995. Horace for Students of Lit­er­a­ture: The ‘Ars Poetica’ and Its Tradition. Gainesville: University Press of Florida. Harrison, Stephen. 2009. “Horace and the Victorians.” In L.B.T. Houghton and Maria Wyke (eds.), Perceptions of Horace: A Roman Poet and His Readers, 290–304. Cambridge: Cambridge University Press. —­—­— (ed.). 2007a. The Cambridge Companion to Horace. Cambridge: Cambridge University Press. —­—­—. 2007b. “Town and Country.” In Stephen Harrison (ed.), The Cambridge Companion to Horace, 235–47. Cambridge: Cambridge University Press.

282  biblio g r a p hy —­—­—. 2007c. “Style and Poetic Texture.” In Stephen Harrison (ed.), The Cambridge Companion to Horace, 262–73. Cambridge: Cambridge University Press. —­—­—. 2007d. “The Reception of Horace in the Nineteenth and Twentieth Centuries.” In Stephen Harrison (ed.), The Cambridge Companion to Horace, 334–46. Cambridge: Cambridge University Press. —­—­—. 2004. “Lyric ­Middles: The Turn at the Centre in Horace’s Odes.” In Stratis Kyriakidis and Francesco de Martino (eds.), ­Middles in Latin Poetry, 81–102. Bari: Levante editori. Hauthal, F. (1866) 1966. Acronis et Porphyrionis Commentarii in Q. Horatium Flaccum. Vol. 2. Berlin: J. Springer. Reprint, Amsterdam: P. Schippers. —­—­—. (1864) 1966. Acronis et Porphyrionis Commentarii in Q. Horatium Flaccum. Vol. 1. Berlin: J. Springer. Reprint, Amsterdam: P. Schippers. Henderson, John G. W. 1989. “Not ‘­Women in Roman Satire’ but ‘When Satire Writes W ­ oman.’ ” In Susanna H. Morton Braund (ed.), Satire and Society in Ancient Rome (Exeter Studies in History), 89–125. Exeter: University of Exeter Press. Hendrickson, George L. 1918. “An Epigram of Philodemus and Two Latin Congeners.” AJP 39.1: 27–43. —­—­—. 1894. “The Dramatic Satura and the Old Comedy at Rome.” AJP 15.1: 1–30. Henriksson, Karl-­Erik. 1956. Griechische Büchertitel in der Römischen Literatur. Annales Academiae Scientiarum Fennicae Ser. B Tom. 102.1. Helsinki: Suomalaisen Kirjallisuuden Kirjapaino Oy. Henry, W. Benjamin. 2009. Philodemus, On Death, Translated with an Introduction and Notes. Atlanta, GA: Society of Biblical Lit­er­a­ture. Hering, Wolfgang. 1979. Die Dialektik von Inhalt und Form bei Horaz: Satiren Buch I und Epistula ad Pisones. Berlin: Akademie-­Verlag. Herrick, Marvin T. 1946. The Fusion of Horatian and Aristotelian Criticism, 1531–­1555 (Illinois Studies in Language and Lit­er­a­ture Volume 32.1). Urbana: University of Illinois Press. Herrnstein Smith, Barbara. 1968. Poetic Closure: A Study of How Poems End. Chicago: University of Chicago Press. Holland, Norman, 1975. Five Readers Reading. New Haven, CT: Yale University Press. Hopkins, David, and Charles Martindale (eds.). 2012. The Oxford History of Classical Reception in En­glish Lit­er­a­ture, Vol. 3: 1660–1790. Oxford: Oxford University Press. Hopman, Marianne Govers. 2013. Scylla: Myth, Meta­phor, Paradox. Cambridge: Cambridge University Press. Horsfall, Nicholas. 2000. Virgil, Aeneid 7: A Commentary. Leiden: Brill. —­—­—. 1981. “Some Prob­lems of Titulature in Roman Literary History.” BICS 28: 103–14. —­—­—. 1979a. “Horace, Sermones 3?” LCM 4.6: 117–19. —­—­—. 1979b. “Horace, Sermones 3; epilegomena.” LCM 4.8: 169–71. Hösle, Vittorio. 2009. “Poetische Poetiken in der Antike: Horaz’ “Ars poetica” und Pseudo-­ Longinos’ Περὶ ὕψους.” Poetica 41.1/2: 55–74. Houghton, L.B.T., and Maria Wyke. 2009. “Introduction: A Roman Poet and His Readers.” In L.B.T. Houghton and Maria Wyke (eds.), Perceptions of Horace: A Roman Poet and His Readers, 1–15. Cambridge: Cambridge University Press. Housman, A. E. 1927. M. Annaei Lucani Belli Civilis Libri Decem. Oxford: Blackwell. Houston, George W. 2014. Inside Roman Libraries: Book Collections and Their Management in Antiquity. Chapel Hill: University of North Carolina Press.

biblio g r a p h y  

283

Hubbard, T. K. 1991. The Mask of Comedy: Aristophanes and the Intertextual Parabasis. Ithaca, NY: Cornell University Press. Huidobro, Vicente. 1976. Obras completas de Vicente Huidobro, Tomo I. Prólogo de Hugo Montes. Santiago, Chile: Editorial Andres Bello. Hunter, R. L. 2009. Critical Moments in Classical Lit­er­a­ture: Studies in the Ancient View of Lit­er­ a­ture and Its Uses. Cambridge: Cambridge University Press. —­—­—. 1985. “Horace on Friendship and ­Free Speech.” Hermes 113.4: 480–90. Hurley, Andrew, trans. 1998. Collected Fictions: Jose Luis Borges. New York: Viking. Huygens, R.B.C. 1970. Accessus ad auctores. Bernard d’Utrecht. Conrad d’Hirsau: Dialogus super auctores. Leiden: E. J. Brill. Ierodiakonou, Katerina. 2005. “Empedocles on Colour and Colour Vision.” OSAPh 29: 1–37. —­—­—. 2004. “Empedocles and the Ancient Paint­ers.” In Liza Cleland and Karen Stears (eds.), Colour in the Ancient Mediterranean World (BAR International Series 1267), 91–95. Oxford: John and Erica Hedges Ltd / Archaeopress. Immisch, O. 1932. Horazens Epistel über die Dichtkunst (Philologus Supplement 24.3). Leipzig: Dieterich’sche Verlagsbuchhandlung. Indelli, Giovanni. 2004. “The Vocabulary of Anger in Philodemus’ De ira and Vergil’s Aeneid.” In David Armstrong et al. (eds.), Vergil, Philodemus, and the Augustans, 103–10. Austin: University of Texas Press. Indelli, Giovanni, and Voula Tsouna-­McKirahan. 1995. [Philodemus] [On Choices and Avoidances]. Naples: Bibliopolis. Innes, Doreen. 2003. “Meta­phor, Simile, and Allegory as Ornaments of Style.” In G. R. Boys-­ Stones (ed.), Meta­phor, Allegory, and the Classical Tradition: Ancient Thought and Modern Revisions, 7–27. Jahn, O. (1843) 1967. Auli Persi Flacci Satirarum Liber cum scholiis antiquis. Hildesheim: Georg Olms Verlagsbuchhandlung. Reprint, Leipzig: Breitkopf & Haertel. Janko, Richard. 2011. Philodemus: On Poems Books 3–4, with the Fragments of Aristotle, On Poets. Oxford: Oxford University Press. —­—­—. 2000. Philodemus: On Poems Book 1. Oxford: Oxford University Press. —­—­—. 1995. “Reconstructing Philodemus’ On Poems.” In Dirk Obbink (ed.), Philodemus and Poetry: Poetic Theory and Practice in Lucretius, Philodemus, and Horace, 69–96. Oxford: Oxford University Press. —­—­—. 1987. Aristotle: Poetics I, with the Tractatus Coislinianus, a Hy­po­thet­ic­ al Reconstruction of Poetics II, the Fragments of the On Poets. Indianapolis, IN: Hackett. Jensen, Christian. 1923. Philodemos über die Gedichte, fünftes Buch. Berlin: Weidman. Jensson, Gottskálk T. 2002. “The Satyrica of Petronius as a Roman Palimpsest.” Ancient Narrative 2: 86–122. Johnson, W. R. 1982. The Idea of Lyric: Lyric Modes in Ancient and Modern Poetry. Berkeley: University of California Press. Jones, Frederick. 2016. The Bound­aries of Art and Social Space in Rome: The Caged Bird and Other Art Forms. London: Bloomsbury Academic. Jong, Erica. 1997. Fruits and Vegetables. Hopewell, NJ: Ecco Press. Kajanto, Iiro. 1965. The Latin Cognomina. Commentationes Humanarum Litterarum 36.2. Helsinki: Societas Scientiarum Fennica.

284  biblio g r a ph y Kalderon, Mark Eli. 2015. Form without ­Matter: Empedocles and Aristotle on Color Perception. Oxford: Oxford University Press. Kaster, Robert A. 1988. Guardians of Language: The Grammarian and Society in Late Antiquity. Berkeley: University of California Press. Katz, Joshua T. 2008. “Dux reget examen (Epistle 1.19.23): Horace’s Archilochean Signature.” MD 59: 207–13. Keane, C. C. 2006. Figuring Genre in Roman Satire. Oxford: Oxford University Press. —­—­—. 2002. “The Critical Contexts of Satiric Discourse.” CML 22.2: 7–31. Keil, H. (1857) 1961. Grammatici Latini, Vol. 1. (Leipzig) Hildesheim: Georg Olms Verlagsbuchhandlung. Keller, William Jacob. 1916. “Goethe’s Estimate of the Greek and Latin Writers as Revealed by His Works, Letters, Diaries, and Conversations.” PhD diss., University of Wisconsin. Kelly, Douglas. 1966. “The Scope of the Treatment of Composition in the Twelfth-­and Thirteenth-­ Century Arts of Poetry.” Speculum 41.2: 261–78. Kemp, Jerome. 2010. “Flattery and Frankness in Horace and Philodemus.” G&R 2nd ser. 57.1: 65–76. Kenney, E. J. 1970. “That Incomparable Poem the ‘Ille ego’?” CR 20: 290. Kermode, Frank. 2000. The Sense of an Ending: Studies in the Theory of Fiction, with a New Epilogue. Oxford: Oxford University Press. Keyser, Paul. 1992. “Review: Shifting Paradigms: New Approaches to Horace’s Ars Poetica by Bernard Frischer.” BMCR 3.2: 118–22. Kiessling, Adolf, and Richard Heinze. 1914 (4th ed., reprinted 1984). Q. Horatius Flaccus: Briefe. Berlin (reprinted, Hildesheim): Weidmann. Kilpatrick, Ross S. 1990. The Poetry of Criticism: Horace, Epistles II and Ars Poetica. Edmonton: University of Alberta Press. —­—­—. 1986. The Poetry of Friendship: Horace, Epistles I. Edmonton: University of Alberta Press. Klebs, Elimar. 1897. Prosopographia Imperii Romani Saec. I, II, III. Pars I (Academiae Scientiarum Regiae Borussicae). Berlin: George Reimer. Klingner, Friedrich. 1937. Horazens Brief an die Pisonen. Leipzig: S. Hirzel. Konstan, David. 1997. Friendship in the Classical World. Cambridge: Cambridge University Press. —­—­—. 1996a. “Friendship, Frankness and Flattery.” In John T. Fitzgerald (ed.), Friendship, Flattery, and Frankness of Speech: Studies on Friendship in the New Testament World, 7–19. Leiden: E. J. Brill. —­—­—. 1996b. “Friendship from Epicurus to Philodemus.” In Gabriele Giannantoni and Marcello Gigante (eds.), Epicureismo greco e romano: atti del congresso internazionale, Napoli, 19–26 maggio 1995, 387–96. Naples: Bibliopolis. —­—­—. 1995. “Patrons and Friends.” CP 90.4: 328–42. —­—­—. 1993. “Foreword: To the Reader.” MD 31 (“Mega nepios: Il destinatario nell’epos didascalico”): 11–22. Konstan, David, Diskin Clay, Clarence E. Glad, Johan C. Thom, and James Ware. 1998. Philodemus: On Frank Criticism. Atlanta, GA: Scholars Press (Society of Biblical Lit­er­a­ture). Körte, Alfred. 1890. “Augusteer bei Philodem.” RhM 45: 172–77. Krenkel, Werner. 1970. Lucilius: Satires. Leiden: Brill.

biblio g r a p h y  

285

Kyriakidis, Stratis. 2006. “Lucretius’ DRN 1.926–50 and the Proem to Book 4.” CQ n.s. 56.2: 606–10. —­—­—. 2004. “­Middles in Lucretius’ DRN: The Poet and His Work.” In Stratis Kyriakidis and Francesco de Martino (eds.), ­Middles in Latin Poetry, 27–49. Bari: Levante editori. LaFleur, R. A. 1981. “Horace and onomasti komodein: The Law of Satire.” ANRW II.31.3: 1790–826. Laird, Andrew. 2007. “The Ars Poetica.” In Stephen Harrison (ed.), The Cambridge Companion to Horace, 132–43. Cambridge: Cambridge University Press. —­—­—. 1996. “Ut figura poesis: Writing Art and the Art of Writing in Augustan Poetry.” In Jaś Elsner (ed.), Art and Text in Roman Culture, 75–102. Cambridge: Cambridge University Press. Landels, John G. 1999. ­Music in Ancient Greece and Rome. London: Routledge. Lanternari, Diana. 1974. “Hor. Epist. II 3: ad Pisones.” RCCM 16: 169–89. Lasky, Dorothea. 2010. Black Life. United States: Wave Books. Lawler, Traugott. 1974. The Pa­ri­siana Poetria of John of Garland, Edited with Introduction, Translation, and Notes. New Haven, CT: Yale University Press. Leach, Eleanor Winsor. 1971. “Horace’s pater optimus and Terence’s Demea: Autobiographical Fiction and Comedy in Sermo I, 4.” AJP 92.4: 616–32. Lee, Rensselaer W. 1940. “Ut Pictura Poesis: The Humanistic Theory of Painting.” Art Bull. 22: 197–269. Lefkowitz, Mary R. 2012. The Lives of the Greek Poets. 2nd ed. Baltimore, MD: Johns Hopkins University Press. Leigh, Matthew. 2004. “The Pro Caelio and Comedy.” CP 99.4: 300–35. Leitch, Vincent B. 2010. The Norton Anthology of Theory and Criticism. 2nd ed. New York: W. W. Norton. Lejay, P. 1911. Q. Horatii Flacci Satirae. Paris: Hachette. Levinson, Jerrold. 1985. “Titles.” JAAC 44.1: 29–39. Lindsay, Wallace M. 1913. Sexti Pompei Festi de verborum significatu quae supersunt cum Pauli epitome. Leipzig: Teubner. Lowe, Dunstan. 2015. Monsters and Monstrosity in Augustan Poetry. Ann Arbor: University of Michigan Press. Lowrie, Michèle. 2014. “Politics by Other Means in Horace’s Ars poetica.” MD 72 (“New Approaches to Horace’s Ars Poetica”): 121–42. —­—­—. 1997. Horace’s Narrative Odes. Oxford: Clarendon Press. Lyons, John. 1999. “The Vocabulary of Color with Par­tic­u­lar Reference to Ancient Greek and Classical Latin.” In Alexander Borg (ed.), The Language of Color in the Mediterranean: 38–75. Stockholm: Almqvist & Wiksell International. MacLaury, Robert. 1999. “Basic Color Terms: Twenty-­Five Years ­After.” In Alexander Borg (ed.), The Language of Color in the Mediterranean, 1–37. Stockholm: Almqvist & Wiksell International. Maltby, Robert. 1991. A Lexicon of Ancient Latin Etymologies. Cambridge: Francis Cairns. Mañas Nuñez, Manuel. 2015. “Los comentarios de Aquiles Estacio al Ars poetica de Horacio.” Ágora 17.1: 147–73. —­—­—. 2012. “La Epistula ad Pisones de Horacio: su normalización como ars poetica hasta el Renacimiento.” CFC(L) 32.2: 223–46.

286  biblio g r a p h y Maresca, Thomas E. 1966. Pope’s Horatian Poems. [Columbus]: Ohio State University Press. Martindale, Charles. 1993. “Introduction.” In Charles Martindale and David Hopkins (eds.), Horace Made New, 1–26. Cambridge: Cambridge University Press. Massaro, Matteo. 1974. “Sapere e i suoi derivati in Orazio.” SIFC 46: 85–128. Maxwell-­Stuart, P. G. 1981. Studies in Greek Colour Terminology. Vol. 1: ΓΛΑΥΚΟΣ. Leiden: E. J. Brill. Mayer, Roland. 1994. Horace: Epistles, Book 1. Cambridge: Cambridge University Press. —­—­—. 1989. “Friendship in the Satirists.” In Susanna H. Morton Braund (ed.), Satire and Society in Ancient Rome (Exeter Studies in History), 5–21. Exeter: University of Exeter Press. McGann, Michael. 2007. “The Reception of Horace in the Re­nais­sance.” In Stephen Harrison (ed.), The Cambridge Companion to Horace, 305–17. Cambridge: Cambridge University Press. McGinn, Thomas. 2001. “Satire and the Law: The Case of Horace.” PCPS 47: 81–102. Meerwaldt, J. D. 1936. “Adnotationes in Epistulam ad Pisones ad picturam praesertim collatam pertinentes.” Mnemosyne 3rd ser. 4.2: 151–63. Mette, Hans Joachim. 1961. “ ‘Genus tenue’ und ‘mensa tenuis’ bei Horaz.” MH 18: 136–39. Michailescu, Calin Andrei. 2000. Jorge Luis Borges: This Craft of Verse. Cambridge, MA: Harvard University Press. Michaelis, A. 1877. “Die Horazischen Pisonen.” In Arnold Schaefer et al., eds., Commentationes Philologae in honorem Theodori Mommseni, 420–32. Berlin: Weidmann. Michels, Agnes Kirsopp. 1944. “Παρρησία and the Satire of Horace.” CP 39.3: 173–77. Miller, John F. 1993. “Apostrophe, Aside and the Didactic Addressee: Poetic Strategies in Ars Amatoria III.” MD 31 (“Mega nepios: Il destinatario nell’epos didascalico”): 231–41. Mitsis, Phillip. 1993. “Committing Philosophy on the Reader: Didactic Coercion and Reader Autonomy in De Rerum Natura.” MD 31 (“Mega nepios: Il destinatario nell’epos didascalico”): 111–28. Möhler, Gabriele. 1989. Hexameterstudien zur Lukrez, Vergil, Horaz, Ovid, Lukan, Silius Italicus und der Ilias Latina. Frankfurt: Peter Lang. Mohr, Paulus. 1895. C. Sollius Apollinaris Sidonius. Leipzig: Teubner. Möller, M. 2004. Talis Oratio—­Qualis Vita: Zu Theorie und Praxis mimetischer Verfahren in der griechisch-­römischen Literaturkritik. Heidelberg: Winter. Money, David. 2007. “The Reception of Horace in the Seventeenth and Eigh­teenth Centuries.” In Stephen Harrison (ed.), The Cambridge Companion to Horace, 318–33. Cambridge: Cambridge University Press. —­—­—. 1998. The En­glish Horace: Anthony Alsop and the Tradition of British Latin Verse. Oxford: Oxford University Press. Most, Glenn. 1996. “Horatian and Pindaric Lyric in E ­ ngland.” In Helmut Krasser and Ernst A. Schmidt (eds.), Zeitgenosse Horaz: Der Dichter und seiner Leser seit zwei Jahrtausenden, 117–52. Tübingen: Gunter Narr Verlag. Moul, Victoria. 2015. “Horace (with a Contribution by Charles Martindale).” In Patrick Cheney and Philip Hardie (eds.), The Oxford History of Classical Reception in En­glish Lit­er­a­ture. Vol. 2: 1558–1660, 539–56. Oxford: Oxford University Press. —­—­—. 2010. Jonson, Horace and the Classical Tradition. Cambridge: Cambridge University Press. —­—­—. 2009. “The Poet’s Voice: Allusive Dialogue in Ben Jonson’s Horatian Poetry.” In L.B.T. Houghton and Maria Wyke (eds.), Perceptions of Horace: A Roman Poet and His Readers, 219–38. Cambridge: Cambridge University Press.

biblio g r a p h y  

287

Muecke, Frances. 1993. Horace: Satires II. Warminster: Aris & Phillips. Nauta, Ruurd R. 2002. Poetry for Patrons: Literary Communication in the Age of Domitian. Leiden: Brill. Nettleship, Henry. 1885. Lectures and Essays on Subjects Connected with Latin Lit­er­a­ture and Scholarship. Oxford: Clarendon Press. —­—­—. 1883. “The de Arte Poetica of Horace.” JP 12: 43–61. Neuburg, Matt. 1993. “Hitch Your Wagon to a Star: Manilius and His Two Addressees.” MD 31 (“Mega nepios: Il destinatario nell’epos didascalico”): 243–83. Newman, John Kevin. 1986. The Classical Epic Tradition. Madison: University of Wisconsin Press. —­—­—. 1967. The Concept of Vates in Augustan Poetry. Brussels: Latomus, Revue d’études latines. Nichols, Marden Fitzpatrick. 2009. “Social Status and the Authorial Persona of Horace and Vitruvius.” In L.B.T. Houghton and Maria Wyke (eds.), Perceptions of Horace: A Roman Poet and His Readers, 109–22. Cambridge: Cambridge University Press. Nietz­sche, Friedrich. Trans. Duncan Large, 1998. Friedrich Nietz­sche: Twilight of the Idols. Oxford: Oxford University Press. Nims, Margaret F. 2010. Geoffrey of Vinsauf: Poetria Nova. Rev. ed. Toronto: Pontifical Institute of Mediaeval Studies. Nisbet, R.G.M. 2007. “Horace: Life and Chronology.” In Stephen Harrison (ed.), The Cambridge Companion to Horace, 7–21. Cambridge: Cambridge University Press. —­—­—. 1999. “The Word Order of Horace’s Odes.” In J. N. Adams and R. G. Mayer (eds.), ­Aspects of the Language of Roman Poetry (Proceedings of the British Acad­emy 93): 135–54. Oxford: Oxford University Press. —­—­—. 1986. “Review: A Rival Teubner Horace.” CR n.s. 36: 227–34. —­—­—. 1961. M. Tulli Ciceronis in L. Calpurnium Pisonem oratio. Oxford: Clarendon Press. Nisbet, R.G.M., and M. Hubbard. 1978. A Commentary on Horace Odes, Book 2. Oxford: Clarendon Press. —­—­—. 1970. A Commentary on Horace Odes, Book 1. Oxford: Clarendon Press. Norden, Eduard. 1957. P. Vergilius Maro Aeneis Buch 6. 4th ed. Darmstadt: Wissenschaftliche Buchgesellschaft. —­—­—. 1905. “Die Composition und Litteraturgattung der Horazischen Epistula ad Pisones.” Hermes 40.4: 481–528. Nussbaum, Martha. 1989. “Mortal Immortals: Lucretius on Death and the Voice of Nature.” Philos. & Phenom. Research 50.2: 303–51. —­—­—. 1986a. The Fragility of Goodness: Luck and Ethics in Greek Tragedy and Philosophy. Cambridge: Cambridge University Press. —­—­—. 1986b. “Therapeutic Arguments: Epicurus and Aristotle.” In M. Schofield and G. Striker (eds.), The Norms of Nature: Studies in Hellenistic Ethics, 31–74. Cambridge: Cambridge University Press. Oakley, S. P. 1998. A Commentary on Livy: Books 6–10, vol. 2: Books 7–8. Oxford: Clarendon Press. Obbink, Dirk. 1995a (ed.). Philodemus and Poetry: Poetic Theory and Practice in Lucretius, Philodemus, and Horace. Oxford: Oxford University Press. —­—­—. 1995b. “How to Read Poetry about Gods.” In Dirk Obbink (ed.), Philodemus and Poetry: Poetic Theory and Practice in Lucretius, Philodemus, and Horace, 189–209. Oxford: Oxford University Press.

288  biblio g r a p h y —­—­—. 1993. “The Addressees of Empedocles.” MD 31 (“Mega nepios: Il destinatario nell’epos didascalico”): 51–98. Oberhelman, Steven, and David Armstrong. 1995. “Satire as Poetry and the Impossibility of Metathesis in Horace’s Satires.” In Dirk Obbink (ed.), Philodemus and Poetry: Poetic Theory and Practice in Lucretius, Philodemus, and Horace, 233–54. Oxford: Oxford University Press. O’Hara, James J. 1996. True Names: Vergil and the Alexandrian Tradition of Etymological Wordplay. Ann Arbor: University of Michigan Press. O’Keefe, Tim. 2003. “Lucretius on the Cycle of Life and the Fear of Death.” Apeiron 36.1: 43–65. Oliensis, Ellen. 1998. Horace and the Rhe­toric of Authority. Cambridge: Cambridge University Press. Olson, S. Douglas, and Alexander Sens. 2000. Archestratos of Gela: Greek Culture and Cuisine in the Fourth ­Century BCE. Text, Translation, and Commentary. Oxford: Oxford University Press. —­—­—. 1999. Matro of Pitane and the Tradition of Epic Parody in the Fourth ­Century BCE. Text, Translation, and Commentary. Atlanta, GA: Scholars Press. Ong, Walter J. 1975. “The Writer’s Audience Is Always a Fiction.” PMLA 90.1: 9–21. Orelli, J. C. 1844. Q. Horatius Flaccus. 2nd ed. Zürich: Orelli. O’­Sullivan, N. 1992. Alcidamas, Aristophanes and the Beginnings of Stylistic Theory. Stuttgart: Franz Steiner Verlag. Ott, Wilhelm. 1970. Metrische Analysen zur Ars Poetica des Horaz. Göppingen: Alfred Kümmerle. Otto, A. 1890. Die Sprichwörter und sprichwörtlichen Redensarten der Römer. Leipzig: Teubner. Palmeri, Frank. 2016. State of Nature, Stages of Society: Enlightenment Conjectural History and Modern Social Discourse. New York: Columbia University Press. Parr, Roger P. 1981. Matthew of Vendôme: Ars Versificatoria (The Art of the Versemaker). Milwaukee, WI: Marquette University Press. Passage, Charles E. 1983. The Complete Works of Horace (Quintus Horatius Flaccus). New York: Ungar. Paule, Maxwell Teitel. 2017. Canidia, Rome’s First Witch. London: Bloomsbury Academic. Pavis, Patrice. Trans. Jaume Melendres. 1998. Diccionario del teatro. Barcelona: Paidós. Pearcy, Lee T. 2102. “Does D ­ ying Hurt? Philodemus of Gadara, De morte and Asclepiades of Bithynia.” CQ 62.1: 211–22. Peerlkamp, P. Hofman. 1845. Q. Horatii Flacci epistola ad Pisones. Leiden: H. W. Hazenberg. Pfaffel, Wilhelm. 1981. Quartus gradus etymologiae: Untersuchungen zur Etymologie Varros in De lingua Latina. Königstein: Hain. Pickard-­Cambridge, Arthur W. 1962. Dithyramb, Tragedy and Comedy. 2nd ed., rev. by T.B.L. Webster. Oxford: Clarendon Press. —­—­—. 1953. The Dramatic Festivals of Athens. Oxford: Clarendon Press. Pickering, P. E. 2003. “Did the Greek Ear Detect ‘Careless’ Verbal Repetitions?” CQ 53.2: 490–99. Pieper, Christoph. 2013. “Horatius praeceptor eloquentiae: The Ars Poetica in Cristoforo Landino’s Commentary.” In Karl Enenkel and Henk Nellen (eds.), Neo-­Latin Commentaries and the Management of Knowledge in the Late M ­ iddle Ages and the Early Modern Period (1400–1700) (Supplementa Humanistica Lovaniensia 33), 221–40. Leuven: Leuven University Press.

biblio g r a p h y  

289

Platt, Verity. 2009. “Where the Wild ­Things Are: Locating the Marvellous in Augustan Wall Painting.” In Philip Hardie (ed.), Paradox and the Marvellous in Augustan Lit­er­a­ture and Culture, 41–74. Oxford: Oxford University Press. Plaza, M. 2006. The Function of Humour in Roman Verse Satire: Laughing and Lying. Oxford: ­Oxford University Press. Pomeroy, Arthur J. 1980. “A Man at a Spring: Horace, Odes 1.1.” Ramus 9.1: 34–50. Pope, Alexander. 1735. An Epistle from Mr. Pope, to Dr. Arbuthnot. London: printed by J. Wright for Lawton Gilliver. Porter, James I. 2016. The Sublime in Antiquity. Cambridge: Cambridge University Press. —­—­—. 1995. “Content and Form in Philodemus: The History of an Evasion.” In Dirk Obbink (ed.), Philodemus and Poetry: Poetic Theory and Practice in Lucretius, Philodemus, and Horace, 97–147. Oxford: Oxford University Press. Putnam, Michael C. J. 2000. Horace’s Carmen Saeculare: Ritual Magic and the Poet’s Art. New Haven, CT: Yale University Press. Quinn, Arthur. 1991. “The Color of Rhe­toric.” In Gert Ueding (ed.), Rhetorik zwischen den Wissenschaften: Geschichte, System, Praxis als Probleme des “Historischen Wörterbuchs der Rhetorik,” 133–38. Tübingen: M. Niemeyer. Rabinowitz, Peter J. 1986. “Shifting Stands, Shifting Standards: Reading, Interpretation, and Literary Judgment.” Arethusa (“Audience-­Oriented Criticism and the Classics”) 19.2: 115–31. —­—­—. 1977. “Truth in Fiction: A Reexamination of Audiences.” Critical Inquiry 4.1: 121–41. Rawson, Elizabeth. 1985. Intellectual Life in the Late Roman Republic. Baltimore, MD: Johns Hopkins University Press. Reckford, Kenneth J. 1969. Horace. New York: Twayne. Reifferscheid, Augustus. 1860. C. Suetoni Tranquilli praeter Caesarum libros reliquiae. Leipzig: Teubner. Reinhardt, Tobias. 2013. “The Ars Poetica.” In Hans-­Christian Günther (ed.), Brill’s Companion to Horace, 499–526. Leiden: Brill. —­—­—. 2002. “The Speech of Nature in Lucretius’ De Rerum Natura 3.931–71.” CQ 52.1: 291–304. Reynolds, L. D. (ed.). 1983. Texts and Transmission: A Survey of the Latin Classics. Oxford: Clarendon Press. Reynolds, Suzanne. 1996. Medieval Reading: Grammar, Rhe­toric, and the Classical Text. Cambridge: Cambridge University Press. Riganti, E. 1978. Varrone: De lingua latina, libro 6. Bologna: Pàtron Editore. Roberts, Deborah H., Francis M. Dunn, and Don Fowler (eds.). 1997. Classical Closure: Reading the End in Greek and Latin Lit­er­a­ture. Prince­ton, NJ: Prince­ton University Press. Rosen, Ralph M. 2008. “Badness and Intentionality in Aristophanes’ Frogs.” In Ineke Sluiter and Ralph M. Rosen (eds.), Kakos: Badness and Anti-­Value in Classical Antiquity, 143–68. Leiden: Brill. —­—­—. 2004. “Aristophanes’ Frogs and the Contest of Homer and Hesiod.” TAPA 134.2: 295–322. —­—­—. 1988. Old Comedy and the Iambographic Tradition. Atlanta, GA: Scholars Press. Rostagni, Augusto. 1930. Orazio: Arte Poetica. Torino: Chiantore.

290  biblio g r a p h y Rowe, Christopher. 1972. “Conceptions of Colour and Colour Symbolism in the Ancient World.” Eranos 41: 327–64. Ruch, M. 1963. “Horace et les fondements de la iunctura dans l’ordre de la création poétique (A.P., 46–72).” REL 41: 246–69. Rudd, N. 2009. “Theme and Variation: Horace in Pope’s Correspondence.” In L.B.T. Houghton and Maria Wyke (eds.), Perceptions of Horace: A Roman Poet and His Readers, 239–55. ­Cambridge: Cambridge University Press. —­—­—. 1989. Horace Epistles Book 2 and Epistle to the Pisones (“Ars Poetica”). Cambridge: ­Cambridge University Press. Ruiz Castellanos, Antonio. 1994. “La tematización en el Ars Poetica de Horacio.” In Dulce Estefanía (ed.), Horacio, el poeta y el hombre, 259–66. Madria: Ediciones Clásicas. Russell, D. A. 2006. “Ars Poetica.” In Andrew Laird (ed.), Oxford Readings in Ancient Literary Criticism, 325–45. Oxford: Oxford University Press. Russell, D. A., and M. Winterbottom (eds.). 1972. Ancient Literary Criticism: The Principal Texts in New Translations. Oxford: Clarendon Press. Rutherford, Richard. 2007. “Poetics and literary criticism.” In Stephen Harrison (ed.), The Cambridge Companion to Horace, 248–61. Cambridge: Cambridge University Press. Salazar, Christine F. 2002–2010. Brill’s New Pauly: Encyclopaedia of the Ancient World: Antiquity. Leiden: Brill. Saller, Richard P. 1989. “Patronage and Friendship in Early Imperial Rome: Drawing the Distinction.” In Andrew Wallace-­Hadrill (ed.), Patronage in Ancient Society, 49–62. London: Routledge. —­—­—. 1982. Personal Patronage u­ nder the Early Empire. Cambridge: Cambridge University Press. Sanabria, Ruth Irupé. 2017. Beasts Behave in Foreign Land. Los Angeles: Red Hen Press. Santirocco, Matthew S. 1986. Unity and Design in Horace’s Odes. Chapel Hill: University of North Carolina Press. Sbordone, Francesco. 1981. “La poetica oraziana alla luce degli studi più recenti.” ANRW II 31.3: 1866–1920. Scaliger, Julius Caesar. 1594. Poetices libri septem. Apud Petrum Santandreanum. Schiesaro, Alessandro. 1993. “Il destinatario discreto. Funzioni didascaliche e progetto culturale nelle Georgiche.” MD 31 (“Mega nepios: Il destinatario nell’epos didascalico”): 129–47. Schwindt, Jürgen Paul. 2014. “Ordo and Insanity. On the Pathogenesis of Horace’s Ars Poetica.” MD 72 (“New Approaches to Horace’s Ars Poetica”): 55–70. Sedley, David. 2014. “Horace’s Socraticae chartae (Ars Poetica 295–322).” MD 72 (“New Approaches to Horace’s Ars Poetica”): 97–120. Seeck, Gustav Adolf. 1995. “Eine satirische Ars poetica für Piso: Bemerkungen zu Form und Absicht von Horaz epist 2,3.” A&A 41: 142–60. Segura Ramos, Bartolomé. 1989. “La Epístola a los Pisones (Hor., Epíst. II,3).” Habis 20: 111–25. Seiffert, Hans Werner. 1968. Christoph Martin Wieland Werke. Vol. 5. Munich: Hanser. Shackleton Bailey, D. R. 2001. Q. Horatius Flaccus: Opera. Munich: K. G. Saur. —­—­—. 1982. Profile of Horace. Cambridge, MA: Harvard University Press. Shapiro, Karl. 2003. Essay on Rime, with Trial of a Poet. Edited with an Afterword by Robert Phillips. Foreword by David Lehman. Ann Arbor: University of Michigan Press.

biblio g r a p h y  

291

Shapiro, Norman R. 2008. One Hundred and One Poems by Paul Verlaine: A Bilingual Edition. Chicago: University of Chicago Press. Sharland, Suzanne. 2009. Horace in Dialogue: Bakhtinian Readings in the Satires. Oxford: Peter Lang. Sharrock, Alison. 1996. “Representing Metamorphosis.” In Jaś Elsner (ed.), Art and Text in Roman Culture, 103–30. Cambridge: Cambridge University Press. Shearin, W. H. 2015. The Language of Atoms: Performativity and Politics in Lucretius’ De rerum natura. Oxford: Oxford University Press. Shorey, Paul. 1933. What Plato Said. Chicago: University of Chicago Press. Sider, David. 2005. The Library of the Villa dei Papiri at Herculaneum. Los Angeles: Getty Publications. —­—­—. 1997. The Epigrams of Philodemos. Oxford: Oxford University Press. —­—­—. 1995a. “Epicurean Poetics: Response and Dialogue.” In Dirk Obbink (ed.), Philodemus and Poetry: Poetic Theory and Practice in Lucretius, Philodemus, and Horace, 35–41. Oxford: Oxford University Press. —­—­—. 1995b. “The Epicurean Phi­los­op­ her as Hellenistic Poet.” In Dirk Obbink (ed.), Philodemus and Poetry: Poetic Theory and Practice in Lucretius, Philodemus, and Horace, 42–57. Oxford: Oxford University Press. Silk, M. S. 2000. Aristophanes and the Definition of Comedy. Oxford: Oxford University Press. Sisson, C. H. 1993. “Deniable Evidence: Translating Horace.” In Charles Martindale and David Hopkins (eds.), Horace Made New, 258–67. Cambridge: Cambridge University Press. Smith, W. K. 1936. “The Date of the Ars Poetica.” CP 31.2: 163–66. Solmsen, Friedrich. 1963. “Nature as Craftsman in Greek Thought.” JHI 24.4: 473–96. Solomon, Daniel. 2004. “Lucretius’ Progressive Revelation of Nature in DRN 1.149–502.” Phoenix 58.3/4: 260–83. Soto, Gary. 1987. “Ars Poetica, or Mazatlan, on a Day When Bodies Wash to the Shore.” Poetry 150.3: 144–45. Sowerby, Robin. 2012. “Horatianism.” In David Hopkins and Charles Martindale (eds.), The ­Oxford History of Classical Reception in En­glish Lit­er­a­ture, Vol. 3: 1660–1790, 255–85. Oxford: Oxford University Press. —­—­—. 2006. The Augustan Art of Poetry: Augustan Translation of the Classics. Oxford: Oxford University Press. —­—­—. 1993. “Figures of Horace in Dryden’s Literary Criticism.” In Charles Martindale and David Hopkins (eds.), Horace Made New, 159–83. Cambridge: Cambridge University Press. Specht, Henrik, and Michael Chesnutt. 1987. The Saint-­Omer Art of Poetry: A Twelfth ­Century Anonymous Ars Poetica from a Manuscript at Saint-­Omer. Denmark: Special-­Trykkeriet Viborg. Stadeler, Anja. 2014. Horazrezeption in der Re­nais­sance: Strategien der Horazkommentierung bei Cristoforo Landino und Denis Lambin. Berlin: De Gruyter. Starkey, Jennifer. 2018. “The Origin and Purpose of the Three-­Actor Rule.” TAPA 148.2: 269–97. Strauss Clay, Jenny. 1993. “The Education of Perses: From ‘Mega Nepios’ to ‘Dion Genos’ and Back.” MD 31 (“Mega nepios: Il destinatario nell’epos didascalico”): 23–33. Sühnel, Rudolf. 1996. “Ars Horatiana in E ­ ngland.” In Helmut Krasser and Ernst A. Schmidt (eds.), Zeitgenosse Horaz: Der Dichter und seiner Leser seit zwei Jahrtausenden, 153–81. Tübingen: Gunter Narr Verlag.

292  biblio g r a p hy Syme, Ronald. 1986. The Augustan Aristocracy. Oxford: Clarendon Press. —­—­—. 1980. “The Sons of Piso the Pontifex.” AJP 101.3: 333–41. —­—­—. 1960. “Piso Frugi and Crassus Frugi.” JRS 50.1/2: 12–20. Syndikus, H. P. 1995. “Some Structures in Horace’s Odes.” In S. J. Harrison (ed.), Homage to Horace: A Bimillenary Cele­bration, 17–31. Oxford: Clarendon Press. Talbot, John. 2009. “A Late Flowering of En­glish Alcaics.” In L.B.T. Houghton and Maria Wyke (eds.), Perceptions of Horace: A Roman Poet and His Readers, 305–23. Cambridge: Cambridge University Press. Tarán, Leonardo, and Dimitri Gutas. 2012. Aristotle: Poetics. Editio maior of the Greek Text with Historical Introductions and Philological Commentaries. Leiden: Brill. Tarrant, R. J. 2007. “Horace and Roman Literary History.” In Stephen Harrison (ed.), The Cambridge Companion to Horace, 63–76. Cambridge: Cambridge University Press. —­—­—. 1995. “Da Capo Structure in Some Odes of Horace.” In S. J. Harrison (ed.), Homage to Horace: A Bimillenary Cele­bration, 32–49. Oxford: Clarendon Press. Tate, J. 1928. “Horace and the Moral Function of Poetry.” CQ 22: 65–72. Taylor, Christopher C. W. 2007. “Democritus and Lucretius on Death and ­Dying.” In Aldo Brancacci and Pierre-­Marie Morel (eds.), Democritus: Science, the Arts, and the Care of the Soul: Proceedings of the International Colloquium on Democritus, Paris, 18–20 September 2003, 77–86. Leiden: Brill. Thomas, Richard F. 2004. “ ‘Stuck in the ­Middle with You’: Virgilian M ­ iddles.” In Stratis Kyriakidis and Francesco de Martino (eds.), ­Middles in Latin Poetry, 123–50. Bari: Levante editori. —­—­—. 1988. Virgil: Georgics. Vol. 1: Books 1–2, Vol. 2: Books 3–4. Cambridge: Cambridge University Press. —­—­—. 1986. “Virgil’s Georgics and the Art of Reference.” HSCP 90: 171–98. Tomlinson, Charles. 1993. “Some Aspects of Horace in the Twentieth C ­ entury.” In Charles Martindale and David Hopkins (eds.), Horace Made New, 240–57. Cambridge: Cambridge University Press. Tompkins, Jane. 1980. Reader-­Response Criticism: From Formalism to Post-­Structuralism. Baltimore: Johns Hopkins University Press. Tracy, H. L. 1948. “Horace’s Ars Poetica: A Systematic Argument.” G&R 17.51: 104–15. Tsakiropoulou-­Summers, Anastasia. 1998. “Horace, Philodemus and the Epicureans at Herculaneum.” Mnemosyne 51.1: 20–29. —­—­—. 1995. “Philodemus’ Περὶ Ποιημάτων and Horace’s Ars Poetica: Adapting Alexandrian Aesthetics to Epicurean and Roman Traditions.” PhD diss., University of Illinois at Urbana-­Champaign. Tsouna, Voula. 2012. Philodemus, On Property Management (Writings from the Greco-­Roman World 33). Atlanta, GA: Society of Biblical Lit­er­a­ture. —­—­—. 2007. The Ethics of Philodemus. Oxford: Oxford University Press. Vahlen, Johannes. 1911. Gesammelte philologische Schriften. Vol. 1: Schriften der Wiener Zeit, 1858– 1874. Leipzig: Teubner. Vance, Norman. 2015. “Classical Authors, 1790–1880.” In Norman Vance and Jennifer Wallace (eds.), The Oxford History of Classical Reception in En­glish Lit­er­a­ture, Vol. 4: 1790–1880. ­Oxford: Oxford University Press.

biblio g r a p h y  

293

—­—­—. 1993. “Horace and the Nineteenth ­Century.” In Charles Martindale and David Hopkins (eds.), Horace Made New, 199–216. Cambridge: Cambridge University Press. Van Rooy, C. A. 1968. “Arrangement and Structure of Satires in Horace, Sermones, Book 1, with more Special Reference to Satires 1–4.” AClass 11: 38–72. —­—­—. 1965. Studies in Classical Satire and Related Literary Theory. Leiden: E. J. Brill. Villa, Claudia. 1994. “I manoscritti di Orazio. III.” Aevum 68.1: 117–46. —­—­—. 1993. “I manoscritti di Orazio. II.” Aevum 67.1: 55–103. —­—­—. 1992. “I manoscritti di Orazio. I.” Aevum 66.1: 95–135. Viola, Alberto. 1906. L’arte poetica di Orazio nella critica italiana e straniera, Vol. 2. Naples: Stabil. tip. Luigi Pierro e figlio. —­—­—. 1901. L’arte poetica di Orazio nella critica italiana e straniera, Vol. 1. Naples: Stabil. tipografico Pierro e Veraldi nell’Istituto Casanova. Vital, Alberto. 2011. “Arte poética en seis poetas latinoamericanos del siglo XX. Alfonso Reyes, Vicente Huidobro, Jorge Luis Borges, Manuel Bandeira, Pablo Neruda y Jaime Torres Bodet.” Literatura Mexicana 22.1: 159–88. Volk, Katharina. 2002. The Poetics of Latin Didactic: Lucretius, Vergil, Ovid, Manilius. Oxford: Oxford University Press. Vollmann, Benedikt Konrad. 1996. “Erziehung zur Humanität: Horaz und die Satire des 11. ­Jahrhunderts.” In Helmut Krasser and Ernst A. Schmidt (eds.), Zeitgenosse Horaz: Der Dichter und seiner Leser seit zwei Jahrtausenden, 36–51. Tübingen: Gunter Narr Verlag. Vollmer, F. 1907. Die Überlieferungsgeschichte des Horaz. Philologus Supplement 10. Leipzig. Wachsmuth, Curtius, and Otto Hense. 1894. Ioannis Stobaei Anthologium, volumen tertium. ­Berlin: Weidmann. Watkins, C. 2000. The American Heritage Dictionary of Indo-­European Roots. 2nd ed. Boston: Houghton Mifflin. Watson, Lindsay C. 2007. “The Bogus Teacher and His Relevance for Ovid’s Ars Amatoria.” RhM 150.3/4: 337–74. Weinberg, Bernard. 1961. A History of Literary Criticism in the Italian Re­nais­sance. Chicago: University of Chicago Press. Wellesz, Egon. 1957. New Oxford History of ­Music, Vol. 1: Ancient and Oriental ­Music. London: Oxford University Press. Wessner, Paulus. 1902. Aeli Donati quod fertur commentum Terenti; accedunt Eugraphi commentum et scholia Bembina. Vol. 1. Leipzig: Teubner. West, David. 1967. Reading Horace. Edinburgh: University Press. White, Peter. 2007. “Friendship, Patronage, and Horatian Socio-­Poetics.” In Stephen Harrison (ed.), The Cambridge Companion to Horace, 195–206. Cambridge: Cambridge University Press. —­—­—. 1993. Promised Verse: Poets in the Society of Augustan Rome. Cambridge, MA: Harvard University Press. —­—­—. 1978. “Amicitia and the Profession of Poetry in Early Imperial Rome.” JRS 68: 74–92. Wickham, E. C. 1891. The Works of Horace. Vol. 2: The Satires, Epistles, and De Arte Poetica. Oxford: Clarendon Press. Wigodsky, Michael. 1995. “The Alleged Impossibility of Philosophical Poetry.” In Dirk Obbink (ed.), Philodemus and Poetry: Poetic Theory and Practice in Lucretius, Philodemus, and Horace, 58–68. Oxford: Oxford University Press.

294  biblio g r a ph y Wilkins, Augustus S. 1896. Q. Horati Flacci Epistulae. The Epistles of Horace. London: MacMillan. Williams, Gordon. 1995. “Libertino Patre Natus: True or False?” In S. J. Harrison (ed.), Homage to Horace: A Bimillenary Cele­bration, 296–313. Oxford: Clarendon Press. —­—­—. 1980. Figures of Thought in Roman Poetry. New Haven, CT: Yale University Press. —­—­—. 1972. Horace (Greece & Rome, New Surveys in the Classics No. 6). Oxford: Clarendon Press. —­—­—. 1968. Tradition and Originality in Roman Poetry. Oxford: Clarendon Press. —­—­—. 1964. “Review: Horace on Poetry by C. O. Brink.” JRS 54.1/2: 186–96. ­Wills, Jeffrey. 1996. Repetition in Latin Poetry: Figures of Allusion. Oxford: Clarendon Press. Wilner, Eleanor. 2016. “Ars Poetica.” Poetry 207.4: 368–69. Wilsmore, S. J. 1987. “The Role of Titles in Identifying Literary Works.” JAAC 45.4: 403–8. Wilson, Penelope. 2009. “Horace and Eighteenth-­Century Commentary.” In L.B.T. Houghton and Maria Wyke (eds.), Perceptions of Horace: A Roman Poet and His Readers, 271–89. Cambridge: Cambridge University Press. Wiseman, T. P. 1988. “Satyrs in Rome? The Background to Horace’s Ars Poetica.” JRS 78: 1–13. —­—­—. 1982. “Pete nobiles amicos: Poets and Patrons in Late Republican Rome.” In Barbara K. Gold (ed.), Literary and Artistic Patronage in Ancient Rome, 28–49. Austin: University of Texas Press. Wood, Allen George. 1978. “Structural Strategies in the Literary Satires and the Theoretical Poems of Horace, Boileau, and Pope.” PhD diss., University of Michigan. Woods, Marjorie Curry. 2010. Classroom Commentaries: Teaching the Poetria Nova across Medieval and Re­nais­sance Eu­rope. Columbus: Ohio State University Press. Yona, Sergio. 2018a. Epicurean Ethics in Horace: The Psy­chol­ogy of Satire. Oxford: Oxford University Press. —­—­—. 2018b. “An Epicurean ‘Mea­sure of Wealth’ in Horace, Satires 1.1.” ClAnt 37.2: 351–78. —­—­—. 2017. “Horace, Ofellus and Philodemus of Gadara in Sermones 2.2.” Mnemosyne 70: 613–30. —­—­—. 2015. “Some Epicurean Aspects of Horace’s Upbringing in Satires 1.4.” CP 110: 227–51. Zechmeister, Joseph. 1877. Scholia Vindobonensia ad Horatii Artem Poeticam. Vienna: Apud C. Geroldum filium. Zetzel, J.E.G. 2018. Critics, Compilers, and Commentators: An Introduction to Roman Philology, 200 BCE–800 CE. Oxford: Oxford University Press. —­—­—. 2009. Marcus Tullius Cicero: Ten Speeches. Indianapolis, IN: Hackett. —­—­—. 1982. “The Poetics of Patronage in the Late First ­Century B.C.” In Barbara K. Gold (ed.), Literary and Artistic Patronage in Ancient Rome, 87–102. Austin: University of Texas Press. —­—­—. 1980. “Horace’s Liber Sermonum: The Structure of Ambiguity.” Arethusa 13.1: 59–77. Ziolkowski, Jan M. 2016. “Virgil.” In Rita Copeland (ed.), The Oxford History of Classical Reception in En­glish Lit­er­a­ture. Vol. 1: 800–1558, 165–85. Oxford: Oxford University Press.

I n de x

Ars Poetica: ages and aging in, 129–37; amici, 106–7, 154, 157–59, 202; appropriateness of ­human be­hav­ior and emotions in, 69–80; Ars Vivendi, 90, 237–43; art of ­doing every­thing in, 211–21; author’s translation of, xiii–­xli; callida iunctura, 90–99; as complete and exceptional literary achievement, 2–3; date of, 17–22, 103, 241; early manuscripts of, 22n99, 25–26; end of (see end of Ars Poetica); father-­teacher in, 138–49; friendship and laughter together in, 159–72, 255–56; genre of, 13–17; Geoffrey of Vinsauf influenced by, 247–51; ­human nature in, 46–61; as inspiration to other writers, 244–71; judges and judging in, 183–89; late antique and medieval scholars and, 251–52; from laughter to madness in, 189–97, 203; laughter ­towards the end of, 197–99; literary significance of, 1–2; living language in, 62–69; Matthew of Vendôme influenced by, 244–47; metrical patterns in, 20–21; name of, 5–12; opening image of, 39–46; Pisones in, 106–29; latticework through, 92–93; risum, 154–57; senex, 80–90; seriousness of, 242–43; as stand-­alone work, 9; text of, xii; unity of, 2–3, 30–35, 45, 99 Ars Poiētikē, 221–37 Ars Versificaria, 244 Ars Versificatoria, 244–47, 250 Arte nuevo de hacer comedias en este tiempo, 252 Art of Cookery, The, 261

Accius, 85, 121 Adelphoe, 140, 141 Aeneid, 12n49, 90, 118, 203n11, 232 Aeschylus, 55, 76, 85, 91, 121, 149, 261 Agamben, Giorgio, 99n257, 210 agent nouns, 75–76, 80, 114, 118, 166 ages and aging, 129–37 Aimeric, 25 Aldus, 28 Alexander, Elizabeth, 269 alliteration, 70, 159, 234n114, 235 amici, 106–7, 154, 157–59, 202 amicitia, 164n38, 179–82 Amphion, 76, 111, 232, 237, 269n112 Andria, 140, 235 Antipater of Thessalonica, 101, 152 arbitrium, 68, 185 Archestratus of Gela, 14, 15, 261 Aristarchus, 169–70, 171, 175, 184, 189 Aristophanes, 13, 13n53, 15, 16, 112, 206n24, 211n41, 237 Aristotle, 1, 13, 15, 16, 24, 26, 28, 29n133, 44n26, 45, 85n197, 90, 154, 251; on appropriate be­hav­ior, 70–71; availability of works of, to Horace and other Romans, 26n117; friendship explored by, 162; instruction by, 117; pathos and, 47; on physis, 60n93. See also Poetics; Rhe­toric Armstrong, David, 32, 34, 48, 104, 120, 141, 172, 205, 227–28 Ars Amatoria, 5, 16 Ars Grammatica, 5 Ars lectoria, 25 295

296 In de x art of ­doing every­thing, 211–21 Auden, W. H., 262 audience, 149–50, 154, 161–62 Augustus, 18–19, 103, 105, 176; Carmen Saeculare, 179n91 Aulus Cascellius, 18, 126 Badius, 27, 44 Bailey, Shackleton, 44 beauty and sweetness, 51, 155 Bogue, Ronald Lynn, 251, 252, 256, 259, 267 Boileau, Nicolas, 1, 16, 29, 112n51, 118–19n81, 251, 252–56, 259, 260, 261, 270 Borges, Jorge Luis, 125n103, 265–68 Brink, C. O., 2, 44, 45, 62, 65, 75, 96, 98n254, 115, 121, 143, 160, 169, 170, 180, 196 Brunt, P. A., 173 building proj­ects, 66–67 Byrom, John, 252 Byron, Lord, 257n54, 260, 262 Caecilius, 63, 68 Caesar, 176 Caesar, Julius, 101, 103 callida iunctura, 90–99 Callimachus, 15, 239, 247n16; Callimachean tradition, 66 Carmen Saeculare, 19, 21, 179n91 castigator censorque, 82, 84, 141, 147 Cato, 64, 68 Catullus, 125n102, 235 Cennini, Cennino, 258 Charisius, 11 Cicero, 23, 101, 148, 163, 164, 165n43, 174, 190, 232, 236n124 collatis membris, 41–42 color, 211–15, 217, 218–21, 264, 269, 270 comedy, 14–16, 49n50, 55, 76, 139–41, 144, 148–49, 259 Commentarii in Artem Poeticam, 9 Commentator Cruquianus, 10–11, 11n40 Conrad of Hirsau, 10, 25 conversation, 17, 63–66. See also sermo countryside, the, 54–55

creatures, hybrid, 39–43, 263 de Crucque, Jacques, 11n40 dactylic and spondaic feet in Ars Poetica, 20–21 Dalton, Roque, 268 D’Anna, Giovanni, 21 Dante, 26, 264 date of Ars Poetica, 17–22, 103, 241 da Vinci, Leonardo, 258 Davis, Carol Ann, 269 Davis, Gregson, 32 De Amicitia, 163, 174, 180 de arte versus ex arte, 24, 81n178, 250–51, 257, 265 De Arte Graphica, 258 De Arte Poetica, 251 death, 66–67, 129–37, 141n157, 205–6, 265, 267, 268 De lingua Latina, 64n109, 97n250, 235 Democritus, 58, 91, 189–92 De Rerum Natura, 14–15, 16, 53n69, 64, 108, 131, 197n171, 231n108 derisum, 170, 197 Dialogues of the Dead, 260 Dialogus super auctores, 10, 25 didactic poetry, 5n18, 13, 14–15, 16–17, 22, 24, 33, 35, 81, 85, 91, 107n34, 108–9, 116, 138, 139, 142, 144, 151, 164, 180n96, 201n6, 242, 247, 250, 261 Dilke, O. A. W., 90, 128 Dillon, Went­worth, 252. See also Roscommon Diomedes, 86, 235–36 Disch, Tom, 269 Divine Comedy, 26, 264 Dolce, Ludovico, 258–59 Donatus, 5, 86–87, 235–36 drama, 53–56, 61, 78–79, 82–91, 94–95, 121, 148–49, 252, 254, 259. See also comedy; tragedy; satyr-­play Dryden, John, 252–53n43, 253, 254, 255, 259–60 Duckworth, G. E., 20–21

In de x   Dufresnoy, Charles, 258 dulcia, 50–51 Eberhard of Béthune, 244–45, 248n24 education, Roman, 50, 128–29, 226 Eliot, T. S., 5 Emerson, Ralph Waldo, 262 emotions, ­human, 46–50, 52n66, 69–70, 161, 250; appropriateness of ­human be­hav­ior and, in Ars Poetica, 69–80 Empedocles, 58, 117n73, 185, 201, 204, 205–6, 206n21–22, 207–8, 220, 232, 161, 262n85 end of Ars Poetica, 200–211; Ars Poiētikē, 221–37; Ars Vivendi, 90, 237–43; art of ­doing every­thing in, 211–21 En­glish Bards and Scotch Reviewers, 260 Ennius, 15, 16, 31, 64, 68, 85, 121 epic poetry, 13–15, 48, 90, 94, 117, 118, 119 Epicurean: philosophical school, 34, 47n42, 130n121, 132n132, 148n177, 201n6; therapeutics, 120n87, 172; poetry and poetics, 43n25, 48, 99n256; friendship, 163–64, 165n43, 174, 178, 180, 181n108, 183n114, 186, 199. See also Philodemus Epicurus, 16, 33n152, 177n84 Epistles, 2, 4, 9n31, 10–11, 13, 15, 17, 19, 20–21, 26, 67–68, 75, 89, 100, 108–9, 116, 130, 152, 205, 233–34, 238, 240, 247; friendship in, 176–77, 180, 182–83 “Epistle to a Friend on the Art of En­glish Poetry, An,” 252 Epistula ad Pisones, 10, 11, 100. Epodes, 21–22n97, 41, 176, 200, 233, 241, 246, 247 Essay on Criticism, An, 94n236, 187n132, 255–57, 260 Essay on Epic Poetry, An, 252 Essay on Rime, 262–64 Essay on Satire, An, 252 Essay on Translated Verse, An, 252, 260n72 Essay upon Unnatural Flights in Poetry, An, 252 Evanthius, 86–87

297 facundia, 54, 55, 62–63 ­fathers and sons, 45, 100, 106, 125–26, 129, 130, 132, 137, 140, 152–53, 158, 232, 244, 249 father-­teacher, Horace as, 138–49 Fresnaie, Jean Vauquelin de la, 252, 258n62 Freudenburg, Kirk, 16, 30, 162, 238 friendship, 76, 106–7, 157–59, 189, 196, 197, 243, 253, 254–56; Epicurean, 163–64, 165n43, 174, 178, 180, 181n108, 183n114, 186, 199; Horace as amicus pisonum and Roman, 173–83; laughter and, 154, 159–72, 198, 201. See also amici; amicitia Friis-­Jensen, Karsten, 24, 28 Frischer, Bernard, 2n12, 20–21n94, 104 Frogs, 13, 15, 112, 237 genre of Ars Poetica, 13–17, 109, 247 Geoffrey of Vinsauf, 1, 16, 244, 245n4, 247–51, 254n48, 256, 270–71 Georgics, 14–15, 16, 32n149, 60n92, 116, 234, 242 Gervase of Melkley, 244, 245n4 Goethe, Johann Wolfgang von, 4, 99, 256n54, 270 Gold, Barbara K., 149–51 Golden, Leon, 256, 262, 264 Goulbourne, Russell, 262 Granville, George, 252 Hardie, Philip, 96, 131 Hardison, O. B., Jr., 256, 262, 264 Harte, Walter, 252 Hayley, William, 252 Heautontimorumenos, 140, 141 Hedypatheia, 14 Hedyphagetica, 15 Helenius Acro, 9. See also pseudo-­Acro Henderson, John, 242 Hesiod, 14, 15, 57n83, 108, 151 hexameter(s) of Ars Poetica, 13, 17, 20–21, 35, 85, 210, 244 Hints from Horace, 260, 262 History of Literary Criticism in the Italian Re­nais­sance, A, 29n133

298 In de x Homer, 48, 59, 65, 118–19, 156, 257, 261 Horace: as amicus pisonum, 173–83; audience of, 154, 161–62; declaring himself a son of a freedman, 139n151; as father-­teacher, 138–49; as “gold” writer, 25; poems of, as mosaics, 31n144; self-­description by, 141–42n159; technical virtuosity of, 32; dating of works of, 16, 17–22; title of Ars Poetica and, 12 Horace’s Narrative Odes, 30 Horatius cum quattuor commentariis, 28 Horatius iudex, 183–89, 225 Horatius pater, 138–49, 144–45, 183 Horatius senex, 80–90, 141–42, 144–45, 148, 183 Horsfall, Nicholas, 8 Huidobro, Vicente, 267, 268 humana natura, 58–60, 72 ­human frailty and failings, 57–58, 61, 72 humano, 39–46, 56, 58, 89, 106–7, 241; humano capiti, 12, 204; ­human nature and, 46–61; living language and, 62–69 ­human speech, 63–66, 68–69, 79 Idea of Lyric, The, 32 Iliad, 65n114, 119, 131n125, imitatio, 240 in medias res, 1, 42, 119 Innocent, Pope, 248–49 In Pisonem, 101, 165n43, 236n124 Institutio Oratoria, 6–7 Irigoyen, Ramón, 268 Isidore of Seville, 235, 236 Janko, Richard, 34 John of Garland, 244, 250n34 Johnson, W. R., 32 Jonson, Ben, 1, 255n51, 260, 270 judges and judging, 56, 183–89 Juvenal, 17, 22n99, 25, 71 Kilpatrick, Ross S., 180 King, William, 261 Konstan, David, 108, 179, 181

Laborintus, 245, 248n24 Laird, Andrew, 211, 240 Landino, Cristoforo, 27–28 language, living, 62–69 L’Art poétique, 253–55, 259, 261 L’Art poétique François, 252 Lasky, Dorothea, 269 laughter, 39, 43, 47, 49, 70, 118, 154–57, 202, 209; friendship and, 45, 159–72, 198, 199, 201; to madness, 189–97 laws, 56, 183–85 Leach, Eleanor Winsor, 139–40, 148 Levinson, Jerrold, 269 life cycles, 3, 45, 60, 65–67, 74–77, 131–33, 137, 141, 157–58, 223, 249, 268 lingua, 52n67, 53, 57, 64, 69 living language, 62–69 Livy, 54n74, 232 Lope de Vega, Don Frey, 252, 259 Lowrie, Michèle, 30, 32, 63, 94 Lucan, 25, 246 Lucilius, 16, 17, 61, 147, 168, 238, 248 Lucretius, 14, 15, 16, 40, 49n54, 53n69, 60, 64–65, 96n242, 122–23n93, 131, 152n189, 197n171, 202n9, 231n108, 232, 261 Lyttelton, George, 260 Macbeth, 5 MacLeish, Archibald, 264–65, 267, 268 MacLennan, Amy, 269 madness, 58, 137, 146, 189–97, 198n174, 199, 203, 208, 209, 220, 225, 232 Maecenas, 105, 107n31, 116, 125n102, 129, 149, 152, 176, 181 Maecius, 18, 135n139, 188 Mancinelli, Antonio, 27n125, 28 Margites, 14 “Materia” commentary, 47n41, 66n120, 70n138, 71n140, 87n205, 89n219, 106n28, 127n110, 160n20, 199n177, 212n45, 213–14n52, 236, 245 Matthew of Vendôme, 244–47, 248n23, 249n29, 250 Maximus, Lollius, 181–83, 234n112

In de x   Maximus of Tyre, 162 ­mental health, 189–97 Messalla, 18, 126 Michaelis, A., 21 Milosz, Czeslaw, 264–65 Molière, 259–60 Mount Vesuvius, 33 Nachleben, 3 naming of Ars Poetica, 5–12 nature/natura, 54, 60, 92, 132, 208, 250; personification of, 52–53, 66–67, 131, 196n170, 223, 231, 249, 254, 257, 265, 270; power of, 53n69, 56, 261 negative ­human nature, 57–58, 61 Neoptolemus of Parium, 1, 13–14, 15, 16, 31, 227, 229–30 Newman, John Kevin, 21, 111, 112 Newton, Isaac, 261 Nietz­sche, Friedrich, 94 norma, 68, 185 Norton Anthology of Theory and Criticism, 1 Notes ­Toward a Supreme Fiction, 264, 269 nouns, agent, 75–76, 80, 114, 118, 166 Odes, 19, 22, 26, 56, 63, 66, 67, 75, 76, 149–50, 194, 197, 200–201, 233, 240; ages and aging in, 129, 130, 132, 146, 149–50; friendship in, 176; Horace’s self-­re-­ presentation in, 192; structures of as analogue for the Ars Poetica, 30–33, 94 On Death, 130n121, 190n140, 201n6, 205 On Flattery, 35, 163–64, 177, 182n110, 190 On Frank Criticism, 156n10, 163–64, 164n38, 167, 171–72, 177, 190 On Greed, 35 On Nature, 117n73, 261 On Poems, 14, 33, 34, 35, 39n1, 91, 117, 227, 236n124 On Property Management, 35n159, 128n118 On Slander, 35, 167n48 On the Sublime, 1, 207, 253 On Vices and Their Opposing Virtues, 35, 113n57, 164n37, 181n108

299 Orpheus, 58, 111–12, 237, 268n108, 269n112 Ovid, 5, 16, 25, 207n28, 234, 246 Pa­ri­siana Poetria, 244, 250n34 paternity, vocabulary of, 64–65, 132, 134, 137, 138, 140, 142, 185 pathos, 47 Pérez del Camino, Don Manuel Norberto, 251, 267 Peri Edesmaton, 15, 16 Persius, 17, 22n99, 25, 49n50, 86–87, 95, 162, 238 Petrarch, 26 Phaedrus, 266 Philodemus, 1, 15, 16, 33–35, 48, 91, 101, 236n124, 242–43; on friendship, 163–64, 165n43, 167–68, 171–72, 177–79, 199; Horace following the poetics of, 33–35, 48, 188; on ­human emotions, 52n66; and identity of the Pisones, 101–2, 104, 105, 151, 152, 171; instruction by, 117; self-­referential closure in On Death, 205; as a source for Neoptolemus, 227–30 Piso, Gnaeus Calpurnius, 19, 101–3, 104, 105, 126, 151 Piso (Augur), Lucius Calpurnius, 103 Piso (Caesoninus), Lucius Calpurnius, 19, 101–2, 104, 105, 126, 169n55, 171, 236n124 Piso (Pontifex), Lucius Calpurnius, 100–5, 126 Pisones, 10, 18, 22, 30, 85, 91, 149–53, 171; ages and aging in, 132–37; in the Ars Poetica, 106–29; Boileau’s understanding of Horace’s, 252–53, 255; conciliatory gestures with critique ­toward, 109–10, 119–20, 122, 127, 186; friendship and, 173–83; Horace’s criticism and instruction ­toward, 115–28; Horatius pater ­toward, 138–49; identity of, 100–106; judging and, 188–89; as vates, 111–12; in versate, 121–22; vestri proavi and, 122–23 Plato, 13, 15, 40, 104n24, 141n155, 162, 163, 164, 266; on physis, 60n93 Plautus, 63, 68, 85, 91, 140n154

300 In de x Plotius (Tucca), 35, 163, 167n48, 181 Plutarch, 162, 220 poeta/poema, 221–37, 267 “Poética,” 251, 267 Poetics, 13, 14n60, 24, 26, 28–29, 42n17, 77, 81n178, 117, 212n46, 223–224n92, 251 Poetria, 10, 236, 248n20 Poetria Nova, 244, 247–51 Poetria Vetus, 248, 252n42 Poetry of Friendship, The, 180 Polyhymnia, 32 Pomeroy, Arthur J., 150 Pope, Alexander, 1, 16, 29, 94, 141n157, 145n169, 187n132, 206n23, 251, 255–57, 260, 261, 263, 270 Porphyrio (Pomponius Porphyrio), 9, 14, 17, 18, 19, 28, 227, 229; on identity of the Pisones, 100–101, 102, 104 Porter, James I., 207–10 Pro Caelio, 148 Propertius, 234 pseudo-­Acro, 9n31, 28, 70, 102, 224 pseudo-­Longinus, 1, 207, 253 pulchra, 50–51, 50n58 purpureus . . . ​pannus, 1, 213–14n52, 215n59, 266 Quintana, Manuel José, 252 Quintilian, 6–9, 11, 12 Quintilius (Varus), 18, 35, 167–69, 171, 175, 181, 189, 226 qui scribitis, 62, 114–15, 135, 216 reader-­response criticism, 43n25 “Reglas del drama,” 252 Republic, 13 Rhe­toric, 13, 49n53, 71, 74n149, 80n175 risum, 154–57, 189. See also derisum; laughter Roman education, 50, 128–29, 226 Roman friendship, 157, 173–83 Roscommon, 252, 260 Rudd, N., 68, 160, 161, 170, 191 Russell, D. A., 99

Sallust, 25 Sanabria, Ruth Irupé, 269 Santirocco, Matthew S., 32 sapientia, 57, 127n111 satire, 15–16, 31, 35, 49n50, 86–87n204, 86–89, 109, 209, 255. See also satura; sermo/sermones Satires, 3–4, 15, 17, 19–20, 22, 31, 34, 45, 56, 59, 61, 63, 85, 94n236, 108, 124, 134, 145, 148–51, 197, 200, 201, 209, 221, 233, 237–44, 247, 259, 264–65; appropriateness of ­human emotions and be­hav­ior in, 69–80; conversations in, 66; ending of, 209; father-­teacher in, 138–42; friendship in, 164, 176–77, 180–81; law in, 183–85, 189n138; satire genre and, 86–90. See also satura; sermo/sermones satura, 15, 17, 86–89, 93, 264 Satyrorum scriptor, 85, 88, 142, 185 satyr-­play, 85–90, 149 Scaliger, Julius Caesar, 2, 29n133, 237–38 senex, 73, 80–90, 82n180, 132, 136, 141–42, 144, 145, 148, 183, 249 sermo/sermones, 15, 17, 19, 35, 63, 64–67, 86, 89, 93, 94, 97, 109, 177, 237, 255, 263 Servius, 9, 108, 213–14n52 Shakespeare, William, 269 Shapiro, Karl, 262, 263, 265 Simonides, 130n124, 220, 221n85, 258 Smith, Adam, 67 Soames, William, 252n43, 253, 254, 255, 259 soul/spirit, ­human, 49–50, 51, 80, 133, 161 Statius, 25, 246 Stephanus, 18 Stevens, Wallace, 264, 267n107, 269 sublime, the, 204, 207–11, 253, 268n108 Suetonius, 18, 19, 232n109, 235–36. See also Vita Horatii suicide, 185, 201n6, 205, 263. See also Empedocles Tarán and Gutas, 28–29 Terence/Terentian, 25, 139, 140–41, 148, 235–36

In de x   Terentius Scaurus, 9 Thespis, 55, 85, 91, 121, 149, 218, 225 This Craft of Verse, 265–66 Thomas, Richard, 16 three-­actor rule, 82–83n186 tibia, 53–54, 83 Tibullus, 234 title of Ars Poetica, 5–12 tragedy, 49, 50, 55, 78, 82n183, 82n186, 90n221, 93, 95, 121. See also comedy; drama Troiano, Marita, 268–69 unity of Ars Poetica, 2–3, 30–35, 45, 99 usus, 67–68, 185 ut pictura poesis, 1, 186–88, 211, 220, 226, 229, 258, 266 Varius, 18, 35, 63, 68, 163, 181 Varro, 15, 16, 54n73–74, 228n102, 235–36. See also De lingua Latina

301 vates, 21, 111–12, 134, 205n20, 225n94 versate, 113–14, 116, 121–22 vestri proavi, 122–23 Vida, Marco Girolamo, 251 Villa dei Papiri, 33, 101–2 Virgil, 14, 15, 16, 18, 25, 35, 40, 63, 65, 68, 118–19, 176, 181, 232, 242, 246, 266n101; Aeneid, 90; living language and, 63, 65; on nature and Homer in Pope, 257; use of poeta, 234. See also Aeneid; Georgics Vita Horatii, 18, 128n117, 139n151 Vital, Alberto, 267n105–6, 268 Wasteland, The, 5 Weinberg, Bernard, 28, 29n133 White, Peter, 173 Whitman, Walt, 262 Wilner, Eleanor, 269 Works and Days, 14n58, 15n65, 108, 117n73, 151n184, 223n92

A NO T E ON T H E T Y P E This book has been composed in Arno, an Old-style serif typeface in the classic Venetian tradition, designed by Robert Slimbach at Adobe.