Arbitri Nugae: Petronius’ Short Poems in the "Satyrica" (Studien zur klassischen Philologie) [New ed.] 9783631605837, 9783653007084, 3631605838

Table of contents :
Table of Contents
Introduction - Petronius’ Poems in the Satyrica 1
Chapter I - The Education of the Orator (Petr. 5) 15
Chapter II - Justice for Sale (Petr. 14.2) 51
Chapter III - Two Views of Success (Petr. 15.9; 18.6) 61
Chapter IV - Petronius’ Sotadeans (Petr. 23.3; 132.8) 73
Chapter V - Trimalchio’s ‘Epigrams’ (Petr. 34.10; 55.3) 91
Chapter VI - Perverted Taste (Petr. 55.6; 93.2) 113
Chapter VII - A Night of Love (Petr. 79.8) 133
Chapter VIII - Role-Playing (Petr. 80.9) 141
Chapter IX - Tantalus and the Miser (Petr. 82.5) 157
Chapter X - Life Choices (Petr. 83.10) 163
Chapter XI - Appeal to Peace (Petr. 108.14) 169
Chapter XII - A Dirge on Lost Hair (Petr. 109.9-10) 177
Chapter XIII - Amorous Blasphemy (Petr. 126.18) 193
Chapter XIV - Homeric Love (Petr. 127.9) 199
Chapter XV - A Vanishing Dream (Petr. 128.6) 211
Chapter XVI - Love in an Ideal Landscape (Petr. 131.8) 225
Chapter XVII - A Literary Vindication (Petr. 132.15) 243
Chapter XVIII - Encolpius and the Role of Priapus (Petr. 133.3; 139.2) 265
Chapter XIX - The Sorceress’ Claim (Petr. 134.12) 285
Chapter XX - Oenothea’s Cottage (Petr. 135.8) 303
Chapter XXI - Mythological Monsters (Petr. 136.6) 317
Chapter XXII - The Omnipotence of Gold (Petr. 137.9) 329
Appendix I - Vegetables and Bald Heads (Petr. 109.10.3-4) 345
Appendix II - Magic at Petr. 131.4-6 357
Appendix III - Novelistic Love between Parody and Ideal: The Greek Novelists and Petronius 369
Works Quoted 391

Citation preview

165 This book aims to provide a comprehensive inquiry into the short metrical intermezzos inserted in the prose narrative of Petronius’ Satyrica. The text of each poem has been thoroughly investigated; in addition, special attention has been devoted to their function in the context and to the aspects connecting Petronius with the literature and culture of his time. Numerous contacts with other ancient authors have been pointed out to illustrate Petronius’ attitude to the cultural and literary heritage on the one hand, and the character of his own work on the other.

www.peterlang.de

SKP 165-Setaioli-260583HCA5-AK.indd 1

Aldo Setaioli · Arbitri Nugae — Petronius’ Short Poems in the Satyrica

Herausgegeben von Michael von Albrecht

LANG

Aldo Setaioli is professor emeritus of Latin Language and Literature at the University of Perugia (Italy). He has published widely in the area of Latin literature, never losing sight of Greek antecedents and devoting special attention to such authors as Seneca, Virgil and his commentators, Horace, Petronius, and several others, as well as to afterlife beliefs in the Greek and Roman world down to late antiquity.

Studien zur klassischen Philologie

Aldo Setaioli

Arbitri Nugae Petronius’ Short Poems in the Satyrica

Peter Lang

165

ISBN 978-3-631-60583-7

12.11.10 11:13:21 Uhr

165 This book aims to provide a comprehensive inquiry into the short metrical intermezzos inserted in the prose narrative of Petronius’ Satyrica. The text of each poem has been thoroughly investigated; in addition, special attention has been devoted to their function in the context and to the aspects connecting Petronius with the literature and culture of his time. Numerous contacts with other ancient authors have been pointed out to illustrate Petronius’ attitude to the cultural and literary heritage on the one hand, and the character of his own work on the other.

Aldo Setaioli · Arbitri Nugae — Petronius’ Short Poems in the Satyrica

Herausgegeben von Michael von Albrecht

LANG

Aldo Setaioli is professor emeritus of Latin Language and Literature at the University of Perugia (Italy). He has published widely in the area of Latin literature, never losing sight of Greek antecedents and devoting special attention to such authors as Seneca, Virgil and his commentators, Horace, Petronius, and several others, as well as to afterlife beliefs in the Greek and Roman world down to late antiquity.

Studien zur klassischen Philologie

Aldo Setaioli

Arbitri Nugae Petronius’ Short Poems in the Satyrica

Peter Lang

165

www.peterlang.de

SKP 165-Setaioli-260583HCA5-AK.indd 1

12.11.10 11:13:21 Uhr

Arbitri Nugae

Studien zur klassischen Philologie Herausgegeben von Prof. Dr. Michael von Albrecht

Band 165

Peter Lang

Frankfurt am Main · Berlin · Bern · Bruxelles · New York · Oxford · Wien

Aldo Setaioli

Arbitri Nugae Petronius’ Short Poems in the Satyrica

Peter Lang

Internationaler Verlag der Wissenschaften

Bibliographic Information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data is available in the internet at http://dnb.d-nb.de.

ISBN 978-3-653-00708-4 (eBook) ISSN 0172-1798 ISBN 978-3-631-60583-7 © Peter Lang GmbH Internationaler Verlag der Wissenschaften Frankfurt am Main 2011 All rights reserved. All parts of this publication are protected by copyright. Any utilisation outside the strict limits of the copyright law, without the permission of the publisher, is forbidden and liable to prosecution. This applies in particular to reproductions, translations, microfilming, and storage and processing in electronic retrieval systems. www.peterlang.de

Table of Contents

Introduction Chapter I Chapter II Chapter III Chapter IV Chapter V Chapter VI Chapter VII Chapter VIII Chapter IX Chapter X Chapter XI Chapter XII Chapter XIII Chapter XIV Chapter XV Chapter XVI Chapter XVII Chapter XVIII Chapter XIX Chapter XX Chapter XXI Chapter XXII Appendix I Appendix II Appendix III Works Quoted

Petronius’ Poems in the Satyrica The Education of the Orator (Petr. 5) Justice for Sale (Petr. 14.2) Two Views of Success (Petr. 15.9; 18.6) Petronius’ Sotadeans (Petr. 23.3; 132.8) Trimalchio’s ‘Epigrams’ (Petr. 34.10; 55.3) Perverted Taste (Petr. 55.6; 93.2) A Night of Love (Petr. 79.8) Role-Playing (Petr. 80.9) Tantalus and the Miser (Petr. 82.5) Life Choices (Petr. 83.10) Appeal to Peace (Petr. 108.14) A Dirge on Lost Hair (Petr. 109.9-10) Amorous Blasphemy (Petr. 126.18) Homeric Love (Petr. 127.9) A Vanishing Dream (Petr. 128.6) Love in an Ideal Landscape (Petr. 131.8) A Literary Vindication (Petr. 132.15) Encolpius and the Role of Priapus (Petr. 133.3; 139.2) The Sorceress’ Claim (Petr. 134.12) Oenothea’s Cottage (Petr. 135.8) Mythological Monsters (Petr. 136.6) The Omnipotence of Gold (Petr. 137.9) Vegetables and Bald Heads (Petr. 109.10.3-4) Magic at Petr. 131.4-6 Novelistic Love between Parody and Ideal: The Greek Novelists and Petronius

1 15 51 61 73 91 113 133 141 157 163 169 177 193 199 211 225 243 265 285 303 317 329 345 357 369 391

V

This book collects and updates essays on the short poems in the Satyrica which were written (for the most part in languages other than English) over a period of several years but were conceived from the beginning as an organic whole. It aims to provide a comprehensive inquiry into the metrical sections inserted in the prose of Petronius’ novel, with the exception of the two most extensive pieces (the Troiae halosis and the Bellum civile), both of which would require a monograph in its own right. All the problems posed by these sections have been addressed, although questions of metrics have been treated only insofar as they are connected with the general interpretation of each poem, in view of the detailed study recently devoted to this aspect by W.-J. Yeh. Special attention has been devoted to the function of the poems in relation to the work as a whole (at least to what we can still read) as well as to aspects connecting Petronius with the literature and culture of his time. Many such connections have been pointed out for the first time and are meant to shed new light not merely on single passages but also on extended portions of Petronius’ work, as well as on his general attitude to the cultural and literary heritage on the one hand, and on the character and aims of his own work on the other. Literary models (or anti-models) belonging to the serious literature of the past can thus be recognized, often functioning as the butt of Petronius’ parody and desecration, whereas contacts with contemporary or later authors and writings, often of a less elevated level, help to see his work in a historical and cultural perspective. The chapters treating the several metrical sections generally follow the sequence in which the latter appear in the text. In some cases two different pieces have been treated in the same chapter, when they are linked by conspicuous likenesses or resemblances. VI

Introduction Petronius’ Poems in the Satyrica* Though occasionally verse may be found in other ancient novels (e.g. Apul. met. 4.33 and 9.8) and quotations from Homer, linguistically integrated in the text, often appear in Chariton, the Satyrica is unique in ancient fiction in that no less than 30 poems (including the Virgilian cento at 132.11) appear in the preserved parts alone. There are then at least as many poems that have been transmitted as fragments under the name of Petronius, mainly in the Anthologia Latina. Though these are anything but devoid of interest and have been rightly revalued1 after a long neglect largely due to mistrust in their Petronian authenticity, we may confidently assume that the lack of a context makes their full appreciation impossible. The poems transmittetd together with the prose by Petronius’ direct tradition are in fact an integral part of the novel2 and acquire full meaning and import only in relation to the prose frame in which they appear. On the one hand literary nods and allusions often interpenetrate both the poems and the surrounding prose; and on the other, as we shall see, several of the poems that can be attributed to Encolpius define and demarcate a level distinct from that of the prose narrative, but inseparable from it, in that verse and prose alternate in expressing the moods of the character, as he goes through his multifarious adventures, or of the narrator, as he reports and recalls his past vicissitudes later in time.3 There are, besides, some poems which are even syntactically connected with the prose context. One is the poem recited by Tryphaena at 108.14,4 which contains a ver* 1 2 3

4

A version of this Introduction will appear with the title Poems in Petronius’ Satyrica in the Wiley-Blackwell Companion to the Ancient Novel. E.g. by Courtney 1991 and Sommariva 2004. This has been recognized by most scholars; cf., for example, Barnes 1971, 6-7; Connors 1998, 1; 4; Yeh 2007, 57. It can be added that the verse pieces tend to pour into the recognized molds of literary convention the often incongruous situations described in the prose and to extract their universal implications. Cf. Connors 1998, 2, 9, 50. It may be said that they have an “organizing” function (Yeh 2007, 57). Cf. ch. XI, with the literature quoted and discussed. For reasons of space, detailed mention of the relevant bibliography is avoided here. References will be made to the follow-

Introduction bum dicendi in the first line, functioning as the main clause of a sentence begun in the preceding prose narrative (‘quis furor’ exclamat ‘pacem convertit in arma?’, significantly quoted by Isidore of Seville, orig. 2.21.19 in this form: quis furor, o cives, pacem convertit in arma?; cf. Lucan. 1.8). Two others, 128.65 and 136.6,6 are often regarded as incomplete, but in reality both are the parabole of similes whose antapodosis rounding off the sentence is contained in the prose. Not unlike Isidore with the poem at 108.14, the excerptor of the Florilegia tradition ( ) has changed the first line of 128.6, which he reports without the prose context, to make the poem syntactically self-standing (veluti cum has been changed to si quando). Even the cento at 132.11 functions as the main clause of a sentence begun in the prose. So, as far as we can see, all the poems in the Satyrica are complete, with the single exception of the two lines at 15.9. Ernout, at the end of his edition, gives a list of nine different meters employed in the poems, including the fragments.7 Hexameters and elegiac couplets are by far the most common, but hendecasyllables are also well represented and two more types are remarkable: the Sotadeans8 used by a cinaedus in a solo performance at 23.3 and again by Encolpius at 132.8, and also the peculiar arrangement of two hexameters and one pentameter employed by Trimalchio in his epigram at 34.10 and then attempted again (unsuccessfully, as it seems) at 55.3. Petronius’ poems have been especially investigated by Heinz Stubbe,9 E.J. Barnes,10 Edward Courtney,11 Catherine Connors,12 and, lately, by Wei-Jong Yeh,13 in a bulky book which addresses especially the problem of the Satyrica’s prosimetric mixture and – exhaustively – Petronius’ metrical technique. Con-

5 6 7 8 9 10 11 12 13

2

ing chapters, in which it is systematically surveyed. I further refer to the bibliography on Petronius’ poems quoted and discussed by Yeh 2007, and, for recent scholarship on Petronius in general, to Vannini 2007. Cf. also Habermehl 2006, for the parts covered by his commentary. Cf. ch. XV, with the literature quoted and discussed. Cf. ch. XXI, with the literature quoted and discussed. Barnes 1971, 317 rightly remarks that “the variety of metres lends corresponding variety in the types of literary caricature and amusement available to the author”. For the two poems in Sotadeans cf. ch. IV, with the literature quoted and discussed. Stubbe 1933, who, though covering all the poems, deals mostly with the Troiae halosis and the Bellum civile. Barnes 1971. Courtney 1991. Connors 1998. Yeh 2007. Yeh’s conclusions, though ostensibly based on a thorough analysis of metrical technique, often appear to be far-fetched and hardly convincing. For example, he dates the Satyrica under the reign of Domitian. However, his work is invaluable as far as metrical questions are concerned and has made further investigation in this area all but superfluous. Among the numerous papers concerned with Petronius’ poems mention may be made of Beck 1973 and Sommariva 1996.

Petronius’ Poems in the Satyrica cerning the former problem, we should at least call attention to the fact that the recent publication of papyri containing scraps of ancient fiction with verse inserted in the prose has cast some doubt on the former classification of the Satyrica as a Menippean satire, a literary genre in which both verse and prose are employed.14 One of these fragments, the so-called Iolaus (POxy 3010), appears to contain a lewd narrative, quite different from the Greek novels of idealized love that have come down to us, and it also features a speech in Sotadeans uttered by a gallus, i.e. a character not too different from Petronius’ cinaedus reciting Sotadeans at 23.3. Another narrative fragment containing verse inserted in the prose has since been published (the so-called Tinouphis), and this too seems to differ from the Greek novels previously known. Though both these papyri go back to a time later than Petronius, it is then not impossible that a type of satiric narrative featuring roguish characters and employing both verse and prose preexisted the Satyrica; but our evidence is too meager to admit of any unqualified statement about the generic affiliation of Petronius’ novel. Surely there is a connection between the verse and the literary affinities displayed – and almost flaunted – in every page of the Satyrica, almost invariably in the form of parody and desecration. This is true for all the poetic forms appearing in the novel, but it is particularly evident in the use of Sotadeans. Each of the two poems is a specimen of the two types of poetry attested for Sotades of Maroneia: song performed by the cinaedi and obscene parody of epic. The Sotadeans at 23.3 are indeed recited by a cinaedus and those at 132.8 cast Encolpius and his treacherous mentula in the roles of Achilles and Agamemnon in Iliad 1: as in Homer, after a frustrated attempt to exact bloody vengeance, the indignant hero is content with hurling verbal abuse at the culprit. As often, the parodic mood overflows into the prose and reaches the climax with the cento (132.11) obscenely desecrating two of Virgil’s loftiest figures: Euryalus and Dido.15 Some of the Satyrica’s verse does in fact contain programmatic references to the author’s literary attitude. The scholars who see a hint at the novel’s closeness to the mime in the second part of the poem at 80.9 (v. 5 grex agit in scaena mimum)16 are probably right – with no need to split the eight lines of the poem in two parts of four lines each, since there are clear structural links uniting the two “quatrains”17 and the poem is unanimously given as a whole in the manu14 15

16 17

The connection of the Satyrica with Menippean satire was roundly denied by Astbury 1977. Recently it has been strongly upheld by Yeh 2007, 20, 37-40. These (132.8 and 132.11) are two of the very few poems that are functional to the narrative, carrying it forward, rather than pausing to reflect on the situation sketched in the prose. For example Connors 1998, 13-14. For 80.9 cf. ch. VIII, with the literature quoted and discussed. It should be emphasized that nomen amicitiae (v. 1) corresponds to nomen divitis (v. 6); vultum (v. 3) corresponds to facies (v. 8); and vultum servatis (v. 3) anticipates both the theatrical metaphor of 5-8 and the emphasis these final lines place on pretence and fictionality. In 3

Introduction script tradition.18 In view of this, assuming that the last four lines were added to those that precede by an excerptor who saw the closeness and similarity of two different poems19 is a clear case of begging the question. Several other programmatic references have been seen in other poems and passages (including such unlikely details as the pastry Priapus at 60.4, the pebble in fr. 40.8 M., and the parrot in fr. 45 M.),20 but, strangely enough, many scholars refuse to recognize them where, in my opinion, they are most evident. At 132.12 Encolpius feigns shame for having addressed his recalcitrant mentula, but immediately after (132.13-14) he adduces the examples of common people cursing their aching limbs (those who complain about their belly remind us of Trimalchio at 57.2) and of epic and tragic heroes addressing parts of their own body (Ulysses is explicitly mentioned, Sophocles’ Oedipus is clearly alluded to), and finally breaks into verse (132.15)21 to vindicate his novae sinceritatis opus and the smiling gratia candidly portraying matters of sex as opposed to the supercilious hypocrisy of his censors, who are also targeted in the prose sentence that follows the verse. The stage illusion is undoubtedly broken here. Encolpius is alone: he would not need to justify his outburst, since there has been no witness, and nobody is there to listen any way. But Encolpius – and Petronius behind him22 – is addressing his readers; the whole context (prose and verse) lucidly lays bare the main components of the work (opus; and cf. the genitive of quality, novae simplicitatis, comparable to 4.5 schedium Lucilianae humilitatis, where a literary composition is surely referred to): “realism”, literary parody,

18 19 20

21 22

4

modern terms, we can say that the “hypocrisy” of false friendship, which is the subject of the first part of the poem, is explained and revealed, as it were, through an etymological association ( is the Greek word for actor) by the description of theatrical pretense in the second. The two parts were first separated by Pithou, in his second edition of 1578. He was followed, among others, by Bücheler and Müller. As done e.g. by Slater 1990b, 160. Cf. Bücheler 1862, 95, who already believed that these verses “ex alio loco inlati sunt”. I am referring to Connors 1998, 30 (pastry Priapus); 79-80 (pebble); 47-49 (parrot). She also sees programmatic references in Encolpius’ Sotadeans (Connors 1998, 33) and, as already hinted, in the verse on the mime (Connors 1998, 13-14). Cf. also Connors 1998, 12 (poem at ch. 5). Cf. ch. XVII, with the literature quoted and discussed. The scholars who admit that this poem deals with expression, but limit it to the character Encolpius, fail to see that the latter’s wording of the story is identical with the text of the Satyrica, which is presented as reported by him. Cf., for example, Barnes 1971, 254274 (he, like others, limits the opus alluded to by Encolpius to the single episode in which the poem appears). Connors 1998, 146 (“in Petronius’ novel, the narrating voice does not seem to consider anything off limits: as Encolpius says [132.15.4], ‘whatever people do, my lucid tongue reports’”) implicitly contradicts her former statement (p. 72 n. 57: “clearly it is best to read this poem within the frame of Encolpius’ character”); it is in fact the “narrating voice” that creates the text of the novel.

Petronius’ Poems in the Satyrica freedom to treat themes like sex, excluded from so-called high literature, sincerity joined with urbanity of expression, and – this is probably the deeper meaning of the final hint at Epicurus – art aiming at pleasure alone, with no pretense of usefulness. The poem was certainly understood as a literary program by Martial, as made clear by the introductory epistle to the first book of his Epigrams.23 All the poems not attributable to Encolpius, the narrating voice, must of course be actually recited by some other character in order to be reported in the narrative (although, as we have seen, the poem recited by Tryphaena at 108.14 contains a verbum dicendi belonging in the narrator’s report, so that its nature of authorial elaboration with no claim to actual mimesis is exposed). This poem, however, is a clear parody of epic speeches, as shown by the multiple allusions in the verse itself and in the whole context, as well as by the formula signaling the end of Tryphaena’s speech in verse, which clearly hints at those employed in epic (of the type haec ubi dicta dedit) and temporarily keeps the metric pattern even in the ensuing prose through the solemn rhythm produced by the first half of a totally spondaic hexameter: haec ut turbato clamore mulier effudit (109.1). One single poem is also recited by Quartilla (18.6: a generalizing conclusion of a speech begun in prose),24 Oenothea (134.12), and the cinaedus (23.3, discussed above). The poem at 14.225 is commonly believed to be recited by Ascyltos, but this is based on an editorial transposition with no support in the tradition. In the mangled text we possess it surely makes more sense if uttered by Ascyltos, but it might equally well be a reflection by Encolpius. Ascyltos never speaks in verse anywhere else in what survives of the Satyrica. The one poem recited by the rhetor Agamemnon (5)26 is very important, in that, in connection with the preceding prose, it presents an educational program which must be assessed against the background of what we know of contemporary literary culture. The poem is made up of two parts in different meters (choliambics and hexameters), which are clearly connected with each other (another single poem in two different meters is very probably 109.9-10: elegiac couplets and hendecasyllables). The first part treats the moral requirements to become a successful orator; the second, the studies necessary to attain the goal. The two parts are clearly connected by the strongly stressed sed opening the section in hexameters, marking the transition from the part in choliambics, cast in the negative form through the mention of the vices the future orator must be free from, to the positive program of literary studies. The poem is not, very probably, the schedium Lucilianae humilitatis (4.5), which must refer to a poem previ23 24 25 26

The verbal parallels are clear and numerous: cf. ch. XVII (also with references to other texts by Martial). Cf. Barnes 1971, 277-278; Yeh 2007, 505-506. See ch. III. Cf. ch. II, with the literature quoted and discussed. Cf. ch. I, with the literature quoted and discussed. 5

Introduction ously recited, possibly by Encolpius. Agamemnon introduces it by stating that he too (et ipse: 4.5) will express his literary ideas in verse (carmine), lest Encolpius should think that he (Agamemnon) did not appreciate (improbasse: notice the past tense) the “improvisation in Lucilius’ simple style”, i.e. the schedium, which, as a consequence, cannot be the poem he has not yet recited. Agamemnon’s questionable morals, as well as his intention and capability to carry out the program he proposes, are immaterial when the program itself is analyzed and evaluated. There are ideas which are clearly paralleled in contemporary or roughly contemporary rhetoric, sometimes with striking verbal simi, Agamemnon larities.27 Like Eumolpus, and like the author of the advocates the harmonization of inspired utterance (13-14 mittat habenas / liber; cf. 118.6 [Eumolpus] praecipitandus est liber spiritus… furentis animi vaticinatio; de subl. 13.2; 15.4; 32.4) and painstaking literary training (21-22 flumine largo / plenus and 4.3 lectione severa irrigarentur; cf. 118.3 [Eumolpus] ingenti flumine litterarum inundata; 118.6 plenus litteris; de subl. 13.2-3). It is of course impossible to ascertain whether Petronius approved or disapproved of this program and whether he judged it appropriate or not appropriate to the situation and needs of Roman culture at the time; but it must be realized that what is presented here are ideas that had full right of citizenship in the contemporary cultural debate. The frequent attempts to pass off Agamemnon’s educational program as risible and absurd in itself are therefore unjustified. The assessment of Eumolpus and his poetry is even more complex. Eumolpus is an important character in what survives of the Satyrica, and a professional poet. We have just sketched some of his theoretical ideas; he makes his most important statements in close connection with his most ambitious poetic essay, the Bellum civile. Like Agamemnon’s, so Eumolpus’ morals are questionable at best; but in my opinion it would be a misunderstanding if we believed, as some scholars do, that he uses poetry as a means to other ends. Quite the opposite: he is a compulsive poet that cannot give up what he believes to be his calling, even though he has long since realized that it offers no practical benefits. No doubt he is a swindler and a rogue; he will wear the mask of the philosopher to seduce the youth of Pergamum, but poetry for him is never roleplaying; even the “mime” he stages at Croton is metaphorically described as literary; and in the shipwreck he turns to poetry not to seek a famosa mors, like Horace’s vesanus poeta28 (he has not planned to die), but to face death in his real and true capacity – as a poet, as Giton and Encolpius face it as lovers. How is this to be reconciled with the actual samples of poetry with which Eumolpus regales the reader in the course of the narrative? And how does 27 28

6

The clearest parallels are perhaps to be found in Quintilian; cf. e.g Quint. 10.5.14-16 (see ch. I). Hor. ars 455-469.

Petronius’ Poems in the Satyrica Petronius judge them? The first poem recited by Eumolpus (83.10)29 is in the form of a Priamel listing different lifestyles, like Horace’s first – and introductory – ode; but, unlike Horace’s, Eumolpus’ choice of a life devoted to poetry has to face a social situation in which all lifestyles are really reduced to one: all activities are pursued with the quest for money as the only goal; all have become a universal . Poetry is different from all other activities in that it does not afford practical rewards. Eumolpus obviously resents this; still poetry is, and will remain, his choice. Just like Horace’s first ode, the poem may be regarded as a sort of introduction to the poetry Eumolpus will produce in the novel, or rather to his whole way of being and acting. The omnipotence of money is of course one of the main themes of the Satyrica, and appears in several poems too,30 but it is also a prominent feature in Eumolpus’ most ambitious poetical essay, the Bellum civile.31 The other two short poems he recites are impromptu compositions; but the first one, 93.2, which, like 83.10, continues a speech begun in prose, and contains a tirade against the craze for the exotic, also features a theme that not only appears in the verse recited by Trimalchio at 55.6, but again presents very close parallels with the beginning of the Bellum civile.32 The second, 109.9-10,33 is a paignion on Encolpius’ and Giton’s temporary baldness. Like the poem at chapter 5, it appears to be made up of two parts, the first one in elegiacs, the second in hendecasyllables. It is probably one poem, although according to several scholars the two parts are really distinct compositions, perhaps originally separated by a prose interlude suppressed by the excerptor. As a professional poet, Eumolpus produces the two ostensibly most ambitious poetic pieces in the novel: the Troiae halosis (89) and the Bellum civile (119-124.1).34 The recitation of the first one is greeted with a throwing of stones by some accidental listeners (90.1). This might be taken as a covert judgment 29 30 31 32

33 34

Cf. ch. X, with the literature quoted and discussed. Cf. e.g. Stöcker 1969, 146-151. Petr. 119.41-55. E.g. Petr. 93.2.3-5 at albus anser / et pictis anas involuta pennis / plebeium sapit; 10 quicqid quaeritur, optimum videtur ~ 119.7-8 non vulgo nota placebant / gaudia, non usu plebeio trita voluptas. Cf. ch. XII and app. I, with the literature quoted and discussed. The problems raised by these two compositions would require a monograph in its own right and can hardly be addressed in this book, which deals with the Satyrica’s short poems. We shall briefly treat them solely in their relation, qua poems, to the prose narrative as well as to the other verse parts of the novel. A clever simile proposed by Connors 1998, 100 should be kept in mind: the Bellum civile is like a rat swallowed by a python; as a part of the Satyrica it produces an impression very different from the one we would get if it were a free-standing piece. For the Troiae halosis and the Bellum civile see lastly Yeh 2007, 116-188, 195-385, 423-470, with the literature quoted and discussed. 7

Introduction passed by the author, but the contemporary public, as portrayed in the Satyrica, hates poetry, be it good or bad (90.2-3; 93.3),35 and Eumolpus has a point when he considers that his poetic calling marginalizes him in contemporary society (84). This is humorously offset through the fleeting appearance of the one character who fervidly admires Eumolpus: Bargates (96.4-7). Also, we should not forget that the Troiae halosis is presented as an impromptu ekphrasis of a painting (89 conabor opus versibus pandere) and the Bellum civile as an unfinished and unperfected poem (115.4; 118.6 – and the last line does indeed have the appearance of a stopgap). The Troiae halosis (the title is curiously reminiscent of the Halosis Ilii reportedly sung by Nero during the fire of Rome: Suet. Nero 38.2 – though in Petronius it is referred to the painting, not to the poem) is often taken to be a parody of Senecan tragedy,36 since it is written in tragic meter (iambic trimeters), and it does somehow resemble the reports of “messengers” common in tragedy. The ekphrastic pretext is totally forgotten; the poem is developed as the report of different moments and scenes by an eyewitness, indeed by a Trojan: there are verbs in the first person plural (11; 35), and even the apostrophe o patria (11). But of course the description, and indeed the reader’s expectations, could hardly prescind from Virgil’s narrative in Aeneid 237 – though with the meaningful shift of attitude we shall presently try to sketch. The other extensive poetic essay, the Bellum civile, is introduced by Eumolpus with clear references to Lucan: when he says (118.6) that historical epic cannot dispense with the traditional mythological machinery and apparatus, he surely has the Pharsalia in mind, a poem that had done away with the mythological gods. Indeed, Eumolpus’ poem covers the same subject as Lucanus’ first book, and there are clear reminiscences of the Pharsalia, including parts that had not been published by Lucan in his lifetime (e.g. 294 ~ Lucan. 7.473; 8.3335

36

37

8

Encolpius himself seems at times to share this attitude (90.3; cf. 93.3), although he is consistently portrayed as compulsively associating his own adventures with those of literary heroes. The impression produced upon him by the Bellum civile is expressed with a somewhat ambiguous formulation (124.2 cum haec Eumolpus ingenti volubilitate verborum effudisset. Such volubilitas was disparagingly associated with the shallow verbosity often attributed to the contemporary Greeks: Plin. ep. 5.20.4; Val. Max. 2.2.2). At 110.1 (plura volebat proferre, credo, et ineptiora praeteritis) Encolpius is simply reacting to Eumolpus’ paignion making fun of his and Giton’s temporary baldness. This is denied, e.g., by Barnes 1971, 92-94 and Yeh 2007, 486-487, though the latter stresses the poem’s metric affinity with Seneca’s tragic trimeter (Yeh 2007, 422-434), which of course does not rule out some differences (Yeh 2007, 434-442, 450 ff.). Stubbe 1933, 31-37, though conceding that Petronius (or rather Eumolpus) tries to emulate Virgil, strangely denies that he directly referred to the Aeneis (p. 31: “es läßt sich jedenfalls deutlich zeigen, daß Petron den zu behandelnden Stoff sich nicht an einem vorliegenden Aeneistext vergegenwärtigte”). Whatever details may come from different sources acquire special significance and relief only in relation to the Virgilian version.

Petronius’ Poems in the Satyrica 34). Given Eumolpus’ attitude (at 118.5 he calls Virgil Rome’s national poet: Romanus Vergilius), the reader expects him to stick to Virgilian expressive modes, and in a way he is not disappointed. There are of course numerous Virgilian reminiscences,38 although Eumolpus’ hexameter may be less close to Virgil’s than has been routinely assumed;39 and at 123.229-237 Eumolpus recasts a simile borrowed from Lucan (1.498-509) into a typically Virgilian mold (parabole introduced by ac velut[i]… cum, with ac and velut reinforcing each other: cf. Verg. Aen. 2. 626 ff.; 4.402 ff.; 6.707 ff.), which is also found in Statius [Theb. 5.599 ff.; 7.436 ff.] and Silius [4.302 ff.], but never in Lucan). This formal detail is already a token of the attitudes of the supposedly traditional, Virgilian branch of Roman epic. The divine figures of the Bellum civile, such as Fortuna or Discordia, are personified allegories more reminiscent of Statius than of Virgil; and in both the Bellum civile and the Troiae halosis we witness an almost morbidly sinister transfiguration of Virgilian elements to fit a dark and hopeless overall picture: e.g. the figure of Laocoon in the Troiae halosis, where no redeeming hope for the future is in view, and some pointed reversals of Virgilian descriptions in the Bellum civile (e.g. 119.53 ~ Verg. Aen. 6.637; 124.253 ac maerens lacera Concordia palla ~ Aen. 8.702 et scissa gaudens vadit Discordia palla; 124.272 extulit ad superos Stygium caput ~ Aen. 1.127 summa placidum caput extulit unda). Ever since Seneca the Elder (contr. 1 praef. 6-7) Roman writers had the feeling of living in an age of epigones. Some, like Lucan, tried to open new paths; some, like Eumolpus and his rightful heirs Statius and Silius, refused to follow them, and though they did attempt some interesting experiments (such as the fusion of epic and tragedy or the “expressionist” exasperation of the received patterns), their poetry was bound to be caught in a dead-end road. In conclusion, though we will never know what Petronius thought of Lucan’s epic and the alternative approach represented by Eumolpus, we should be aware of the fact that, like in the case of Agamemnon’s ideas on rhetoric, he presents us with a position that had full right of citizenship in the contemporary debate, and was actually to gain the upper hand in the following generation. Eumolpus surely represents a recognized cultural trend; the correspondence of the judgment he passes on the historical epic with no mythological machinery (118.6 quod longe melius historici faciunt) with the evaluation of Lucan as a historian rather than a poet that obtained down to the end of antiquity and beyond 38 39

Cf. e.g. Barnes 1971, 143. Stubbe 1933, 103, Barnes 1971, 128, Connors 1998, 116 n. 47 all have stressed the closeness of Eumolpus’ hexameter to the Virgilian pattern. On the basis of a thorough metrical analysis, however, Yeh 2007, 259 concludes that the meter of the Bellum civile differs both from Virgil’s and from Lucan’s. It rather shows archaizing leanings and is close to Silius Italicus’ hexameter (Yeh 2007, 385, 575; etc.). 9

Introduction (Serv. Ad Aen. 1.382 Lucanus… in numero poetarum esse non meruit, quia videtur historiam composuisse, non poema, almost literally repeated by Isid. orig. 8.7.10; Comm. Bern. in Lucan. 1.1 Lucanus dicitur a plerisque non esse in numero poetarum, quia omnino historiam sequitur; Iordanes, Get. 5.43 Lucano plus historico quam poeta testante; and cf. already Mart. 14.194) cannot surely be coincidental. Another character in the novel is portrayed as customarily composing poetry: the rich but ignorant Trimalchio. He epitomizes a social phenomenon exposed by Horace (epist. 2.1.117 scribimus indocti doctique poemata passim). For the former slave this is a way to assert his social success. Interestingly enough, not all of Trimalchio’s poetry that is recited in the novel is actually reported (41.6; neither is Trimalchio’s recital of mime at 35.6),40 but only a few specimens recited by himself, obviously sufficient to show haw bad a poet he is. The two “epigrams” at 34.10 and 55.341 contain the whole, very commonplace “philosophy” of the author: enjoyment – especially food and drink – is the only remedy against death and misfortune. The first one has been obviously prepared beforehand and is introduced with a careful stage direction: 100-year-old wine, a prose speech opening with the same word as the poem (eheu) and containing its most conspicuous one (homuncio), the silver skeleton. The meter (two hexameters followed by one pentameter) is typical of uncultured literary production (it is commonly found in funerary inscriptions), and, with Trimalchio’s failure to keep correct grammar (sic erimus cuncti postquam nos auferet Orcus, instead of abstulerit), it neatly characterizes his pretentious ignorance. The second epigram (55.3) is composed at the moment, after the fall of an acrobat, which could hardly have been planned in advance.42 Trimalchio asks for writing material, but cannot concentrate for too long and comes up with three metrically very defective lines, a vain attempt to reproduce the meter of the previous poem. The third poem is recited by Trimalchio immediately after (55.6) and attributed to him to the mimic poet Publilius Syrus. These lines, however, are probably by Petronius, and therefore, in the novel’s fiction, by Trimalchio.43 The poem, at any rate, aims to pursue his characterization. Trimalchio’s extravagant pairing of Cicero and Publilius, which introduces the verse, is surely meant as a sign of his ignorance. That these lines mean to parody Seneca’s use of quotations from Publilius

40 41 42

43

10

Similarly, only the first line of the recitation by Habinnas’ slave is reported (68.6); but it is immediately followed by Encolpius’ expression of disgust. Cf. ch. V, with the literature quoted and discussed. Yeh 2007, 96 unplausibly thinks that the acrobat’s fall has been planned beforehand, and even less plausibly that 34.10 and 55.3 really constitute one single poem (Yeh 2007, 97-98, 521-522). Cf. ch. VI, with the literature quoted and discussed.

Petronius’ Poems in the Satyrica in his moral writing44 is surely possible, but it is hard to believe that the author has totally superimposed himself upon his character. The poems that can be attributed to Encolpius are both the most interesting and the most difficult to assess. He is the narrator, and does not need to utter his poems as an acting character in order to report them. Actually, aside from the prayer to Priapus at 133.3, in the surviving parts of the novel no poem appears to be recited by the character Encolpius in the situation reported, including 126.1845 (the words at 127.1 delectata risit tam blandum probably appear after a lacuna and are hardly the description of Circe’s reaction to the recital of Encolpius’ poem); and we are not sure whether the schedium Lucilianae humilitatis (4.5) was recited by him to Agamemnon, even though, as we saw, this appears to be possible. Poems by Encolpius appear at 79.8, 80.9, 126.18, 127.9, 128.6, 131.8, 132.8, 132.11 (Virgilian cento), 132.15, 133.3, 135.8, 136.6, 137.9, 139.2. He is also probably the author of 82.5,46 15.9 (two isolated lines, the only surely incomplete poem), and possibly also of 14.2, which, as we have seen, most editors attribute to Ascyltos. Two of these poems, 133.3 and 139.2,47 are crucial for the interpretation of the novel as a whole. In both the central figure is the god Priapus. The first one is a prayer addressed to him by Encolpius after being affected by sexual impotence, modeled after well-known patterns of such invocations, and containing the confession of a “minor fault” (culpaeque ignosce minori: v. 11) as well as a declaration of innocence as far as such grievous faults as murder and sacrilege are concerned. The second is a list in the form of a Priamel of heroes who were persecuted by some god, culminating with Encolpius himself, hounded, as he says, by the wrath of Priapus. Over 120 years ago Elimar Klebs48 based himself on 139.2 to support his theory that the wrath of Priapus was the leitmotif granting unity not merely to the parts that came down to us, but to the novel as a whole, ever since its beginning (probably in the city of Massilia, modern Marseilles), and fulfilling the same function as the wrath of Poseidon in the Odyssey. Klebs’ theory enjoyed great success, and is accepted even today by several scholars, though the idea of Priapus’ wrath persecuting Encolpius from the very beginning hardly seems to be well-founded. It is based on the notion that Encolpius’ impotence should be regarded as a punishment, whereas, in his prayer to Priapus, Encolpius himself sees it as a fault, even though a minor one, for which he must ask for the god’s forgiveness, as confirmed by his statement that this fault was committed with only a part of his body, which hardly needs to be specified (133.3.9). It is also 44 45 46 47 48

Cf. e.g. Courtney 1991, 21. Cf. ch. XIII, with the literature quoted and discussed. Cf. ch. IX, with the literature quoted and discussed. Cf. ch. XVIII, with the literature quoted and discussed. Klebs 1889. 11

Introduction clear, despite the efforts of some scholars, that Encolpius was not impotent before his adventure with Circe (cf. 11.1, 24.7, 79.8, 129.1),49 and that he, formerly an Achilles in sex-play, as he says (129.1), is not only saddened, but also surprised at his failure with her. His declaration of innocence as far as murder and sacrilege are concerned, however, does not necessarily imply that he never committed these crimes. The confession of minor faults, while disclaiming more grievous ones, obeys a wellattested pattern of ancient prayers. Clearly, Encolpius is referring to the present situation, in which the fault to be forgiven is only his impotence with Circe, not any crime he may have committed in the past. In preceding parts of the novel we come across at least three passages (9.8 [Ascyltos], 81.3, 130.1-2) which lead us to believe that Encolpius is really guilty of murder and sacrilege. The efforts of some scholars to have us regard these three passages as unreliable, and thus clear Encolpius of these crimes, are unconvincing; it is difficult to understand why Encolpius (as well as Ascyltos) should always lie, and tell the truth only once – furthermore in his very prayer to Priapus, in which such a declaration of innocence was required by the accepted patterns of religious speech. Besides, Encolpius does not come immediately, but rather slowly and gradually, to the conclusion of being persecuted by a god, like the mythological heroes of 139.2. At first he thinks of a spell (128.8; and still at 138.7), and even of psychological and pathological reasons (130.5, 7-8); but, whatever the cause, it is clear that impotence is an offense against the god of sex, Priapus, for which forgiveness must be asked. Finally, since nothing avails, be it magic, medicine, or prayer, Encolpius comes to the conclusion that he must be persecuted by a god, and that his impotence is not merely a fault, but also and primarily the effect of Priapus’ wrath (139.2.7-8); and he sticks to this idea in the few more pages that survive (140.11, 12). In this whole episode Encolpius’ mind comes to be more and more dominated by Priapus – which means that the god plays an important role in this part of the story, but hardly warrants the attribution to him of such a central function from the very beginning of the novel. Not rarely Encolpius’ poems are presented as his own reflections, which “universalize” the specific situation being narrated; but neither do they all belong to this type nor do they invariably define the moods of the acting character as opposed to the allegedly more detached attitude of the narrator at the moment he reports his past adventures. Some are mere descriptions (131.8, 135.8);50 in one case (132.8; cf. 132.11)51 the poem continues the narrative. It is clearly a fallacy to believe that the poems invariably represent the literary transfiguration through which reality is perceived by the naïve, or “mythomaniac” character Encolpius – and that it is invariably exposed as delusion by the older and more 49 50 51

12

At 20.2 and 23.5 Encolpius’ failures are due to exhaustion and disgust respectively. Cf. chapters XVI and XX respectively. Cf. ch. IV.

Petronius’ Poems in the Satyrica mature narrator in the prose. Sometimes the verse clearly reveals that the literary transfiguration continues in the narrator’s present: the poem at 136.652, for example, proves that the narrating voice still believes that the geese with which he has fought in the narrated adventure were like mythological monsters, as shown by the present tense (136.6.1-2 tales Herculea Stymphalidas arte coactas / ad caelum fugisse reor). In reality the moments in time defined by the poems are interchangeable: 126.18 and 139.2 appear to be the acting character’s reflections, 80.9 and 137.9 the narrator’s. At times the narrator vividly relives in verse the emotions he experienced as a character: 79.853 (qualis nox fuit illa, di deaeque) clearly reflects the narrator’s standpoint, as shown by the perfect tenses; but in the prose he has his sobering present violently break in upon his cherished past: sine causa gratulor mihi (notice the present tense, as opposed to the past tenses in the poem). In general, when verse and prose define two different moments, it is always the narrator – and of course Petronius behind him – who contrives this effect for artistic purposes; the two moments can never be separated, if we are not to miss the author’s subtle play. In the cycle of poems connected with Encolpius’ adventure with Circe (126.18, 127.9, 131.8, with 128.6 in between)54 the first two poems hark back not merely to the Odyssey, but to an even greater extent to Iliad 14 (Zeus and Hera making love on mount Ida).55 The scene is transposed from nature to a garden (cf. also 131.8). Save perhaps 131.8, the poems taken in and by themselves appear totally serious; but Encolpius’ impotence, narrated in the prose, forms the other panel of the diptych, which can only be understood and enjoyed as a whole. In the Oenothea episode, while the prose consistently depicts a sordid reality, the four poems (including the first one, 134.12, recited by the priestess) record the changing moods of Encolpius and hark back to different literary models: Theocritus (134.12),56 Callimachus and Ovid (135.8, in which Callimachus’ Hecale is expressly quoted),57 Hellenistic epic (136.6).58 The fourth poem, 137.9,59 in elegiac couplets, whereas the previous ones are in hexameters, is a reflection on the omnipotence of money, prompted by the conclusion of the episode. The last two poems reflect the standpoint of the narrator; both contain a verb in the first person of the present tense (136.6.2 reor; 137.9.9 multa loquor), 52 53 54 55 56 57 58 59

Cf. ch. XXI, with the literature quoted and discussed. Cf. ch. VII, with the literature quoted and discussed. Cf. chapters XIII, XIV, XVI, and XV respectively. But with a pointed reversal of the sexual roles, made explicit in the prose context: cf. ch. XIV. Cf. ch. XIX, with the literature quoted and discussed. Cf. ch. XX, with the literature quoted and discussed. Cf. ch. XXI, with the literature quoted and discussed. Cf. ch. XXII, with the literature quoted and discussed. 13

Introduction but they define two different attitudes: the continued literary transfiguration on the one hand, the disenchanted assessment of reality on the other: obviously, not even the sadder but wiser Encolpius has entirely lost his irremediable literary “idealism”.

14

Chapter I The Education of the Orator (Petr. 5)* 4.5 Sed ne me putes improbasse schedium Lucilianae humilitatis, quod sentio et ipse carmine effingam: 5. artis severae si quis ambit effectus mentemque magnis applicat, prius mores frugalitatis lege poliat exacta. Nec curet alto regiam trucem vultu cliensque cenas impotentium captet, 5 nec perditis addictus obruat vino mentis calorem, neve plausor in scaenam sedeat redemptus histrioniae addictus. Sed sive armigerae rident Tritonidis arces seu Lacedaemonio tellus habitata colono 10 Sirenumque domus, det primos versibus annos Maeoniumque bibat felici pectore fontem. Mox et Socratico plenus grege mittat habenas liber et ingentis quatiat Demosthenis arma. Hinc Romana manus circumfluat et modo Graio 15 exonerata sono mutet suffusa saporem. Interdum subducta foro det pagina cursum et fortuna sonet celeri distincta meatu; dent epulas et bella truci memorata canore grandiaque indomiti Ciceronis verba minentur. 20 His animum succinge bonis: sic flumine largo plenus Pierio defundes pectore verba. 4.5 L(=ldmrtp)O(=BRP) 5 LO 4.5 schedium lmgtmgp: schadium B: studium Bmgcett. Lucil(l)ian(a)e lmgtpB: Lucianae lRP Samb. Puteol.: Lucinianae dmr humilitatis dmrtBmg: improbitatis pO: vel improbitatis humilitatis l 5.1 ambit tmg: amat LO : hamat lmgE(Messanensis deperditus) 2 mores BR: more cett.

Chapter I 3 poliat Heinsius: polleat dmrtpO : palleat ltvl 4 nec LP : ne BR 5 cliensve p2 Scaliger 7 scaenam Heinsius (sce-): scena 8 histrioniae lmgtp Samb.: histrionei drmgB: histrioni ltvlRP : histriones m Memm.: histrionis r Turneb. Scaliger: histrion G(Guelferbytanus extravag. 299) ad rictus Ribbeck: (histrionis) ad dicta Bücheler1 (ubi ad dicta perperam impressum est) 11 Sirenumve Bücheler 15 hinc m Samb. Puteol.: huic O: hunc lmgdrtp : huc l 16 exornata R: alii aliter emendaverunt 22 defundes lmgE Scaliger: diffundes LBR : diffundens P

1. It has become customary to identify the poem recited by the rhetorician Agamemnon, which takes up the whole chaper 5 of the Satyrica, with the schedium Lucilianae humilitatis1 mentioned by him immediately before. This being so, it is necessary, before analyzing these lines, to check whether it is indeed justified to do so. There are, it is true, a few scholars who observe that this identification is hardly warranted by the Petronian text. Collignon2 was the first to realize that the words et ipse carmine effingam appear to show that, before Agamemnon’s speech, a schedium in the Lucilian manner had been recited by somebody else, probably Encolpius, in a lost part of their dialogue, which opens the part of the novel that has come down to us. This idea was later picked up by Ciaffi,3 who also calls attention to the perfect infinitive improbasse, which appears to refer to the past. Barnes4 believes, in fact, that with the words schedium Lucilianae humilitatis Agamemnon is referring to a lost poem recited by Encolpius, and stresses the fact that the rhetorician refers to his own poem with a different term (carmine effingam). Cosci5 develops Ciaffi, in that she contends that improbasse can hardly be taken as an aoristic perfect defining no temporal relation * 1

2 3 4 5

16

A version of this chapter has appeared with the title La poesia in Petr. Sat. 5, “Prometheus” 28, 2002, 253-277; 29, 2003, 65-78. This is the text which can be reconstructed with all but absolute certainty: see the critical apparatus. The term schedium obviously suggested Lucilian associations: cf. Lucil. 1279 M. = 1296 Krenkel = H 14 Charpin; Apul. de deo Socr. prol. 104, p. 164 Beaujeu (in reality to be referred to the Florida) etiam in isto, ut ait Lucilius, schedio; Festus p. 450, 16-21 L. (cf. Paul. exc. p. 451, 9-11 L.). See, among others, Collignon 1892, 229; Fiske 1909, 122-123; Ingersoll 1912; Mariotti 1960, 17, 78-79; Flores 1982, 77-88; Aragosti 1995, 140 n. 8; Scarsi 1996, 6 n. 5. Nothing new in Cavalca 2001, 149-150. Collignon 1892, 228. Ciaffi 1955, 24-25 and n. 20. Barnes 1971, 27-28. Cosci 1978, 202-203. Before Cosci, the idea that the schedium had been recited by Encolpius in a lost part can also be found in Coffey 1976, 192. It has also been adopted by Courtney 2001, 59.

The Education of the Orator (Petr. 5) with the situation described.6 Aragosti too7 realizes that the words et ipse seem to refer to a poem previously recited by Encolpius and prompting Agamemnon to utter a poem too; but the conditioned reflex impelling to identify the schedium Lucilianae humilitatis with the preserved verse is so rooted and powerful that Aragosti takes it for granted that these words are nothing but Agamemnon’s way to characterize his own poem, though in an antiphrastic fashion.8 Long before, even Collignon, who had been the first to envisage the concrete possibility that the schedium might be not Agamemnon’s poem, but a lost one uttered by Encolpius, later devoted several pages to searching (in vain, by his own admission) for Lucilian elements in the verse at chapter 5, thus contradictorily identified, après coup, with the schedium Lucilianae humilitatis.9 There are, at any rate, three unequivocal textual data that call in doubt this identification, even though it is almost universally accepted: the fact that Agamemnon refers to his own poem with the term carmen, the perfect infinitive improbasse, and the words et ipse. It is undoubtedly possible to try to limit the import of each, but the heaping of the three in the short sentence preceding Agamemnon’s poem should at least give us pause. Carmine might indeed simply mean “in verse”,10 and therefore not refer to a poetic composition distinct from the schedium. We should however bear in mind that an unfortunately rather mangled passage of Festus and the corresponding abridgment of Paulus – both of which have been quoted in the footnotes11 – make it clear that this term applied not merely or even mainly to improvisation, but, in general, to poemata mala.12 It is true that judgments concerning the poetic worth of Agamemnon’s verse differ widely: some scholars find a clear trace of improvisation in the numerous repetitions found in it, whereas according to 6 7 8

9

10 11 12

As is the case with the perfect infinitive at 2.2: pace vestra liceat dixisse, primi omnium eloquentia perdidistis. Aragosti 1995, 132-133 n. 1. Aragosti 1995, 140-141 n. 8. On the other hand, even Cosci 1978, 202 and n. 5 expresses not too different an idea, thus contradictorily supposing the identification of the schedium with the verse at chapter 5. Flores 1982, 74 is more consistent, in that, though stressing that Agamemnon’s poem is anything but an improvisation (schedium), he rejects the idea that a previous poem recited by Encolpius may be alluded to in the context. See also Soverini 1985, 1731 n. 118. Collignon 1892, 232-235. The verse at chapter 5 is identified with the schedium also on pp. 229 and 230. On p. 234 Collignon admits that the comparison of this poem with Lucilius’ fragments does not yield any result. The failure of Collignon’s efforts along this line is stressed by Brugnoli 1963, 258; Soverini 1985, 1733 n. 124. Cf. the numerous examples collected in TLL III 470, 63-471, 11. Cf. above, note 1. This was perhaps the meaning the term had in Lucilius, if we may trust Lindsay’s integration in Festus’ text: us quoque poemata tis perfecti qui essent. Paulus has unde mala poemata schedia appellantur. For schedium as “poor poem” cf. also Scarsi 1996, 6 n. 5. 17

Chapter I others it is poetically refined and far removed from any humilitas.13 In any case, it seems to me, Agamemnon would hardly have referred to his own verse with a term carrying disparaging nuances (schedium), when he states immediately after that his carmen means to express his thought – and, what’s more, doing so by resorting to a term (effingere) which is often charged with nuances connected with a lofty and even sublime type of expression.14 As far as the perfect infinitive improbasse is concerned, obviously those who uphold the identity of Agamemnon’s verse with the schedium have no other resort but taking it as aoristic,15 though there are some who are not at ease with this interpretation.16 Concerning et ipse, finally, the prevailing interpretation among those who share the common opinion is that these words imply a reference to Lucilius himself; Agamemnon would say: “I will also express my thinking in verse, just like Lucilius did”.17 Let us now try to analyze Petronius’ text with no preconceptions. If we take the schedium Lucilianae humilitatis (which is the object of improbasse) to be one and the same with the poem appearing at chapter 5, that infinitive, even if understood as aoristic, hardly seems to be an appropriate expression, if Agamemnon is assumed to say, as this interpretation does, not that he does not or did not disapprove a schedium Lucilianae humilitatis, but rather that he shall

13

14

15 16

17

18

Sullivan 1968, 191 believes that the numerous repetitions (on which we will dwell later on) are meant to create the impression of an improvised composition. On the other hand, as we have seen (cf. note 8), Cosci, Flores, Soverini, and Aragosti believe Agamemnon’s lines to be anything but a schedium. Slater 1990a, 30 is isolated in thinking that the choliambics aim to create the effect of improvisation, whereas the hexameters strive to appear accurately thought out and polished. Cf. e.g. Cic. nat. deor. 1.47 soletis… cum artificium effingitis fabricamque divinam… describere; Macr. in somn. Scip. 1.17.5 totius mundi… collecta descriptio est et integrum quoddam universitatis corpus effingitur. Besides, effingo is often used in connection with literary imitation of lofty models: e.g. Gell. 8.8, 17.20.8 (Plato); 9.9.3 (Homer and others); Quint. 10.1.108 (Demosthenes, Plato, Isocrates); Plin. ep. 1.10.5 (Plato); 9.22.2 (Horace). As dixisse at 2.2 (see above, note 6). Cf. e.g. Pellegrino 1986, 143-144. Cf. Van Thiel 1971, 26 n. 2: “dem Infinitiv improbasse wird von allen präsentische Bedeutung beigelegt, was möglich, aber nicht notwendig ist”. Van Thiel believes the verb to refer to a previous pronouncement of Agamemnon’s about Lucilius’ poetry (and et ipse to have Lucilius as referent: “I too, in the manner of Lucilius”; but see below for the difficulties posed by this interpretation). Walsh 1996, 3 translates: “but I would not have you think that I have been carping at the impromptu commonplace utterances of a Lucilius, so like him I will express my feelings in verse”. Cf. e.g. Flores 1982, 72; Pellegrino 1986, 144, and Walsh’s translation quoted in the preceding note. Paratore 1933, II, 19 n. 1 takes et ipse to mean “io come tanti maestrucoli”.

The Education of the Orator (Petr. 5) shortly present such a schedium himself, as in fact will happen.18 Clearly, the reference must be to a poem which is different from the one he too (et ipse) will recite. Besides, the referent cannot be Lucilius himself, as many scholars have suggested, but a poem in the Lucilian style. It is true that, if we take Agamemnon to say “I will also express my thinking in verse, just like Lucilius did”, it necessarily follows that the schedium of which he does not disapprove cannot be a simple allusion to a general literary typology but must be specifically referred to Lucilius’ poetic production;19 but it is not difficult to realize that such a reference, and therefore the interpretation connected with it, is not supported by the text.20 In fact, the adjective derived from the name of an author applied to the literary qualities of a writing in a construction like the one appearing in our text generally refers to a work written in the manner of that author, not to an authentic work of his. An oratio Ciceronianae eloquentiae is not by Cicero, just like a “sonnet of Petrarchan musicality” is not by Petrarch. It follows that the schedium Lucilianae humilitatis cannot be by Lucilius. Besides, if we surmise that Agamemnon means to say that the schedium is his own verse, the verb he uses would once more be hardly appropriate. Aside from the obstacle of the perfect tense of the infinitive, it is not easy to imagine that anybody would announce a speech he is about to present by stating that he does not disapprove of the form in which he will cast what he is going to say. Therefore, if the schedium is neither by Lucilius nor by Agamemnon, the most probable alternative is that the rhetorician is referring to a poem previously recited, probably by Encolpius, in a lost part, to which his own poem is meant to correspond and reply.21 I do not presume that the preceding remarks have unquestionably attained the truth; but I do believe them to point out the riskiness, or even the downright vanity, of searching for Lucilian elements in Agamemnon’s verse. We have al18

19 20 21

The expression carmine effingam does of course contain a future, but it can be understood in the way stated above, in the text, only if this carmen is taken beforehand to be identical with the schedium. Cf. the interpretations of Van Thiel and Walsh reported above, note 16; the same is true for those of Flores and Pellegrino: note 17. The reference is to a schedium Lucianae humilitatis, not to a humile Lucilianum schedium, nor to humilitas Lucilii. The plausibility of this approach is confirmed by a comparison with the ingenious, but needlessly complicated argument put forward by Flores 1982, 74 in order to deny the necessity to suppose a reference to a lost part. He understands Agamemnon’s words sed ne me putes improbasse schedium Lucilianae humilitatis as an ironical praise for plain improvisation in the manner of Lucilius, as opposed to the serious study he has previously advocated; on the other hand, however, the poem at chapter 5 will be anything but an improvisation. According to Flores, this alternation of assertion, negation, and final reassertion of artistic engagement would explain Agamemnon’s expression with no need to suppose a reference to a poem previously recited by Encolpius. 19

Chapter I ready mentioned Collignon’s failure in this attempt,22 even though his posture has been picked up by Sullivan,23 with no new arguments. Here we may add that it is equally vain – or rather apt to produce an effect opposite to the one intended by those who identify the schedium with Agamemnon’s poem – to pursue the contrary path, i.e. searching these lines in quest of elements contradicting the definition of schedium and the Luciliana humilitas.24 If we believe that such elements do exist, it would be more natural to admit that the poem at chapter 5 is not the schedium than to look for more or less far-fetched explanations crediting the author with this or that recondite intention on the only basis of the preconception that the definition of the poem as a schedium must be at odds with its actual literary quality. At any rate, Schönberger’s accurate analysis25 clearly shows that the model of Agamemnon’s poem is not Lucilius – neither in a positive nor in a negative, antiphrastic sense, – but rather Virgil, Horace, and Cicero. It is even more ineffectual and dangerous, then, to seek an explanation to the metrical variety of our poem (eight choliambics followed by fourteen hexameters) in Lucilian precedents for which all documentation is lacking, as done by some scholars.26 Such a conclusion is prompted by the natural association27 of Petronius’ chapter 5 with Persius’ Choliambics, which are very probably a proem written in a meter different from the hexameters of the following first satire, which also presents some of the traits of a proem.28 This pairing led Reitzenstein29 to believe that Persius was imitating Lucilius in juxtaposing two poetic

22 23 24

25 26

27 28

29

20

See above, note 9. Sullivan 1968, 191. See above, note 8, for the interpretations of this type proposed by Cosci (selfcontradictory), Soverini, Aragosti, and Flores (for the latter see also note 21). Those who find an element of stylistic humilitas in the frequent repetitions of the poem at ch. 5 (identified with the schedium) are at least more consistent: Collignon 1892, 230 (see note 9 for Collignon’s ambiguous position); Sullivan 1968, 191. Schönberger 1939, 509-510. Cf. also Barnes 1971, 19-21 (adding Propertius and Ovid to Virgil, Horace, and Cicero). Collignon 1892, 229 already believed that the metrical variety proved Lucilian derivation; in this too he was followed by Sullivan 1968, 191. Cf. also Fiske 1909, 122 and the justified reservations on the latter’s (and on Sullivan’s) position in Barnes 1971, 24-26, 38 n. 29. See below, note 43. The connection has been made by numerous scholars. Among those closer to our time I will only mention Flores 1982, 75-76; Soverini 1985, 1733 n. 123 (with some reservations); Connors 1998, 12; Courtney 2001, 59; Yeh 2007, 514-515. Reitzenstein 1924, 4-5. Reitzenstein’s approach, besides being based upon the unproven Lucilian precedent, is clearly biased: he believes Agamemnon’s verse to be made up of two different poems and posits a strong affinity between Persius’ and Petronius’ subject matter, when in fact only some general similarities can be found (both criticize flattery of the powerful).

The Education of the Orator (Petr. 5) compositions in different meters. Based on this, Flores30 strives first to find in Persius’ Choliambics and first satire parallels to our Petronian poem, then to identify the referent of Agamemnon’s et ipse with some unattested Lucilian composition metrically analogous to the poem at chapter 5. It goes without saying that this is a totally hypothetical construction built on the flimsy foundation of the presence in Persius of a composition in choliambics,31 that I too take to be a proem, though according to others it is an envoy. Persius’ Choliambics, it is true, have the proemial function in common with his first satire, but the two remain nevertheless distinct and separate compositions,32 and are therefore not a real parallel to Petronius 5, where, as we shall see, the two parts, though differing in meter and content, are fused and integrated to form an organic unity. Much less, then, will it be justified to seek a parallel to Agamemnon’s verse in Lucilius, on the basis of his hypothetical influence on Persius. This preliminary paragraph has attained its goal if, aside from the acceptance or refusal of the interpretation of the words preceding the poem proposed here as the most likely, it will at least allow an unbiased analysis of the latter, deliver the interpreter from the obligation to look for either positive or antiphrastic references to Lucilius in the verse, and permit to look at the text with no preconceptions. 2. As already hinted, the two parts of our poem, though different in meter, function, and content, undoubtedly form an organic whole. Suffice it to remark the strongly adversative sed at the beginning of the hexameters, which can only be understood in opposition to what has been said in the choliambics.33 This has led some scholars to see the first eight line as the negative part of Agamemnon’s program, and the hexameters that follow as the positive didactic paraenesis.34 From the strictly formal point of view there is no doubt that the sed opening the hexameters marks a strong contrast to the negative precepts that

30 31 32 33

34

Flores 1982, 75-77. Justified reservations concerning alleged ideological, thematic, and verbal references to Persius are expressed by Soverini 1985, 1733 nn. 124-125. As rightly remarked by Barnes 1971, 25. Hardly anybody has doubted the unity of the composition, in spite of the change in the meter. Cf. e.g. Aragosti 1995, 141 n. 9: “i due gruppi di versi, da considerare senza soluzione di continuità” etc. Barnes 1971, 13 asks himself “whether we are dealing with a single poem or with two (or fragments of two)” but naturally concludes that we are faced with one poem; he actually goes as far as as regarding the mixed meter as “the only original touch in the whole poem” (p. 26). The manuscript A has versus XXII as a subscriptio to the poem (cf. Bücheler 1862, in the apparatus) So, for example, Stubbe 1933, 157; Schönberger 1935, 1242; Barnes 1971, 14; 22; Yeh 2007, 410; 513. 21

Chapter I precede.35 We cannot agree with Barnes, however, when he states36 that Fuchs’ remark37 on the positive import of the first precept given in the choliambics (prius mores frugalitatis lege poliat exacta) is nothing but a sophism. In fact, irrespective of the positive or negative form in which they are expressed, the precepts given in the first eight lines all advocate the observance of a high ethical standard, seen as indispensable to reach success in literature. From the conceptual point of view, therefore, the scholars who recognize in the choliambics the formulation of the necessary moral requisites, and in the hexameters the strictly didactic part of Agamemnon’s program, are surely in the right.38 Some interpreters rightly assume that this twofold feature of Agamemnon’s educational program supposes the traditional conception of the orator as vir bonus dicendi peritus.39 In fact, the rhetorical teaching of the I century A.D. not rarely supplemented the strictly technical and literary instruction with a strong recommendation to live according to high moral standards.40 This first instance of agreement between Agamemnon’s program and the leading representatives of contemporary, or quasi-contemporary, rhetorical teaching must be borne in mind when the import and validity of the former will be assessed. Cato’s traditional conception of the orator may be supplemented, in our lines, through an idea that was quite widespread at the time: the idea that the decline of eloquence was intimately connected with moral decay.41 The choliambics’ recommendations aim precisely to obviate the deterioration of contemporary ethical standards. Owing to this attitude the traditional theme of the incompatibility of profligacy and eloquence42 is enhanced through hints at contempo35 36 37 38

39 40

41

42

22

Walsh 1996, 3 is fully justified when, in his translation, he substitutes a pregnant question for Petronius’ sed (“what must he do?”). Barnes 1971, 22. Fuchs 1938, 173 n. 2; Stubbe’s and Schönberger’s interpretation (cf. above, note 34) was also rejected by Kissel 1978, 321 (among others). As already noted by Collignon 1892, 229. Cf. Walsh 1970, 86: “the change of the metres signifying the progression from moral to literary precepts”. See also Soverini 1985, 1713; Aragosti 1995, 141 n. 9. Cf. already Sage 1915, 48. Later especially Pellegrino 1986, 105-106; 113-114; 147. See, for example, Quintilian’s whole ch. 12.1, which opens precisely by reminding Cato’s maxim on the orator as a vir bonus dicendi peritus; besides, Quint. 1.11.2; 12.2.1; 12.11.18; Tac. dial. 28-29. These passages list several ethical behaviors to be avoided, inasmuch as they are harmful for the orator-to-be, sometimes with formulations that are very close to Petronius’ choliambics. Cf. Schönberger 1938, 174. Ogrin 1983, 49 proposes the interesting observation that Agamemnon’s poem presents the pairing of literature and life that can often be found in contemporary culture (cf. Sen. ep. 114.1 talis hominibus fuit oratio qualis vita, and see Setaioli 2000, 111-217). It is however difficult to follow Ogrin when she assumes a double level of meaning in the poem. A theme developed e.g. by Cic. Cael. 45-46, a passage that Schönberger 1929, 1199 believes to be the source of Petronius’ lines.

The Education of the Orator (Petr. 5) rary social phenonmena (such as the claque in theaters, very probably the mention of the regia, etc.) and betrays an attitude not far removed from that of contemporary diatribe’s moralizing.43 This also explains the recourse to the choliambic meter, which, after Hipponax and Callimachus, had been employed not only by Herodas, but also by poets who wrote compositions of the very type we find here, such as Cercidas and Phoenix of Colophon.44 It is not surprising, on the other hand, that the literary precepts should be given in didactic hexameters not unlike those of Horace’s Ars poetica. These remarks, in my opinion, may explain our poem’s metric variety much better than the intent to expose Agamemnon’s feigned literary and rhetoric ability that some scholars attribute to Petronius.45 But this reference leads us to tackle the main problem posed by our poem: the meaning and the import of Agamemnon’s verse. 3. It seems to me all too easy to dismiss the educational program sketched in this poem by pointing out that it is put forward by a disreputable character, who can be caught in blatant contradiction with it, especially with the high moral standards advocated in the choliambics.46 This attitude, however, is not uncommon in Petronian scholarship and has spanned the whole of the twentieth century, from Monti47 down to several recent interpreters.48 Clearly, such an approach can easily slip into a convenient excuse for the lack of an in-depth analysis of the ideas expressed in the poem, which, not surprisingly, are often dismissed as hackneyed or even risible,49 or at best as Petronius’ way to expose Agamemnon’s ignorance and hypocrisy.50 There is undoubtedly a striking contrast between Agamemnon’s precepts and his practical behavior, which has of course been pointed out by several 43

44 45 46

47 48 49 50

The often repeated pairing of Petronius’ and Persius’ Choliambics is hardly coincidental. This association is justified if limited to the critical attitude towards contemporary ways (e.g. as far as flattery of the powerful is concerned); much less, if exact parallels are sought for: see above, text to notes 29-32. This “diatribic” attitude easily explains che choliambics’ aggressiveness as opposed to the hexameters’ quiet rhythm stressed by Paratore 1933, II, 19-23. Cf. Barnes 1971, 23; Flores 1982, 73 n. 36. E.g. Cugusi 1967, 88; also Panayotakis 1995, 8 n. 3. Just about all scholars have stressed the fact that Agamemnon, while expecting the future orator not to cadge free meals from the wealthy nor to serve as claque for theatrical stars (5.5; 5.7-8), is not above flattering Trimalchio in order to be invited to his banquets (52.7) and slavishly applauds his theatrics in the Cena. Monti 1907, 14-15. I will mention just a few: Barnes 1971, 26-27; 30; Pellegrino 1986, 119-120; Panayotakis 1995, 8-9; Sommariva 1996, 59-60; Conte 1996, 118-119. After Monti, cf. e.g. Tandoi 1965, 324 n. 3 (“il programma e le risibili idee… del… personaggio”); Walsh 1970, 85-86; Heldmann 1982, 254. Cf. e.g. Tandoi 1965, 324; Barnes 1971, 31; Pellegrino 1986, 119-120. 23

Chapter I scholars who do not believe in the seriousness of the educational program presented in the poem.51 Many have also remarked that Agamemnon is only playing a role: the role of the scrupulous schoolteacher he likes to be taken for.52 Roleplaying on the part of the several characters is in fact ubiquitous in the Satyrica, and can easily be assumed here too.53 Nevertheless, the very fact that Agamemnon is playing the role of a scrupulous schoolteacher, caring for his pupils’ education, makes it worthwhile – indeed necessary – to examine his program with great attention. Its seriousness must be assessed in itself, quite irrespective of Agamemnon’s obvious moral shortcomings, and even of his actual intention and capability to carry it out. Its validity can only be judged in the frame of an indepth analysis of the Petronian text in relation with what we know of the contemporary cultural debate. We should not forget that, aside from, and independent of, playing a role in order to attain particular goals,54 Agamemnon, like all other characters in the Satyrica, is mainly playing the primary role allotted to him by the author: that of the teacher of rhetoric.55 It can hardly be surprising, and is actually to be expected, then, that, as a literary character, he should be modeled after the typical teacher of the time.56 It follows that, in view of Agamemnon’s primary role in the novel – representing a definite professional group –, his program must necessarily possess a certain degree of seriousness.57 It may be added that the interpretations that regard the program sketched in our poem as the complete curriculum of an accomplished education58 do not seem to be unjustified. It must also be stressed that this program appears from the start to be closely linked with the actual problems outlined in the preceding prose text; it 51

52

53 54

55

56

57

58

24

Cf. e.g. Nelson 1956, 204; Kissel 1987, 327; the latter, however, rightly remarks that the inconsistency of Agamemnon’s behavior as compared with his precepts is irrelevant as far as the evaluation of his educational program is concerned. Suffice it to quote Slater 1990a, 32: “Agamemnon is a professional, if limited, poseur. Even his name denotes an assumed role”. Slater, however, rightly remarks that Agamemnon’s role-playing does not rule out a serious identification with the role. Panayotakis 1995, 9 is undoubtedly right on this point. Slater 1990a, 31 remarks that Agamemnon’s object in our context is not clear. The opinion of Kennedy 1978, 176, who believes he aims to obtain sexual favors from Encolpius, does not seem to be supported by the text. Walsh 1970, 86 thinks Agamemnon to be “the typical teacher of the age”, proposing an educational program “which every pedagogue approved”, though not believing it to be serious (cf. above, note 49) and fancifully looking for hints at Petronius’ personal situation in our poem. Sane remarks in Soverini 1985, 1734 and Barnes 1973, 797-798. As already observed by Monti 1907, 22-31, though he believed the ideas expressed by the character Agamemnon to be hackneyed. More recently Soverini 1985, 1726 n. 93 remarks that Agamemnon’s speech is in accord with the role he plays in the whole novel. This is recognized even by Tandoi 1965, 324 (Agamemnon’s speech is Petronius’ way to sketch him as a schoolteacher typically unable to get rid of commonplaces, but is serious “nelle intenzioni dell’espositore”). As done especially by Nelson 1956.

The Education of the Orator (Petr. 5) closely linked with the actual problems outlined in the preceding prose text; it is not an ideal program to be carried out, if at all, later in time, as maintained by one of the scholars who take it to be serious.59 Quite the other way: our poem aims, in its two parts, to obviate exactly the evils pointed out in the previous prose speech: ambition at the ethical level, which entails, at the specifically literary level, superficiality and slovenliness, consequent to the haste with which practical success is sought and pursued.60 The morally oriented choliambics and the technical and didactic hexameters reveal their organic unity from this point of view too. This particular consonance with the preceding prose, as it seems, has never been pointed out before, but it seems undisputable that on this point Agamemnon’s poem picks up ideas he had just expressed, even though scholars disagree on the exact relation of the poem as a whole to the rhetorician’s prose speech, in spite of the conspicuous formal parallels.61 In the light of these remarks it seems to me that we are hardly entitled to dismiss the program proposed in our poem by declaring it insignificant or even risible. It is necessary, however, to avoid the opposite extreme, as represented by those who sever the poem and the prose context from the narrative situation and from the characters of Petronius’ novelistic fiction, to search for the author’s genuine literary ideas in the opening sections of the Satyrica. In this relation, the sanest attitude seems to be Soverini’s, who stresses Petronius’ detachment from the ideas and behaviors he lends the characters of his novel,62 even though, later on, he too does not resist the temptation to isolate some critical concepts allegedly corresponding to the author’s own ideas.63 In 59 60 61

62 63

I am referring to Kissel 1978, 320. Cf. Petr. 4.2-4. It is therefore incorrect to assume that the choliambics correspond to ch. 3 and the hexameters to ch. 4. See following note. These have been already listed by Collignon 1892, 230 (3.3 cum cenas divitum captant ~ 5.5 cenas impotentium captet; 4.1 severa lege proficere ~ 5.1 artis severae; 4.3 irrigarentur ~ 5.12 bibat… fontem and 5.21-22 flumine largo / plenus; 4.3 grandis oratio ~ 5.20 grandia… verba). Schönberger 1929, 1199 lists more parallels (also from chs. 1-2) and believes the choliambics to correspond to the ideas expressed in ch. 3, the hexameters to those in ch. 4 (in particular to 4.3: cf. also Schönberger 1938, 175 and Fuchs 1938, 173 n. 32); the correspondence of the choliambics to ch. 3 is rightly rejected by Kissel 1978, 321 n. 51. Many scholars, nevertheless, believe the whole poem to develop themes of the preceding prose chapters: among others Brugnoli 1963, 257; Cugusi 1967, 87; Barnes 1971, 29; Flores 1982, 73. Nelson 1956, 221 n. 5 seems isolated in warning that only sporadically do the former chapters provide material that can be useful for the interpretation of the poem. Soverini 1985, 1728; 1731; Cf. also Codoñer 1990, 73. Soverini 1985, 1735 n. 1. Cf. also Pellegrino 1986, 119, who believes it possible to pick out some ideas possibly corresponding to the author’s convictions. I am far from ruling out that Petronius may agree with some, or even with several, ideas on literature expressed by Encolpius, Agamemnon, or Eumolpus; I only wish to stress that his portrayal of these characters is totally artistic, not at all meant to convey the author’s critical 25

Chapter I my opinion these will for ever escape us,64 inasmuch as Petronius’ intention is not to express his own thinking but rather to sketch two fictional characters discussing literature on the basis of the guidelines commonly accepted in the contemporary cultural debate. This takes place, besides, in a largely elusive fictional context, owing to the state of the novel’s tradition. It is therefore hardly warranted both to contend that the ideas expressed in the Satyrica’s opening chapters are totally unrelated to Petronius’ own65 and to assert that they reflect them more or less faithfully,66 with the characters simply playing the role of his spokespersons. It is then all the more unjustified to inquire whether Petronius is closer to Agamemnon’s or to Encolpius’ ideas,67 even if this is done within the frame of a perfectly acceptable investigation of the relation of Encolpius’ and Agamemnon’s speeches to each other.68 It is more justifiable to compare the rhetorician’s literary ideas with those of another characther of the novel: the poet Eumolpus. It should be noted that they express them with similar formulations.69 Like Agamemnon, Eumolpus behaves

64 65

66 67 68

69

26

ideas; cf. Barnes 1973, 797; and already Collignon 1892, 62-63, even though he aims to glean Petronius’ literary credo from his characters’ utterances. Heldmann 1982, 244 rightly remarks that it is impossible to establish how much of the opinions stated by Encolpius, or others, is shared by the author. So Kennedy 1978, 176 (referring to Encolpius’ ideas expressed in chs. 1-2). A variant of this attitude is represented by those who believe that the ideas on eloquence expressed by Petronius’ characters are only the author’s way to portray them negatively: Tandoi 1965, 324; Barnes 1971, 27, 31; Panayotakis 1995, 8-9. Pellegrino 1986, 119120 proceeds farther than anybody else: he believes that Cicero’s quotations which Petronius places on Agamemnon’s lips (but that Petr. 5.13 Socratico… grege alludes to Cic. de orat. 1.42 philosophorum greges does not seem certain) mean to expose the superficiality of the rhetorician’s culture and, through him, the decay of eloquence. So Sage 1915, 48; Scheidweiler 1922, 1052; Schönberger 1929, 1200; Sullivan 1968, 164; Kissel 1978, 317-319. Rightly Barnes 1973, 797: “it is irrelevant to ask, with whom does Petronius agree?”. Cizek 1975a remarks that Encolpius launches a general assault against the decay of eloquence and of all the arts, whereas Agamemnon restricts himself to the former and identifies some faults which he tries to obviate (it is hard to understand why, on p. 199, Cizek ascribes the poem, which proposes the remedies, less credit than to the previous chapters); he believes Petronius to be closer to Encolpius. Similar ideas are expressed in Cizek 1975b, with a stronger emphasis on the compatibility of the two characters’ speeches. According to Kissel 1978, 318-318 the two speeches are perfectly complementary. Cf. e.g. Petr. 5.6-7 nec… obruat vino / mentis calorem ~ 88.6 at nos vino scortisque demersi; 5.13-14 mittat habenas / liber ~ 118.6 praecipitandus est liber spiritus; 5.2122 flumine largo / plenus ~ 118. 3 ingenti flumine litterarum inundata and 118.6 plenus litteris. Both Agamemnon and Eumolpus associate literary expression with philosophy (4.3 and 5.13; 88.7) and mention bona mens (3.1; 88.8). Both identify the cause of the decline with greed, foreign and insensitive to artistic perfection (4.2; 88.8). And, as we shall see, both try to reconcile enthusiastic inspiration with painstaking literary training.

The Education of the Orator (Petr. 5) as a rogue; he is nevertheless sincere in his poetic vocation, irrespective of the literary level he attains.70 He has a deeper and more complex relationship with poetry than Agamemnon has with rhetoric; he practices poetry in person, and has actually adopted it as a lifestyle, whereas Agamemnon is, more than anything else, a teacher, and, even though he may theoretically believe that the true orator must be a vir bonus dicendi peritus, appears to be content with teaching this concept in theory; he does not live as an orator as Encolpius lives as a poet. 4. We may discuss, then, whether Agamemnon’s educational program amounts – or not – to a proposal that can be considered adequate to face the unanimously recognized crisis affecting Roman eloquence, but we can hardly dismiss it as an impostor’s absurd brainchild. It will be necessary, therefore, to examine it in detail and compare it with what the leading representatives of contemporary teaching were able to offer. Before proceeding on this path, however, it is necessary to refute some interpretatations that superpose whimsical patterns upon the reality of the text and can only divert attention from its real content and import.71 By sticking closely to the text we shall be able to avoid the misunderstandings to which many interpreters of our poem have fallen victim to. The poem’s last two lines72 confirm beyond any doubt what should be clear from the very beginning to the unprejudiced reader, i.e. that Agamemnon is concerned with the education of the future orator, with no reference to the activity of one that is already active and practicing.73 Consequently, subducta foro… pagina (v. 16) must be referred to literary but not specifically oratorical exercises, though ob70 71

72

73

See ch. X, not overlooking the final remark on the total separation between Eumolpus’ representation – or self-representation – and Petronius’ own opinions. According to Flores 1982, 71-72 the poem at ch. 5 explains and justifies the prosimetric nature of Petronius’ work. Ogrin 1983, 49-51, though starting from the interesting observation hinted at above (note 41), sees a double level of meaning in the choliambics, and proposes several idiosyncratic interpretations, which, by the way, suppose that Agamemnon be aiming at the education of the poets, when it is obvious that he has the orators in mind. Connors 1998, 12 suggests that the distinction of various literary genres in Agamemnon’s poems (but she overlooks the fact that they are supposed to lend their various virtues to the orator’s expression) may be pointedly opposed to the fusion – or confusion – of several genres in the novel. This is ingenious, but it is also a way to superimpose modern fashionable patterns upon the reality of the text – with the aggravation of looking for a nonexistent programmatic import where none is clearly visible, while rejecting it where, in my opinion, it is evident (at 132.25). Cf. ch. XVII and Introd. note 20. The imperative (succinge) followed by the future (defundes) corresponds to a conditional clause (cf. Setaioli 2000, 54). The acquisition of the bona listed in the preceding line is an indispensable condition in order to attain eloquence. This reference is supposed by Kissel 1978, 324-325. He has been rightly refuted by Pellegrino 1986, 163. 27

Chapter I viously aiming to acquire expressive ranges and powers the pupil will later be able to employ when he becomes a practicing orator. There is no hint at a temporary pause in forensic practice on the part of an already active orator.74 It is also self-evident, unless we lose sight of the reality of the text, that Agamemnon recommends not only reading the models, but also unremittingly carrying out written exercises, in reference both to Greek oratory of the Demosthenic type (v. 14) and to the Latin literary genres mentioned later (vv. 17-20). It is indeed difficult to explain otherwise the stubborn persistence with which the idea that our poem only proposes a canonical reading list keeps reappearing in Petronian scholarship,75 even though the correct meaning was already clear to Petronius’ early interpreters.76 In fact, almost all translations,77 obviously obliged not to lose contact with the reality of the text much more than critical essays are, correctly take it for granted that our poem refers to written exercises too.78 Some scholars propose a needless compromise: though they realize that our text implies something more than mere reading, they assume that it recommends the recital (almost a “performance” or “staging”) of the models rather than an attempt to imitate and reproduce them through written exercises.79 In reality, though we shall leave for the last paragraph some remarks on specific linguistic details, it is easy to notice as of now that Petronius’ formulation, after the mention of Homer and the Socratic philosophers, presents us with a dynamic view of literature: it is a living and acting might, which at first is put in 74

75

76

77

78

79

28

The reference to Sen. tranq. 7.2 Isocrates Ephorum iniecta manu a foro subduxit proposed by Friedrich 1935, 496, to prove that Petronius has the already practicing orator in mind, is hardly fitting. Frankly outlandish is the interpretation of Nelson 1956, 211, who wishes to lend an active meaning to subducta. See the rightful criticism of Müller 1957, 504-505. E.g. in Nelson 1956, 211, 214; Müller 1957, 505 (approving Nelson and expressly ruling out any reference to written exercises); Courtney 1970, 65; Walsh 1970, 86; Barnes 1971, 17; Kissel 1978, 325; Pellegrino 1986, 162; Courtney 2001, 60-61. Palmerius, Barthius, and Bourdelot already realized that this poem also refers to written exercises: cf. Burman 1743, I, 34-35. In more recent times the correct meaning was grasped by Sage 1915, 48; Schönberger 1935, 1243; Schönberger 1938, 175 (with fitting references to Cic. de orat. 1.95, 2.90, 2.96, 2.131, Quint. 10.1.2-3), 222 (with reference to Quint. 10.5.15); Tandoi 1965, 323-325; Ogrin 1983, 56 n. 39; Slater 1990a, 30. With the exception of Ehlers’ ap. Müller-Ehlers 1983, 15, who at v. 17 renders det pagina cursum with “soll er sich… der Lektüre befleißen” and translates vv. 18-19 with “lauschen dem ewig wechselnden Gang des Erdengeschehens, und wo ein epischer Dichter vom Krieg singt, sei er zu Gaste”. So, for example, Heseltine 1913, 9; Ernout 1923, 5; Aragosti 1995, 141; Reverdito 1995, 7. A case in point is Walsh 1996, 4 (a translation), who hints at written exercises (“let his page run free”), contrary to Walsh 1970, 86 (an essay): cf. above, note 75. Scheidweiler 1922, 1052 and Fuchs 1938, 173 mention the Vortrag of model texts (in reference to Demosthenes). Nelson 1956, 213-214, 215, 216 believes the Vortrag to apply to all model texts. This was, no doubt, a widespread practice in the schools of rhetoric (cf. e.g. Quint. 2.5.6-9).

The Education of the Orator (Petr. 5) motion by the pupil (v. 14 quatiat), but then operates of its own accord (v. 15 circumfluat; 16 mutet; 18 sonet; 19 dent; 20 minentur). After the preliminary preparation accomplished by resorting to Homer and the philosophers,80 then, Agamemnon’s verse assumes as a matter of course that the study of the models should not be separated from the active attempt at imitation and re-creation by the pupil through written exercises. In the first stage, which is devoted to Greek literature, he must try to reproduce the eloquence81 of Demosthenes.82 The shift to Latin models is also made in view of the pupil’s literary activity: the sapor which they must affect is the “flavor” of his writing, which is one and the same with the pagina of line 17. And, as we shall see more in detail later on, lines 1720 undoubtedly allude to literary exercises carried out by the pupil. There can be no doubt about this: Agamemnon himself has expressly stated, a few lines before, that the purpose of studying the models is to attain the ability to imitate them.83 According to some scholars the educational program proposed by the poem differs from the one that was commonly adopted at Rome at the time.84 We may counter this opinion by objecting, in the first place, that the two evils that the rheorician clearly diagnoses in his prose speech and tries to obviate in the poem – i.e. the parents’ ambition and their consequent indifference to the real learning of their children, if not conducive to immediate practical success85 – had been identified generations before as hindrances to the progress of Roman elo80

81

82

83

84 85

It should not escape attention that Homer must be “drunk” (bibat) and that one must have “one’s fill” (plenus) of the philosophers. In both cases there is no hint at any active “re-creation”. In view of this, Pellegrino 1986, 158 is probably right when he regards these stages as propaedeutic to the strictly technical study of eloquence. The importance of the reading of Homer for future literary activity is stressed by felici (v. 12), which must be taken as predicative: the Homeric source fertilizes the pectus of those drinking at it: a metaphor to say that it provides the necessary equipment for expression. Notice the complementary image (“food”, like “drink” here) in reference to the exercises in Latin epic (v. 19 dent epulas, for which see below). See below for the value of philosophical preparation. The image of weapons in reference to eloquence was traditional: cf. Schönberger 1939, 510. The expression Demosthenis arma is attested before Petronius: Prop. 3.21.27 (already quoted by Burman 1743, I, 33). Exercises in Greek with the intent of attaining eloquence in Latin were common in the orator’s apprenticeship; cf. Cic. Brut. 310 idque faciebam multum etiam Latine, sed Graece saepius… quod Graeca oratio plura ornamenta suppeditans consuetudinem similiter Latine dicendi adferebat. Petr. 4.3 quod vellent imitari diu audirent. It is hardly necessary to remind that the alternation of reading and writing was taken for granted; cf. e.g. Quint. 1.12.4 et stilus lectione requiescit, et ipsius lectionis taedium vicibus levatur. Cf. Sen. ep. 84.1-2. This is the opinion of Nelson 1956, 207 and Sullivan 1968, 165 (who at any rate does not see this program as revolutionary) See above, text to note 60. 29

Chapter I quence86 and continued to plague its teaching even after Petronius.87 The very structure of our poem, then, places it in the line of the most serious and unadulterated tradition of Roman teaching. Several scholars have maintained that the Campanian (and therefore Greek) Agamemnon, who places the study of Greek models before that of the Latin ones, must be addressing pupils whose mother tongue was Greek,88 or at least that he is snobbishly aping courses of studies meant for young aristocrats.89 This interpretation has rightly been refuted by several scholars.90 It is in fact well known that in Roman teaching Greek came before Latin at every level, including elementary education. It is hardly necessary to dwell on the numerous testimonies to this effect, which have already been adduced by many scholars.91 I will only call attention to the fact that among these witnesses we find Petronius himself92 and his contemporary Seneca.93 In Agamemnon’s curriculum Greek readings precede those in Latin – which of course mark a level that is closer to the final goal of the teaching, i.e. producing a Roman orator94 – just like in the former (Greek readings) poetry comes before prose, and in both areas the final and highest level is allotted to oratory, emblematically indicated through the names of Demosthenes and Cicero.95 This study program is in fact carefully graduated, with stages and passages which are punctually marked, even at the linguistic level: primos… annos, mox,

86

87

88 89 90 91

92

93 94 95

30

As attested in reference to Orbilius, Horace’s teacher, by Suet. de gramm. 9.2 librum etiam cui est titulus Peri algeos edidit continentem querelas de iniuriis quas professores neglegentia aut ambitione parentum acciperent. Cf. Tac. dial. 28.2; 28.5-6. Cf. Schönberger 1940, 624 for the difference between Petronius, who stresses the parents’ blameworthy haste, and Tacitus, who puts more emphasis on their indifference. So Nelson 1956, 206; Aragosti 1995, 141 n. 9. So Nelson 1956, 207. Cf. e.g. Müller 1957, 504; Kissel 1978, 322 n. 54; Pellegrino 1986, 154-155. Cf. e.g. Schönberger 1938, 222; Pellegrino 1986, 154-155; and already Burman 1743, I, 32-33. See, among others, Quint. 1.1.12; Plin. ep. 2.14.2. Ogrin 1983, 52 n. 24 rightly takes it for granted that Petronius is following the established didactic tradition. This was so ingrained that Quintilian (1.1.13) protests excessive delay in the study of Latin. Petr. 46.5 ceterum iam Graeculis calcem impingit et Latinas coepit non male appetere. The reference has been made by Schönberger 1935, 1243. This passage is correctly illustrated by Tandoi 1968, 77-78. Sen. nat. 6.23.4 quisquis primas litteras didicit scit illum (Neptunum) apud Homerum vocari. In the same way, in Quintilian (10.1.46-131) the list of Greek authors precedes the Latin ones. For the propaedeutic character of the exercises in Greek cf. above, note 82. It is hardly necessary to remind that both were regarded as the exemplary standard in Greek and Latin oratory respectively. Cf. e.g. de subl. 12.4-5 ; Quint. 10.1.39, 105-106; Iuv. 10.114; etc.

The Education of the Orator (Petr. 5) hinc, modo, interdum, sic.96 Nothing different was to be expected from a teacher who shortly before had criticized those parents whose ambition and superficiality prevented the educator from carrying out the necessary laborum gradus.97 Interestingly enough, Agamemnon agrees with the best teachers of his time on this point too; actually, he places himself in the most authentic strand of the Ciceronian tradition, as interpreted by the finest representatives of rhetorical teaching in the I century A.D.98 Gradualness in teaching, on the other hand, does not rule out directing all levels of studies to the production of the accomplished orator, and identifying rhetoric with culture tout court,99 or at least with the climactic level of culture, in accordance with the conception of eloquence formulated by Agamemnon shortly before,100 again in agreement with what we find, for example, in Quintilian.101 As we have already remarked, the curriculum proposed in the hexameters is comprised of two inseparable and complementary components: reading and written composition. As far as the first one is concerned, it has been remarked102 that Petronius’ poem anticipates, under Nero, the effort to create a canon of standard models, which will be fully developed under the Flavians in Quintilian’s tenth book.103 We may add that Agamemnon’s canon, which is certainly much more meager than Quintilian’s,104 agrees with him in placing Demosthenes and Cicero at the summit of oratory; these are actually the only orators he expressly names. This remark (Nelson’s) is correct, but neglects the important circumstance that catalologs of this type had been put forward long before the Flavians. We 96

97 98

99

100

101 102 103 104

Cf. Barnes 1971, 35 n. 12; Connors 1998, 12 registers a looser network of stages and passages. The gradualness in study marks the whole structure of the poem and strongly supports the reading hinc at v. 15, as against hunc (accepted by Collignon 1892, 231 (other attested readings are huic and huc). Petr. 4.3. See e.g. Tac. dial. 30. 3 (Cicero) sua initia, suos gradus, suae eloquentiae velut quandam educationem refert; and cf. e.g. Cic. Brut. 232 gradus tuos (= Ciceronis) et quasi processus dicendi studeo cognoscere. Nelson 1956, 218 rightly remarks that Agamemnon recommends one single type of education: the rhetorical type, and that “«rhetorisch» ist identisch geworden mit «allseitig und allgemein gebildet»”; cf. Nelson 1956, 201. According to Pellegrino 1986, 156-158, the program sketched in our poem must be referred to the higher-level curriculum, in which the authors are studied in order to enhance one’s rhetorical capabilities. Petr. 4.2 eloquentiam, qua nihil esse maius confitentur. The same idea is expressed by Quintilian just in connection with the need for encyclopedic education on the part of the orator: Quint. 1.10.7 mirabimur, si oratio, qua nihil praestantius homini dedit providentia, pluribus artibus egeat…? Cf. also the following note. Cf. e.g. Quint. 1 prooem. 9-20. By Nelson 1956, 208. The rejection by Kissel 1978, 325 n. 67 is unjustified. Quint. 10.1.46-131. As rightly observed by Nelson himself (Nelson 1956, 202). 31

Chapter I may confidently state that in Petronius’ time they had been a relevant feature, at least in Greek teaching, for several decades. We may get an idea from the preserved abstract of a lost work by Dionysius of Halicarnassus: the De imitatione.105 Once more we should notice that Agamemnon follows the main strand of ancient rhetorical teaching: in both Dionysius and Quintilian the list is in fact comprised of poets, philosophers, historians, and orators – the very same literary genres which are mentioned in our poem.106 As far as written composition is concerned, it is hardly necessary to dwell on the numerous texts which, from Cicero on, recommend it as an essential part of the future orator’s apprenticeship.107 I will only point out a text by Quintilian,108 which, as far as I know, has escaped the attention of all scholars, but sheds clear light upon lines 17-20 of our poem, which have caused Petronius’ interpreters a great deal of problems. The comparison with Quintilian’s passage shows that Agamemnon is thinking of exercises through which the orator-to-be will undertake the writing of history, poetry (epic), and finally oratory of the Ciceronian type – the highest level of expression. In all three cases these exercises are different and distinct from those directly aiming to prepare for the activity in the forum: sometimes,109 says Agamemnon, the pupil’s writing must fling itself110 into a wider range than strictly required by his future profession: it must be sub-

105 106

107

108

109 110

32

Dion. Hal. de imit. fr. VI, II, pp. 202, 18-214, 2 U.-R. We shall point out – in part anticipating what will be developed later on – that poetry appears in lines 11-12 and 19; philosophy at 13; oratory at 14 and 20; history at 18. For the identification of the literary genres see below. The comparison with Dionysius’ and Quintilian’s lists is a further argument against different identifications. It goes without saying that even for the genres mentioned only in reference to writing reading is naturally required too. E.g. Cic. de orat. 2.90, 96 ; Quint. 10.1.1-3, 10.1.20. These texts are referred to by Schönberger 1938, 175, and prove that the reading of the models and written composition were inseparable. Quint. 10.5.14 alitur vero atque enitescit velut pabulo laetiore facundia et adsidua contentionum asperitate fatigata renovatur. 15 Quapropter historiae nonnumquam ubertas in aliqua exercendi stili parte ponenda et dialogorum libertate gestiendum. Ne carmine quidem ludere contrarium fuerit, sicut athletae, remissa quibusdam temporibus ciborum atque exercitationum certa necessitate, otio et iucundioribus epulis reficiuntur. 16 … nam si nobis sola materia fuerit ex litibus, necesse est deteratur fulgor et durescat articulus et ipse ille mucro ingenii cotidiana pugna retundatur. Shortly before (10.5.4-5) Quintilian encourages the pupil to write paraphrases of Latin orations too, which corresponds to line 20 of Agamemnon’s poem, just like the passage transcribed above corresponds to lines 17-19. Petr. 5.17 interdum; cf. Quint. 10.5.15 nonnumquam. This is how det pagina cursum should be understood. We are faced with a full-fledged metonymy: the pagina symbolizes the orator-to-be and the eloquence he deploys in his writing.

The Education of the Orator (Petr. 5) ducta foro, that is not limited to judicial quarrels (ex litibus, says Quintilian).111 Only thus – so can be inferred from Quintilian – will it be possible to raise eloquence above the shallows inevitably bound with its practical function and endow it with vigor and swing. That the literary genres alluded to at lines 17-20 are history, epic poetry and Ciceronian oratory is recognized by some scholars,112 though no one, as far as I know, refers to Quintilian’s text we have just quoted. Many, however, propose different interpretations, in some cases also resorting to whimsical corrections and/or transpositions of lines.113 In reality Petronius’ text is clear enough in itself and the parallels in Quintilian and Dionysius of Halicarnassus we have pointed out114 help remove any reasonable doubt. The historiography line 18 refers to is the opus oratorium maxime of Hellenistic mold, in which fortuna ( ) played a crucial role with the swift inter111

112

113

114

Quintilian probably has both apprentices and practicing orators in mind. By contrast, the text of our poem does not allow to see a reference to the latter: cf. above, beginning of § 4, and note 74. E.g. by Nelson 1956, 221-217; Barnes 1971, 17; Kissel 1978, 325; Pellegrino 1986, 163-164. Courtney 2001, 60 and nn. 12-13 believes a reference to epic poetry not to be certain, though he does not rule it out. This also applies to preceding verses, starting with line 15 (for which see below, § 6). I do not mean to give a complete list of all the numerous (and equally unwarranted) corrections. They may be found in the works quoted in this note. I only point out that already Burman 1743, I, 39 states that, starting with line 15, “mihi hic sunt merae tenebrae”. The same remark is found in Bücheler 1862, 9: “hunc versum (15) et qui sequuntur ita dedi ut in libris leguntur, quoniam probabiliter emendati non sunt”. Müller 1961, 7 writes in the apparatus, in reference to vv. 18-19: “versus perobscuri”. Burman 1743, I, 36 placed v. 20 after v. 16. He was followed by Scheidweiler 1922, 1053; Fuchs 1938, 174 (who also transposes vv. 18-19); Courtney 1970, 65; Courtney 2001, 60 n. 11; and regrettably even by Müller 1995, 4. That this transposition is groundless is well demonstrated by Kissel 1978, 325 n. 65. These lines are supposed to be corrupt also by Schönberger 1935, 1243; Schönberger 1939, 511 (he also unconvincingly takes sonet to be transitive). Friedrich 1935, 496-497 believes that only historiography is alluded to (his interpretation of epulae and bella as the subjects it treats is unacceptable, though it has been recently picked up by Rimell 2002, 25, 96, 132, 167; cf. also Walsh 1996, 4: “recounting feasts and war’s harsh blasts / with lofty Ciceronian force”; Heseltine 1913, 9: “tell tales of feasts”); the syntactical structures of vv. 19-20 he proposes (Friedrich 1935, 496 n. 5) cannot be accepted either. By contrast, Collignon 1892, 232 believes lines 18-19 to refer solely to epic. Walsh 1970, 86 n. 1 refers line 18 to tragedy; a possible hint at tragedy is seen in line 19 by Slater 1990a, 30 (he too believes epulae and bella to be parallel). Schönberger 1935, 1243 (cf. Schönberger 1939, 511) corrects dent epulas to regum epulas, with Thyestes’ tragic meal in mind. Even more radical corrections to a similar effect were proposed by Birt 1928, 44. Quint. 10.5.14-16 and 10.5.4-5 (above, note 108); and the canons of models at Quint. 10.1.46-131 and Dionys. Hal. de imitat. fr. VI (notes 103 and 105). 33

Chapter I change of highs and lows.115 Quintilian not only recommends to practice it – together with poetry – as a literary exercise in the passage previously quoted,116 but elsewhere too he stresses its similarity to poetry, which makes it a veritable carmen solutum.117 Agamemnon places himself along the same line. It cannot be doubted, in fact, that at line 19 he implies that the future orator’s training includes exercises in poetry (concretely, in epic) after those in historiography.118 The expression dent epulas, which has sparked so many discussions,119 becomes perfectly intelligible only in the light of Quintilian’s passage we have often referred to, where very similar expressions appear.120 Quintilian’s context makes it clear that the epulae are to be identified with the exercises in poetic (and no doubt historiographical) composition themselves, even though this obvious meaning does not rule out, and actually supposes, the reading of texts of this type. Agamemnon’s lines must undoubtedly be interpreted in the same way: eloquence is fed (epulae) not merely through the study of the models, but also, and primarily, through written exercise. That historiographic and poetic writing should be called “removed from the forum” is hardly surprising; but even line 20 certainly hints at exercises in oratory of the Ciceronian type that are still directed to the proficiency of the oratorto-be in the name of pure and disinterested eloquence, with no reference yet to the practical success of the accomplished orator already practicing in the forum.121 As we have already remarked, Cicero represents the summit of eloquence for Agamemnon as for Quintilian and the other teachers of the time;122

115 116 117 118 119

120

121 122

34

Suffice it to refer to the clear treatment by Nelson 1956, 211-214. Quint. 10.5.15. Quint. 10.1.31 est enim proxima poetis et quodam modo carmen solutum. For parallels to bella truci memorata canore see Pellegrino 1986, 164. The most radical position consists in getting rid of dent, or epulas, or both words, through totally unwarranted corrections: see some in the apparatus of Bücheler 1862 and in Nelson 1956, 214, and see above, note 113. Regrettably, even Courtney 1970, 65 proposes to substitute per pugnas for dent epulas. Nelson 1956, 216 understands dent epulas as “let them please” the ear (his proposal to interpret et as a case of postposition is equally unacceptable: this et surely means “also”: cf. Müller 1957, 505); a similar interpretation in Kissel 1978, 325 n. 64. Both are criticized by Pellegrino 1986, 163, who understands “let them nourish the spirit” [cf. Tandoi 1965, 323 n. 1: “forniscano alimento, materia” (scil. to your oratorical exercises – which, therefore, are not identical with them)]. More interpretations above, note 113. Quint. 10.5.14 pabulo laetiore; 15 iucundioribus epulis. The complete text above, note 108. For epulae used metaphorically cf. Plaut. Poen. 1175 epulas dare (same iunctura as in Petronius) and the texts quoted at TLL V 2, 701, 11-25. See above, notes 73 and 74 for Kissel’s and Friedrich’s erroneous interpretation. The grandia verba of line 20 correspond in fact to the grandis oratio which is the ideal of both Encolpius and Agamemnon (2.6 and 4.3 respectively).

The Education of the Orator (Petr. 5) but in this poem Cicero is more important as a model of style than as a pattern of behavior in court.123 We can confidently assume, then, that line 20 alludes to exercises in oratory of the Ciceronian type. What remains to be ascertained is their specific nature. According to an appealing suggestion124 these would be suasoriae in the form of prosopopoeias, in which the pupil made the great orator speak, lending him firm words and attitudes against Mark Antony’s threats: a theme that had long since become commonplace in the schools of rhetoric.125 In reality, though, the reference to a genre that never attained autonomy and dignity out of the schools seems rather out of place after the hints at lofty genres such as epic and historiography.126 What is probably meant are reworkings of some of Cicero’s speeches, writings, that is, that formally belonged in the same genre of lofty oratory as those of the great Arpinas. As we have already seen, this type of exercise was expressly recommended by the teachers of rhetoric.127 An exercise described as subducta foro will probably not aim to reproduce Cicero’s judicial speeches; in all likelihood the models will be political orations, such as the Philippics (but also the Catilinarians, which fit the verb minentur even better), uttered by the great orator with strong political determination and indomitable civil passion (indomiti),128 pugnacious attitude (minentur) and lofty style (grandia… verba).129 In conclusion, it seems unquestionable to me that, after the study of the Latin models alluded to at lines 15-16, already combined with written exercises,130 Agamemnon’s program expects the pupil to try himself in the reproduc123 124

125 126 127 128 129

130

I am not the first to express this opinion: see Kissel 1978, 325 n. 65. By Tandoi 1965, 322-325 [p. 325: “tuonino declamati i grandia verba (da te rifatti per esercizio) di Cicerone che non intende piegarsi (ad Antonio)”]. Tandoi refers to Pers. 3.45 grandia… morituri verba Catonis. I had independently thought of a rhetorical exercise in prosopopoeia, before reading Tandoi; but later I have given up the idea. Cf. e.g. Sen. rhet. suas. 6 and 7. The same is true for the grandia verba, which echo the previously stated literary ideal (cf. above, note 122). Cf. Quint. 10.5.4-5 (above, note 108). Tandoi 1965, 324-325 refers indomiti precisely to the Philippics. According to Ogrin 1983, 56 n. 41 the epithet can only refer to Cicero’s political behavior. We gather from Quintilian that the orator must read and practice poetry and history in order to acquire the expressive ranges fitting the purpose of delectare and docere (Quint. 10.1.28 solam petit voluptatem; 31 scribitur ad narrandum, non ad probandum). Probably Agamemnon’s intention is the same, when he recommends exercises in historical and poetic writing; the formulation of line 20 seems to imply that the imitation of Ciceronian speeches was encouraged in order to acquire the expressive range aiming to movere. Possibly the stylisic sapor of line 16, which is surely to be referred to the pupil’s written exercises, also applies to his oratorical style, whose “flavor” will be changed by his oc35

Chapter I tion of the three genres alluded to at lines 17-20, oratory of the Ciceronian type occupying the top.131 In the same way, the reproduction of Demosthenes’ oratory is placed at top of the preceding stage devoted to Greek literature and writing.132 At any rate, in my opinion the great amount of parallels we have adduced in this paragraph, especially in connection with the part in hexameters of our poem (in addition to those previously pointed out for the choliambics)133 prove that Agamemnon’s educational program cannot be dismissed as lacking any seriousness, and that it actually belongs in the most engaged strand of contemporary teaching.134 Establishing whether Roman traditional teaching could adequately cope with the challenges faced by contemporary culture is quite another problem. This falls outside the scope of our inquiry. I only add that those who tax Agamemnon’s educational program as conventional and inadequate should do the same not merely with the degraded current practice of rhetorical teaching,135 but also with the attempts at obviating the crisis devised by the best teachers of the time.136 5. We have already repeatedly encountered the author that had risen to be the living symbol of Roman eloquence: Cicero.137 Agamemnon mentions him ex-

131 132 133 134

135

136

137

36

casional (interdum) forays into more spacious literary fields (the exercises mentioned at lines 18-20) At line 19 his… bonis is of course instrumental ablative, not dative, as interpreted by Rimell 2002, 72: “for these great things”. Cf. above, text to notes 81-82. Cf. above, note 40. Even the mention of Cicero at line 20 we have just discussed is best understood in the light of a text by Quintilian. I am referring to Quint. 12.1.16-17 (but cf. also Iuv. 8.236244; 10.114-126). If Cicero is indomitus according to Agamemnon, Quintilian mentions his neque spe neque metu declinatus animus and represents the great orator as a veritable vir bonus dicendi peritus (see the whole chapter 12.1). It should not escape us that, at the end of the technical and literary part of the poem, Agamemnon reintroduces an ethical theme that harks back to the choliambics, thus placing the seal of unity upon this metrically composite poem. According to Barnes 1971, 30 Agamemnon “cannot conceive of an educational system which would dispense with the ills of the one in vogue”. According to Ogrin 1983, 57 “il risultato al quale giunge non è la restaurazione della disciplina, bensì la raffigurazione della pratica”. Such is in fact the conclusion of Pellegrino 1986, 113 and 120, according to whom Agamemnon is the representative of a decay we can follow through those that denounce it: the elder Seneca, Quintilian, the Dialogus de oratoribus. Cf. note 42 for the association of the choliambics with Cic. Cael. 45-46; note 82 for Cicero’s oratorical exercises in Greek (Cic. Brut. 310); note 98 for the basic concept of the gradus in rhetorical education (Cic. Brut. 232); note 107 for the importance of written exercises (Cic. de orat. 2.90 and 96).

The Education of the Orator (Petr. 5) plicitly both in prose138 and in the poem, where he represents the highest model for the orator-to-be.139 Though stating that Cicero is the only source for the opening chapters of the Satyrica140 is no doubt an exaggeration, it is nevertheless assured that he has been kept in mind by Petronius – also in the capacity of expressive model – not only in Agamemnon’s, but in Encolpius’ speech as well.141 The opinion that “il modello al quale Agamennone si ispira è… quello proposto da Cicerone”142 is therefore justified, to a certain extent. It is necessary to add, however, that this Cicero is not the one of history, but rather the one the teachers of rhetoric of the I century A.D. had raised to the status of epitome and ideal of oratory.143 A decisive factor in evaluating the difference, if any, between Agamemnon’s and – say – Quintilian’s attitude concerning this point may be sought in an aspect we have deliberately left in the background in the preceding paragraph, owing to an extremely succinct formulation which makes it difficult to propose an unequivocal interpretation. I am referring to the attitude to philosophy as part of the orator’s education and training. According to Nelson144 Agamemnon agrees with Quintilian in regarding philosophy as useful for the orator-to-be, in that it can restore unity and harmony in his upbringing by obviating an unjustified but established dissociation in educational curricula. In the great teacher of the Flavian era145 we can easily grasp the idea that – according to him – philosophy has usurped an area which by right belonged to eloquence, while the latter has unduly given it up; if at present eloquence cannot do without philosophy, it is also true that the right of primogeniture belongs undoubtedly to the former. Nelson has been followed by Kissel146 and, with more attentive and more developed arguments, by Pellegrino,147 who believes the study of philosophy, both for Agamemnon and for Quintilian, to be 138

139 140 141 142 143 144 145 146 147

At Petr. 3.2 Agamemnon quotes Cic. Cael. 41 (ut ait Cicero ‘soli in scholis relinquentur’). According to Pellegrino 1968, 104-105 the quotation is reduced to a mere learned reminiscence. This Ciceronian reference is in fact detached from the original context and acquires a different meaning. The attempt of Pellegrino 1986, 119 to discover in Socratico… grege (5.15) a debased allusion to Cic. de orat. 1.42 philosophorum greges is less convincing. Cf. also above, note 65. Petr. 5.20, for which see above. As done by Schönberger 1939, 511. Cf. Petr. 2.6 (Encolpius) grandis et, ut ita dicam, pudica oratio ~ Cic. Brut. 203 grandis et, ut ita dicam, tragicus orator. So Ogrin 1983, 52 n. 25. This process is already apparent in the elder Seneca (contr. 1 pr. 6-7) and Velleius Paterculus (1.17.3) Nelson 1956, 209. Cf. Quint. 10.1.35, quoted by Nelson; but we can add the whole proem to the first book (especially 1 prooem. 10-20) and the whole chapter 22. Kissel 1978, 323. Pellegrino 1986, 116-119. 37

Chapter I reduced to a “semplice componente del cursus di studi prescritto all’aspirante oratore”.148 Even though the idea of the primordial unity of eloquence and philosophy on which Quintilian insists149 is undoubtedly Ciceronian,150 it is not difficult to realize that the Flavian rhetorician is far from assigning to philosophy the crucial importance it has for Cicero,151 which we hardly need to stress here. The conciseness of Agamemnon’s reference to philosophy makes it difficult to determine whether he allocates it the same ancillary function it has in Quintilian152 or is closer to Cicero’s authentic position. Yet, aside from Agamemnon’s traditional designation of philosophy through the turn Socratico… grege,153 even his few words may provide a spur for fruitful reflection. It must in the first place be emphasized that the adjective liber at the beginning of line 14 is linked by enjambment with the preceding mittat habenas, though not a few scholars prefer to relate it to the following quatiat, thus assuming the postposition (anastrophe) of the et that follows liber,154 even though there are no comparable instances in Petronius.155 It follows that the pupil can only be free after feeding his soul with the study of philosophy.156 148 149

150 151

152

153 154

155 156

38

Pellegrino 1986, 116-117, Also by referring to Cicero himself: Quint. 1 prooem. 13 (the reference is to the De oratore passages quoted in the following note). Quintilian (12.2.5) also refers to Cicero (de orat. 3.76, 107, 122-123) to reaffirm that the investigation of general (i.e. “philosophical”) principles pertains by right to eloquence. Cf. e.g. Cic. de orat. 3.57, 60, 72. Pellegrino 1986, 117-118 rightly remarks that Quintilian’s interest in philosophy is merely practical (a philosophical upbringing may be of help for the orator-to-be), while little or no room is allotted to pure and disinterested speculation. An emblematic passage for Quintilian’s posture on philosophy can be found at 12.11.17, where the philosophorum… opiniones are part of a list of utilia, together with those of the consulti and historical and rhetorical exempla. Cf. Schönberger 1935, 1243; Schönberger 1938, 222. Among the translators e.g. Heseltine 1913, 9; Walsh 1996, 4; also Kissel 1978, 323 n. 60. The majority of interpreters correctly refers liber to mittat habenas, but cases of ambiguity or vacillation are not lacking: Schönberger 1935, 1243 links liber with mittat habenas, but Schönberger 1939, 510 appears to refer it to quatiat. Ogrin 1983, 55-56 and n. 38 places a comma after liber, but self-contradictorily adopts Kissel’s interpretation. Pellegrino 1986, 118 also places a comma after liber, but on pp. 158-159 refers the adjective to quatiat. At any rate, he has the merit of stressing that this libertas is the result of the study of philosophy. As remarked also by Nelson 1956, 208-209 (but in connection with an unacceptable interpretation of line 19). Nelson 1956, 208-209 is completely mistaken when he takes the adjective to hint at the pupil’s independent attitude to philosophical sects (“ohne dogmatische Zwang”). More unconvincing interpretations have been proposed: Stubbe 1933, 66 paraphrases liber with generosior; Rimell 2002, 25, 81, 116, 182, 199 takes liber to mean “as a free man”.

The Education of the Orator (Petr. 5) We have already remarked that this study, just like the reading of Homer mentioned in the preceding line, marks a propaedeutic stage which still prescinds from the active imitation and reproduction of the model. But, if only after having imbibed philosophy can the pupil gain free swing in eloquence, it will not be unwarranted to surmise that the former, though preceding the strictly technical teaching, amounts, in Agamemnon’s view, to the necessary basis for the attainment of eloquence, not merely another branch of instruction in the orator’s curriculum equal in rank with the other branches of the that is required of anyone who wishes to become an orator. If this is so, then Agamemnon is closer to the Dialogus de oratoribus than to Quintilian. In the Dialogus Messala tackles the problem in two whole chapters,157 in which, to be sure, Quintilian’s conception of philosophy’s practical usefulnss for the orator does make its appearance, but philosophy is also unambiguously declared to be eloquence’s indispensable foundation. A Ciceronian reference made by Tacitus’ Messala is emblematic. Quintilian, as we remarked in a footnote,158 quoted Cicero’s De oratore to assert that philosophy is only a branch of eloquence which has unduly separated itself; Tacitus, by contrast, quotes the famous statement in the Orator by which Cicero avows to owe his eloquence to philosophy rather than to rhetoric.159 In the Dialogus there is also an expression referring to philosophy which is very close to Agamemnon’s verse: pectus implerent160 reminds us of our poem’s formulation, Socratico plenus grege, implying, it seems to me, that Tacitus’ and Petronius’ characters share the same attitude. Both react to pure and strict technicality in the name of a more comprehensive education and of the “humanistic” ideal of founding the necessary specific competence upon a cultural basis of wider scope. The comparison with Agamemnon’s previous prose speech161 allows us to conclude that a philosophical preparation is an integral and indispensable complement to those laborum gradus which are often regrettably forgone owing to the shortsighted ambition of the parents (no doubt affecting the pupils too), who are only interested in a teaching conducive to immediate practical re-

157 158 159

160 161

Tac. dial. 31-32. Above, note 149. Tac. dial. 32.6 Cicero his, ut opinor, verbis refert, quicquid se in eloquentia effecerit, id se non rhetorum , sed Academiae spatiis consecutum. Cf. Cic. orat. 12. Quintilian too (12.2.23) quotes, among others, this Ciceronian passage, but in a context that vindicates for oratory the functions usurped by philosophy (cf. above, note 149). Tac. dial. 31.1 ut iis artibus pectus implerent, in quibus de bonis ac malis, de honesto et turpi, de iusto et iniusto disputatur; cf. 32.4 pectora implebat. Petr. 4.3 ut sapientiae praeceptis animum componerent.

39

Chapter I sults,162 thus causing the irremediable decay of eloquence. The closeness of Messala’s and Agamemnon’s ideas is by now apparent.163 But we must additionally notice that in the short reference our poem makes to philosophy three expressions succeed one another164 that appear to refer to the enthusiastic type of literary inspiration, an idea, at any rate, that can be made out in other parts of the poem as well.165 In this connection some scholars have upheld the presence of the stylistic theory of the sublime in all the first five chapters ,166 not merely in Agamemnon’s poem, while others believe the text to reflect Atticist or “classicist” positions.167 To determine the literary leaning prevailing in the opening chapters of the Satyrica – granting such an inquiry to be indeed warranted – falls definitely outside and beyond the scope of our study. There is, at any rate, a definite link between Agamemnon’s (and Eumolpus’ too)168 literary doctrine and what we find in the anonymous treatise On the sublime: a link scholars seem not to have remarked. Both Agamemnon and Eumolpus advocate, almost with the same words,169 an ideal literary writing in which the swing of enthusiastic inspiration is reconciled with a solid and uninterrupted connection with the great literary

162 163

164

165 166

167

168 169

40

By contrast, Agamemnon recommends not merely long poetical and philosophical studies, but also written exercises “removed from the forum” (v. 17). For further contacts between Petronius’ opening chapters and the orator’s ideal education sketched by Tacitus in dial. 30-36 see Collignon 1892, 95-98; Kissel 1978, 319 n. 39. Plenus; mittat habenas; liber. For this implication in plenus cf. Pellegrino 1986, 118, who quotes Hor. c. 2.19.6 plenoque Bacchi pectore; 3.25.1-2 quo me, Bacche, rapis tui / plenum? The reference to Hor. ars 337 omne supervacuum pleno de pectore manat is less convincing. Cf. plenus and defundes… verba at line 22; also mentis calorem in the choliambics (v. 7). Schönberger 1935, 1248; Schönberger 1938, 175-176 (he also understands in the same way grandis at 2.6; 4.3; 5.20; his interpretation of 4.3 verba atroci stilo effoderent – cf. also Schönberger 1938, 221 – is unacceptable). Alfonsi 1948 believes the first five chapters to follow the theories of Theodorus of Gadara, which were close to those of the ; Flores 1982, 69-70 agrees; cf. also Soverini 1985, 1717. According to Sage 1915 the text reflects Atticism; Nelson 1956, 217 prefers the term “classicism”; Kissel 1978, 315 sees a contrast not between Atticism and Asianism, but rather between artificiality and naturalness. See above: § 3, end. Cf. above, notes 69, 164, 165. In detail: 5.13 mittat habenas ~ 118.6 praecipitandus... spiritus; 5.14 liber ~ 118.6 liber; 5.21-22 flumine largo / plenus ~ 118.3 ingenti flumine litterarum inundata and 118.6 plenus litteris. In Eumolpus’ words there are more hints at enthusiastic inspiration: 118.6 furentis animi vaticinatio; 118.3 neque generosior spiritus sanitatem (adopting the reading given by LO) amat.

The Education of the Orator (Petr. 5) tradition. It is remarkable that the teacher of rhetoric and the practicing poet170 concur on the need not to sever study and scholastic and technical preparation from a trend laying stress on enthusiastic inspiration. Clearly, both endeavor to save both ars and ingenium; and this is where they coincide with the position defended in the treatise On the sublime. There is more: the attempt at reconciling enthusiastic swing with a close link with the great classics is significantly parralleled in the same work, down to the employment of metaphors that closely resemble those we find in Petronius.171 It seems to me, then, that the pairing of Petronius and the literary trend represented for us by the author of the 172 can be justified with more punctual and cogent arguments than has hitherto been done.173 We must of course be aware of the fact that the literary doctrines we have recognized in Petronius’ text should be referred to the novel’s

170

171

172

173

It is hardly surprising that Eumolpus, the practicing poet, insists on enthusiastic inspiration more than Agamemnon (cf. preceding note). The latter, as a teacher, must lay more stress on the pupil’s moral and literary education. Cf. also above: § 3, end. For the image of the river and irrigation as a metaphor of the connection with literary tradition (Petr. 4.3 lectione severa irrigarentur; 5.21-22 flumine largo / plenus; 118.3 ingenti flumine litterarum inundata) cf. de subl. 13.3 . As far as enthusiastic inspiration is concerned, the references in the are almost countless; we can quote, at any rate, de subl. 15.4 ! (cf. Petr. 5.13 mittat habenas); 32. 1 and 32.4 !" !" # ! (cf. Petr. 118.6 praecipitandus… spiritus); 13.2 $ % & !" (cf. Petr. 118.6 furentis animi vaticinatio). If, as we have argued above (note 164) plenus implies an enthusiastic connotation, in flumine largo / plenus (5.21-22) and plenus litteris (118.6) the ecstatic fervor of inspiration is directly connected with the link with literary tradition, just like at de subl. 13.2 !" (cf. spiritus)… ! ! ! ' ! ( ! ! ! ! ) !* (cf. spiritus) !" ! + ! We may add that the choliambics which open the poem have much in common with the complaint we find in the about moral decay as the cause of the decline of literature (de subl. 44.6-12). Scholars have debated whether the grandis oratio which is Agamemnon’s ideal (4.3; 5.20), as well as Encolpius’ (2.6), should be taken as a technical reference to the genus grande or, as I had rather believe, in a more general sense. The first alternative is preferred by Schönberger 1929, 1199; Schönberger 1938, 221; Sullivan 1968, 164. According to Sage 1915, 50 it is what Petronius regards as the true grandis oratio, not the one thus defined by the rhetoricians. For Nelson 1956, 201 both Encolpius and Agamemnon oppose the true genus grande to the false one advocated by the moderns. Cf. also Tandoi 1965, 322-325. According to Alfonsi 1948, 48 the reference is to “sublime” eloquence. See also Kissel 1978, 313. 41

Chapter I characters only, not to the author; it is nevertheless significant that Agamemnon the rhetorician and Eumolpus the poet move on substantially common ground. 6. I will now discuss a few particular problems raised by our poem which have not been tackled in the previous pages.174 V. 1. ambit] The manuscript tradition is all but unanimous in offering amat,175 which is rejected by nearly everyone because of the meter, which requires a long first syllable.176 The only acceptable correction is ambit,177 a verb which is attested, followed by an accusative denoting the object of the ambition, in the meaning of “to aspire to”, since Plautus and down to late Latin.178 It is hardly possible to accept hamat,179 though it was adopted by Burman180 and has been defended even in recent times.181 Only for completeness’ sake shall I mention Stubbe’s conjecture navat,182 which has been accepted by Nelson.183 V. 3. frugalitatis lege poliat exacta] Poliat is a correction by Heinsius, which has been generally accepted against the transmitted readings polleat and palleat. It is useless to correct exacta to exactae, as already proposed by Gon174

175 176

177

178

179 180 181

182 183

42

The text of our poem is one of the parts of the Satyrica that most have prompted conjectures and corrections (cf. above, note 113). It is no exaggeration to say that sometimes it has been downright rewritten. Cf. e.g. Mössler 1891, 723-726. See Bücheler 1862, ad loc.; Cugusi 1967, 87; and the detailed discussion in Pellegrino 1986, 144-147. Amat is accepted by Bücheler 1862, 7 (as already González de Salas ap. Burman 1743, II, 92). Marmorale 1948, 292 also preserves amat (he is influenced by his late dating of the Satyrica, and believes Petronius not to perceive syllable quantity any more). Barnes 1971, 13 expresses no preference between amat and ambit. Kissel 1978, 321 n. 4 believes all proposed corrections to be unsatisfactory. Ambit appears in the margin of the editio Tornaesiana and has been accepted in Bücheler’s editions later than his first one of 1862, and by most subsequent editors. This reading was at first rejected by Müller 1957, 504; but he changed his mind since his first edition (Müller 1961). Ambit is favored also by Tandoi 1968, 77; Flores 1982, 68; Pellegrino 1986, 144-147. TLL I 1850, 46-67 registers Plaut. Amph. 69 sive qui ambissit palmam histrionis; 74 quasi magistratum sibi alterive ambiverit; then our Petronian passage, followed by numerous late instances. Attested by the manuscripts E (Messanensis deperditus) and l (margin). Burman 1743, I, 27-28. See Brugnoli 1963, 257-259 (he refers to 3.4 nisi tamquam piscator… imposuerit hamis escam; but in the poem, contrary to what Brugnoli says on p. 259, the subject is not the teacher of rhetoric, as at 3.4, but the aspiring orator); also Cugusi 1967, 87 (see the convincing refutation of Tandoi 1968, 77). Cf. Nelson 1956, 203. Hamat is still defended by Schnur 1992, 170. Stubbe 1933, 156. Nelson 1956, 202.

The Education of the Orator (Petr. 5) zález de Salas184 (with a reference to Sen. brev. 18.4 frugalitatis exactae homines), and recently by Müller, in the apparatus.185 V. 4. Nec curet alto regiam trucem vultu] Some interpreters refer alto… vultu to regiam.186 In this case it would be a useless duplicate of trucem. These words must of course be taken with nec curet,187 as demonstrated beyond doubt by an exact Horatian parallel.188 V. 5: cliensque] The all but unanimously transmitted reading has been needlessly corrected to cliensve.189 The ideas connected by the conjunction are not mutually incompatible, but rather linked with each other.190 V. 7. in scaenam] Heinsius’ correction of the transmitted in scaena (scena) is necessary and accepted by nearly everybody.191 It is however rejected by Burman192 and Sommariva.193 V. 8: histrioniae addictus] The manuscripts are unanimous in transmitting addictus, whereas the preceding word varies greatly in the tradition: histrionis, histriones, histrion , histrionei, histrioni, histrioniae are attested.194 The last form195 offers a satisfactory meaning: “enslaved (or ‘addicted’) to theatrical 184 185 186

187

188 189 190 191

192 193 194 195

Ap. Burman 1743, II, 93. Müller-Ehlers 1983, 14; Müller 1995, 4. E.g. Heseltine 1913, 7: “nor must he care for the lofty frown of the tyrant’s palace”; Ernout 1923, 4: “qu’il dédaigne le palais insolent au front altier”; Walsh 1970, 85: “he should disregard the harsh palace with its naughty air”; Walsh 1996, 3: “let him not aspire to insolent palace with its lofty stare”. As done, for example, by Cesareo-Terzaghi 1950, 3: “a viso aperto sfidi la reggia truce”; Canali 1990, 23: “a fronte alta non si curi della torva reggia”; Aragosti 1995, 140: “trascuri a fronte alta l’arroganza di palazzo”; Reverdito 1995, 7: “sdegni con viso aperto”; Scarsi 1996, 7: “a fronte alta”. Hor. c. 4.9.42-43 reiecit alto dona nocentium / voltu (already quoted by Schönberger 1939, 509). By Bücheler 1862, 8, followed by many editors; also by Nelson 1956, 7; Flores 1982, 68); but not by Müller (besides his editions, cf. Müller 1957, 504). Rightly Pellegrino 1986, 147. As remarked by Pellegrino 1986, 150, in scaena governed by sedeat should mean in theatro – which is nowhere documented. In scaenam is obviously governed by redemptus (“paid in view of the theatrical performance”, to act as claque). Burman 1743, I, 30. Sommariva 1996, 59 n. 16. Cf. the critical apparatuses, especially those of Bücheler 1862 and Müller 1961; besides, Nelson 1956, 203. As remarked by Nelson 1956, 203, histrioniae also lurks behind histrionei, which can be explained as a wrong transcription of histrionie, that is histrioniae. 43

Chapter I art”.196 The term histrionia poses no problems;197 as far as the iunctura is concerned,198 the idea of being enslaved to the theater and its representatives is exactly paralleled in Petronius’ contemporary Seneca.199 An intervention in the text destroying the reference to a well-attested social phenomenon of the time seems to me to be a priori unlikely. Nevertheless, Ribbeck’s conjecture histrionis ad rictus is adopted by the majority of editors and scholars.200 One of the main arguments of those who favor the correction is the repetition of addictus, which already appears in line 6.201 This, however, can hardly carry conviction, in view of the numerous repetitions, which scholars have not failed to point out, and may be said to be a veritable stylistic mark of this poem,202 and, as we shall see, also of other poems in the Satyrica. This being so, the repetition of the same word should be taken as a proof that the transmitted text is correct, rather than the other way around.203 Another argument deserves more careful consideration. I am referring to Müller’s remark204 calling attention to the fact that in Petronius’ preserved verses a cretic, or a word ending in a cretic, is never elided, as it happens here, if we take histrioniae to be the correct reading. This cannot be lightly dismissed. The weight carried by this observation, however, appears to be more than bal196 197 198 199 200

201 202

203

204

44

It is hardly necessary to remind of the rhetoricians’ stern criticisms against fanatic addiction to theater and actors: cf. e.g. Tac. dial. 29.3; Quint. 1.12.18; 12.1.6. See the occurrences registered in TLL VI 2846, 77-79 (Plaut., Sen. Rhet., Macr.). We may refer to Fronto, p. 141, 16 v. d. H. histrioni studiosus, where the correction is all but certain. Sen. ep. 47.12 ostendam nobilissimos iuvenes mancipia pantomimorum. So all of Bücheler’s editions later than the first one (1862), and all of Müller’s. Histrionis ad rictus is also adopted by Heseltine 1913, 9; Giardina-Cuccioli Melloni 1995, 5; Schönberger 1939, 509; Schönberger 1940, 624; Walsh 1970, 85; Walsh 1996, 3; Kissel 1978, 321 n. 52. Ogrin 1983, 47 transcribes Ernout’s text (histrioniae addictus), accepts later histrionis ad rictus (p. 48 n. 8), and then translates “asservito all’arte teatrale” (p. 50). Histrioniae addictus is accepted by Burman 1743, I, 30; Bücheler 1862, 8; Cesareo-Terzaghi 1950, 3; Nelson 1956, 202; Ciaffi 19672, 74; Pellegrino 1975, 49; Pellegrino 1986, 70; Aragosti 1995, 140; Reverdito 1995, 7; Scarsi 1996, 7. Ernout 1923, 4 has histrioniae addictus in the text, but supposes histrionis ad rictus in the translation (“prodiguer des bravos aux grimaces d’un histrion”); the same in Yeh 2007, 512. Other, less successful conjectures are histrionis ad nutus (Anton) and histrionis ad dicta (Bücheler 1862, 8, in the apparatus, where ad picta is surely a misprint). See e.g. Sommariva 1996, 59 n. 16. Besides addictus at lines 6 and 8, cf. mentem / mentis (vv. 2 and 7); trucem / truci (vv. 4 and 19); det / det / dent (vv. 11, 17 and 19); pectore (vv. 12 and 22); plenus (vv. 13 and 22); sono / sonet (vv. 16 and 18); verba (vv. 20 and 22). Cf. already Collignon 1892, 230; then Sullivan 1968, 191. Or, at least, it should reinforce Burman’s judicious remark about this passage (Burman 1743, I, 30): “si vero toties medica manus esset adhibenda quoties eadem vox parvo intervallo recurrat apud auctores, excedet, credo, medicina modum”. Müller 1957, 504.

The Education of the Orator (Petr. 5) anced if we consider the implausibility of an intervention doing away with a perfectly fitting parallel with Seneca and an allusion to a well-documented and widespread social phenomenon of the time.205 Vv. 9-11. Sed sive armigerae… Sirenumque domus] In my opinion, the correct interpretation is found already in Burman’s commentary.206 He sees in these lines allusions to Athens, Taranto, and Naples, in this order, and has been followed by several scholars.207 Among the numerous different interpretations208 one that insistently keeps coming back is the identification of the Lacedaemonio tellus habitata colono with Sparta.209 The supporters of this way to understand these words must of course take the colonus to be not a “colonist” (Taranto was in fact Sparta’s colony), but rather a “farmer” in his own country.210 This semantic connotation is undoubtedly anything but foreign to the term colonus; but a look at the influence of this line’s formulation in subsequent literature suffices to assure beyond doubt that what the ancient reader saw in it was a hint at a colonist in a far-off land. We are led to this conclusion by two poetical texts whose conspicuous linguistic parallels suggest a derivation from our Petronian line – or, in case the first one were proved to be authentic, we might regard it as 205

206 207

208

209

210

At ant rate, even if we accept histrionis ad rictus, the reference cannot be to the theatrical mask, as asserted by Schönberger 1939, 509. Aside from the fact that in mime no masks were used, the plural ad rictus should refer to the shifting expressions of a real face. Rictus appears in the singular in Petr. 55.6.1. Burman 1743, I, 31. Bücheler 1862, 8; Collignon 1892, 234; later Müller 1957, 504; Kissel 1978, 322; Aragosti 1995, 141 n. 9; Reverdito 1995, 270 nn. 10-12. A detailed discussion in Pellegrino 1986, 151-154, whose conclusion, however, is a non liquet. Cf. also Pellegrino 1975, 221-222. Codoñer 1990, 67 believes the places alluded to to symbolize politics, jurisprudence, and poetry respectively. Some of which are manifestly groundless. Schönberger 1935, 1242-1243 places the first location too in southern Italy, referring to Sen. ep. 77.2; he is followed in this by Nelson 1956, 205, who takes the second location to be Sparta. The second location is identified with Rhodes by Paratore 1933, II, 22-23, who thinks the third location to be Sicily. Sicily is taken into account also by Nelson 1956, 206 (with an extension to all of southern Italy), Barnes 1971, 16 (who also sees a general allusion to the Greek islands), and Codoñer 1990, 67. Cf. e.g. Heseltine 1913, 9; Nelson 1956, 205; Walsh 1970, 86 (but Walsh 1996, 3 seems to have changed his mind: see following note); Barnes 1971, 16; Ogrin 1983, 55 (in the frame of a questionable interpretation, which identifies Troy as the first place mentioned: p. 54). Heseltine 1913, 9 renders colonus with “farmer”; Nelson 1965, 205 quotes Verg. ecl. 9.4 veteres migrate coloni. All the scholars mentioned in the preceding note understand colonus as “farmer”; but Walsh 1996, 3 translates (correctly in my opinion), “the Spartan immigrant”, thus changing his previous interpretation. The supporters of the interpretation of colonus as “farmer” could have quoted a text quite close to Petronius’: Sil. 7.160 tellus mentita colono. 45

Chapter I Petronius’ model. I am referring to the opening line of the second epigram attributed to Seneca,211 whose overall structure is parallel to, and whose second part is perfectly identical with, Petronius’ verse; and to a passage of Silius Italicus,212 where not only do we find a similar construction in the same position at the end of the verse, but at the beginning of the following line tellus is preceded by adridet, obviously echoing Petronius’ rident. In this second text the term colonus partially preserves the semantic connotation of “farmer” (cf. sulcata), but that of “colonist” is no doubt prevalent, in that Silius hints at the Trojan immigrants who settled in Latium; as for Seneca’s (or Ps. Seneca’s) text, the allusion to the Phocaean colonists who settled in Corsica is apparent. These parallels prove that in Petronius too the reference is to Sparta’s colony, i.e. Taranto. Burman already remarked that Sparta, whose fame was most of all associated with military power, would be out of place in a list of locations fit for studious leisure.213 By contrast, Athens, Taranto, and Naples214 perfectly become the idea manifestly expressed in the context: Athens was in fact the city of election for those who wished to pursue higher studies; and Taranto and Naples were a traditional buen retiro for Roman intellectuals.215 That Taranto, not Sparta, is meant is also confirmed by the clear echo of the lines describing that city in a famous Horatian ode;216 and elsewhere Horace refers the same epithet as Petronius – Lacedaemonius – to the same city.217 So Agamemnon opposes the educational voyage to Greece (Athens), favored by many Roman intellectuals, to the studious quiet in peaceful towns of southern Italy.218 This being so, the correction of the unanimously transmitted Sirenumque to Sirenumve is needless and misleading, though it has been ac211 212 213 214

215

216

217 218

46

[Sen.] epigr. 2.1 Corsica, Phocaico tellus habitata colono. Sil. 9.203-204 seu Laurens tibi, Sigeo sulcata colono / adridet tellus. Burman 1743, I, 31. Naples was the place where the Siren Parthenope was buried, and some reefs located not far away, near the Sorrento peninsula, were traditionally identified with the Sirens’ abode. Athens is denoted through an allusion to the Acropolis, which was sacred to Athena. Suffice it to remind that Taranto was the ideal place of retirement for Horace (c. 2.6.1020; epist. 1.7.45; cf. also Sen. tranq. 2.13); and Virgil spent many years at Naples. The two cities are mentioned together as places fit for retirement by Sen. ep. 68.5 ille Tarentum se abdidit, ille Neapoli inclusus est. The idea that locations far away from Rome’s hustle and bustle were favorable to study was traditional: see the texts collected by Schönberger 1935, 1248; Schönberger 1938, 221-222; Schönberger 1939, 510; Barnes 1971, 33 n. 4. Hor. c. 2.6.13-14 ille terrarum mihi praeter omnis / angulus ridet is picked up by rident in line 9 of our poem. The interpretation of Plaza 2000, 56-58 is barely worth mentioning. She sees in rident of the first hexameter an opposition to the last choliambic, where she reads histrionis ad rictus with no discussion at all. Hor. c. 3.5.56. So, rightly, Kissel 1978, 322.

The Education of the Orator (Petr. 5) cepted by some editors.219 The second and third location, in fact, amount to the two faces (connected through -que) of a single unit, which is opposed as a whole to the place mentioned first, sive and seu marking the only real alternative: Greece on the one hand, southern Italy on the other.220 Vv. 15-16. Romana manus circumfluat et modo Graio / exonerata sono mutet suffusa saporem] The express reference of our text to the student of rhetoric rather than to the accomplished orator that we have previously illustrated rules peremptorily out the interpretation of those who see in the Romana manus an audience made up of quick and kicking Romans surrounding the orators who speak in the forum.221 It is obvious that Agamemnon is thinking of the Roman literary models,222 to which the pupils must now turn, after studying the Greek ones. From the linguistic point of view this is confirmed by clear parallels in Quintilian, who employs both manus and the verb circumfluo223 in reference to literary models and resources.224 The words that follow have caused many problems to the interpreters, not a few of whom regard the text as irremediably corrupt from this point on.225 In my opinion this attitude is hardly justified; and rather than reporting and discussing 219

220

221

222

223 224 225

Since Bücheler 1862, 8. Later, all of Müller’s editions (though Müller 1957, 504 doubted the need to correct -que to -ve both here and at v. 5); Giardina-Cuccioli Melloni 1995, 5; Aragosti 1995, 140; also Schönberger 1935, 1243; Nelson 1956, 202. Sirenumque is preserved by Ernout 1923, 4; Pellegrino 1986, 154; Reverdito 1995, 7; Scarsi 1996, 7; also Kissel 1978, 322 n. 55. It is therefore needless (and misleading) to defend the tradition by reminding that -que may at times assume the meaning of -ve, as done by Müller 1957, 504 and Pellegrino 1986, 154. The enclitic must receive disjunctive meaning only if we assume that the two first locations are opposed to the third (this is in fact the reason adduced by Nelson 1956, 220 n. 20 to justify his correction of -que to -ve: he opposes Athens and Sparta to southern Italy), or at any rate that the second is opposed to the third. This interpretation is old: cf. already Palmerius ap. Burman 1743, I, 34: “Romani homines coronam faciant” (also correcting circumfluat to circumstruat). Later it was picked up by Schönberger 1935, 1243 (though he had the public listening to rhetorical declamations in mind); Fuchs 1938, 174 n. 34. More recently it is somehow surprising to find it in Müller-Ehlers 1983, 15: “nun sei er mitten im Treiben von Rom”. As already correctly understood by Burman 1743, I, 34: “debet intelligi de scriptoribus Latinis”; later, among others, Scheidweiler 1922, 1052; Nelson 1956, 209-211; Kissel 1978, 324 and n. 61; Pellegrino 1986, 159-160. Circumfluat continues the metaphor pervading the whole context: 4.3 irrigarentur; 5.12 bibat; 5.16 suffusa; 5.21 flumine largo. Cf. 118.3 ingenti flumine litterarum inundata. Quint. 10.1.76 sequitur oratorum ingens manus; 12.10.78 circumfluentibus undique eloquentiae copiis. These passages are already quoted by Collignon 1892, 231. Cf. above, note 113, for the remarks of Burman, Bücheler, and Müller. Similar comments reappear in several of the scholars who have studied our text. Among editors, Pellegrino 1986, 70, places the whole v. 16 between cruces. Müller in all his editions and Giardina-Cuccioli Melloni 1995, 5 do the same with the word exonerata. 47

Chapter I all the corrections and interpretations (the former often whimsical, the latter often far-fetched and questionable) of lines 15-20, I will try a linguistic illustration of the controversial points, which I hope will provide a satisfactory explanation of Petronius’ meaning, which in my opinion is far from being so obscure as is often assumed.226 The word that seems to have posed the greatest problems is exonerata.227 If the Romana manus is to be identified with the great Roman writers whom the orator-to-be takes as his models, some have found it difficult to understand why it should be said to be “free from Greek sound”.228 This approach, however, ignores an important signal offered by the text: the adverb modo.229 Who has “recently” got rid of the Graius sonus is in reality the pupil, who up to then had fed himself on Greek models, following the didactic method previously illustrated. The passage is a natural and easy one, inasmuch as the models in the two languages are regarded as parts of the curriculum proposed by the teacher of rhetoric: to a certain extent, then, it is the subject of study itself that, due to the change of language, gets rid of the old sound. It is a figure of speech no more daring than the shifting of the oratorical swing from the pupil to his written page in line 17. Both the passive moment of the pupil’s exposition to the models and the active stage which brings about the production of his own writings are illustrated in reference to the literary text: on the one hand the new (modo) models proposed by the teacher get rid of the Greek sound, and, by fusing and mixing with the results of the pupil’s previous preparation, change his stylistic “flavor”; on the other hand his pagina – the new writings he will produce – will now be able to launch into free swing. In both cases the real protagonist that gets rid of 226 227

228

229

48

For dent epulas (v. 19) I refer to what already remarked in § 4 (and notes 113, 119, 120). I will only mention, as a specimen, a few corrections: Scheidweiler 1922, 1053: vox onerata (accepted by Stubbe 1933, 156 and Schönberger 1935, 1243); Fuchs 1938, 173: vox ornata; Müller 1961, 6 (in the apparatus): vox operata (accepted by Courtney 1970, 65; cf. also Schnur 1992, 170); Harrison 2003, 128: vox imbuta. Some accept the reading exornata of the manuscript R: so Burman 1743, I, 34 (in the commentary), followed by Kissel 1978, 324 n. 62. On this reading see Pellegrino 1986, 160. Concerning the questionable interpretations we have hinted at, cf. e.g. Nelson 1956, 210 (who takes both exonerata and subducta to have an active meaning: he is rightly refuted by Müller 1957, 504; Pellegrino 1986, 160-161); Barnes 1971, 17 (exonerata is regarded as a neuter plural and as the object of mutet, together with suffusa, which governs saporem); Pellegrino 1986, 161 (exonerata as a neuter plural governed by mutet, and correction of saporem to sapore). I will quote one for all: Pellegrino1986, 160: “dal momento che ci si libera solo di qualcosa che si ha con sé, non si capisce che senso abbia dire questo degli autori latini in relazione al sonus Graecus”. Which as a consequence is not rarely omitted in translations (e.g. Ernout 1923, 5; Canali 1990, 23; Reverdito 1995, 7; Scarsi 1997, 9).

The Education of the Orator (Petr. 5) the old sound and tries himself in literary production is the young aspirant to eloquence. V. 17. det pagina cursum] From what we have just said and from the interpretation proposed in § 4230 it follows that in my opinion these words mean “let it dash, or rush” as in a race;231 but not a few scholars interpret “let it give the thrust”,232 obviously in the frame of the interpretation we have shown to be inacceptable: the one that only takes into account reading, rules out writing exercises on the part of the pupil, and sees in the pagina the works of the models, not the pupil’s own writing. But, even aside from this, a clear Virgilian parallel confirms the interpretation we have proposed.233 It is irrelevant to observe234 that in Virgil the dative equo must be supplied (Mezentius flings himself in the midst of the enemies on horseback). It is in fact probable, or at least possible, that Petronius may be using the Virgilian image to formulate his literary metaphor. But this in turn is only continuing mittat habens (v. 13); we can therefore be certain that, like there (v. 13), so here too (v. 17) the pupil is actively springing up of his own accord rather than receiving the thrust from outside. Nor can it be any other way, since, as we have seen, he is not merely reading, but also trying himself in written composition. V. 22. defundes] The correctness of this reading235 is guaranteed by a poetic parallel in Petronius himself.236

230 231

232

233 234 235 236

Above, note 110. This is also the interpretation of TLL V 1, 1686, 61-62, though secondarily it does not rule out the other interpretation we mention in the text. The interpretation proposed here is also shared by Heseltine 1913, 9; Ernout 1923, 5; Aragosti 1995, 140; Reverdito 1995, 7; Scarsi 1996, 9. So Stubbe 1933, 156 (he refers to Verg. georg.1.40 da facilem cursum); Friedrich 1935, 496; Müller-Ehlers 1983, 15 (“soll er sich... der Lektüre befleißen”); Pellegrino 1986, 162 (“sono queste letture [pagina] a conferire vim orandi… al giovane studente”). Verg. Aen. 10.870 cursum in medios rapidus dedit. This parallel is already pointed out by Schönberger 1929, 1200. As done by Pellegrino 1986, 162. Defundes appears in the margin of l (and apparently was in E). The most authoritative tradition has diffundes. Cf. Nelson 1956, 203. Petr. 121.102 defudit pectore voces. 49

Chapter II Justice for Sale (Petr. 14.2)* Quid faciunt leges, ubi sola pecunia regnat aut ubi paupertas vincere nulla potest? Ipsi qui Cynica traducunt tempora pera non numquam nummis vendere verba solent. Ergo iudicium nihil est nisi publica merces, atque eques in causa qui sedet, empta probat.

5

L(= ldmrtp)O(=BRP) Voss.(=Leidensis Vossianus Latinus F 111) 1-2, 5 5-6 Ioh. Sarisb. Pol. 5.15 1 faciunt ldmrtvlO Voss.: faciant tp Samb. Scaliger 2 nulla lmgdmrtRP : nuda ltvlB Voss. 3 pera Heinsius: cera 4 vendere vera solent Voss.: verba solent emere O 5 ergo LO: iam nunc

1. The first problem posed by the three elegiac couplets at Petr. 14.2 concerns their exact position. Nearly all editors place them after Ascyltos’ prose speech at the beginning of the chapter: the verse would then be its conclusion and, so to speak, the universalization of the idea which, in the prose, only concerns the particular situation narrated in the context. In fact, scholars generally take it for granted that this poem is uttered by Ascyltos, even though he nowhere else speaks in verse.1 In reality, however, this arrangement is the result of a transposition carried out by Bücheler in his 1862 edition of Petronius. Bücheler followed a suggestion made by K.G. Anton in his 1781 Leipzig edition, and was in turn followed by the overwhelming majority of editors and scholars.2 In the manuscript tradition * 1 2

A version of this chapter has appeared with the title La poesia in Petr. Sat. 14.2, “Prometheus” 24, 1998, 152-160. Cf. Stöcker 1969, 146-147; Aragosti 1979, 108; Slater 1990a, 163 (who expressly remarks that “nowhere else does Ascyltos speak in verse”). See the deatiled treatment in Sommariva 1997, 8-9.

Chapter II (L) the verse appears before Ascyltos’ speech, immediately after the end of what is now chapter 13. It must be added that the Florilegia tradition ( ) transmits three lines of our poem3 followed by the last words of Ascyltos’ speech. The reasons which prompted this transposition are clear, and seemingly justified: as our poem is an attack against the justices’ and the courts’ corruption, it appears to be in agreement with Ascyltos’ mistrust of justice more than with Encolpius’ intention to resort to the law in order to recover the lost tunic in whose hems a small treasure has been hidden.4 Nevertheless, the arguments put forward by Pellegrino, the one recent editor who keeps the traditional order, assuming of course that some parts of the text have been lost, cannot be easily dismissed.5 This being so, it is difficult, in my opinion, to award full confidence to the generally accepted transposition, even though it must be admitted that it is quite attractive. Stöcker,6 though ascribing the poem to Ascyltus, remarks that it might equally well be explained as a comment to the situation7 made by the narrator Encolpius or by the author himself. What is certain is that the theme of the omnipotence of money connects it with other poetic compositions on the same subject, which are presented as general reflections, even though sparked by particular narrative situations.8 Our poem, however, differs from those which develop the same theme in that it specifically targets the greed of the representatives of justice.9 Just like the

3

4 5

6 7 8

9

52

Lines 1-2 and 5 of the poem, followed by Ascyltos’ slightly adapted words tutius est parvo aere rem perditam recuperare quam in ambiguam litem descendere (14.1). Cf. Hamacher 1975, 126. In O and in Voss. (Leidensis Vossianus Latinus F 111) the poem is transmitted with no context. See lastly Sommariva 1997, 21-24, who concludes accepting the transposition. See Pellegrino 1975, 231-232; Pellegrino 1986, 28-29 (L offers the correct reading at line 4; if it had misplaced the poem because it has taken it from O, which offers no context, it would probably have picked up O’s wrong reading too; Ascyltos seems to be more concerned with his and Encolpius’ lack of social relations than with the corruption of justice). Cèbe 1966, 314 attributes the poem to Encolpius, seemingly rejecting the transposition (but it is only a slip, as he lends to Encolpius the rejection of a lawsuit); cf. also Jensson 2004, 9, 223. Slater 1990a, 163 attributes the poem to Ascyltos, while remarking that nowhere else does he speak in verse, adding that we cannot be sure “whether these lines are his own or another’s”. Barnes 1971, 276 states that 14.2 explains 14.1 and naturally leads to what follows, with no intimation that what is now 14.2 precedes 14.1 in the manuscript tradition. Stöcker 1969, 146 n. 1. The verse is regarded as such a comment by Sullivan 1968, 189. I am thinking in particular of Eumolpus’ poem at 83.10 (for whose “universal” import see ch. X) and of the verse at 137.9 (see ch. XXII). See Stöcker 1969, 146-151; also Fröhlke 1977, 70; Yeh 2007, 508. Debray 1919, 61-70 offers a detailed analysis of all the juridical implications of the scene in the forum (Petr. 12-15), but with no specific reference to our poem.

Justice for Sale (Petr. 14.2) more general topic of the omnipotence of money, this particular instance too had become commonplace in literature, no doubt after the pattern of what happened in reality.10 A great number of Latin writers pitilessly describe the corruption and greed of all the groups anyone involved in a lawsuit had to deal with – justices, lawyers, witnesses –, often with expressions close to Petronius’.11 Petronius insists on the greed of the first and the last of the groups now mentioned (justices and witnesses), but naturally supposes a general corruption of justice.12 An unexpected detail, which allows the author to widen the target of his polemic, can be spotted in his choice of the Cynic philosophers to represent the group of corrupt witnesses. This is not always duly stressed, partly owing to the existence of another reading – vendere vera13 –, which is accepted by several scholars,14 partly because sometimes verba is wrongly taken to refer to the philosophers’ teaching activity, in the meaning of “lessons” imparted for money.15 But the parallels in Martial, Seneca, and especially Ovid and Juvenal we just quoted in a footnote make it all but certain that what is meant here is bought and sold testimony16 – an interpretation which is the only one in accordance with the poem’s general context, where laws, lawsuits and courts feature prominently. 10 11

12 13

14

15 16

Cf. already Burman 1743, I, 67-68; more recently Stubbe 1933, 157; Kelly 1966, 31 ff. Justices: Varro Men. 499 Astbury; Cic. Cluent. 102; Att. 1.18.3. Lawyers: Mart. 5.17.6 sollicitis… vendere verba reis; Sen. Herc. fur. 173-175 clamosi rabiosa fori / iurgia vendens improbus iras / et verba locat; Tac. ann. 11.5.2 nec quicquam publicae mercis tam venale fuit quam advocatorum perfidia. Witnesses: Iuv. 14.218-219 falsus erit testis, vendet periuria summa / exigua. Justices and lawyers: Mart. 2.13. Justices, lawyers, witnesses: Ov. am. 1.10.37-40 non bene conducti vendunt periuria testes; / non bene selecti iudicis arca patet. / Turpe reos empta miseros defendere lingua; / quod faciat magnas turpe tribunal opes. The final stage is reached with Plin. ep. 2.14.4, from whom we gather that even the claque for young lawyers had to be bought with money. Witnesses are alluded to at line 4 (vendere verba), justices at line 6 (empta probat); but see below, note 53. Attested by the Leidensis Vossianus Latinus F 111. Sommariva 1990a, 28 convincingly argues in favor of verba and against vera. O offers verba solent emere, plausibly explained by Courtney 1991, 17 with the assumption that vendere disappeared because of the homoearchon with the following verba, with a subsequent integration (emere) based on empta (v. 6). E.g. by Bücheler 1862; Ernout 1923; Cesareo-Terzaghi 1950. Cf. also Heseltine 1913; Aragosti 1979, 108 n. 18; Aragosti 1995, 158; Canali 1990, 30. It must be remarked that some, though writing verba in the text, translate as if they had accepted vera: Reverdito 1995, 16-17; Scarsi 1996, 14-15. So Barnes 1971, 275-276; Slater 1990a, 163 n. 12; cf. also Ehlers ap. Müller-Ehlers 1983, 27: “nimmt... für den Vortrag bares Geld”. Lastly Sommariva 1997, 26. Cf. above, note 11. This interpretation is also accepted by Courtney 1991, 17; Walsh 1996, 10: “perjure themselves at times for sordid gains”. See also Sommariva 1990a, 27-28; Sommariva 1997, 15-16, 20, though she does not rule out that in vendere verba a reference to an unlawful activity as patronus may also be implied. 53

Chapter II The Cynic philosophers did not enjoy good reputation in Petronius’ time. The extremism of their nonconformist attitudes had caused from the very beginning the criticism implied in the very name of the movement; the Latin language, in which its connection with the word for “dog” is not immediately apparent, allows Martial to play with the concurrent pairing and opposition of Cynicus and canis,17 in an epigram which already implies that the outward attitude and appearance does not suffice to make a real Cynic. The theme is developed by Epictetus, who opposes the idealized figure of the real Cynic to the lowliness of the so-called Cynics of his own time, one of whose most characteristic marks was an insatiable beggarly greed.18 But the author who most pitilessly unmasks the hypocrisy of the contemporary Cynics’ attitudes is Lucian, who describes them as gluttonous, lascivious, deceitful, and – above all – greeedy, in total opposition to the principles they flaunted.19 This background helps us realize that Petronius’ Cynic, ready as he is to perjure himself in the courts of justice, is anything but an unlikely figure. It is not by chance, however, that he has chosen to describe in such a desecrating way precisely those philosophers who had always flaunted poverty as a proud symbol of freedom: for Petronius’ time we may mention the Cynic philosopher Demetrius, whom Seneca describes with the highest admiration.20 We have already remarked that this poem shares with others the theme of the omnipotence of money. It may even be regarded as complementary to one of these: the first poem uttered by Eumolpus at 83.10. In this poem all the traditionally recognized ways of life aim at the one and only goal of making money; , to the triumph each life choice is reduced to a universal and worship of wealth all men bow to, thus causing the poet, whose vocation prevents him from joining the common trend, to be marginalized and actually be regarded as an outcast.21 The two poems’ similarity and complementarity is emphasized even from the formal point of view, through the close correspondence of the verse at 14.2 with the immediate prose context of 83.10: in both places

17 18

19

20 21

54

Mart. 4.53.7-8 esse putas Cynicum deceptus imagine ficta: / non est hic Cynicus, Cosme; quid ergo? canis. See Epictetus’ whole diss. 3.22: mark in particular the reference to the false Cynic’s “big jaws” ( : § 50). Unlike Epictetus, who opposes this to the idealized real Cynic, Petronius makes no such distinction; for him all Cynics (actually, all philosophers) are “false”, in that they hypocritically seek what they ostensibly despise. See especially Lucian’s Fugitivi, where Philosophy itself laments being injured by them. More attacks can be found in the De morte Peregrini and in the Symposium, where the contemporary representatives of all philosophical schools are mocked. Cf. e.g. Paratore 1933, II, 43. Already Mössler 1891, 729 believed that Petronius meant to criticize Demetrius. Cf. also McMahon 1997. For this interpretation of 83.10 see ch. X.

Justice for Sale (Petr. 14.2) money is described as men’s one goal and one value.22 If even (ipsi) the representatives of Cynicism, which of all philosophies was the most outspokenly opposed to social rules and conventions, bow to greed , which in the meantime has become the rule and principle of society, it is clear that even what had once been the has been swallowed up by the , like all other traditional mentioned by Eumolpus at 83.10. The reader who remembers the verse at 14.2 understands why at 83.10 Eumolpus follows Horace in placing the ideal of a life devoted to poetry instead of the at the summit of the Priamel: the elegiac couplets of 14.2 make it clear that the philosopher’s life is subject to greed no less than the other , which means that it cannot be opposed to them any more; this leaves poetry as the one vocation free from greed and compromises with what contemporary society has accepted as its sole guiding principle and goal – in spite of all the material inconveniences it entails. That, unlike poetry, philosophy can lower itself to the point of becoming a hypocritical front aiming to attain the material advantages poetry does neither offer nor seek, is confirmed shortly after by Eumolpus himself, when he recounts how he got his way with the youth of Pergamum by putting on the mask of the philosopher.23 Only a life devoted to poetry is an end to itself; philosophy, like all other , is subservient to material goals. 2. V. 1. Quid faciunt leges] O, , the Leidensis Vossianus Latinus F 111, and the majority of the L witnesses read faciunt, whereas the reading faciant found in some Renaissance editions is much less authoritatively attested.24 Nevertheless, it is preferred by several scholars,25 though the parallels adduced for the construction quid faciant… ubi hardly carry conviction. The correct reading is no

22

23

24 25

Petr. 14.2.1 sola pecunia regnat ~ 84.2 solas extruere divitias curant. The affinity of the two poems was grasped by the compiler of the manuscript Leidensis Vossianus Latinus F 111, which records the two poems one after the other (fol. 38: cf. Peiper 1886, XXVII). Petr. 85.2 ut me mater praecipue tamquam unum ex philosophis intueretur. It is hardly surprising that a no-nonsense character like Trimalchio wishes this inscription to be engraved on his tomb: nec umquam philosophum audivit (71.12). For traditional support and reception of faciunt and faciant cf. Sommariva 1997, 9-10. Cf. e.g. Burman 1743, I; Ernout 1923; Cesareo-Terzaghi 1950; Aragosti 1979, 108 n. 18; Slater 1990a, 163; Canali 1990; Reverdito 1995; Scarsi 1996. Heseltine 1913 writes faciant in the text, but translates as he had accepted faciunt: “of what avail are laws…”. The parallels adduced are Val. Max. 5.3 ext. 2 quid aliae faciant urbes, ubi etiam illa…?; and Petronius himself (107.10): quid debent laesi facere, ubi rei ad poenam confugiant? It should not escape us, however, that in these instances facere denotes an action performed by a subject which is opposed to others that “do”, or “should do”, less. Cf. also Catull. 66.47 quid facient crines, cum ferro talia cedant? In our text, by contrast, it is clear that facere means “to be of avail” and ubi sola pecunia regnat does not describe the same action performed, or to be performed, to a lesser degree. 55

Chapter II doubt quid faciunt.26 The verb has the meaning of prodesse, borrowed from the technical language of medicine, where it applies to healing substances “being “effective” or “being of avail”;27 This employment is rather common in Silver Latin.28 Our line means, then: “to what avail are laws29 where only money is king?” V. 2. paupertas vincere nulla potest] Nuda given by the Leidensis Vossianus Latinus F 111 is surely wrong.30 We are surely faced with the colloquial use of nullus in lieu of non, which appears several times in comedy and in Cicero’s letters, and in isolated instances in Catullus and the Augustan poets.31 By contrast, it is rather rare in Silver Latin, and therefore it is all the more significant to encounter it in this poem. By this device Petronius probably meant to bestow a tone of directness and closeness to reality to the narrative situation described. V. 3. Ipsi qui Cynica traducunt tempora pera] Pera is the obvious correction, due to Heinsius, of cera, given by the majority of witnesses.32 We hardly need to call attention to the innumerable texts in which a satchel (pera) is attributed to the Cynic; it may be of interest that even wanderers like Encolpius and Ascyltus have one.33 The key to a correct understanding of this line lies rather in the alliterating words traducunt tempora. Many scholars34 take this expression to mean “they spend time”, or “their life”.35 But Gronovius already saw a much

26 27 28

29

30 31 32 33 34

35

56

Accepted by Müller 1961; Müller-Ehlers 1983; Müller 1995; Courtney 1991, 17; Giardina-Cuccioli Melloni 1995; Aragosti 1995; Walsh 1996 (“what point have laws”). Cf. TLL VI 1, 122, 12 ff. See the instances collected by Citroni 1975, 197-198, especially Quint. 10.7.4 quid porro multus stilus et adsidua lectio et longa studiorum aetas facit, si manet eadem quae fuit incipientibus difficultas? Rightly Courtney 1991, 17: “what good are laws?”. Sommariva 1997, 13 suggests quid faciunt to mean “what is the fate” of laws in a society ruled by greed? But the very parallel she shrewdly points out (Hor. c. 3.24.35-36 quid leges sine moribus / vanae proficiunt: p. 11) rather favors the interpretation we have accepted, as Sommariva herself realizes. Though it is accepted by Mössler 1891, 728, who places a question mark at the end of the first and the second line. Cf. Löfstedt 1933, II, 370-371; Leumann-Hofmann-Szantyr 1965, 205; Hofmann 1980, 208-209. Another reading is cena, adopted by Burman. Cf. Mössler 1891, 729. Petr. 11.4. Cf. Sommariva 1997, 26. Among others Ernout 1923; Cesareo-Terzaghi 1950; Ehlers ap. Müller-Ehlers 1983; Canali 1990; Slater 1990a, 163; Aragosti 1995, 159; Reverdito 1995, 17; Scarsi 1996, 15; Sommariva 1997, 14 and n. 22. It is possible that the reading cena has interfered here: the “satchel” would contain the Cynic’s poor food (“qui vilissimis cibariis se sustentat, more Cynicorum”: Burman 1743, I, 68).

Justice for Sale (Petr. 14.2) more pregnant meaning in the verb: the Cynics’ critical attitude to their times, symbolized precisely by the satchel in which they carried their humble belongings.36 This interpretation37 is no doubt the correct one. It is true that tempora may refer to the time of one’s life,38 but the meaning of “times”, “epoch” is very common for this word in the plural;39 and – most important – in Petronius traduco almost invariably means “to treat with contempt or derision”, and thus “defile” or “abuse”.40 In this context traducunt tempora opposes the Cynics’ hypocritical reproof of contemporary customs to their practical behavior: after ostensibly condemning their contemporaries’ greed, they are not above selling their testimony for money in the courts of justice. Criticism of the times is only a pretense on the part of philosophers, which Petronius means to expose. The meaning upheld by others (“spend their time”) is duller and more hackneyed, besides being linguistically less probable. V. 4. vendere verba] Cf. above (text to notes 13-16) for text and interpretation. Notice the marked alliterations in this line. V. 5 Ergo] This reading is found in LO ( has iam nunc).41 It is confirmed by a quotation of this line in Iulianus Toletanus.42 iudicium] The meaning of this term (“lawsuit”, “court action” – more generally “justice” –, as understood by most interpreters, or “sentence”, “verdict”43)

36 37 38 39 40

41 42 43

Cf. Burman 1743, I, 68: “infamant, culpant, allatrant et convitia faciunt suis temporibus, ostentatione frugalitatis”. Accepted by Heseltine 1913; Courtney 1991; Walsh 1996. This meaning is also found in a poetic fragment attributed to Petronius (fr. 33.12, 14 Müller); and see OLD s.v. tempus 4b. Cf. OLD s.v. tempus 4a, where this meaning, that we believe to be the correct one, is also assigned to tempora in our Petronian passage. See also s.v. tempus 12. Petr. 17.9 neve traducere velitis… secreta; 87.4; 126.6 scaenae ostentatione traductus; 132.10 traduceres annos primo florentes vigore; perhaps also 41.6 poemata domini sui acutissima voce traduxit (Trimalchio’s already awful poems are made even worse by the reciter’s shrill voice: “sviolinò poesie del suo padrone”: Canali 1996; in the same way some Virgilian verses are ruined too: 68.4-5). The only different meaning is found at 121.118 (Bellum civile), where the verb describes Charon’s bark ferrying the souls across the infernal waters. Sommariva 1997, 14 n. 22 believes the abstract object (tempora) and the instrumental ablative (Cynica… pera) to rule out the meaning of traduco we have accepted, but is refuted by the passages quoted at the beginning of this note. Cf. Hamacher 1975, 126. GL VIII, p. CCXXXIV 3 Keil. As interpreted by Debray 1919, 167: “judicium, mot qui prend ici évidemment, par la qualification de publica merces que l’auter lui donne, le sens de jugement”. 57

Chapter II depends on the sense we attribute to publica merces at the end of the line. See immediately below. publica merces] Most interpreters44 believe merces to be the equivalent of merx in this passage: a well-attested meaning.45 This interpretation is seemingly supported by a Tacitus passage we have already quoted; but it should not escape us that there merx, not merces is employed.46 It is not necessary, however, to assume that in Petronius merces does not keep its more common meaning of “reward”: the judge pronounces a verdict favoring those who bribed him as a reward for the money received.47 It is certainly attractive to assume that Petronius may have willingly expressed himself in an ambiguous fashion: justice would be a good that anyone can purchase, for the very reason that a favorable verdict may be accorded to anyone who can afford to pay for it (for this reason justice is a publica merces). We are not forced, however, to believe that merces has taken on the primary meaning of merx (“good”, “merchandise”). By contrast, the wellknown conservatism of juridical language rather leads us to believe that Petronius may be echoing a specific and already existing technical formula,48 even though, at the state of our knowledge, it does not appear until much later, in a decree by Diocletian and Maximianus.49 In this decree the term merces is applied precisely to the rulings pronounced by corrupt judges to favor those who bribed them, “as a reward” (in mercedem) for money received. Petronius’ ex-

44

45 46

47

48

49

58

E.g. Burman 1743, I, 69; Bücheler 1862, 15; Heseltine 1913; Ernout 1923; CesareoTerzaghi 1950; Ehlers ap. Müller-Ehlers 1983; Canali 1990; Courtney 1991, 17; Aragosti 1995; Reverdito 1995; Scarsi 1996; Sommariva 1997, 17-18. Cf. TLL VIII 2, 797, 82-798, 12. Tac. ann. 11.5.2 nec quicquam publicae mercis tam venale fuit quam advocatorum perfidia (cf. above, note 11). Sommariva 1997, 17-18 arbitrarily declares Tacitus’ and Petronius’ expressions to be identical: “l’espressione publica merces… è attestata… soltanto in un altro autore latino (Tac. ann. 11, 5, 1-3)…; Tacito usa la medesima espressione petroniana publica merx…”. She goes on to say that both are drawing on the same topos. The use of merces in the meaning of “praemium operarum illicitarum” is common: see TLL VIII 2, 795, 24-56, and cf. Aragosti 1979, 109 n. 22. Slater 1990a, 163 renders publica merces with “public salary”, but rather inconsistently translates iudicium with “trial” For the numerous juridical elements in the episode of the forum, besides Debray 1919, 61-70 (cf. above, note 9), see Aragosti 1979, 107-109 (n. 22 for the juridical expressions in our poem). Cod. Iust. 7.64.7 venales sententias, quae in mercedem a corruptis iudicibus proferuntur.

Justice for Sale (Petr. 14.2) pression should probably be interpreted in the light of this parallel; therefore, in our poem, iudicium is primarily the verdict pronounced by the judge.50 V. 6. eques] The equites were the class that provided the majority of the members of courts of law in the first century of the empire,51 even though the removal of senators from the functions of justices ever since the beginning of the principate is no longer considered to be so well established as Mommsen believed.52 empta probat] “Makes the purchases legal”, “does nothing but legalize the market”.53

50

51 52

53

For this meaning of iudicum see TLL VII 2, 607, 62-74. Among the many possible instances we might quote Serv. ad Aen. 6.743 ut si quis dicat ‘iudicium patimur’ et significet ea quae in iudicio continentur. See Jones 1972, 89. Mommsen 1887, 535; but see Sherwin-White 1966, 309 (ad Plin. ep. 4.29.2). As remarked by Aragosti 1979, 109 the verse at 14.2 emphasizes the corruption of justice portrayed in the prose context and extends the criticism to both civil and criminal procedure. So, respectively, Canali 1990 (“legalizza gli acquisti”) and Scarsi 1996 (“non fa che legalizzare il mercato”). See lastly Sommariva 1997, 19. Not only is the judge personally corrupt; he sanctions the general corruption of justice. For empta cf. Cic. Att. 1.18.3; Ov. am. 1.10.39. 59

Chapter III Two Views of Success (Petr. 15.9; 18.6)* 15.9

Nolo quod cupio statim tenere nec victoria mi placet parata

L(=lmrtp) 2 placet michi

18.6

Contemni turpe est, legem donare superbum: hoc amo, quod possum qua libet ire via. Nam sane et sapiens contemptus iurgia nectit, et qui non iugulat victor abire solet.

L(=lmrtp)O(=BRP) 3 nectit E(Messanensis deperditus)tmg: flectit ceteri, defendit Petrone: plectit i. nectit Samb. 4 at Scheidweiler, Tandoi iurgat Scheidweiler, Nisbet victus Nisbet dolet Heinsius(‘in vet. cod.’ Boschii), Burman

1. In the two Phalaecean hendecasyllables at 15.9 and in the two elegiac couplets at 18.6 the speaking character takes the cue from the particular situation being narrated to express, in general terms, his idea of success and the ways to make it most desirable. The first person singular appears in both texts, but in the first one it merely refers to the speaker’s preference, whereas in the second it is accompanied by sentences which, in the speaker’s intention, should be universally valid, and at any rate impart the short poem a gnomic tone. The two lines of 15.9 follow the forum episode, in which Encolpius and Ascyltos, who had come to sell a cloak they had apparently stolen, unexpectedly recover a tunic they had lost, as it can be evinced from some hints that have survived in the text as we have it. In it some gold coins had been concealed, which,

Chapter III as it seems,1 are still in place. Our heroes come back to their abode in high spirits, and at this point the two Phalaecean hendecasyllables occur. There is nothing in the tradition to make us think that something has been lost between these lines and the preceding prose, which should suggest the verse to be a comment attributable to the narrating voice – Encolpius –, as is often the case with the Satyrica’s poetic inserts, in which, like here, the first person singular is frequently found.2 Not a few scholars, in fact, ascribe these lines to Encolpius.3 In Italy, however, a strong influence has been exerted by the opinion of Paratore,4 who, without any solid argument, assigned them to Ascyltos. The scholars who follow him, however, behave inconsistently, inasmuch as they do not mark a lacuna before the verse. We have seen that no lacuna is signaled in the manuscript tradition; but it would be necessary, if we are to assume that there was a caption in the text ascribing these lines to Ascyltos.5 As the suspicions of those who think them to be out of place seem groundless,6 they must then be allocated with certainty to Encolpius. * 1

2 3

4 5

6

62

A version of this chapter will appear in “Prometheus” with the title Due concezioni del successo (Petr. Sat. 15.9, 18.6). Many scholars have seen in the parenthesis at 15.8 et recepto, ut putabamus, thesauro, and even in the one that appears in Ascyltos’ previous words (13.3 illa est tunicula adhuc, ut apparet, intactis aureis plena) the foreshadowing of a future disappointment caused by the gold coins’ either not being there or going to be immediately lost again. Cf. e.g. Ciaffi 1955, 37; Van Thiel 1971, 30 and n. 2, with the bibliography (though Van Thiel does not rule out that the coins might actually be there); Aragosti 1979, 113-114 and n. 30; Slater 1990a, 36 n. 25; Plaza 2000, 70; Courtney 2001, 64-65; Patimo 2001, 193. The forum episode, which leads to the recovery of the tunic, is examined in detail by Focardi 1986; Patimo 2001; Patimo 2002. E.g. Petr. 79.8; 132.8; 132.15; 136.6; 139.2. E.g. Ernout 1923, 12, who in the translation places a colon at the end of the words preceding the verse (the same is done by Bracht Branham-Kinney 1997, 12-13); Barnes 1971, 311 n. 29 (doubtfully); Slater 1990a, 183; Patimo 2001, 193 n. 73. Paratore 1933, II, 47: “i due versetti… vanno riferiti evidentemente (!) ad Ascilto che si vanta della sua presenza di spirito, col solito tono parodisticamente esagerato”. Cf. Pellegrino 1975, 234 (no mark of a lacuna before the two lines: p. 5); Pellegrino 1986, 195 (no mark of a lacuna in the text: p. 79); Scarsi 1996, 18 and n. 1. Walsh 1996, 11, 160, is at least more consistent; he expresses no opinion concerning the speaker of these lines, and proposes to transpose them before chapter 11, in view of the topos they develop, which has predominantly erotic implications (see below, notes 12-14); in the translation he poses a colon at the end of the words preceding the hendecasyllables. Already González de Salas, ap. Burman 1743, II, 101, 181 joined these lines with those at 93.2; but though the topos is the same, the application is different: the contempt for the victoria… parata is not easily reconciled with that for ordinary fare, even though the scope of the poem at 93.2 widens at the end (see ch.VI). Something similar is also found in Burman 1743, I, 77. Bücheler 1862, 17 remarks in the apparatus: “hos versus… non haesisse credo superioribus”, without any further explanation. For Walsh 1996, 11, 160 see preceding note.

Two Views of Success (Petr. 15.9; 18.6) In my opinion this is the only poem transmitted by Petronius’ direct tradition to be surely incomplete.7 Several witnesses of the L tradition, which, with the Florilegia ( ), is the only one transmitting these two lines, place a mark of lacuna after them.8 It is true that according to some scholars these marks deserve no credit and nothing has been lost between the second hendecasyllable and the prose that follows.9 But even if we assume that nothing has disappeared in the prose, our two lines give the reader the impression that he is faced with the mere beginning of a poem, in which the theme is stated, but is followed by no development. I am not the only one to believe that the development has been lost.10 Our two lines, in fact, state, but do not develop, a widespread topos of ancient literature: the scant value of anything that is acquired with no strife, patience, and toil. The image of the effortless victory probably originates in reference to sports races and competitions.11 In Hellenistic, and Latin, poetry it often receives an erotic slant, in order to depreciate lawful love or too easy amorous conquests.12 The fountainhead of this attitude can be recognized in a well-known epigram by Callimachus13 which strongly influenced the subsequent poetic tradi-

7 8

9

10 11

12 13

Neither Petr. 128.6 (cf. ch. XV) nor 136. 6 (cf. ch. XXI) are incomplete. Neither is any other poem transmitted with the prose of the Satyrica. More precisely, one or more asterisks appear in p (Pithou’s two editions), in t (the editio Tornaesiana), and in l (the cod. Leidensis Scaligeranus 61), which also has the words desunt multa in the margin. So Cosci in Aragosti-Cosci-Cotrozzi 1988, 45, followed by Aragosti in Aragosti-CosciCotrozzi 1988, 1, 14 n. 1; Aragosti 1995, 162-163 n. 34. By doing so, Aragosti contradicts what he had upheld in Aragosti 1979, 113-114 and n. 30, where, as we saw (above, note 1), he maintained that the parenthesis ut putabamus at 15.8 foreshadows a future disappointment, which, of course, had to be described in a lost portion of the text (Aragosti rightly refuses to see a hint at this in our very hendecasyllables, against Sage 1969, ad loc.). Plaza 2000, 70 makes it clear that, if ut putabamus refers to Encolpius’ and Ascyltos’ disappointment, then this denouement has undoubtedly been lost. Cf. eg. Courtney 1991, 18: “the rest of the poem has been cut out by the excerptor”. Cf. Otto 1890, 290, s.v. pulvis (2), who quotes Hor. epist. 48-50 (quis circum pagos et circum compita pugnax / magna coronari contemnat Olympia, cui spes / cui sit condicio dulcis sine pulvere palmae?) and the corresponding Greek turn . Two Christian writers employ the sports image in a similar way, to stress that there is no victory without fight: Ambros. ep. 10.73.29 nulla enim sine adversario corona victoriae; Hieron. ep. 14.10 nemo athleta sine sudoribus coronatur. The Greek expression was already metaphorically applied to war, as was later done at Rome, as we shall see: cf. e.g. Thuc. 4.73.2; Demosth. 18.200. In Xen. Agesil. 6.3 it still applies to sport, but a sport likened to war. As we have seen (above, note 5) Walsh 1996, 160 suggests transposing our lines before chapter 11, because he believes the topos they formulate to have erotic implications. Callim. epigr. 31 Pfeiffer (AP 12.102) / / ! " # / ‘ $ ’ % / & ! 63

Chapter III tion. Horace clearly alludes to it in a satire, and the theme reappears several times in the same erotic application in the Latin love poets, from Ovid down to Ausonius.14 But the erotic application was not the only one. In Petronius himself the topos reappears later in reference to food: simple and common fare is depreciated in comparison with what is rare and expensive. The theme is introduced in prose,15 and then developed in a poem16 which, in the last lines, expands the particular application to conclude with a more general statement: quicquid quaeritur optimum videtur.17 In our hendecasyllables what is depreciated and rejected is the victoria… parata. The metaphor might originate in the area of sport,18 but the parallel with a text of Livy leads me to believe that in our Petronian lines, where the topos appears to be universalized to define a veritable life choice, we should rather recognize a metaphor drawn from warfare.19 This text of Livy’s was first pointed out by Burman and later referred to by Stubbe,20 but neither remarked that, though the historian is talking about war,

14

15 16 17

18 19

20

64

" ! / ' ! ! ( % Hor. sat. 1.2.105-108 ‘leporem venator ut alta / in nive sectatur, positum sic tangere nolit’ / cantat, et apponit ‘meus est amor huic similis ; nam / transvolat in medio posita et fugientia captat’; Ov. am. 2.19.3 quod licet ingratum est, quod non licet acrius urit; ars 3.473-474 mora semper amantes / incitat; 579 quod datur e facili longum male nutrit amorem; Mart. 4.38.1 satiatur amor nisi gaudia torquent; Auson. epigr. 22.1 hanc amo, quae me odit, contra hanc, quae me amat, odi; 56.1 hanc volo, quae non vult; illam quae vult, ego nolo; etc. Petr. 93. 1 vile est quod licet, et animus errore lentus iniurias diligit. Petr. 93.2. Cf. ch. VI. Petr. 93.2.10 (at vv. 8-9 we also find the common application: amica vincit / uxorem). We have seen (above, note 6) that González de Salas went as far as joining our lines with those of 92.3. The two poems are paired by Habermehl 2006, 248 too. Cf. Ambros. ep. 10.73.29 corona victoriae, quoted above (note 11). A military application of this image originating in sports competitions is in Gell. 5.6.21 impulverea, ut dici solet, incruentaque victoria obvenit (cf. the Greek usage: above, note 11). A passage in Tryphiod. 134 ! ! , containing an expression similar to Petronius’ (cf. victoria… parata), also belongs in the field of war (the speaker is Ulysses, and the victory is the Greeks’ over the Trojans). In Petronius, of course, the reference to war is only metaphorical. Liv. 5.6.1 si medius fidius ad hoc bellum nihil pertineret, ad disciplinam certe militiae plurimum intererat insuescere militem non solum parata victoria frui, sed etiam si res lentior sit, 2 pati taedium et quamvis serae spei exitum exspectare et si non aestate perfectum bellum, hiemem opperiri nec sicut aestivas aves statim autumno tecta et recessum circumspicere. 3 Obsecro vos, venandi studium ac voluptas homines per nives ac pruinas in montes silvasque rapit: belli necessitatibus eam patientiam non adhibebimus quam vel lusus vel voluptas elicere solent?… 7 An mediocre discrimen opinionis secuturum ex hac re putatis, utrum tandem finitimi populum Romanum eum esse putent cuius si qua urbs primum illum brevissimi temporis sustinuerit impetum, nihil deinde timeat, 8 an hic sit terror nominis nostri ut exercitum Romanum non taedium longin-

Two Views of Success (Petr. 15.9; 18.6) there can be no doubt that the way he expresses himself supposes Callimachus’ epigram on love,21 and, more important, that Livy mistrusts any soldier who seeks only a parata victoria – an expression that can be easily paired with Petronius’ victoria… parata – and states that wars are not won by storm (impetu), but rather with persistence (perseverantia): an idea that is very close to what we find in Petronius’ first hendecasyllable. I believe, then, that our poem is employing a metaphor drawn from the military application of the topos in order to bestow universal validity to the idea that, in any field, what is promptly and easily attained does not amount to a real success.22 2. The two elegiac couplets at 18.6 are no doubt uttered by Quartilla.23 There is no explicit caption to mark the transition from her prose speech to the verse, but this is often the case. 24 The poem is actually intended to round off the speech she began in prose, as though aiming to illustrate and justify, through a conception of success expressed by resorting to gnomic sentences of supposedly universal import, her decision not to give up her claim to redress, if Encolpius and Ascyltos had not been willing to offer it of their own accord – which does not rule out a show of generosity, once her goal has been reached.25 There is nothing, then, to make us think that anything has been lost before the verse,26 though

21 22

23 24

25

26

quae oppugnationis, non vis hiemis ab urbe circumsessa semel amovere possit, nec finem ullum alium quam victoriam noverit, nec impetu potius bella quam perseverantia gerat? Cf. Burman 1743, I, 77; Stubbe 1933, 157. Both refer to hunting and employ very similar expressions (Liv. in montes ~ Callim. ; Liv. per nives ac pruinas ~ Callim. ! )% The idea keeps resurfacing down to the end of antiquity. Cf. e.g. Cassiod. var. 9.24.10 diutius quidem differendo pro te cunctorum vota lassavimus, ut et benivolentiam in te probaremus generalitatis et cunctis desiderabilior advenires. Habet enim hoc humana condicio, ut celerius adepta fastidio sint, dum omne pretiosum vilescit oblatum et contra dulcius accipitur, quod sub aliqua dilatione praestatur. Their attribution to Encolpius in Sullivan 1968, 194 and n. 1 is probably a slip. Cf. Barnes 1971, 284 n. 7. E.g. Petr. 14.2 (if this poem’s transposition is correct: see ch. II); 15.9; 79.8; 80.9; 82.5 (though this poem has no context); 83.10; 126.18; 127.9; 131.8; 132.15; 135.8; 137.9; 139.2 (no context). This is acknowledged by the numerous editors who place the quotation mark closing Quartilla’s speech at the end of the poem (the quotation mark opening her speech being in the prose at 18.5: ‘facio’ inquit ‘indutias vobiscum… Cf. e.g. Bücheler 1862, 20, and his later editions; Cesareo-Terzaghi 1950, 12; Müller 1961, 17, and his later editions. Also Aragosti 1995, 168; Reverdito 1995, 22; Scarsi 1996, 20-22. Cf. also Slater 1990a, 162: “she (Quartilla) has delivered her set speech to persuade the two and caps it with verse”. This is not ruled out by Cosci, in Aragosti-Cosci-Cotrozzi 1988, 63 and Sommariva 1990b, 78 n. 1. 65

Chapter III very probably something has been lost after it.27 As for the text, I believe the one found in most authoritative editions to be satisfactory.28 The only real problem appears at line 3, where nectit, at any rate, seems to be a better reading than flectit, given by the majority of the tradition.29 Some proposed corrections have been mentioned in the apparatus, but in my opinion should not be accepted.30 A detailed account of the textual problems may be found in Sommariva’s excellent essay, which is also the most in-depth inquiry into Quartilla’s poem to date.31 Sommariva starts by stressing its gnomic character, in that Petronius’ priestess is trying to draw universal rules of conduct from the specific situation in which she is acting32 (we should add that Quartilla is thus trying to define her idea of success); Sommariva then strives to explain the epigram by regarding it as an instance of a pattern she defines “of the happy medium”, in which two contrasting alternatives are discarded in favor of a third, intermediate one.33 Sommariva’s interpretation had been anticipated, to a certain extent, by some of the earliest modern interpreters of Petronius, such as González de Salas and Burman.34 According to Sommariva the two contrasting terms that are rejected in view of the happy medium are not primarily those pointed out in the first line, i.e. the acceptance of the others’ contempt (contemni) on the one hand and the 27

28 29

30

31 32

33 34

66

After the poem l and t mark a lacuna; the former also has desunt in the margin. According to Van Thiel 1971, 12, 33, the poem at 18.6 was originally followed by 19.2. He believes the whole passage 18.7-19.1 to have been wrongly placed by L after the poem. At line 2 qua libet must of course be written as two separate words: cf. Sommariva 1990b, 79. Cf. Ov. am. 2.9.45 iurgia nectat, already pointed out by Sambucus (cf. Burman 1743, I, 90); also Ov. am. 2.2.35; and see TLL VII 2, s.v. iurgium, 666, 4-6. Petrone 1993-1994, 98-99 defends flectit by quoting several texts, the most fitting being Petr. 58.1 flexit convicium in puerum. In our poem, however, there is no hint at anyone to whom the iurgia would be addressed. At line 4 et is corrected to at by Scheidweiler 1922, 1056 and Tandoi 1992b, 647; iurgia to iurgat by Scheidweiler 1922, 1056 and Nisbet 1962, 230; victor to victus by Nisbet 1962, 230; and solet to dolet by Heinsius (on the authority of a lost manuscript used by Bosch). Sommariva 1990b. Sommariva 1990b, 78. The poem’s gnomic character is also stressed by Paratore 1933, II, 56; Cosci, in Aragosti-Cosci-Cotrozzi 1988, 63; Petrone 1993-1994, 98; Yeh 2007, 506. Barnes 1971, 277 is not unjustfied in applying the terms “obscurity” and “sententiousness” to this poem. Yeh 2007, 506 opines that the poem is held together more by alliteration than by logical links. Burman 1743, I, 89 already remarked: “varie hoc epigramma exercuit eruditos”. This pattern had already been discovered in some of Martial’s epigrams by Citroni 1975, 191-192 (on Mart. 1.57). González de Salas, ap. Burman 1743, II, 107: “medium vero inter utrumque [i.e. between contemni and legem donare] viam tenere ipsam (= Quartillam) discupere”; Burman 1743, I, 89: “servandam, docet Quartilla, esse mediocritatem”. Sommariva 1990b, 79-82 does not miss these precedents.

Two Views of Success (Petr. 15.9; 18.6) behavior described by legem donare on the other. In her opinion, rather, the latter behavior is opposed to iurgia nectit (line 3). At this point we must emphasize that Sommariva is surely right in her interpretation of legem donare superbum as meaning: “to forgo law (i.e. to renounce legal action and the redress that can be obtained that way) is a sin of pride”.35 Burman thought that the sin of pride consisted in believing not to be in need of the law,36 and we shall soon see that, in my opinion, he has grasped an essential point of this poem. We can further agree with Sommariva when she rejects Burman’s interpretation according to which Quartilla approves of both the behaviors she implies (to react to contempt or give up legal action), with the provision of choosing one according to circumstances.37 I would specify, however, that what Quartilla rejects, besides legem donare, is not reaction to, but rather passive endurance of, other people’s contempt. When this is understood it seems incontrovertible to me that both attitudes sketched in the first line are rejected: both turpe and su35

36

37

For this meaning of donare (cf. “to condone”) see TLL V 1, s.v. dono, 2014, 57-74, where our Petronian passage is listed too. Petronius’ expression was correctly understood by Burman 1743, I, 89: “remittere actionem lege datam”; Mössler 1891, 729; Scheidweiler 1922, 1055-1056; Ehlers ap. Müller-Ehlers 1983, 35: “verzeihn” (sic); Courtney 1991, 18: “to give up one’s legal right”; Aragosti 1995, 169: “praticare l’indulgenza”; Reverdito 1995, 23: “perdonare”; Petrone 1993-1994, 97: “concedere indulgenza”. Nevertheless, not a few scholars prefer to understand “impose (one’s own) law”: González de Salas ap. Burman 1743, II, 107 (who pairs legem donare with 25.3 legem… accipere); Heseltine 1913, 27; Ernout 1923, 15; Cesareo-Terzaghi 1950, 12; Ciaffi 19672, 94; Canali 1990, 39; Cicu 1992a, 19; Scarsi 1996, 23; Bracht BranhamKinney 1997, 16; Yeh 2007, 506. As for superbum, many scholars, including some who correctly understand legem donare, regard it erroneously as signifying “beautiful”, “splendid”, “wonderful”: e.g. Heseltine 1913, 27; Scheidweiler 1922, 1056; Ernout 1923, 15; Cesareo-Terzaghi 1950, 12; Ehlers ap. Müller-Ehlers 1983, 35; Cosci, in Aragosti-Cosci-Cotrozzi 1988, 63-64 (as Cosci takes legem donare as meaning a constituta lite dimittere, the meaning of superbum must be positive); Courtney 1991, 18; Cicu 1992a, 19 (“Quartilla… ama imporre la sua legge”); Petrone 1993-1994, 97 (“con la grandezza dei vincitori”); Tandoi 1992b, 646 (“Quartilla rinunzia volentieri a valersi della legge”); Aragosti 1995, 169; Reverdito 1995, 23; Scarsi 1996, 23; Bracht Branham-Kinney 1997, 16; Yeh 2007, 506. In my opinion, though, the meaning of superbum we have accepted in the text is incontrovertible: cf. OLD s.v. 2a (which includes our Petronian passage), and the parallel at Petr. 30.10-11 superbus ille sustulit vultum et… inquit… ‘dono vobis eum’. Burman 1743, I, 89: “neque etiam ita inciviliter nos efferre debere, ut inferioribus ita superbe et arroganter gratiam faciamus delictorum, quasi lege nobis non opus esset, quae unicuique civi remedium praestat contra iniuriam”. Cf. Mössler 1891, 729: “qui poenam lege irrogatam remittit, legem despicatui habere videatur”. Burman 1743, I, 89: according to him, line 2 implies that Quartilla “dicit se probare viam utramque, et pro re nata posse vel intendere actionem et litem, vel remittere”. Sommariva 1990b, 81-82 rejects this interpretation, inasmuch as it cannot be reconciled with the pattern of the happy medium. 67

Chapter III perbum are words that, in different ways, emphasize rejection. But can we really be sure that these two behaviors are unconditionally opposite and incompatible? We shall soon answer this question. At line 2 Sommariva, following Tandoi,38 detaches quod possum from the preceding amo, and takes these words to be a restrictive parenthesis meaning “as far as I can”. Amo would then govern the infinitive ire: “I take pleasure in treading (in the sense of ‘I am used to tread’) the path I like, as far as I can”. Quartilla would describe her usual behavior,39 the one she follows as long as she can, i.e. clemency, which is the happy medium between iurgia nectere and legem donare.40 Sommariva believes the preceding prose to confirm her interpretation: Quartilla first takes pity on the culprits’ young age,41 then envisages their punishment.42 In the verse she would state – according to Sommariva – that her clemency is not abolished, but certainly limited, by the respect she owes to the law. Se likes to tread the path she prefers – clemency – but only insofar as it is permitted by the law. The pivot around which Sommariva’s exegetical approach revolves – it is clearly to be seen – is her interpretation of quod possum as a restrictive parenthesis independent from the preceding amo. It is certainly true that the Thesaurus linguae Latinae only records three instances, including this Petronian passage, of amo governing a quod clause;43 but what both Tandoi and Sommariva have failed to remark is that in Petronius quod is preceded by the demonstrative hoc, which is surely governed by amo and can hardly be detached from the following quod, which it surely anticipates.44 The fact that there are only a few instances of amo governing a quod clause is irrelevant in view of the fact that hoc is the object of amo and anticipates the following quod.45 Quod possum, then, is governed by amo, and governs, in turn, the infinitive ire. These two words, far from being idle, as some believe,46 are meant to stress Quartilla’s unhindered 38 39 40 41 42 43 44

45

46

68

Tandoi 1992b, 646. This is a posthumous writing that Sommariva was able to see before publication; cf. Sommariva 1990b, 82-83. Cf. Gr. . For amo followed by the infinitive in this meaning cf. TLL I, s.v. amo, 1956, 35-59. Sommariva 1990b, 83-84. Petr. 17.6. Petr. 18.5. TLL I, s.v. amo, 1956, 32-34; the two other instances are Lucan. 8.78 ipsum quod sum victus ama; Sidon. epist. 2.10.1 amo in te quod litteras amas. Notice that also in Lucan’s text quoted in the preceding note quod is preceded by the demonstrative ipsum. For quod used epexegetically with a demonstrative see KühnerStegmann 1971, 270-271; for hoc… quod Leumann-Hofmann-Szantyr 1965, 574. Cf. OLD s.v. quod 2a. Among the various examples that could be quoted, I select one which appears to be close to our Petronian passage: Hor. sat. 1.6.62-63 magnum hoc ego duco / quod placui tibi. Quod possum is not translated, e.g., by Reverdito 1995, 23 and Yeh 2007, 506.

Two Views of Success (Petr. 15.9; 18.6) freedom,47 not the limits the law imposes to her leaning toward clemency, as Sommariva would have it. Does Quartilla, by declaring her total and unhindered freedom, mean to oppose herself to someone who, in her opinion, does not enjoy equal freedom? The answer may be sought in the following line, where – at first glance unexpectedly – the figure of the wise man, the mythical sapiens, suddenly appears. According to Sommariva48 the sapientia here advocated by Quartilla is meant as a caricature of the Stoic sapiens’ wisdom, as described by Seneca. She believes our verse to be particularly close to a passage in the De clementia.49 Although the parallels she points out do not seem to me to be particularly close,50 I do believe that, in Petronius’ intention, the sapiens Quartilla refers to should be taken to be the Stoic wise man. Sommariva adds51 that Quartilla pinpoints and disputes an aspect of Seneca’s thinking that seems questionable to her: the imperturbability of the sapiens, who is superior to, and unattainable by, all injuries and offenses, as he is described in the philosopher’s works, and especially in the De constantia sapientis. In reality, though, Quartilla’s verse does not say that it is the wise man’s duty to protect his dignity and react to other people’s contemptus, as Sommariva would have it. It only remarks that he does so in real life (contemptus iurgia nectit).52 If there is a polemic against Seneca and the Stoics, therefore, it rather concerns the unlikelihood of the sapiens’ unresponsiveness to offense and immunity from passion, as shown by the fact that in Quartilla’s description 47

48 49

50

51 52

The meaning of quod possum is effectively conveyed by Ernout’s translation (Ernout 1923, 15: “j’aime pouvoir, comme je le peux, suivre librement ma voie”. We shall presently see that Quartilla probably means to oppose her freedom to the dogmatism of philosophic utopias. Sommariva 1990b, 86-88. Sen. clem. 2.7.1 ei ignoscitur qui puniri debuit; sapiens autem nihil facit quod non debet, nihil praetermittit quod debet: itaque poenam quam exigere debet non donat, 2 sed illud quod ex venia consequi vis honestiore via tribuet. Parcet enim sapiens, consulet et corriget, idem faciet quod si ignosceret, nec ignoscet, quoniam qui ignoscit fatetur aliquid se quod fieri debuit omisisse. Aliquem verbis tantum admonebit, poena non adficiet, aetatem eius emendabilem intuens… 3 Haec omnia non veniae sed clementiae opera sunt…Ignoscere est quem iudices puniendum non punire, venia debitae poenae remissio est. It is true that in Seneca we read poenam… donat (close to Petronius’ legem donare) and that he hints at the culprits’ young age, as Petronius does at 17.6; but unlike Seneca’s sapiens Quartilla does not mean to correct the culprits, but openly declares to seek personal redress (18.5). The culprits’ youth is seen as an extenuating circumstance, not one favorable to their reformation. The so-called sapiens’ widely advertised “pedagogic” intention is implicitely denounced as hypocrisy. See below. Sommariva 1990b, 87-88. Petrone 1993-1994, 98 correctly remarks: “facendo citare a Quartilla… il caso… del saggio che, contravvenendo ai suoi principi, anch’egli infine reagisce, Petronio scherza con la logora chiacchiera dello stoicismo a lui contemporaneo”. 69

Chapter III he too – or, like the priestess seems naughtily to imply, whoever proclaims himself to be a sapiens – does react to contemptus.53 The two alternatives posed in line 1 can now be seen in a new light. We have seen that for Quartilla both are to be rejected;54 but enduring other people’s contemptus and giving up legal action are not necessarily incompatible alternatives: anyone who – like the imaginary Stoic sapiens – fancies his own superiority to passions and offences, and therefore believes to be exempted from resorting to the law’s protection, would be, in Roman traditional conception, both contemptus, inasmuch as he does not care to protect his dignity according to the generally accepted social behavior, and superbus, inasmuch as he places himself outside and above the law; he would thus be doubly regarded as a social outcast. The ethics advocated by Stoicism, Quartilla seems to say, is incompatible with Roman customs – actually with reality itself, in that line 3 proves that a sapiens like the one idealized by the Stoics is not found in real life.55 Petronius’ (and his character’s) skepticism toward those who professed more or less utopistic philosophical doctrines is hardly surprising; and that he has contemporary figures in mind (Seneca? Demetrius?) is also suggested by a poem not far removed from this one, in which the Cynics of the time are portrayed as incapable to behave according to their lofty ideals in real life.56 The Stoics’ ideal sapiens, then, does not exist. But Quartilla proceeds beyond: whereas the so-called sapiens reacts aggressively and in a way far removed from his much-vaunted imperturbability,57 Quartilla, who is free from such hypocrisies and unambiguously declares that her aim is not to effect the moral reform of her offenders, but merely to obtain redress, is capable, if she chooses to, to be moderate; but, most inportant, she will follow a path she has

53

54 55

56 57

70

Petronius’ attitude to sapientia may be illustrated by his use of this term. At 91.9 the pectus sapientia plenus is Giton’s, who has chosen the stronger Ascyltos over his former lover Encolpius. Cf. also 94.1, where Giton’s sapientia is paired with his beauty. And that, as a consequence, superbum cannot mean “beautiful”, “splendid”, wonderful”, or the like: cf. above, note 35. Notice the formulation of this line: nam sane et sapiens contemptus iurgia nectit. Quartilla is saying this: the two attitudes sketched at line 1 are both to be rejected; this is proved (nam) by the fact that in reality (sane) even (et) the [so-called] sapiens (who in theory should not care about contempus and should forgo legal action) resorts to litigation when he is affected by contemptus”. Clearly Stoic imperturbability is regarded as unreal, and just as clearly Quartilla has had the figure of the Stoic sapiens in mind from the very beginning. Petr. 14.2.3-4 ipsi qui Cynica traducunt tempora pera / non numquam nummis vendere verba solent. Cf. ch. II. As implied by iurgia nectit. Iurgium seems to be a technical term to refer to litigation, but its connotatation appears to be more negative, as compared with other technical terms: cf. Sommariva 1990b, 85.

Two Views of Success (Petr. 15.9; 18.6) freely chosen, not abstract philosophical dogmas removed from real life;58 as she seems to suggest, she is the real sapiens, inasmuch as she is intellectually free. It goes without saying that, if the two alternatives of line 1 are not opposite and incompatible, it is no more necessary to seek a structure founded on the principle of the happy medium in the epigram.59 The whole poem hinges on a pervasive juridical metaphor:60 besides legem donare and iurgia, the last line too contains an image drawn from the same domain. The verb iugulo, as a matter of fact, often appears in a particular metaphorical meaning expressing the utter defeat of the opponent in court.61 It must, however, be strongly stressed, that this is only a metaphor. Before the verse Quartilla says that, if Encolpius and Ascyltos had not acceded of their own accord to her request for redress, she would have resorted not to law, but to violence.62 Only the pseudo-sapiens, perhaps, is so naïve as to become hoarse in courts of justice in order to protect a dignity, which, if he were consistent with his principles, he should not care for. He too, however, like Quartilla, is not willing to endure contemptus nor to legem donare. Quartilla does not give up the right which, in theory, the law should guarantee; but she knows that she can hardly trust the law; here too do we encounter ideas close to those expressed in the poem we referred to above.63 Though she is seemingly heedful of law (donare legem is superbum), in practice Quartilla recognizes only two ways to obtain personal redress, both totally foreign too law: personal agreement, or, in case this cannot be reached, resort to violence. Only when redress has been obtained in the first way can she afford a beau geste: non iugulare, i.e. not press her ad58 59

60 61

62

63

Quartilla’ clemency is anything but “nutrita di cultura filosofica”, as contended by Sommariva 1990b, 88; quite the other way: she opposes herself to philosophic utopias. We may continue to regard, with Sommariva, the last line (et qui non iugulat victor abire solet) as an intermediate term between legem donare and iurgia nectit, but without identifying, as she does, the formulation of the happy medium in line 2; this line, as we have seen, expresses Quartilla’s unhindered liberty (including intellectual freedom, as we have just discovered), not her leaning toward clemency, which is curbed by the respect due to law, as Sommariva maintains. It begins in the preceding prose: Petr. 18.5 a constituta lite dimitto. Cf. the numerous instances in TLL VII 2, s.v. iugulo, 636, 58-77. So, correctly, also Courtney 1991, 18. By contrast, Sommariva 1990b, 86 believes the image of line 4 to come directly from gladiatorial jargon, not philtered through the language of the forum (cf. also TLL VII 2, 636, 51-52). I do not believe that (unless we place a comma after victor) the idea expressed in line 4 may be paired with Sen. epigr. 6.3-4 (desere confossum! Victori vulnus iniquo / mortiferum impressit mortua saepe manus), as done by Burman 1743, I, 88-89. Petr. 18.5 ‘facio’ inquit ‘indutias vobiscum et a constituta lite dimitto. Quod si non adnuissetis de hac medicina quam peto, iam parata erat in crastinum turba quae et iniuriam meam vindicaret et dignitatem’. Petr. 14.2. Cf. ch. II. 71

Chapter III vantage to the hilt, refrain from dealing the finishing blow.64 The scholars who see in Quartilla’s prose speech and in the verse that rounds it off a clear illustration of the priestess’ domineering character are anything but wrong.65 The last line expresses Quartilla’s idea of success; it does not signify any renunciation on her part. She would never be content with a mere moral victory, if it did not entail the full attainment of her goals; but,66 once they have been attained, victory can be considered complete if it allows the beau geste of not inflicting the finishing blow on the fallen enemy.

64

65

66

72

In fact, she gives up, metaphorically, the constituta lis (18.5). Quartilla does not say that real victory is obtained by yielding (as Sommariva 1990b, 86 and n. 41 would have her say), but that real victory is obtained by not being uselessy cruel. It is not possible, therefore, to regard non iugulare as similar to legem donare, as maintained by Sommariva 1990b, 86 and Petrone 1993-1994, 98. The association might be justified if superbum is understood as “beautiful”, “splendid”, “wonderful”, as done by Petrone; much less if, like Sommariva does, it is interpreted (correctly in my opinion) as “sin of pride”. It is easier to agree with Sommariva when she takes non iugulare as an intermediate term between legem donare and iurgia nectere. Quartilla is never willing to legem donare, and describes non iugulare as an act due to the conqueror’s generosity (only after full victory has been gained). Cf. e.g. Cicu 1992, 19; Yeh 2007, 505-506, who lays emphasis on the poem’s metrical structure: the first couplet (whose hexameter is holospondaic) aims to make the general sentence’s authoritativeness subservient to Quartilla’s personal interests. The et opening line 4 has a clearly adversative meaning (“the so-called sapiens resorts to litigation, but only those who give up needless cruelty always – solet – obtain real victory”). So Heseltine 1913, 27 (“while”); Ernout 1923, 15 (“mais”); Pellegrino 1986, 207; Sommariva 1990b, 84 and n. 34; Tandoi 1992b, 646-647 (who proposes to correct to sed or at, in order to avoid two non correlative et in lines 3 and 4); Reverdito 1995, 23 (“ma”). For adversative et, also in the age of Petronius, cf. Setaioli 2000, 64-65; for Petronius Petersmann 1977, 241 (quoting Petr. 83.4; 63.6). According to Burman 1743, I, 90 and Mössler 1891, 729 the two et are correlative (denoting two parallel, not opposed, situations: as the sapiens only punishes through his word – so Mössler – so Quartilla too limits herself to reproaches and refrains from extreme measures).

Chapter IV Petronius’ Sotadeans (Petr. 23.3; 132.8)*

23.2 intrat cinaedus, homo omnium insulsissimus et plane illa domo dignus, qui ut infractis manibus congemuit, eiusmodi carmina effudit: 3

huc huc convenite nunc, spatalocinaedi, pede tendite, cursum addite, convolate planta, femore facili, clune agili et manu procaces, molles, veteres, Deliaci manu recisi.

23.3 L(=lmrtp) 1 add. L. Müller nunc lrp2, om. mtp1 3 add. Fraenkel et om. l, del. Fraenkel

132.8

Ter corripui terribilem manu bipennem, ter languidior coliculi repente thyrso ferrum timui, quod trepido male dabat usum. Nec iam poteram, quod modo conficere libebat; namque illa metu frigidior rigente bruma confugerat in viscera mille operta rugis. Ita non potui supplicio caput aperire, sed furciferae mortifero timore lusus ad verba, magis quae poterant nocere, fugi.

L(=lrtp)O(=BRP) 1 terribilem Opt Samb.: terribili lr 3 male dabat Orp: dabat male lt

5

Chapter IV 9 Erectus igitur in cubitum hac fere oratione contumacem vexavi… 11 Haec ut iratus effudi, illa solo fixos oculos aversa tenebat, nec magis incepto vultum sermone movetur quam lentae salices lassove papavera collo. 1. The two poems at 23.3 and 132.8 are the only surviving verse in the Satyrica to be written in Sotadeans. Besides this conspicuous metrical similarity, the two poetical inserts share the lewdness in content, as many scholars have not failed to remark.1 We too, in the course of our analysis, shall point out several aspects closely linking the two poems; it should not be overlooked, however, that, though both belong in the tradition initiated by the poet Sotades of Maroneia, they are also remarkably different. This point was best illustrated by Maurizio Bettini,2 who calls attention to the fact that the poem at 23.3, which is recited by a cinaedus, is a clear and explicit specimen of cinaedic poetry, whereas the composition at 132.8 is a case of epic parody like that attested for the poet of Maroneia.3 The two poems, then, reproduce the two types of poetry Sotades devoted himself to. This diversity is reinforced by the different function allotted to the two poems in the prose context. The first one is recited by a character and aims to represent him as a member of the group this type of poetry particularly suited, the

* 1

2 3

74

A version of this chapter has appeared with the title Le due poesie in sotadei di Petronio (Sat. 23.3; 132.8), “Cuadernos de Filología Clásica. Estudios Latinos” 23, 1, 2003, 89106. Barnes 1971, 290-291 stresses the affinities between these poems and the Priapea. Aragosti, in Aragosti-Cosci-Cotrozzi 1988, 105 only emphasizes the lewd subject of both. Others, like Stephens-Winkler 1995, 367 remind that the Sotadean meter was often associated with the galli, the castrated priests of Cybele (they refer to Demetr. de eloc. 189 and Syrian. in Hermog. I, p. 47, 9-11 Rabe, who in reality only emphasize the Sotadeans’ effeminacy and the Ionic meters’ softness respectively). The galli, in turn, were often paired with the cinaedi: cf. Parsons 1971, 62 (he quotes Mart. 2.86 and Athen. 14, 620E, but only in the former text is cinaedic poetry associated with the gallus Attis); Parsons 1974, 39; Bettini 1982, 91-92; Merkelbach 1973, 92 (who stresses the metric affinities of the Sotadean and the galliambus); Stephens-Winkler 1995, 359. The theme of castration, at any rate, links the two Petronian poems (at 132.8 Encolpius attempts to castrate himself; 23.3 is recited by a cinaedus addressing his castrated equals: Deliaci manu recisi). Bettini 1982, 86-87. For Sotades’ “rewriting” of the Ilias with the change of hexameters to Sotadeans, see Sotad. frs. 4a-c Powell and cf. Aly 1927, 1208; Bettini 1982, 67-69.

Petronius’ Sotadeans (Petr. 23.3; 132.8) cinaedi;4 the second, by contrast, is one of the few poetic inserts that actually moves the narration forward:5 it is therefore “narrative” poetry, just like serious epic is.6 2. This is not the place to investigate the prosimetric nature of Petronius’ novel and its relation to Greek and Latin precedents. It is obviously understandable why, after the discovery of the fragment of the so-called “Iolaus-novel”, the traditionally assumed kinship linking the Satyrica to Menippean satire is no longer accepted, or at least drastically reduced, by several scholars – even though an outright separation between “Menippean” and “novelistic” prosimetric narrative is hardly justified by the scantness of the material at our disposal.7 Nevertheless the pairing of Petronius’ first poem in Sotadeans (and its context) with the socalled Iaolaus is natural, and – hopefully – profitable. The Iolaus fragment has been preserved by a papyrus going back to the II century A.D.8 It is a prose narrative, but it contains a speech in Sotadeans addressed to a certain Iolaus and uttered by a character who apparently has undergone some type of initiation and has become a gallus. As we have seen,9 this term and cinaedus have similar connotations and are even interchangeable, to a certain extent. At any rate, the word appears several times in the text.10 The narrative was no doubt lewd, as we may infer with certainty, though the papyrus is mangled. This general character, added to the specific detail of the recitation of Sotadeans by a character somehow resembling Petronius’ cinaedus, led Parsons to pair the Iolaus with Petronius’ Satyrica even before the

4 5

6

7

8 9 10

That cinaedic poetry suited the cinaedi well was of course well-known even before the discovery of the Iolaus fragment; cf. e.g. Barnes 1971, 288. It is the only one to do so according to Beck 1973, 56; one of the few for Bartonková 1976, 82. That 132.8 contributes to the story’s development is also recognized by Barnes 1971, 289, 292. It is no chance that the same applies to Petronius’ Virgilian “cento” too (132.11). This “cento” appears soon after the Sotadeans at 132.8; it is part of a sustained epic parody along with the Sotadeans and the short prose intermezzo. Such a separation is upheld by Astbury 1977, who rules out any relationship between Petronius and Menippean satire. This seems too radical: cf. Barchiesi 1986, 232 n. 27 (= Barchiesi 1999, 136 n. 27); Stramaglia 1992, 137-138; Conte 1996, 166-167, whose whole fifth chapter offers a detailed analysis of the question (also with extensive literature), while at any rate reducing the traditionally assumed influence of Menippean satire. Yeh 2007, 20, 37-40 sticks to the traditional connection with the latter. POxy 3010. The Sotadeans appear at lines 14-33. Each verse occupies the space of two prose columns. See the text in Stephens-Winkler 1995, 368-370. Above, note 1. Three times: POxy 3010.14, 26, 27 (p. 368 Stephens-Winkler) 75

Chapter IV publication of the papyrus.11 It is indeed quite possible that the new fragment may belong to a strand of Greek narrative fiction different from the previously known novels of idealized love, and close to the ambiance and atmosphere of the Satyrica; this type of fiction might even have been already established before Petronius’ times, though the Iolaus papyrus must be dated in the II century A.D. Parsons, however, is commendably cautious when he warns that, in view of the shortness of the fragment, the Iolaus’ prosimetric form may be mere appearance, although a quotation from Euripides’ Orestes at the end of the preserved text once more introduces meter into the prose narrative.12 Quotations from Greek poets also appear in the novels preserved in their entirety, and the speech in verse uttered by the gallus may be due to the wish to characterize the speaker, and thus to a narrative contingency, rather than signaling a constant charachteristic of this work of fiction, which is totally lost except for this short fragment.13 However, another novelistic fragment featuring a metric section has since been discovered;14 the story it told apparently takes place in a fanciful Orient, and both characters and situations seem to suggest an ambiance removed from that of the novels of idealized love. It is therefore reasonable to suppose, with Stephens and Winkler,15 that already in Petronius’ time there existed not merely a novelistic tradition that featured roguish characters and lewd situations but also a prosimetric type of narrative. If this is so, then the similarities between Petronius and the Iolaus16 should be in all probability explained through the common generic affiliation rather than by supposing a direct contact. Nevertheless what strikes the reader’s attention is not so much the general similarity as the circumstance that in the Greek fragment the Sotadeans are recited by a character who has undergone an initiation17 and has also learned to dress like a 11

12

13 14 15 16 17

76

Parsons 1971, 52, 65-66. Shortly after Parsons the Iolaus’ affinity with Petronius (especially with the Quartilla and Oenothea episodes) was emphasized by Merkelbach 1973, 89. POxy 3010.39-43, p. 370 Stephens-Winkler quotes Eur. Orest. 1155-1157, with a slight change. This citation is not evidenced by writing one verse in each line, as was the case with the speech in Sotadeans addressed to Iolaus. The quotation, however, is preceded by a blank space. Parsons 1971, 65. We may add that verse appears occasionally in Apuleius’ Metamorphoses too (4.33; 9.8), though nobody would regard this novel as prosimetric. The so-called Tinouphis (published in 1981). See lastly Stephens-Winkler 1995, 400408. Stephens-Winkler 1995, 365-366. Cf., besides, Aragosti 1995, 35 n. 38; 45 n. 46; Conte 1996, 166; Connors 1998, 16-18; Petersmann 2000, 379. Similarities and differences of the two texts have been carefully surveyed ever since Parsons 1971, 65. Stephens-Winkler 1995, 360-361 emphasize the fragment’s unreliability as far as the ritual connected with Cybele’s cult is concerned. At any rate, this is of no great impor-

Petronius’ Sotadeans (Petr. 23.3; 132.8) woman;18 we should remember that in Petronius the poem is recited by an effeminate cinaedus taking part in a ceremony connected with mysteries. 3. The most obvious difference between the Sotadeans recited by the gallus in the Greek fragment and those uttered by Petronius’ cinaedus lies in the fact that the former amount to a speech addressed to Iolaus, with details possibly referring to previous adventures and also to Iolaus’ lewd intentions.19 By contrast, Petronius’ poem at 23.3 is a sort of solo aria recited by a character who suddenly occupies the scene and for a moment captures all attention.20 Bettini rightly remarks21 that the details preceding these Sotadens function like directions in an opera score, so that the reader is enabled to get a clear idea of how this solo is performed; and Aragosti22 emphasizes the neurotic atmosphere of this scene, with the cinaedus addressing an imaginary multitude of his equals,23 as if wishing to expand his own frenzy to infinity. The main problem posed by this poem is the conspicuous metric irregularity of the first line, where two short syllables are wanted in the first foot, and of line 3, where something is missing after femore. Bettini accepts the transmitted text. He thinks that Petronius, whose negative attitude toward the cinaedus is evident, from the esthetic point of view,24 wished to emphasize his dullness (homo omnium insulsissimus) even through his metrical faults in the type of poetry that should best suit him.25 This is anything but impossible; even in the Iolaus vulgarisms and incorrectnesses concentrate in the verse, perhaps to characterize the speaker;26 nor are metrical problems missing.27 Most editors, however, insert L. Müller’s after huc huc in the first line28 and accept Fraenkel’s femore

18 19 20 21 22 23 24 25

26 27 28

tance for us; what matters is the narrative fiction, regardless of its correspondence to reality. POxy 3010.7-8 [ (p. 368 Stephens-Winkler). POxy 3010.30 (p. 370 Stephens-Winkler). E.R. Dodds, ap. Parsons 1974, 35 n. 1 believes the plot to be similar to Terentius’ Eunuchus’. Cf. Panayotakis 1995, 46. Bettini 1982, 87-89. Aragosti, in Aragosti-Cosci-Cotrozzi 1988, 9, 104. Cf. Plaut. Stich. 772 omnis voco cinaedos. Cf. e.g. Aragosti, in Aragosti-Cosci-Cotrozzi 1988, 9. Bettini 1982, 89-90 (he refers to Trimalchio’s verse at 55.3 and 34.10); cf. Aragosti, in Aragosti-Cosci-Cotrozzi 1988, 109. Cugusi 1967, 89-90 already thought of an intentional mistake in order to further debase the character, but mistakenly believed the first line to be regular; he doubtfully suggested the prosody f more at line 3. The attempt made by Yeh 2007, 104, 527 to prove that these verse is not improvised but prepared beforehand (not by the cinaedus) appears to be groundless. Cf. Parsons 1971, 63. Cf. Bettini 1982, 90 n. 104. Cito is accepted, in the editions following his first one, by Bücheler too, though elsewhere he had previously made a different proposal. 77

Chapter IV in the third.29 These additions,30 it cannot be denied, convey the impression of idle fillers;31 but it is not easy to admit such an incorrectness in Petronius, although some parallels might be invoked.32 The frenzied opening of the verse reminds of certain patterns found in lofty poetry, especially tragedy,33 but it should not escape us that only another poem, as far as we know, opens with the same gemination; and this poem is part of the collection of the Priapea,34 which means that its atmosphere is closely related to the one we find in Petronius’ poem. Priapus plays an important role in the whole of Petronius novel, which shall be discussed at length in one of the following chapters. For the moment it must be stressed that, as we shall see, the links with the type of poetry represented by the Priapea is one of the common traits connecting 23.3 and the other poem in Sotadeans at 132.8.35 The wildly excited opening appeal reaches the highest pitch with the long compound in the vocative case which fills the whole final part of the first line. It denotes the addressees of the cinaedus’ poem and has been chosen to capture the listener’s (and the reader’s) attention through its size, position, and peculiarity: spatalocinaedi. Scholars have long since called attention to a similar compound employed by Lucilius: moechocinaedi.36 This may have suggested to Petronius 29 30

31

32

33 34 35

36

78

Bücheler prefers femore in his editions following the first one. In this (1862) he acepted the traditional reading of both line 1 and line 3. Nobody, except Barnes 1971, 306 n. 2, has accepted another of Fraenkel’s suggestions: convertite in lieu of convenite at line 1 (cf. Müller 1961, 211). The expunction of et at line 3 is needless. As rightly remarked by Bettini 1982, 89, in reference to the inserted in the trikolon in line 3. He also adduces several instances of huc huc immediately followed by an imperative (cf. v. 1). One would be at 93.2.4 et pictis anas renovata pinnis (a line close to a Sapphic hendecasyllable in a series of Phalaeceans). Cugusi 1967, 86 accepted the transmitted text here too, but in this case the irregularity could hardly be explained assuming the speaker’s ignorance, since these lines are uttered by the professional poet Eumolpus. The text must probably be corrected. Cf. ch. VI. For Trimalchio’s doggerels cf. above, note 25, and ch. V. Sen. Med. 980; Phaedr. 1247; also Catull. 64.195. In an erotic context, but far removed from Petronius’ paroxysm, Catull. 61.8-9. Priap. 14.1-2 huc huc, quisquis es, in dei salacis / deverti grave ne puta sacellum. Barnes 1971, 307 n. 5 saw the parallel, but thought it might be coincidental. Another detail connecting 23.3 with the Priapea and related poetry is the imperative convenite, following the geminated huc huc. Courtney 1991, 19 adduces an epigraphic poem in which the same imperative is twice used as an appeal to honor Priapus: CE 1504.13-17 convenite simul quot est[is om]nes / … / convenite quot estis atque [be]llo / voce dicite blandula [Pria]po. For the parallel of v. 3 (clune agili) with Priap. 83.23 see below, note 53. A Priapus, by the way, is also attested among the titles attributed to Sotades (cf. fr. 5 Powell). Lucil. 1058 Marx = 994 Krenkel = 30.89 Charpin. The reference already appears in Burman 1743, I, 107; then in Collignon 1892, 235 n. 1. Subsequently e.g. in Pellegrino

Petronius’ Sotadeans (Petr. 23.3; 132.8) the flamboyant term with which he has chosen to hint at the cinaedi’s lewdness.37 The poem’s first line is the kernel from which the whole composition develops. The first line’s opening appeal is in fact repeated and nuanced in the second, while the third and fourth lines amount to a characterization of the spatalocinaedi. The three lines developing the first one are all molded in the pattern of the trikolon, but whereas in the first two (i.e. lines 2 and 3) each of the three members of the trikolon is made up by two words which are placed in parallel or at most chiastic order (cursum addite, convolate planta), in the last line the cola are arranged in a pattern of progressive increase: the third member consists of three words, the first and the second of one.38 Incidentally, this rules out the interpretation of the first two words of the last line given by Courtney,39 who places no comma betwwen molles and veters and understands “effeminates of long standing”. Mollis can of course be employed as a substantive in the meaning advocated by Courtney,40 but here an adjective is needed; otherwise the triadic pattern, which is clearly the pivot around which the whole structure of the poem revolves, would be destroyed. As for veteres, most scholars take it to mean “expert”, “experienced” (in homosexual practice),41 just like Courtney, but not to be an epithet specifying molles understood as substantive, as Courtney does. This meaning of vetus is well attested in Latin. It is surely the meaning intended by the speaker, but we cannot rule out that in the the narrator’s (and behind him the author’s) intention the primary meaning of the word may still lurk behind this specific one – a possible reference to the disparaging topos of the eunuchs’ and the cinaedi’s early aging.42 The reciting cinaedus himself is marked by wrinkles, which are still visible through a garish makeup,43 and Cleopatra’s eunuchs described by Horace are wrinkled too.44 If, as I believe, this nuance cannot possibly reflect the

37

38 39 40 41 42 43 44

1986, 209; Courtney 1991, 19. For the formation and the meaning of this Greek-style compound cf. Marbach 1931, 137; Page 1978, 46; Cavalca 2001, 158-159. This is the meaning of the Greek word , which is also confirmed by the Latin glossarists: CGL V, p. 151, 38 spatale, deliciae. Aragosti’s suggestion, in AragostiCosci-Cotrozzi 1988, 106, who refers to Varr. Men. 275 and to boys supposedly castrated by the Spatale mentioned in Varro (but the transmitted text is spatule – spatula Mercerus – eviravit omnes Veneri vaga pueros), appears to be groundless, though Iunius’ correction Spatale is now accepted by Astbury too. Cf. Aragosti, in Aragosti-Cosci-Cotrozzi 1988, 108. Courtney 1991, 19. Cf. TLL VIII 1379, 38-52; also González de Salas ap. Burman 1743, II, 119; Pellegrino 1986, 209. See e.g. Schönberger 1935, 1244. Cf. Burman 1743, I, 108; Aragosti, in Aragosti-Cosci-Cotrozzi 1988, 108. Both refer to Trimalchio’s puer vetulus (28.4) and to Mart. 1.41.13 vetuli… cinaedi. Petr. 23.5 inter rugas malarum tantum erat cretae… Hor. epod. 9.13-14 spadonibus… rugosis. 79

Chapter IV speaker’s intention but must be seen as the narrator’s (or the author’s) implied judgment, the opposition assumed by Walsh45 to exist between veteres and the preceding molles (in the sense of “young”) must also be ruled out. This opposition would in the first place mess up the last line’s trikolon, whose members do not identify distinct groups, but are referred to all the spatalocinaedi regarded as one group. 46 Secondly, mollis in the sense of “effeminate” is too well-attested in Latin47 for doubts to arise concerning the meaning of the word in a context like ours. Resuming the analysis of the poem, which has been interrupted in order to anticipate that of the first two words of the last line, it is only fair to state that not much can be added to Aragosti’s exhaustive commentary to be found in the work often referred to.48 As far as the peculiarities of language, style, and content of the parallel trikola in lines 2-3 are concerned, we can unhesitantly refer to his work.49 I limit myself to a few additional remarks. The second member of the trikolon in line 3, clune agili, is rightly understood by Aragosti50 and numerous other interpreters, including some early ones whose explanations were collected by Burman,51 as a deliberately ambiguous allusion to ability in dance, which of course includes a hint at sexual availability.52 In my opinion, though a double entendre cannot in any way be ruled out, the sexual meaning is undoubtedly in the foreground. This will appear all but certain if we compare a Priapeum which, as we shall see, also has much in common with Petronius’ other poem in Sotadeans (132.8). In this Priapeum we read an

45 46

47 48 49

50 51 52

80

Walsh 1996, 17: “Come, tender youths, and you in later life, / and lads castrated by the Delian’s knife”. Notice that Walsh must force Petronius’ text, by inserting two “and” which have no counterpart in the Latin, in order to create three distinct groups. Not to mention the fact that, after the polar opposition he poses between the first two groups, it is difficult to understand their relations to the third group, the one made up of the “lads castrated by the Delian’s knife”. It is quite clear that all spatalocinaedi are molles, veteres, and recisi. Cf. TLL VIII 1379, 26-52; Sandy 1971, 51 and n. 2. Aragosti, in Aragosti-Cosci-Cotrozzi 1988, 105-109. For cursum addite we may add to Verg. georg. 1.512 addunt in spatia the reference to the texts collected in TLL I 587, 22-25, and especially to Plaut. trin. 1010 adde gradum, adpropera. Cf. also Liv. 3.27.6; 10.20.10; 26.9.5. Aragosti, in Aragosti-Cosci-Cotrozzi 1988, 107. Burman 1743, I, 106-107. Besides CGL V, p. 654, 7-8 and Iuv. 6.464, Aragosti quotes Petronius himself (21.2 extortis nos clunibus cecidit; 24.4 clunibus eum basiisque distrivit: in the same homoerotic context of Quartilla’s orgy). In Burman reference is made to Arnob. 2.42, Iuv. 11.162-164, and Iuv. 2.20-21, where a sexual meaning is unequivocal.

Petronius’ Sotadeans (Petr. 23.3; 132.8) expression close to Petronius’ clune agili which has no reference at all to dancing.53 One last remark on the poem’s closing words. The interpretations collected by Burman54 already offer the two alternatives developed in later exegesis, based on the interpretation of Deliaci either as a vocative plural or as a genitive singular.55 In my opinion the first alternative must be ruled out, since it would leave manu without a referent,56 but in either case a metaphorical reference to the castration57 of roosters to produce capons may be surmised. Poultry raising was in fact a widespread activity on Delos.58 A non-metaforical reference to capons can indeed be found in another poem of Petronius’.59 Be that as it may, I believe Deliaci (suerely a genitive) to refer not to a Delian poultry raiser but to the Delian par excellence: Apollo, whose medical and healing functions are well known; or, alternatively, to his son, Asclepius, the god of medicine.60 In my opinion Courtney is right,61 when he asserts that in this way the speaker means to ennoble, and bestow prestige upon, his group.62 4. As we have already pointed out, Petronius’ other poem in Sotadeans (132.8) relates to the other strand of Sotades’ production: irreverent epic parody entailing both metrical alteration and a desecrating travesty of content.63 In fact, the connection of this poem with epic is unanimously recognized by all scholars. All, however, limit themselves to emphasizing the Virgilian paro53 54 55

56 57 58 59 60 61

62 63

Priap. 83.21-23 tener puer / … imminente qui toro / iuvante verset arte mobilem natem. Burman 1743, I, 108-109. Among recent interpreters Deliaci is understood as vocative by Pellegrino 1986, 209 and Bruneau 1985, 547-549. Aragosti, in Aragosti-Cosci-Cotrozzi 1988, 109 does not take either position. This is rightly pointed out by Courtney 1991, 20, who adds that the meaning “artificially”, often attested for manu, can hardly apply to castration. For recisus in the sense of “castrated” Ov. am. 2.3.3 qui primus pueris genitalia membra recidit is usually referred to. I forgo adducing once more the numerous ancient testimonies, which are collected and discussed by Burman 1743, I, 108-109; Barnes 1971, 306 n. 4; Bruneau 1985, 548. Petr. 55.6.4 gallus spado – a text which is not pointed out by those who see a reference to capons at 23.3.4. Asclepius actually performs castration in Stat. silv. 3.4.68-70 (quoted by Courtney 1991, 19). Courtney 1991, 19. The Deliacus is taken to be Apollo also by Walsh 1996, 161-162; Aragosti, in Aragosti-Cosci-Cotrozzi 1988, 109. Aragosti 1995, 180 n. 53 believes that both the poultry raiser and Asclepius may be alluded to. This is the case in Statius’ passage (cf. above, note 60). As Yeh 2007, 529 rightly remarks, this poem is metrically more regular than the one recited by the cinaedus. Possibly, this is so because Encolpius is not so ignorant as the cinaedus is. 81

Chapter IV dies.64 This was partially to be expected, and Bettini has a point when he states that for Petronius the obvious epic referent was Virgil.65 But though this is certainly true for the textual and stylistic parallels – strongly encouraged by the fact that the two writers shared the same language –, we must not underestimate the circumstance that, as far as content is concerned, the irreverent degradation which lowers an epic situation to a desecrating level of sexual innuendo supposes a reference that has apparently escaped all scholars. I am thinking of the famous episode of Achilles’ and Agamemnon’s quarrel at the beginning of the Iliad. There the former reaches for the sword in order to kill the latter, but then, after being reproached by Athena, only assails him with words.66 Not long before Encolpius has confessed to Giton the mishap that has affected the part of his body which once made him feel like an Achilles.67 Now, in spite of his impotence, he continues to play the role of the hero, while the member that has betrayed him – from which he means to exact bloody vengeance – takes on the role of Agamemnon. As in Homer, though, the cruel intention is not carried out and all boils down to verbal abuse.68 Besides the obviously desecrating transfer of the quarrel from Achilles and Agamemnon to Encolpius and his recalcitrant member,69 another factor of degradation may be seen in the cowardice of both parties,70 as opposed to the epic heroes’ pride and boldness. Besides the desecration of Homeric epic, it is also possible that our poem may be meant as an irreverent caricature of mythological tales connected with the gods of tradition (not merely of poetry), as is the case in other Petronian po-

64

65

66 67 68

69

70

82

The one exception is Fedeli 1988, 77, who, however, limits himself to the very general statement that we may observe “la compresenza in Petronio di due codici epici, quello omerico e quello virgiliano”. Bettini 1982, 86: “se per Sotade il codice epico di riferimento (o meglio di mascheramento?) era costituito da Omero, per Petronio esso non poteva che esser sostituito dal suo equivalente latino, Virgilio”. Hom. Il. 1.189-224. Petr. 129.1 funerata est illa pars corporis, qua quondam Achilles eram. We shall return later to the image of impotence as “death”. Petr. 132.8.9 ad verba, magis quae poterant nocere, fugi ~ Hom. Il. 1.223-224 / ! " # $ Analogous desecrations are certainly not wanting in the Satyrica. A similar scene of castration not actually carried out (in this case not intended in the first place) can be found at 108.10-11, with an explicit reference to another lofty literary genre: tragedy (108.11 audacius tamen ille tragoediam implebat…). The protagonist is Giton, who gives the cue to tragic parody elsewhere too (cf. 80.3). Encolpius: languidior repente thyrso / ferrum timui, quod trepido male dabat usum (vv. 2-3); his member: namque illa metu frigidior rigente bruma / confugerat in viscera mille operta rugis (vv. 5-6) and mortifero timore (v. 8). See also below, note 96.

Petronius’ Sotadeans (Petr. 23.3; 132.8) ems too.71 According to a legend reported by Clement of Alexandria, Zeus performed a very realistic pretense of castration, as atonement for raping Demeter. Here, as elsewhere in Petronius, we witness a complete reversal: Encolpius really attempts to castrate himself, but fails; and he does so for the opposite reason, i.e. for not having made love to Circe. A parodic reversal of this type is not without counterparts: according to a burlesque version of Attis’ legend – once more significantly transmitted by Christian authors72 – the young man rejected Cybele’s advances, and for this reason the goddess impelled him to castrate himself, not because of his unfaithfulness, as recounted in the traditional version.73 It must be stressed, then, that not only does Encolpius compare himself to Jupiter in the poems he recites at the moment of amorous elation;74 he continues to do so in this poem too, when he describes the punishment he tries to inflict on himself, to make amends for a sexual act that he, unlike the god, has not been able to perform. In many details the epic parody of our poem depends of course on the Virgilian model. The opening image (corripui terribilem manu bipennem) has exact counterparts in the Aeneid,75 and the pattern ter… ter is also clearly epic (both Homeric and Virgilian). The majority of interpreters76 refer to a couple of verses that appear unchanged in two distinct passages of the Aeneid, when Aeneas vainly tries to embrace the ghosts of his wife and of his father;77 in reality, however, there is a much more fitting parallel: the description of Dido’s attempt to rise from her bed, which she vainly repeats for three times after striking herself

71

72 73 74

75

76

77

E.g. in the poem at 126.8, which describes a gray-haired and impotent Jupiter. Cf. ch. XIII. We may ask ourselves if it is by mere chance that in both cases Petronius is close to the criticism of traditional gods we find in Clement of Alexandria. For Petr. 126.18 cf. Clem. Alex. protr. 2.37.1 ff. For our verse at 132.8 cf. Clem. Alex. protr. 2.15.2. Tertull. ad nat. 1.10.45; apol. 15.2; and especially Min. Fel. Oct. 22.4. Cf. e.g. Ov. fast. 4.223-244. Cf. Petr. 126.18 and 127.9 (chs. XIII and XIV respectively). As we shall see, at 127.9 Encolpius places himself on a par with Zeus in the famous love scene of Iliad 14, but the sexual roles are reversed. As we have seen, there is a similar reversal in 132.8 too. Verg. Aen. 2.479 correpta dura bipenni is referred to by almost all scholars; by contrast, Verg. Aen. 11.651 validam dextra rapit indefessa bipennem is adduced, if I am not mistaken, only by Bettini 1982, 85-86 and Fedeli 1989, 213. E.g. Barnes 1971, 295; Zeitlin 1971a, 71; Blickmann 1988, 8; Slater 1990a, 185; Adamietz 1995, 323; Connors 1998 31; Rimell 2002, 154-155. Courtney 1991, 34 also quotes Verg. Aen. 10.685-686 (ter conatus utramque viam, ter maxima Iuno / continuit), the only parallel quoted by Merkelbach 1973, 84 n. 12. Verg. Aen. 2.792-793 = 6.700-701 ter conatus ibi collo dare bracchia circum; / ter frustra comprensa manus effugit imago. Cf., of course, Hom. Od. 11.206-208. 83

Chapter IV with Aeneas’ sword.78 As it seems, only two scholars have called attention to this passage.79 Both rightly point out that the allusion to Virgil’s queen anticipates in this poem the cento closely following at 132.11, whose first two lines are taken from Virgil’s description of Dido in the lower world in Aeneid 6.80 In the latter irreverent pastiche Dido is directly transformed, with no sheltering screen whatsoever, into Encolpius’ recalcitrant mentula; but we may add that this development is already foreshadowed in the Sotadeans, in which the wayward member’s retreat has been paired by some with the dead queen disdainfully turning away in the same Virgilian passage.81 In this context the allusion to the dying Dido through the poem’s opening anaphora (ter… ter) is an additional touch which, in the most genuine spirit of Sotadic poetry, contributes to prepare, and give more point to, the Virgilian queen’s equation with Encolpius’ member. Like Dido, it too is unable to rise and deaf to all appeals. That the figure of Dido was particularly fit for this type of literary operation had already been made clear in the famous story of the widow of Ephesus, in which some lines addressed to her by her sister Anna at the beginning of Aeneid 4 are recycled by the widow’s handmaid, when she urges the lady to become the soldier’s lover;82 In other words, these lines are trasferred from a noble and tragically intense love story to the despicable affair recounted by Eumolpus.83 Possibly Petronius is taking pleasure in reversing an established feature of ancient fiction; that this may be so may be inferred not merely from the quota78

79 80

81

82 83

84

Verg. Aen. 4.690-691 ter sese attollens cubitoque adnixa levavit, / ter revoluta toro est. By contrast, Encolpius in his bed (conditus lectulo: 132.9) is erectus in cubitum (132.9). The type of “erection” he is not capable of is quite different! Bettini 1982, 86 and, more recently, Mutgatroyd 2000, 350. Petr. 132.11.1-2 = Verg. Aen. 6.469-470 illa solo fixos oculos aversa tenebat / nec magis incepto vultum sermone movetur. The third line (quam lentae salices lassove papavera collo) is made up by an expression drawn from Verg. ecl. 5.16 = 3.83 (lenta salix) and the final part of Aen. 9.436. Many scholars remark that it is naughtily substituted for the one that follows the first two in Virgil (Aen. 6.471 quam si dura silex aut stet Marpesia cautes), which certainly does not fit Encolpius’ present state: e.g. Fedeli 1989, 212; Vergé-Borderolle 1999, 5; lastly Rimell 2002, 156. Zeitlin 1971a, 71 n. 2 notes that the three Virgilian passages making up this cento have all to do with premature death (Dido committed suicide; ecl. 5.16 comes shortly before Daphnis’ epicedium; Aen. 9.436 describes Euryalus’ death). This is hardly coincidental, in view of the importance, in the context, of the theme of impotence equated with death, as we shall soon see. Like Zeitlin, Fedeli 1989, 211 also prefers to see a hint at Verg. ecl. 5.16, rather than to ecl. 3.83, since in the former passage the expression appears in a comparison, like in Petronius. Cf. also Crogliano 2003b, 145-147. Petr. 132.8.6 confugerat in viscera mille operta rugis is paired with Verg. Aen. 6.471472 tandem corripuit sese atque inimica refugit / in nemus umbriferum by Zeitlin 1971a, 71 and Fedeli 1989, 213, 220. Petr. 111.12 (Verg. Aen. 4.34); 112.2 (Aen. 4.38). Cf. e.g. Cutolo 1986, 62.

Petronius’ Sotadeans (Petr. 23.3; 132.8) tions we find in Greek novels (for example in Chariton), but from a particularly fitting comparison with a passage of the Historia Apollonii regis Tyri, in which numerous expressions drawn from the same Virgilian context are in all seriousness incorporated in the story in order to describe the rise of a love as noble as in Virgil.84 So, the parody of epic overflows from the Sotadic context85 to include the Virgilian cento. It is true that Encolpius’ prose address to the mentula, which fills the intervening space, after the solemn opening apostrophe,86 does not keep the mock loftiness humorously aping the poetic model. It should not escape us, however, that the end of this prose address, right before the cento, is marked by an eminently epic formula signaling the end of the speech,87 not unlike that which had been employed to signal the end of Tryphaena’s poetic speech88 and was part of an elaborate literary experiment, as we shall see in detail in another chapter.89 It is worth remarking that a comparable formula is employed by Petronius in a third passage,90 but with a variation that allows him to use it to mark not the conclusion but the beginning of a speech in verse – none other than the other poem in Sotadeans recited by the cinaedus we have previously discussed. One might think that the author has meant to place a specular mark on the two contexts, thus emphasizing the closeness of the two poems in Sotadeans, but also their different type and purpose. The epic parody extending organically over the whole passage from 132.8 to 132.11 produces a natural result: the poem in Sotadeans does not need to be placed next to, and compared with, the prose context for the reader to grasp its parodic nature, as is the case, for example with the poem at 127.9;91 I do not share Bettini’s opinion on this point.92 True, the reader knows beforehand whom this speech in verse is directed to; but in the verse itself the feminine (illa… operta… furciferae) leaves no possible doubt: the speech is directed to the mentula: Petronius’ finesse in thus indirectly alluding to the part of the body he 84

85 86 87 88 89 90 91 92

Hist. Apoll. reg. Tyr. 18, p. 13, 19-20 Schmeling (Redactio A) sed regina gravi iamdudum saucia cura (~ Verg. Aen. 4.1) Apolloni figit in pectore vultus verba (~ Aen. 4.12), cantusque memor credit genus esse deorum (~ Aen. 4.12). Nec somnum oculis nec membris dat cura quietem (~ Aen. 4.5). To employ Bettini’s felicitous expression (Bettini 1982, 86). Petr. 132.9 omnium hominum deorumque pudor. Petr. 132.11 haec ut iratus effudi. Petr. 109.1 haec ut turbato clamore mulier effudit. Cf. ch. XI. On formulas of this type signaling the end of an epic speech see Setaioli 1985, 105. Petr. 23.2 eiusmodi carmina effudit. Cf. ch. XIV. Bettini 1982, 86: “se il carme fosse dato senza indicazioni contestuali più ampie, difficilmente si potrebbe arguire chi sia mai il timido e vile personaggio con cui l’eroe tenta invano di scontrarsi”. 85

Chapter IV never mentions by name, though it plays such an important role in his work, should not escape the reader.93 There are, besides, several elements whose mold is different from, or even opposed to, the epic manner, which would prevent the reader from taking this poem seriously, even independently from its relation to the prose. These elements have not been duly stressed, and we shall presently try to point them out. If we add the numerous alliterations and assonances so conspicuously present in this poem (ter… teribilem… ter… timui… trepido, etc.)94, the picture of Petronius’ ways to attain epic parody will be complete. 5. The tone of epic parody is surely prevalent in the Sotadeans at 132.8, but it does not exhaust the poem’s expressive range. Its very subject connects it with a theme often treated in love poetry – in epigram and elegy, but also, as we shall see, the Priapea – both before and after Petronius: the theme of impotence. Among the several epigrams in the Anthologia Palatina which might be quoted,95 one offers close correspondences with our Sotadeans both in content and expression.96 The author, the epigrammatist Automedon, is surely earlier than Petronius; probably, however, both drew on a common repertoire. At Rome, after some hints in Horace’s Epodes,97 a famous model existed: the seventh elegy of the third book of Ovid’s Amores, a poem that Petronius surely has in mind.98 In Ovid, however, the main element of Petronius’ Sotadeans is missing: I mean the threat of punishment to the recalcitrant member. This appears, by contrast, in another poetic text centered on the theme of impo93 94 95 96

97 98

86

Vergé-Borderolle 1999, 5 n. 25 blatantly misunderstands the passage: he believes Encolpius’ mentula to correspond to the bipennis mentioned in v. 1. This aspect has been stressed by Barnes 1971, 293. AP 5.47; 11.29; 12.216; 12.232. Automed. AP 11.29.3-4 # † ! † % / & ! " ~ Petr. 132.8.2 languidior coliculi… thyrso (cf. also Catull. 67.21 languidior tenera… sicula beta: actually in Petronius the expression refers to Encolpius, not to his member; but, once the latter has lost its vigor, he becomes equally weak: the former “Achilles” is now as enervated as his treacherous “Agamemnon”: cf. above); 6 confugit in viscera. Both parallels are pointed out by Courtney 1991, 34. Automedon’s date is not sure (cf. Gow-Page 1968, II, 186), but Degani 1997 places him in the Augustan age. At any rate, he is earlier than Petronius, since Philip of Thessalonica included him in his Garland: cf. Gow-Page 1968, I, xlv-xlix. Hor. epod. 8; 12. Themes similar to those in Horace reappear in Mart. 6.23; 11.29. Murgatroyd 2000, 346 n. 2 believes the contacts between Ovid and Petronius to be rather loose (he polemicizes with Walsh 1970, 42 n. 2; Barbieri 1983, 28-29). But we have already remarked that the context – prose and verse – must be regarded as an organic whole. The influence of Ov. am. 3.7 is well illustrated by Barbieri in the whole impotence episode; cf. also Fedeli 1989, 214. For our specific context (Encolpius’ speech at 132.9-10, between the two parts in verse), Ov. 3.7.67 and 3.7.17-20 are particulary close to Petronius.

Petronius’ Sotadeans (Petr. 23.3; 132.8) tence, namely poem 83 in the collection of the Priapea, which tradition attributes to Tibullus. It can hardly be determined whether this particular Priapeum is earlier or later than Petronius; at any rate the contacts between the two are evident, involving not merely Encolpius’ speech in prose,99 but the Sotadeans too: witness the insistence on the guilty member’s caput,100 which is referred to three times101 – the last two joined with a threat of punishment on the part of the speaker102, who, like Encolpius in his prose speech, addresses the offending part in the second person. In Petronius’ Sotadeans the theme of “beheading” the culprit hinges precisely on the metaphor which attributes a caput to the male member. The punishment threatened in the Priapeum is not so cruel, but it still affects the caput of the offending part; and the theme of its punishment, which does not appear in the numerous texts featuring sexual impotence103 and, like in Petronius, is not actually carried out, is the the most obvious link between Encolpius’ poem and this Priapeum. If we now recall the contact with the Sotadeans recited by the cinaedus at 23.3,104 we will recognize a further link uniting Petronius’ two poems in this meter, and perhaps even a clue pointing to the Priapeum’s chronological anteriority. The primary, physiological meaning of caput aperire (v. 7) is self-evident, but as this expression is joined with the mention of a supplicium, it will also imply a parodic allusion to the procedure of executions, as several scholars have not failed to remark.105 But the preceding mille operta rugis (v. 6) not only anticipates the impossibility to uncover the culprit’s head to perform its intended decapitation: it may also antiphrastically allude to a peculiarity of the god who is constantly in a state opposite to Encolpius’ present impotence: namely Priapus, who, according to Klebs’ over-a-century-old interpretation,106 directs the whole action and will soon be addressed by Encolpius in a prayer in verse.107 In a poem collected in the Priapea the god asks, in fact, not to be blamed because of his

99 100 101 102 103 104 105

106 107

Priap. 83.4-5 nec viriliter / iners senile penis extulit caput ~ Petr. 132.10 senectae… ultimae mihi lassitudinem imponeres. For caput = glans cf. Adams 1982, 72. In connection with the theme of impotence, the same metaphor also appears in Maximian. 5.98. As already noted by González de Salas ap. Burman 1743, II, 272. Priap. 83.32 voret profunda fossa lubricum caput; 37 vagum sonante merseris caput luto. The other occurrence of caput is in the passage quoted above, note 99. With the exception of the unexplained hint in AP 12.232.6 " % $ Cf. above, text to note 53. Cf. already González de Salas ap. Burman 1743, II, 271: “ei igitur supplicio necessario caput denudari aut aperiri debebat. Quod cum fieri nequisset, supplicii quoque exsecutio pariter non potuit”. More recently e.g. Stöcker 1969, 141; Courtney 1991, 34. Klebs 1889. Petr. 133.3. Cf. ch. XVIII. 87

Chapter IV mentula semper aperta,108 an expression which, besides primarily meaning “not covered by clothing”, will probably also contain a hint at a physiological state that is the opposite to the one described in Petronius’ Sotadeans.109 With the expression languidior coliculi… thyrso (v. 2), which may be a burlesque variant of the allusion to Bacchus’ thyrsus in the Priapeum we have just quoted,110 we touch the element nuancing and enriching our Sotadeans’ epic parody, which is the only aspect usually noticed.111 I am referring to language and style. We hardly need to emphasize this expression’s remoteness from epic solemnity, which is made clear by the use of the diminutive and given further emphasis by both the closing of the diphthong, typical of colloquial language (coliculi),112 and the transfer to humble horticulture of a term (thyrsus) which, when referred to Bacchus, enjoyed right of citizenship in lofty poetry. At line 3 trepido male dabat usum is also colloquial in itself, but a significant interference of the Virgilian model supporting the parody of epic may be assumed.113 A proverbial114 (and therefore colloquial) flavor may be detected at line 5 in the expression metu frigidior rigente bruma,115 which is significantly paralleled in prose in Quartilla’s episode, not long before the Sotadeans at 23.3.116 The idea of cold is often paired with sexual impotence, and it also appears in this connection in some of the poetical compositions we have quoted for their closeness to Encolpius’ Sotadeans.117 The idea of cold fitted numerous associations, all connected with, or at least not distant from, that of impotence: with fear (explicitly 108 109

110 111 112 113 114 115 116

117

88

Priap. 9.13 nec mihi sit crimen, quod mentula semper aperta est. This is hardly the place to discuss the date of the Priapea. I will only say that, like Rankin and Richlin, I am not convinced by the conclusions drawn by Buchheit 1962; more on this in ch. XVIII. It is probably no coincidence that shortly before the passage quoted in the preceding note Bacchus’ thyrsus is mentioned (Priap. 9.11 quis Bacchum gracili vestem praetendere thyrso). The thyrsus mentioned by Encolpius in his Sotadeans (132.8.2) is much less noble – a cabbage stem. As we have already remarked (above, note 96), though the reference is to Encolpius’ person, it must surely be extended to his recalcitrant member. Cf. preceding note. Scholars almost invariably limit themeselves to emphasizing the opposition of epic tone and burlesque content. This attitude is typically represented by Sandy 1969, 297-298. Cf. Adams 1982, 26-27 for caulis (colis) in the sense of mentula. I will stress once more that in Petronius the reference is to Encolpius, not merely to his member. Cf. Verg. Aen. 3.735 trepido male numen amicum, quoted by Bettini 1982, 86. Cf. Otto 1890, 59. Cf. Dehon 2001, 316 and n. 6. Petr. 19.3 frigidior hieme Gallica (here because of fear). Cotrozzi, in Aragosti-CosciCotrozzi 1988, 73 correctly remarks: “l’immagine del frigus prelude a quella della morte su cui si insisterà nei frammenti successivi”. Ov. am. 3.7.13 tacta tamen veluti gelida mea membra cicuta; Priap. 83.29 annuo gelu (referred to an old woman); Maximian. 5.35 derigui; 60 frigus.

Petronius’ Sotadeans (Petr. 23.3; 132.8) mentioned here), sterility, old age, death.118 The association of cold with impotence and death appears twice in Petronius: here (frigidior rigente119 bruma / confugerat in viscera… motifero timore) and once more, significantly, in Quartilla’s episode;120 besides, the equation of impotence with death is a truly pervasive leitmotif in the whole context in which Encolpius’ Sotadeans appear.121 This theme, well-attested both before and after Petronius,122 plays a crucial role in the novel, especially in the Circe episode, where the opposite and complementary association of recovered virility with resurrection also appears: Encolpius will be brought back to life by Mercury psychopomp himself, just like the mythological hero Protesilaus.123 Finally, it is hard to miss the stylistic level of furciferae (v. 8), which, despite the bold transfer to the feminine, is strongly reminiscent of colloquial language and comedy.124 Its pairing with another compound word of similar formation (mortifero), which enjoys right of citizenship in high-style poetry,125 may be taken as a symbol of the masterly blend of literary levels and tones found in this poem.

118 119

120 121 122

123 124

125

For these associations see Dehon 2001. It seems excessive to see in rigente a mischievous allusion to the impotent Encolpius’ incapability to become “rigid”, as done by Murgatroyd 2000, 348 and Dehon 2001, 317. It is true that rigeo, rigidus often refer to erection (also in Petr. 134.11 nisi illud tam rigidum reddidero quam cornu); these and related words are also employed, however, to convey the idea of a “cold” that extinguishes all erotic ardor. Cf. e.g. Hor. epod. 8.17 num minus nervi rigent (misunderstood by Murgatroyd 2000, 348 n. 11); Maximian. 5.35 derigui, quantusque fuit calor ille recessit. Petr. 20.2 inguina mea mille iam mortibus frigida. As rightly emphasized by Murgatroyd 2000, 347-348, 350, 351. Cf. e.g Ov. am. 3.7.65 nostra tamen iacuere, velut praemortua membra; Maximian. 5.83 ast ubi dilecti persensit funera membri; 103 hinc velut exposito meritam te funere plango; 154 me velut expletis deserit exequiis. Also Automedon’s epigram quoted in note 96. Petr. 140.12. Cf. app. II, with the texts and the reference to Bowersock 1994, 111-113. The attempts to find different nuances in the word are hardly convincing: cf. e.g. González de Salas ap. Burman 1743, II, 272: “quippe quae egeat furca, qua sublevetur”; Murgatroyd 2000, 348 n. 17: “furcifera might play ironically on the sense of ‘groin’ (only at schol. Hor. S. 1.2.9)”. Cf. e.g. Verg. Aen. 6.279 mortiferum... Bellum. 89

Chapter V Trimalchio’s Epigrams (Petr. 34.10; 55.3)* 34.7 … Complosit Trimalchio manus et ‘eheu’ inquit ‘ergo diutius vivit vinum quam homuncio; quare tangomenas faciamus. Vita vinum est…’ 8 Potantibus ergo et accuratissime nobis lautitias mirantibus larvam argenteam attulit servus sic aptatam, ut articuli eius vertebraeque laxatae in omnem partem flecterentur. 9 Hanc cum super mensam semel iterumque abiecisset et catenatio mobilis aliquot figuras exprimeret, Trimalchio adiecit: 10

‘eheu nos miseros, quam totus homuncio nil est! sic erimus cuncti, postquam nos auferet Orcus. Ergo vivamus, dum licet esse bene.

35.1 Laudationem ferculum est insecutum plane non pro expectatione magnum. 34.10 HL(=lmrtp) 2 sic : nil

55.1 Comprobamus nos factum et quam in praecipiti res humanae essent vario sermone garrimus. 2 ‘Ita’ inquit Trimalchio ‘non oportet hunc casum sine inscriptione transire’ statimque codicillos poposcit et non diu cogitatione distorta haec recitavit: 3

‘quod non expectes ex transvero fit et supra nos Fortuna negotia curat. Quare da nobis vina Falerna, puer’.

4 Ab hoc epigrammate coepit poetarum esse mentio… 55.3 H 2 supra Heinsius: super negotium L

1-2 L(=lmrtp)

Chapter V 1. The poems recited by Trimalchio during the Cena may be regarded as specimens of the wealthy freedman’s literary production, since, in a passage occurring half way between the first and the second, Petronius drops the hint that he harbored literary pretensions.1 These, then, are not poems that can be attributed to Encolpius, either in his capacity of acting character or of narrator expressing poetical comments on the facts and situations he recounts, and at times playing the role of intermediary between the author and the reader.2 Even though this is just one of the many brush strokes that contribute to paint the nature and personality of this unforgettable character,3 we are therefore justified in regarding these poems as an expression of their author’s individual temperament and social condition;4 actually, given their undoubtedly low poetic worth, which certainly corresponds to Petronius’ intentions, they should primarily be regarded and analyzed in the frame of a widespread social phenomenon, not merely from the strictly literary point of view. Trimalchio’s longest poem5 will be studied in the next chapter; in this we shall analyze the two shorter compositions, which in many ways are those that best illustrate the author’s personality. The first point to be made is that they can hardly be read in isolation, since they are part of an intricate web connecting them with each other and with the overall architecture of the Cena. These two short poems express, we might say, Trimalchio’s whole philosophy of life: his unreserved acceptance of the carpe diem attitude, in the face of overhanging and inescapable death and of unpredictable and unresistable fortune. The foremost means to exorcise the resulting distress is wine, and, as we shall see, food too. The theme of death permeates the whole Cena from end to end;6 but the urge to a vital reaction marks the funda* 1

2 3

4 5

6

92

A version of this chapter has appeared with the title I due epigrammi di Trimalchione (Petr. Sat. 34.10, 55.3), “Prometheus” 30, 2004, 43-66. Petr. 41.6 puer speciosus… poemata domini sui acutissima voce traduxit. It should not escape us that only the poems recited by Trimalchio himself are reported. The hint at the page boy reciting more of his verse aims to let the reader know that his poetrywriting was habitual. As I believe him to do at 132.15. See ch. XVII. Eumolpus’ poetic compositions are also introduced as specimens of his literary production, but unlike Trimalchio he is more than anything a poet, though his figure is also finely nuanced. See ch. X. Occasionally poems are recited by other characters too (e.g. Agamemnon, Quartilla, Tryphaena), with no hint at a habit to compose poetry. As correctly remarked by Slater 1990a, 161. The senarii at 55.6 As rightly remarked by Aragosti 1995, 258 n. 53, the sententious character of the “epigram” at 55.3 anticipates the closely following longer poem, which Trimalchio ascribes to Publilius Syrus (cf. Publ. Syr. sent. 171 fortuna vitrea est, tum cum splendet frangitur, conceptually close to Petr. 55.3). See, among others, Arrowsmith 1966; Grondona 1980, 9-75; Herzog 1989; Gagliardi, 1989.

Trimalchio’s Epigrams (Petr. 34.10; 55.3) mental passages of the episode, also defining its structural cornerstones. At the beginning of the Cena the poem at 34.10 and its immediate context introduce the theme of the ineluctability of death, and the consequent stimulus to enjoy the fleeting moment, with linguistic signals which will reappear in the middle, in connection with the verse at 55.3, and once more at the end, in chapters 72 and 73.7 Within the narrower frame of the prose context in which the first poem is embedded, equally clear linguistic signals emphasize the close connection between Trimalchio’s words that precede the appearance of the silver larva and the verse that follows it: both the prose speech and the poem begin with the same word (eheu) and both contain a peculiar and eye-catching word: homuncio.8 Actually, Petronius explicitly presents the poem as a continuation of, and an addition to, the prose speech (adiecit). 2. After Trimalchio’s prose speech a servant brings in a miniature silver skeleton with disjointed articulations, which the host repeatedly throws on the table, thus making it take on different postures, before reciting the verse at 34.10 This was a current custom in Petronius’ time, as amply confirmed by archaeological evidence. All the scholars that have studied our text have referred to archaeology to some extent, but the one who has most organically investigated the theme of the skeleton at the banquet is undoubtedly Katherine Dunbabin, whose essay amounts to an exhaustive monograph on the subject, taking into account the testimonies not merely of archaeology, but also of literature.9 In the Greco-Roman world this idea is closely connected with the widespread theme of the carpe diem, understood as a vital reaction to death’s inescapability. In Petronius himself it is clearly expressed in the words the handmaid directs to the inconsolable widow of Ephesus: ipsum te iacentis corpus admonere debet ut vivas.10 The corpse, or – in Trimalchio’s case – its representation, reinforces the drive toward the enjoyment of life, as long as it lasts.

7

8

9 10

Petr. 34.7 quare tangomenas faciamus ~ 55.3.3 quare da nobis vina Falerna ~ 73.6 itaque tangomenas faciamus; 34.10.3 ergo vivamus 72.2 cum sciamus nos morituros esse, quare non vivamus? The theme of wine as vital remedy against death is introduced from the very beginning: 34.7 vita vinum est. On the theme and the relations linking the quoted passages cf. Deonna-Renard 1961, 99-104; Grondona 1980, 14-17; Gagliardi 1989, 20 n. 35. Baldwin 1979, 145 appears to be the only scholar to realize that the verse picks up the preceding prose speech; Grondona 1980, 15 only notes that both begin with eheu. Connors 1998, 52 lays emphasis on a questionable connection of auferet (34.10.2, referred to Orcus) with the ferculum at 35.1. This, according to her, would be a symbol of mortality, in that it represents the Zodiac. She does not mention the much clearer connection the author himself signals between Trimalchio’s prose speech and his verse. Dunbabin 1986. Petr. 111.12; cf. Mart. 5.64.5-6. 93

Chapter V Herodotus relates an Egyptian custom which consisted in bringing the image of a corpse into the banqueting hall, while at the same time urging to enjoy the fleeting moment through the pleasure afforded by wine and food.11 His testimony is confirmed by later sources, particularly by Plutarch, who twice hints at the same Egyptian custom.12 In one of these texts13 we read the expression , which makes us think that the image of the corpse shown to the banqueters might have been in the shape of a skeleton.14 The main Latin literary testimony of this custom is precisely our Petronian text; but, as already hinted, it is amply confirmed by archaeological evidence. In the first place we shall mention some miniature human skeletons with disjointed articulations, and so able to take on different postures, like the one described by Petronius. Dunbabin lists seven such miniatures;15 of these only one can be dated, since it comes from Pompeii (it is now in the Arhaeological Museum at Naples): its terminus ante quem brings us close to Petronius’ times. It is also made of the same metal as Trimalchio’s: silver, the others being made of a baser material: bronze.16 Though only the model skeleton from Pompeii can be dated with any accuracy, Dunbabin concludes her analysis of all the available material (including vases, gems, and other supports) by fixing with no uncertainty the pe11

12

13 14 15 16

94

Herod. 2.78 ! " ! # " $ Plut. Is. et Os. 17, 357F ; conv. sept. sap. 2, 148AB. Dunbabin 1986, 209-210 believes Herodotus’ and Plutarch’s testimony to reflect a real Egyptian custom. Wöhrle 1990, 293-294 adduces and discusses two ancient Egyptian texts in which similar ideas may be detected, but warns against their interpretation along the lines of the Greco-Roman carpe diem. They advocate no so much the enjoyment of life in view of impending death, but rather the consciousness of one’s mortality even in the midst of pleasures. Wöhrle (p. 295) recognizes a similar attitude in Plutarch’s second passage (on which more below). As for Sil. It. 13.474-476, which Dunbabin 1986, 208 refers (with Lucian. de luctu 21, already quoted by Burman 1743, I, 194) to the same Egyptian custom, Wöhrle (pp. 298-299) persuasively contends that it should be rather connected with the custom to offer food to the dead. Plut. conv. sept. sap. 2, 148A. Cf. Wöhrle 1990, 296. % may be a dried up corpse (in this case a mummy), but also a skeleton. Dunbabin 1986, 196. In connection with Petronius see also Maiuri 1945, 165-166. The fact that only one of these miniature skeletons is made of silver confirms that Trimalchio is ostentatiously displaying his wealth. In his house even the humblest articles are made of silver: 27.3 (matella); 28.8 (lanx); 29.8 (Lares); 31.11 and 70.7 (craticula); 33.1 (pinna); 35.6 (clibanus); 50.1 (corona); 70.8 (pelvis). These are listed by Panayotakis 1995, 70 n. 43. More passages should be added: 31.10; 34.3; 56.8; 67.2; 73.6. Trimalchio himself confesses his predilection for silver (52.1 in argento plane studiosus sum); and there is more silver in his ostiarius’ cell than in anybody else’s estate (37.8).

Trimalchio’s Epigrams (Petr. 34.10; 55.3) riod in which the representation of the skeleton at the banquet was in vogue in the two centuries astride the birth of Christ.17 Which was the meaning of the skeleton for the ancients, particularly in this connection? A still basic essay on this topic was written in the XVIII century by Lessing, who maintained that it represented not Death in itself, but rather the larva, i.e. the spirit of the evil dead, represented as a skeleton, which scares and persecutes the living.18 This is undoubtedly the term’s primary meaning, documented by Petronius himself, who in two passages of his work uses the word in a sense not far removed from the original one.19 Lessing maintains that in time the word ended up meaning simply “skeleton”; he was followed by Dunbabin, who at any rate correctly remarks that it never became a purely technical word with the strictly anatomical meaning the Greek acquired.20 I think it possible to take a further step by asserting that, even when larva denotes the skeleton, the original meaning never completely disappears. In reality the authors who use this term in the sense of “skeleton” are no more than two:21 Petronius in our passage, and Apuleius in his Apology.22 Trimalchio’s miniature skeleton has disjointed articulations and can take on different postures, like living people. Peculiarly enough, however, Dunbabin, after establishing a distinction between the representations of the human skeleton as motionless and inanimate and those that impart to it attitudes and activities of the living, places Trimalchio’s model in the first class, and refers it to the Egyptian custom to introduce the image of a corpse into the banquet hall;23 she actually goes as far as expressing the opinion that the custom attested by Petronius was probably derived from Egypt.24 As for the skeletons represented as perform17 18 19

20 21

22

23 24

Dunbabin 1986, 193. Lessing 1983, 68-71. For this meaning of larva see TLL VII 2, 977, 82-978, 46. Cf. also Thaniel 1973, 186-187. Cf. especially Petr. 62.10 in larvam (“with the appearance of a ghost”). At Petr. 44.5 larvas means rather “puppets”, “clowns”. For these two meanings see TLL VII 2, 978, 47-55. Dunbabin 1986, 187-188. Even though their usage is upheld by the Greek equivalent given (with others) for larva in the Glossaries. Cf. TLL VII 2, 978, 57-67. We may add, at most, Priap. 32.12 pallorem maciemque larvalem (of a girl all skin and bones); Apul. met. 1.6 larvale simulacrum (of a man ad miseram maciem deformatus). Apul. apol. 63 (where sceletus and larva apparently are indifferently used. At ch. 61 the same object is called only sceletus). Shortly after (cap. 64) Apuleius uses larva in the sense of “evil spirit”. Dunbabin 1986, 208. Dunbabin 1986, 210-211. Friedlaender 1891, 214, followed by Ernout 1923, 30 n. 2, already remarked that in a bustling port like Pozzuoli (if indeed it is the Graeca urbs referred to at 81.3) it would be hardly surprising to encounter an Egyptian custom. 95

Chapter V ing various activities, for example those on the famous cups from Boscoreale, now at the Louvre, that we shall presently discuss, Dunbabin regards them as a grotesque parody of the living and their actions, and – rather inconsistently with what she had previously stated – contends that the various postures taken on by skeleton models like the one described by Petronius were intended to serve the same parodical purpose.25 She rules out, a priori, the possibility that the idea of the animated skeleton may be rooted in popular tradition and that the evil spirits called larvae might have been imagined to have a skeleton-like appearance.26 This latter statement seems untenable, at least as far as Petronius’ time is concerned, in view of a contemporary’s testimony. As already remarked by Lessing,27 there is a passage in Seneca which clearly proves that at the popular level the larvae frightening the superstitious were believed to look like skeletons.28 The same idea is attested in the Greek-speaking area too, under the empire.29 The previously quoted passage from Apuleius’ Apology proves beyond doubt that, even when larva refers to the human skeleton, the negative nuance connected with the original meaning has not completely disappeared. One of the charges imputed to Apuleius by his opponents concerned the fact that he possessed a macilentam vel omnino evisceratam formam diri cadaveris… prorsus horribilem et larvalem; and he does not deny that such an object would amount to an evidens signum magiae.30 From this passage Dunbabin draws the correct conclusion that in the II century A.D. the original meaning of larva, “evil spirit”,

25

26 27 28

29

30

96

Wöhrle 1990, 298 thinks that Petronius alluded to the passage of Herodotus quoted above (note 11). Dunbabin 1986, 215. Kämpfer 1994, 239 goes even farther: he believes that Trimalchio means to amuse his guests by having the skeleton take on the attitudes and movements of a drunk man. See my objections below, note 35. Dunbabin 1986, 189-190. Lessing 1983, 69. Sen. ep. 24.18 nemo tam puer est ut Cerberum timeat et tenebras et larvalem habitum nudis ossibus cohaerentium. The fears mocked by Seneca do not concern skeletons in themselves i.e. inert bone remains, but evil beings associated with the lower world (as Cerberus, mentioned at the same title), whose description leaves no doubts as to their skeleton-like appearance. The same idea is found already a few generations before Seneca: cf. Ov. Ibis 144 insequar et vultus ossea forma tuos. A part of the tradition offers larva instead of forma; but the passage is pertinent with either reading. Lucian. philops. 32 tells a story about some young men who wanted to look like ghosts to frighten Democritus, and put on skull-like masks for this purpose ( ). This does not necessarily mean that the idea went back to Democritus’ time, but it was clearly established in Lucian’s. The latter often represents the souls of the dead dwelling in Hades in the shape of skeletons: Men. sive nekyom. 15; dial. mort. 25. AP 11.92.4 (Lucillius) employs the expression ! in reference to spirits of the dead (looking like skeletons), not to bone remains. Apul. apol. 63.

Trimalchio’s Epigrams (Petr. 34.10; 55.3) had unquestionably resurfaced;31 it must be added, however, that this very text unambiguously confirms that evil spirits were thought to look like skeletons. In Petronius’ time Trimalchio’s skeleton model surely did not raise the suspicion of being used for black magic. We may add that, if our Petronian passage can be associated with the roughly contemporary cups from Boscoreale, the skeleton does not represent the soul (much less the evil spirit) of the deceased, but simply his mortal remains.32 Is then this Petronian text the only one in which the original meaning of larva has totally disappeared? We should not forget that Trimalchio’s model skeleton can be moved.33 Dunbabin, as we saw, believes its movement to be a parodical imitation of the motions and attitudes of the living;34 it is easy to retort, however, that from Herodotus through Petronius and down to Plutarch what is emphasized is not that skeletons are like us (if anything, they were), but that we shall be like them.35 We may also refer to Lucian’s text we have already quoted in a footnote:36 some young men, who want to pass themselves off as ghosts, surround the philosopher Democritus jumping and dancing with skull-like masks on their faces; and a moving skeleton-like ghost is described by Ovid too, in another passage already quoted in a footnote.37 It is not impossible, then, that in Trimalchio’s skeleton’s acrobatics there might lurk a weak, residual trace of the original idea conveyed by the term larva, which survives to some extent at all times. If, as we have seen, this nuance clearly resurfaces in the only other author that uses the word in 31 32

33

34 35

36 37

Dunababin 1986, 248-249. In the Boscoreale cup A, indeed, a skeleton holds a butterfly in its hand, with the inscription & . The butterfly is a current symbol for the soul. See Dunbabin 1986, 226 and fig. 37 (p. 225). According to Panayotakis 1995, 68 Trimalchio’s larva functions as a puppet drawn by strings. He refers to Cic. de orat. 3.220 for exprimo in the meaning of “take on a theatrical posture” (in reality, however, in that passage the object of the verb is verba). Rosati 1983, 215 n. 7 limits himself to the remark that the larva changes postures “quasi un neurospaston”. Dunbabin 1986, 215. Cf. above, text to note 25. Herod. 2.78 ; Petr. 34.10.2 sic erimus cuncti; Plut. Is. et Os. 17, 357F ! ; conv. sept. sap. 2, 148B . Cf. also AP 11.38.6 , on which see below, note 56. If anything, the skeletons performing human activities represented on vases and other supports mean to offer a lively symbolism of man’s mortality, which he can never shake off in all his activities (“questa morte che ci accompagna”, in Cesare Pavese’s poetic words). This is explicitly conveyed by inscriptions like the one on the Boscoreale cup B: . It is a thinly veiled , similar to the explicit one in the famous mosaic of the Museo Nazionale Romano, which has this inscription under the figure of a reclining skeleton. Lucian. philops. 32 (cf. above, note 29) ! ' $ Ov. Ib. 144 insequar et vultus ossea forma tuos (above, note 28). 97

Chapter V the sense of “skeleton” – Apuleius in his Apology –, we may be authorized to suppose that it never completely disappeared. Surely, the use of a term such as larva, which carried overtones unmistakbly connecting it with native folklore, makes it more likely that Trimalchio’s act is somehow reminiscent of ancient beliefs – even though transformed into an appeal to the carpe diem –, rather than merely aping a custom imported from Egypt, where we should not forget that the corpse brought into the banquet hall did not move, but lay rigid and inanimate in its coffin.38 According to Dunbabin39 this custom (a corpse or skeleton shown to the banqueters) was followed in time by the representations of skeletons – inanimate at first, animated later – on banquet chattels. In her opinion Trimalchio’s act (which she connects with the Egyptian custom) was by then out out of fashion. What she does not see is that, even if we admit some relation with the Egyptian custom attested by Herodotus and Plutarch, the animated skeleton model amounts to an epochal innovation. It seems more justified to surmise that the fad of the skeleton (or corpse) model in the banquet hall underwent an inner evolution, which, in the Roman area, was influenced by the beliefs connected with the term larva. Indeed, the only movable miniature skeleton which can be dated – and the only one made of silver – is the one from Pompeii, which is nearly coeval with Trimalchio; Petronius’ freedman then, was anything but old-fashioned. Dunbabin asserts40 that elaborate representations like those of the skeletons on the Boscoreale cups were too complex for the “old-fashioned” Trimalchio. This is not manifestly impossible, in view of the way the wealthy but ignorant freedman is characterized in Petronius’ novel. We should not forget, however, that some scenes represented on the Boscoreale cups were hardly foreign to Trimalchio’s circle. On cup B a standing skeleton pours a libation on a lying skeleton, which is reduced to a heap of bones.41 Trimalchio’s friend, Habinnas, reports taking part in a funeral banquet in which one half of every drink was poured on the dead man’s bones.42 The scene is then a banquet, the same the two cups refer to, as Dunbabin herself has clearly shown. We should finally add that there are further contacts between the Cena and the treatment of the skeleton theme in the visual arts.43

38 39 40 41 42 43

98

Cf. Herod. 2.78 (above, note 11). Dunbabin 1986, 233. Dunbabin 1986, 230. Cf. Dunbabin 1986, 226 (fig. 39, p. 227). Petr. 65.11. Cf. Dunbabin 1986, 206-208 (a sarcophagus from Herakleion, I century A.D., on which, besides a skeleton and other figures, there is a sundial: Trimalchio has a horologium in the triclinium [29.6] and another one, probably a sundial, on his tomb [71.11]); cf. also Dunbabin 1986, 228 (on the Boscoreale cup A, next to the skeleton representing the

Trimalchio’s Epigrams (Petr. 34.10; 55.3) 3. The three lines of 34.10 are clearly part of the accurate stage direction governing the Cena: they have certainly been prepared beforehand, to be recited after the appearance of the one-hundred-year-old wine and the silver skeleton.44 This is confirmed by the occurrence of the opening eheu and the term homuncio both in the verse and in the prose speech that Trimalchio utters after the “first act” (the arrival of the one-hundred-year-old wine). At this point he clearly has his next line, i.e the poem prepared for the “second act” (the play with the miniature skeleton), already in mind. The final seal is provided by the caption adiecit, which introduces the verse as a continuation of the previous speech. We have already called attention to these details. In the past the metrical pattern, joining two hexameters and one pentameter, has been regarded as problematic, to the point that some interpreters tried to “normalize” the poem either by expunging the last line or by inserting a pentameter after the first one;45 and even in times close to us this metrical sequence has been dubbed an “incongruous mixture”.46 In reality this combination is anything but rare in Greek and Latin metrical funerary inscriptions;47 it is normally

44

45

46

47

tragic poet Moschion, there is the inscription !' : cf. Petr. 80.9, and see ch. VIII). Cf. Sandy 1974, 332; Panayotakis 1995, 70. Walsh 1970, 128 equates the verse at 34.10 with that at 55.3; through both, he says, Trimalchio “consigns to immortality two incidents in the dinner” – in other words, he thinks both poems to be improvisations. This is surely untrue as far as 34.10 is concerned. Cf. also Courtney 2001, 85; Jensson 2004, 11. Slater 1990a, 161-162 rightly regards the first poem as the triumph of Trimalchio’s stage direction, the second as marking the beginning of his failure to control the events. See below. The opinion of Yeh 2007, 95-98, according to which the two “epigrams” recited by Trimalchio at 34.10 and 55.3 are part of one and the same poem, is totally groundless. The third line is graphically presented as prose by H, the only manuscript offering the complete text of the Cena. For the attempts at metrical “normalization” see Burman 1743, I, 196-197. Scaliger refrained from inserting a pentameter after the first hexameter (Tornaesius inserted quam fragilis tenero flamine vita cadit), but marked its alleged disappearance with asterisks. See also the apparatus of Bücheler 1862. Walsh 1970, 128. According to Paratore 1933, II, 106 the coupling of two hexameters and one pentameter is a consequence of improvisation and reveals the “crassa ignoranza di Trimalchione”. For the former see Kaibel 1878, 701, who, in the tabula metrica, lists nine inscriptions formed by two hexameters and one pentameter, plus one (no. 823) in which this pattern is doubled. Cf. also AP 13.15 (in AP 13.16 three hexameters are followed by one pentameter); in another inscription (no. 327 in Peek 1960) a normal elegiac couplet is followed by two hexameters and one pentameter. Cf. also Keil 1865, 559. For the latter (the Latin inscriptions) see e.g. CE 1105, 1179, 1260, 1292. Indeed, in many Latin epigraphical poems both hexameters and pentameters, which are often metrically defective like Trimalchio’s at 55.3, alternate without following a regular pattern: e.g. CE 880, 914, 924, 947, 965, 971, 986, 1007, 1009, 1019, 1039, 1058, 1081, 1187, 1188, 1190, 1206, 1216, 1223, 1224, etc. In view of this the opinion expressed by Yeh 2007, 524, 99

Chapter V not employed in literature and is generally regarded as denoting a low cultural level,48 and therefore fitting Trimalchio. This can undoubtedly be accepted, even though the level of ignorance betrayed by this metrical pattern should not be exaggerated. As pointed out by Smith,49 it appears even on tombs certainly not belonging to the most uncultured layers of the empire’s population. What can be reasonably inferred from Trimalchio’s lines is rather their closeness to funerary poetry, as represented not so much in literature as in the epitaphs for not necessarily uncultured people. This closeness concerns form, but, as we shall see, content too. As we have already hinted, the intertwining of social and literary aspects in Trimalchio’s “poetry” must be carefully kept in mind. If there is mockery on the part of the author, it is a subtle one; what would be perfectly acceptable in a poem so close to funerary inscriptions, becomes a conspicuous mark of defective culture and poetic ability shortly after, when Petronius, seemingly in passing, drops the hint at Trimalchio’s literary ambitions.50 With this the reader is warned, this time in advance, as to the way he should read and judge Trimalchio’s next “epigram”.51 The relations linking convivial and funerary poetry are indeed tight, as several scholars have not failed to remark,52 and as confirmed by the epigraphic material that has come down to us.53 Some of the scholars who correctly grasped the link connecting our poem with funerary poetry, however, have proceeded too far, in that they have seen in the word following the verse – laudationem – a reference to Trimalchio’s lines, regarded as so close to epitaphs as to amount to a veritable laudatio funebris.54 This is an untenable interpretation, both because it is difficult to imagine such a

48

49

50 51 52 53 54

100

according to which Trimalchio is trying to write metrically hybrid, but in all respects literary, poems, like those at Petr. 5 and 109.9-10, is untenable. After Bücheler 1862, 27 (“carmina… in sepulcris plebeiorum hominum inscripts”), cf., among others, Friedlaender 1891, 214 (“Grabinschriften ungebildeter Leute”); Cugusi 1967, 90 (“metro tipicamente popolare, comune fra le persone incolte”); Barnes 1971, 303-304 (speaking of “Trimelegiacs”); Slater 1990a, 162; etc. Smith 1975, 74. He calls attention to the fact that the sequence hexameter-hexameterpentameter is found on the tomb of a physician and his wife (558 Kaibel 1878 = 1736 Peek 1955, vv. 5-7). Their names were respecively Claudius Agathinus and Felicitas. Petr. 41.6. See above, text to note 1. The one at 55.3. See below. E.g. Grondona 1980, 16-17; Dunbabin 1986, 208. Cf. Lier 1904, 56-63; Lattimore 1942, 260-263. Not a few of the epigraphical poems quoted and discussed by these scholars are very close to Trimalchio’s verse. Already Burman 1743, I, 197: “possit hic capi pro laudatione funebri, quam stolidus Trimalchio, ad larvae conspectum, versibus istis peregit”; more recently Stöcker 1969, 93, who even supplies a word: laudationem ferculum est insecutum… This interpretation is upheld also by Grondona 1980, 17. Müller 1961 writes in the apparatus: “ante laudationem aliquid deesse suspicor”; these words disappear in his subsequent editions.

Trimalchio’s Epigrams (Petr. 34.10; 55.3) laudatio without someone in whose honor it is uttered, and because the guests’ praises to Trimalchio are a matter of course in the whole Cena, and are even denoted at times with the very term laudatio.55 As we said, the reader will learn, a few chapters later, that Trimalchio harbors poetic and literary ambitions. This implies that, besides the contacts with poetic forms intermediate between popular and literary production, like an epigram in the Anthologia Palatina already pointed out by Burman,56 we will be justified in seeking the influence of great literature too. Many of the most representative poets of ancient literature did not indeed disdain developing the topic of Trimalchio’s poem: the carpe diem theme, with an explicit or covert reference to impending death. The most obvious contacts have been repeatedly pointed out: for example, those with Catull. 5 or with Horace;57 other scholars refer to a famous Lucretian text in which man’s weakness and mortality are expressed with a term (homullis) which possibly suggested homuncio to Petronius.58 Contacts with poetry of a lower level, but still fully belonging in artistic literature, are not wanting either: suffice it to refer to the pseudo-Virgilian Copa’s last verse, where Death itself utters the same advice as Trimalchio’s poem.59 But Petronius may have meant to conceal a more mischievous blow to his character’s literary pretensions in a seemingly cryptic expression he has Trimalchio repeat at the beginning and at the end of the Cena, which is obviously made significant as a structural post by this very arrangement and, like his poem, may be said to synthesize the host’s whole philosophy of life: tangomenas faciamus.60 Burman already referred to a fragment of Alcaeus61 urging to drink, in 55

56

57 58

59 60 61

Cf. e.g. Petr. 34.5; 36.4 (plausum); 39.6; 41.8; 48.7 (laudationibus); 52.8. Fuchs’ supplement laudationem is unnecessary; there is no need to specify who bestows the praise – who else but Trimalchio’s guests? Burman 1743, I, 194; more recently Stöcker 1969, 121; Wöhrle 1990 n. 30: I am referring to AP 11.38, attributed to Polemon, king of Pontus (vv. 5-6 / " . Dunbabin 1986, 251 informs that these lines appeared engraved on a gem next to the figure of a skull and a table loaded with food). I refer to Cugusi 1967, 90 and Gagliardi 1989, 14 n. 16, as representative of a great number of scholars. Since Bourdelot (see Burman 1743, I, 197); then, e.g., Ernout 1923, 30 n. 2, who mentions Anacreontic themes already parodied by Lucretius. The passage is Lucr. 3.912915 hoc etiam faciunt ubi discubuere tenentque / pocula saepe homines et inumbrant ora coronis, / ex animo ut dicant ‘brevis est hic fructus homullis; / iam fuerit neque post umquam revocare licebit’. Lucretius is undoubtedly referring to impending death, as made clear by what follows. Copa 38 mors aurem vellens ‘vivite’ ait ‘venio’. Cf. Franzoi 1988, 98-99. Petr. 34.7 = 73.6. Cf. above, note 7. Burman 1743, I, 192. The reference is to Alc. 347.1 Voigt . Burman suggests that Trimalchio, whose knowledge of 101

Chapter V which a verb appeared – – which may have caused the semiliterate Trimalchio to make up this pseudo-Greek verbal form.62 But the Greek poet, immediately after the imperative , expresses the reason for this recommendation: . If Trimalchio has understood these words, which primarily hint at the heat of the hour and of the season, as a reference to the heavenly revolutions ( ), and thus to the inexorable passing of time, it would amount to a strong support of the interpretation of tangomenas faciamus as “let’s drink”; and in turn, such an allusion to Alcaeus at this point would foreshadow the poem at 34.10, which urges to enjoy the pleasure of the banquet while we still have the time. With a single stroke, Petronius would have illustrated both Trimalchio’s lofty poetic ambitions and his complete lack of the necessary linguistic and literary preparation. Many of Petronius’ interpreters insist on the poor literary worth of Trimalchio’s poems. As far as the meter of 34.10 is concerned, we have already observed that the target is not so much the popular pattern in itself, as the ambition of someone who wishes to pass himself off as a poet in spite of the Muses, though not being able to rise above the amateurish level that metric pattern betrays. Trimalchio incarnates Horace’s complaint: scribimus indocti doctique po-

62

102

Greek was rudimentary, meant tangomenas to be the participle of , and an allusion to Alceus’ verse, which was in fact an invitation to drink. He was followed by Jahn 1867, 427 n. 3. Bücheler 1862 goes as far as writing in the text (following Müncker: see the apparatus); but his later editions have tengomenas. See lastly Cavalca 2001, 172, with further bibliography. Possibly the Greek form crossed itself with the Latin verb tangere (cf. 66.3 de melle me usque tetigi). Cf. Perrochat 1939, 20. Even Cicero was not above using humorous Greco-Latin hybrid forms (e.g. Cic. Att. 1.16.13 facteon). Not all have accepted this explanation, and the passage remains a difficult one. We should at any rate remember that the Asian Trimalchio (75.10) was probably not completely ignorant of Greek, at least at the colloquial level (cf. indeed 48.8), even though command of literary language was surely something else, as proved by the peculiar references to Homer made by Trimalchio in the very context in which he uses Greek (48.7). At 34.7, which, unlike the later passage (73.6), is transmitted not merely by H but also by L, this branch of tradition omits the sentence containing tangomenas faciamus. Alcaeus’ passage was at any rate well-known: according to Macr. Sat. 7.15.13 it was often sung and had been imitated by Eratosthenes (Macr. 7.15.23). Cf. indeed Eratosth. Erigone fr. 2 Rosokoki ' # (Rosokoki 1995, 41). Tangomenas would then mean “wet”, “drenched” (scil. epulas or potiones, as suggested by Smith 1975, 73). It is not necessary to correct in order to obtain a plural masculine form. What Trimalchio means is no doubt “let’s drink” (“madidemus nos”, as paraphrased by Burman 1743, I, 192). Page 19653, 303 believes that Alcaeus’ is echoed not merely by Petronius’ tangomenas, but also by Hor. c. 4.12.22-23 non ego te 3 meis / immunem meditor tinguere poculis. The same scholar (Page 1965 , 302) translates with “for the Star is coming round”. Besides the prevailing theme of the heat of the hour a hint to heavenly revolutions may in fact be there too.

Trimalchio’s Epigrams (Petr. 34.10; 55.3) emata passim.63 But, aside from the meter, it is not difficult to find fault with this “wonderful awful poem”, as Slater calls it,64 to the point that the expression “marked literary inferiority” has been employed in reference to Trimalchio’s poetic compositions.65 One more momentous linguistic censure can be added to the numerous blames addressed by scholars to this poem, which we have reported in a footnote. Trimalchio employs, in reference to death – actually to all-abducting Orcus – a verb (aufero) which is often found in high poetry in this connection.66 But we should not miss his inability to keep the correct verbal relations: sic erimus cuncti, postquam nos auferet Orcus, instead of abstulerit.67 It is true that poetry enjoys more freedom than prose, but the contrast between Trimalchio’s ambition to attain lofty style and the linguistic level betraying his lowly origin and defective culture should not be missed.68 Another detail that places Trimalchio half way between lofty poetry and archaic notions of undoubtedly popular and to some extent preliterary nature is his reference to Orcus.69 In literature Orcus tends to be identified with the Greek Hades, or even with the lower world itself, whereas in native conception he is a demon of death, who not only receives people in the lower world but forcibly carries them off. In itself this would be no unequivocal sign of the popular mold 63 64

65

66

67

68 69

Hor. epist. 2.1.117. Slater 1990a, 161. Slater lays emphasis on the low literary level of homuncio and on the displeasing effect of two monosyllables at the end of the first hexameter. Cf. also Connors 1998, 52. This detail had already been noticed by Cugusi 1967, 90 n. 32, who also stresses the inelegance of the pyrrhic foot b n at the end of the pentameter and of the spondee heu at the beginning of the first hexameter. The pyrrhic foot at the end of the pentameter is regarded as a metric shortcoming by Sochatoff 1969-1970, 342 too. See the polemic reply of Barnes 1970-1971, 255 and of Woodall 1970-1971, 257. The latter calls attention to the fact that pyrrhic feet are found at the end of pentameters in Ovid too. The pentameter is metrically correct for Baldwin 1970-1971, 254 too. Sochatoff 1969-1970, 342, with immediate reference to the poem at 55.3, but clearly inclusive of the one at 34.10 too. See also the debate among Canadian and American scholars referred to in the preceding note. As already remarked by González de Salas ap. Burman 1743, II, 148, who quotes Prop. 2.20.18 and 2.11.4, as well as Verg. Aen. 6.429. Cf. also Huxley 1970-1971, 69, who quotes Hor. c. 2.16.29. Postquam is seldom employed with the future; but, in the few cases it is, a future in the main clause has a future perfect as its counterpart in the clause introduced by postquam: Cato de agr. 65.1 post diem tertium quam lecta erit, facito; cf. Liv. 31.7.7 quinto die quam (= postquam) ab Corintho solverit naves, in Italiam perveniet. Cf. KühnerStegmann 1971, II, 358. The usage in this Petronian passage does not seem to be paralleled in all the span of Latinity: cf. TLL X 2, 251, 34-35 (recording a very late case of postquam with the future, but with the present in the main clause). It should be added that even the verb aufero referred to Orcus might be the echo of a popular cliché. See below, text to note 71. For more details see Setaioli 1997, with the literature quoted and discussed. 103

Chapter V of Trimalchio’s poem, since the conception of Orcus as an abducting demon often appears in Horace too. He even depicts Orcus while mowing off the high as well as the low – an idea seemingly very close to the Medieval notion of the scythe of Death.70 In all probability, however, Trimalchio is not harking back to literary models like Horace, or at least he is influenced not merely by these, but also – and probably to a greater extent – by popular ideas connected with old native beliefs. This is confirmed by the fact that his friend Echion, in a long speech which amounts to a veritable concoction of popular expressions and ideas, twice refers to an Orcus conceived as a power annhilating and carrying off everything – both the evil and the good that people have attained in their lifetimes; he even employs the same verb as Trimalchio’s poem: auferre.71 4. The incitement contained in the final pentameter of 34.10 ends with an expression that some of Petronius’ early interpreters regarded as an intentional double entendre; the infinitive in esse bene can indeed be understood as a form either of sum or of edo.72 In times closer to us this interpretation was picked up by several scholars, none of whom refers to those who had already proposed it centuries before.73 I would like to point out, as a personal contribution, that an athematic form of edo does appear in the Satyrica, and – what is more – it is placed in Trimalchio’s mouth. Shortly after his next poetic performance he says: aliquis ovillam est et tunicam habet.74 There is no doubt that the epigrams and the funerary inscriptions developing the same theme as Trimalchio’s poem hint at drinking more often than they do at eating; but eating appears several times too, joined with drinking,75 or even by itself. A Greek metrical epitaph, to which 70

71 72 73

74

75

104

Hor. epist. 2.2.178-179 si metit Orcus / grandia cum parvis, non exorabilis auro. In Horace the image of mowing appears to be suggested by the context, where granaries and fields are mentioned; it is probably accidental, rather than intrinsically linked with Orcus. Cf. Setaioli 1997, 448. Petr. 45.9 quamdiu vixerit, habebit stigmam, nec illam nisi Orcus delebit; 46.7 artificium… quod illi auferre non possit nisi Orcus. Cf. the interpretations collected in Burman 1743, I, 69. Burman himself rejected the idea. Huxley 1970-1971, 69; Baldwin 1970-1971, 254; Baldwin 1979, 145; Bracht BranhamKinney 1997, 31 n. 3; Connors 1998, 52. Stöcker 1969, 120 and n. 3 takes esse to be only the infintive of edo, as shown by his placing a comma between esse and bene. Though H, the most authoritative manuscript, punctuates between these two words, this interpretation cannot be accepted; if it were so, bene should necessarily be taken with vivamus, but this would destroy the pregnant meaning of this verb, which will be shown to have a pivotal function in this poem. See below. Petr. 56.5. Edo appears two more times: at 66.3 (edi) and again at 141.9 (ed re: third person plural of the perfect: a parody of the historians’ style). Not surprisingly, comedo is used more often (seven times). E.g. AP 11.38.5-6 $$$ (cf. above, note 56). Cf. also several of the epitaphs discussed by Lier 1904, 56-63 and Lattimore 1942, 260-263 (above, note 53).

Trimalchio’s Epigrams (Petr. 34.10; 55.3) we shall come back below, urges to live (in the same pregnant meaning as in Trimalchio’s verse), because in the world of the dead it is not possible to eat well.76 Esse bene in the sense of “enjoying oneself” is a colloquial expression,77 which appears elsewhere in Petronius with no double meaning;78 but it is not disdained by artistic poetry,79 and therefore it does not contribute to the undoubtedly low literary level of the composition. What prevents it from rising above the hackneyed production known to us primarily through the funerary inscriptions is rather the accumulation of features like the esse double entendre, the metrical pattern, and the other details we have illustrated. A clear confirmation is offered by Seneca, in which the philosopher – a contemporary of Petronius’ – humorously apes those who, like Trimalchio,80 regard philosophy as no more than empty verbiage.81 In a few lines all the elements that make up Trimalchio’s poem reappear: esse as the infinitive of edo82 (here with no double entendre, but closely followed by another esse, this time the infinitive of sum); the appeal to the carpe diem justified by man’s mortal status; the verb vivere in the same pregnant meaning as it has in Trimalchio’s verse.83

76

77 78 79 80 81

82

83

Also Sen. ep. 123.10 (cf. below, note 81). For Isaiah 22.3 and Paul. 1 Cor. 15.3 see below, text to notes 97-98. I am referring to the inscription no. 480 in Peek 1960, 288 (no. 1367, Peek 1955, 409): ( !( " ( ( / $ / ) / & "/ (Rome, II/III century A.D.). The mention of eating, not accompanied by drinking, appears in “Sardanapalus’ epitaph” as reported by Athen. 8, 336A. Eating and drinking are mentioned in the version of the Anthologia Palatina (see below). Cf. Hofmann 1980, 337-338. Petr. 65.9 bene fuit; 92.4 oportet hodie bene sit. E.g. Hor. c. 3.16.43-44 bene est cui deus obtulit / parca quod satis est manu; Ov. fast. 4.399 bene erat iam glande reperta. Both are quoted by Burman 1743, I, 197. Trimalchio’s tomb must bear the inscription nec umquam philosophum audivit (Petr. 71.12). Sen. ep. 123.10 virtus et philosophia et iustitia verborum inanium crepitus est; una felicitas est bene vitae facere; esse, bibere, frui patrimonio, hoc est vivere, hoc est mortalem se esse meminisse. And farther down: non est istud vivere sed alienae vitae interesse. Here esse is accompanied by bibere. Cf. above, text to note 75. Trimalchio has abundantly developed the drinking theme in the speech preceding the poem: quare tangomenas faciamus (34.7: cf. above, text to notes 61-62); vita vinum est (ibid.: above, note 7). It may be not merely coincidental that in a letter (116.7) coming shortly before the one quoted in note 81 Seneca has the opponents of philosophy say: nos homunciones sumus; omnia nobis negare non possumus. 105

Chapter V Vivere in the pregnant sense of “enjoying life”84 amounts to a veritable leitmotif in the texts which develop the theme of Trimalchio’s verse. In the Senecan passage referred to it is opposed to the preceding vitae, which is devoid of pregnancy: a pattern also frequently found in funerary inscriptions.85 In Petronius’s context it appears shortly before the poem, in the host’s prose speech: eheu… ergo diutius vivit vinum quam homuncio… vita vinum est.86 There is another Senecan text closely reminding our passage, and the whole atmosphere of the Cena: the philosopher’s decription of the banquets daily given by Pacuvius, a governor of Syria, which, like Trimalchio’s, always ended with the host’s feigned obsequies and with a song repeating the refrain ' ' # '' .87 5. Thanks to Seneca, as we have seen, we can listen to those who, like Trimalchio, placed the fullness of life (“living it up”, so to speak) in the joys of the banquet, exemplified through eating and drinking; but the philosopher, on his own part, charges the verb vivere and the noun vita with a very different, actually with an opposite, pregnancy: according to him only philsophy can help us attain the fullness of life.88 The pattern opposing two words stemming from the same root (vita, vivere, etc.) is often found in him too, but what is advocated as a life fully lived is now conceived in the terms of the Stoic philosophical ideal.89 Seneca takes the cue from Pacuvius’ banquets, overhung by the shadow of death like Trimalchio’s is, to change the meaning of the refrain dictated by the governor of Syria’s mala conscientia (' ' '' ) by imparting to vivere the sense it acquires when man is directed by bona conscientia. To do so, he resorts 84 85

86 87

88 89

106

See Setaioli 1988, 482. Cf. e.g. the famous vivamus, mea Lesbia, et amemus (Catull. 5.1); Copa 38 (above, note 59); CE 1231.4 qui legitis, moneo: vivite, mors properat. E.g. CIL XII 4548 amici, dum vivimus, vivamus. Cf. also inscription no. 480 Peek 1960 (above, note 76), v. 2 ( !( ; and especially CE 190.7 (I transcribe the whole inscription, which can be usefully compared with the ideas we are discussing: adeste, amici, fruamur tempus bonum. / Epulemur laeti, vita dum parva manet, / Baccho madentes, hilaris sit concordia. / Eadem fecerunt hi cuncti cum viverent, / dederunt acceperunt, dum essent, fruniti sunt. / Et nos antiquorum emitemur (sic) tempora. / Vive dum vivis, nec quidquam denegaveris / animo indulgere, quem commodavit deus). Petr. 34.7. Sen. ep. 12.8. In this verb (' ' ) the nuance of the “conclusion” of life probably prevails: cf. Cic. Att. 12.2.2; 14.21.3. The Greek counterpart of Latin vivere in the sense of “enjoying life” is ( , that can be used with equal pregnancy. Cf. Setaioli 1988, 482 n. 2272. But ' too, in the sense of successfully concluding one’s life, was liable to acquire a similar meaning and could also receive a philosophical sense. See below, note 90 I refer to Setaioli 1988, 473-488 for further details. Sen. ep. 49.10 saepissime fieri, ut qui diu vixit parum vixerit; 93.3 non vixit ille, se in vita moratus est ; brev. 2.2 exigua pars est vitae qua vivimus (the famous sentence of the maximus poetarum, on which see my pages quoted in the preceding note).

Trimalchio’s Epigrams (Petr. 34.10; 55.3) to a famous Virgilian verse, while enhancing the egressive meaning of Virgil’s vixi by adding a nuance fitting his Stoic attitude.90 It is possible to follow the gradual transformation of the incitement to enjoy life in its fullness, in view of impending death, into an increasingly “philosophical” (or should we say Stoic?) rule of conduct. Seneca is of course a cornerstone in this evolution, and a further development appears in one of Plutarch’s texts we have previously quoted91 in connection with the Egyptian custom to bring the image of a corpse into the banquet hall. In this text92 that custom is explained not as an urge to enjoyment, but rather as a spur to virtue. But the theme’s transformation begins even before the official outset of Stoicism: with Crates of Thebes, one of whose disciples was Zeno of Citium, the founder of the Stoa. We know two lines attributed to him93 which are a reworking of the famous “epitaph of Sardanapalus”, whose Greek metrical version was attributed to Choerilus;94 but in lieu of the pleasures the king of Assyria regards as the only check against overpowering death, Crates allocates this function to learning and thought. Later on, Stoicism’s great systematizer, Chrysippus, rewrote the whole epitaph: the conscience of his own mortality, he warns, must not induce man to seek material pleasures, but learning and self-improvement.95 90

91 92 93

94

95

From Pacuvius’ “Epicurean” point of view having lived his life to the end meant having enjoyed all the pleasures it could offer. In Petronius himself this “philosophy” is expressed by Eumolpus: 99.1 ego sic semper et ubique vixi, ut ultimam quamque lucem tamquam non redituram consumerem (cf. Lucr. 3.915: above, note 58). Seneca shifts the meaning of the theme in a Stoic direction immediately after his report on Pacuvius: ep. 12.9 hoc quod ille ex mala conscientia faciebat nos ex bona faciamus, et in somnum ituri laeti hilaresque dicamus ‘vixi et quem dederat cursum fortuna peregi’ (Verg. Aen. 4.653) …; quisquis dixit ‘vixi’ cotidie ad lucrum surgit. Seneca quotes the same Virgilian verse also at beat. 19.1 (again with a reference to bona conscientia) and ben. 5.17.5. Cf. Setaioli 1965, 149-150 (also for the meanings acquired by Virgil’s verses when the philosopher detaches them from the original context). Above, note 12. Plut. conv. sept. sap. 2, 148AB. AP 7.326 * / # " ' & . The preceding epigram (AP 7.325) is a slightly altered version of lines 4-5 of Sardanapalus’ epitaph (cf. following note). Cf. Athen. 8, 336A + / " $ / , ' $ / , ' / " ' $ / [ . ' / $ ) ! ]. For the attribution of the Greek version to Choerilus, cf. Athen, 12, 529F; Strabo 14.5.9. Chrysipp., ap. Athen. 8, 336F / + !0 % " 337A ‘ + / " $ / , ! ! ! $ / 1 107

Chapter V The incitement to a purely sensual carpe diem in view of the ineluctability of death, which elicited these philosophical responses, is older than any philosophy96 and is found in all peoples and in all ages. Outside Greece, a much older testimony than Herodotus’ on the Egyptians appears in Isaiah, in reference to the Jewish people’s refusal to obey God’s call to penance.97 “Let us eat and drink, for tomorrow we die” were the words they uttered. The prophet’s condemnation is clear, but it is equally clear that Trimalchio could entirely underwrite these words. An expression identical with the one that, in the Septuagint, Isaiah places in the people’s mouth appears in Paul’s First Letter to the Corinthians.98 But for the promise of resurrection, Petronius’ great Christian contemporary would agree with Trimalchio and the cheerless wisdom of pagan epitaphs. A Greek funerary inscription we have repeatedly referred to closes on this gloomy notion: “from there nobody, once dead, wakes up to life”.99 But for Paul redemption has defeated death, so that it is no more necessary to exorcise it through sensual pleasure. Only a few lines after he will be able to utter the joyous cry: “Death, where is thy victory? Death, where is thy sting?”.100 Trimalchio’s “epigram” must then be considered in the context of its times. In this unquiet historical moment, as the crisis affecting old values was becoming more and more evident, our Petronian character blatantly ignores the new stimuli which led on the one hand the pagan Seneca to rejuvenate the old idea by filling it with a new philosophical content, and on the other the Christian Paul to transcend it by appealing to the sphere of the divine. Trimalchio obdurately sticks to the old patterns and ideas, but cannot escape the terror – and the fascination – of death, whose shadow threateningly overhangs his sumptuous banquet.101

96 97

98 99 100 101

108

/ " ! ’. Dunbabin 1986, 194 correctly remarks that this attitude is not at all restricted to the followers of Epicureanism. Is. 22.12 ' ! # ( 13 ( ' + 2 $ Paul. 1 Cor. 15.32 + $ No. 480 Peek 1960, v. 5 (cf. above, note 76). Notice the identical verb here and in Paul’s passage ( / ). Paul. 1 Cor. 15.54 355 3 I regard as a both idle and insoluble problem the attempt to establish whether, and to what extent, Petronius has sketched his character – and our scene in particular – taking into account the evolution of the old theme prompted by both Stoicism and Christianity. We can only say that he was surely acquainted with Stoic doctrines and that notions of Judaic, or even Judaic-Christian mold may have been not totally unknown at Nero’s

Trimalchio’s Epigrams (Petr. 34.10; 55.3) 6. We have already pointed out the structural import of Trimalchio’s second “epigram” in the whole architecture of the Cena, and the links connecting it with the first one.102 These unquestionable connections prompt me to assume that, in the fictional author’s intention, it was meant to repeat the same metrical pattern too. In the poem at 55.3, however, serious metrical flaws are apparent, to the point that both L and the Florilegia ( ) not only skip the final pentameter, bu present the first two lines as prose. By contrast, in H we read, after the first twot lines, a copyist’s gloss (distichon Trimalchionis cum elego suo) followed by the pentameter, and immediately after, when the prose resumes, the important words ab hoc epigrammate – omitted by L –, which provide Encolpius’ definition of the host’s short poem: “epigram”.103 The first knot to be untangled concerns the state of the text. Many scholars believe that in origin the verses were complete and correct and that the metrica irregularities should be imputed to faulty transmission;104 quite a few, thoughl ,

102 103

104

court, especially if Poppaea’s Judaic leanings are not an invention. For possible traces of Christian notions in the Satyrica cf. app.II. Cf. above, note 7. Stöcker 1969, 92 stresses the irony implied in Encolpius dubbing an “epigram” a poem in which two hexameters and one pentameter are substituted for the normal elegiac couplet. When Stöcker reads Trimalchio’s words at 55.2 as expressing the intention to write an epigram, he seems to take the word used by Trimalchio (inscriptio) as a calque of the Greek , which appears as a loan word in the words preserved by H immediately after the poem (epigrammate). The Latin and the Greek word seem indeed to be interchangeable in Petronius: cf. 103.2 frontes notans inscriptione sollerti ~ 103.4 fugitivorum epigramma ~ 106.1 inscriptione frontis… adumbrata inscriptione; 71.12 inscriptio (on Trimalchio’s tomb) ~ 115.20 dum epigramma mortuo facit. Heinsius already reconstructed two hexameters in the following way: quod non expectes, ex transverso fit et supra nos fortuna negotia curat (cf. Burman 1743, I, 319; Heinsius also corrected the transmitted super to supra: cf. Sen. ep. 95.57 haec supra nos natura disposuit; also nat. 2.32.7, 2.42.3. Cf., however, [Quint.] decl. 8.13, p. 163, 18-19 Håkanson volvitur super nos haec caeli siderumque compago). Heinsius’ reading is printed in the text in Müller’s editions after his first, and is accepted with no reservations by Connors 1998 52 and Rimell 2002, 192. Other scholars believe the lines should somehow be integrated: e.g. Cugusi 1967, 91; Barnes 1971, 312 n. 34; Courtney 2001, 99. Campanile 1974, 45-46 proposes the following in line 2: et supra nos Fortuna negotia curat. This proposal fails to carry conviction, because Fortune, far from minding her own business, constantly interferes with ours. Smith 1975, 147 thinks it probable that the text should be integrated, but subordinately admits that metric irregularities may be meant to convey a vivid idea of the throes of poetic composition Trimalchio is going through (this would be hinted at by non diu cogitatione distortus, 55.2: Smith accepts Fuchs’ correction of distorta to distortus). In my opinion the strongest argument of those who wish to integrate the text is the lack of any determination in reference to negotia. I believe, however, that both this and the metrical irregularities may be explained with the improvised character of these lines and with Trimalchio’s ignorance. 109

Chapter V contend that the text should not be touched,105 and I believe they are right. Most scholars agree that the meter was meant to duplicate that of 34.10, but it is clear for all to see that the goal was not reached. If the metrical faults are not due to transmission, then, it is clear that Trimalchio was not able to come up with correct verses on the spur of the moment. In this, we should add, he does not differ from many metrical inscriptions, whose verses are often gravely defective. Some have proposed a different metrical pattern, but these attempts fail to carry conviction.106 We should not forget the different situations in which Trimalchio recites the two short poems. The first one, as we have seen,107 is part of the host’s accurate stage direction, is in some way announced by his previous prose speech (which opens with the same word, and contains the most eye-catching term that will later appear in the poem), and is effortlessly recited by heart, no doubt at a prearranged moment, signaled by the appearance of the silver skeleton, just like the prose speech took the cue from the serving of the one-hundred-year-old wine. The occasion for the second poem, by contrast, is an acrobat’s fall, which injures Trimalchio himself. If the accident is anything but unexpected in Petronius’ narrative strategy and in the readers’ expectation (Horatius’ cena Nasidieni is clearly hovering in his mind),108 it can hardly be denied that it is probably the only not carefully prepared “number” in the Cena.109 This is confirmed by the fact that Trimalchio does not have a ready-made epigram fitting the situation, 105 106

107 108

109

110

E.g. Walsh 1970, 128; Baldwin 1970-1971, 254-255; Slater 1990a, 161-162 and n. 11. Lastly Jensson 2004, 10-11. More will be mentioned in the following notes. Marmorale 1947, 94 believes the three lines to be a defective senarius, a limping hexameter, and a pentameter (but on p. 24 he states that they are two hexameters and a pentameter). The first line is taken to be a senarius also by Tandoi 1967, 77 and Gagliardi 1989, 20 n. 34. According to Woodall 1970-1971 Trimalchio meant to compose a poem consisting of three pentameters, of which only the last scans correctly (he does not adduce any metrical parallel; we should remember, however, that Kaibel 1887, 702 lists three metrical inscriptions made up of two pentameters). Bücheler 1862 acknowledges in the apparatus the attempt by L. Schmidt to read the first two lines as a quotation from a more estensive composition including the last part of an hexameter and an almost complete one. But it is clear that the poem is an impromptu production by Trimalchio. Above, § 3, beginning. Cf. Hor. sat. 2.8.55-77 (the fall of the curtain, the scared guests, the commonplaces on fortune’s blind omnipotence, etc.). The contacts with Petronius’ Cena are numerous, despite the reservations expressed by Smith 1975, 146-147. Cf. Révay 1922, 209; Perrochat 1939, 76-77; Sullivan 1968, 211. Gagliardi 1989, 19 and n. 31 correctly points out that it would have been all but impossible to prearrange the acrobat’s impact on the landlord so as to avoid serious injuries. It should rather be noticed that Trimalchio readily takes advantage of this accident and succeeds in including it, a posteriori, in the Cena’s careful stage direction through his attempt at literary transfiguration and the double entendre which precedes it (hunc casum). According to Courtney 2001, 99 and n. 38 this accident was probably prearranged too. Cf. lastly Yeh 2007, 96.

Trimalchio’s Epigrams (Petr. 34.10; 55.3) and not only must he ask for the codicilli to write down an appropriate composition, but is also forced to undergo the throes of poetic meditation, though he is not able to concentrate for long.110 It is hardly surprising that the resulting verse is anything but impeccable; but it cannot be doubted that, in his intention, it was meant to duplicate the metrical pattern of 34.10 (which is, then, another “epigram”). Aside from the metrical irregularities, the short composition does not offer insurmountable problems. In the first line the expression ex transverso111 is originally a military metaphor,112 commonly referred to something unexpected.113 The idea expressed by Trimalchio is naturally a hackneyed one.114 We hardly need to follow the history of the widespread theme of fortune’s blind omnipotence. Suffice it to remark that neither here nor in the previous poem does Trimalchio succeed in rising above the commonplace, much less in attaining the level of poetry, despite the last line’s Horatian echoes,115 not unlike those of famous poets in the previous epigram.116 Between 55.3 and 34.10 there is, however, an unquestionable difference. All in all, the theme of fortune remains an occasional theme, linked to a particular accident. The poem at 34.10, by contrast, though not rising above the level to which its author’s modest poetic endowments confine it, develops a topic which Trimalchio profoundly feels: death is his obsession, which casts its shadow over the whole Cena. Unlike Seneca and Paul, he has no other resort but a precarious escapism through pleasure and 110

111 112 113 114

115

116

Petr. 55.2 nec diu cogitatione distorta. Surprisingly, some interpreters regard distorta as an accusative plural of the neuter gender and refer it to the composition’s metric irregularities (distorta haec recitavit). So Baldwin 1970-1971, 254-255; Woodall 1970-1971, 256 and 257; Barnes 1971, 312 n. 34. Sochatoff 1969-1970, 342 cannot decide whether distorta should be referred to cogitatione or to haec. Baldwin 1970-1971, 254-255 states that if distorta were referred to cogitatione, it would be idle after non diu. How does he explain, then, non diu cogitatione, that with his interpretation is left hanging in the air? Fuch’s correction distortus is needless, though accepted by Smith 1975 (above, note 104). Meaning “suddenly”, “unexpectedly”, not “a rovescio”, as translated by CesareoTerzaghi 1950, 43. Cf. Liv. 1.13.1; 2.20.3; 3.62.8; 10.41.5; 33.18.18; 37.42.5. E.g. Cic. acad. 2,121; Sen. beat. 15,6; vd. OLD s.v. transversus (to which Sen. ep. 117.21 and Querolus p. 21, 6 Ranstrand may be added). Barnes 1971, 304-305 refers to the ending of some Euripidean dramas (Alcestis, Andromacha, Bacchae, Helena, Medea). These all end with the same five lines (in the Medea the first of the five is different). Lucian. symp. 48 (already quoted by Stubbe 1933, 164) adduces three of these lines: / / $ Petr. 55.3.3 quare da nobis vina Falerna, puer ~ Hor. c. 2.11.18-20 quis puer ocius / restinguet ardentis Falerni / pocula praetereunte lympha? Cf. also Catull. 27.1 minister vetuli puer Falerni. Cf. above, text to notes 57-59. 111

Chapter V wine. In his eyes this is the universal remedy, the only one enabling man to forget all threats: impending death as well as fortune’s whims: quare da nobis vina Falerna, puer.

112

Chapter VI Perverted Taste (Petr. 55.6; 93.2)* 55.6

Luxuriae rictu Martis marcent moenia. Tuo palato clausus pavo pascitur plumato amictus aureo Babylonico, gallina tibi Numidica, tibi gallus spado; ciconia etiam, grata peregrina hospita, pietaticultrix gracilipes crotalistria, avis exul hiemis, titulus tepidi temporis nequitiae nidum in caccabo fecit tuae. Quo margaritam caram tibi, bacam Indicam? An ut matrona onerata phaleris pelagiis tollat pedes indomita in strato extraneo? Zmaragdum ad quam rem viridem, pretiosum vitrum? Quo Carchedonios optas ignes lapideos? Nisi ut scintillet aequitas e carbunculis. Aequum est induere nuptam ventum textilem, palam prostare nudam in nebula linea?

HL(=lmrtp)O(=BRP) 2 pascitur Scaliger: nascitur 3 aureo L: auro HO Babylonicus Fraenkel 6 pietaticultrix L: pietatis cultrix HO 7 hieme Fraenkel 8 tuae Fraenkel: tuo Heinsius: modo Bücheler: meae Dell’Era: meo codd. Marmorale 9 margaritam caram Ribbeck: -a cara tibi bacam Indicam Heinsius: tibi bac(c)a Indica Ol: tribac(c)a Indica Hmrtp 10 onerata mrtp: ornata HO 14 e B: est HLP: et R carbunculis Bücheler: -lus Hlmt : -os Orp 15 induere lp Samb. Scaliger: inducere 16 linea lmtp Samb. Scaliger: lun(a)e rHO

5

10

15

Chapter VI 93.2

Ales Phasiacis petita Colchis atque Afrae volucres placent palato quod non sunt faciles, at albus anser et pictis anas involuta pennis plebeium sapit. Ultimis ab oris attractus scarus atque arata Syrtis siquid naufragio dedit, probatur; mullus iam gravis est. Amica vincit uxorem. Rosa cinnamum veretur. Quicquid quaeritur optimum videtur.

5

10

L(=lrtp) 2 Afrae Putean.: aeriae 4 involuta Busche: innovata seu ingravata seu aemulata Bücheler: renovata codd., defendit Cugusi: alii alias 7 probatur : -amus L

1. These two poems, which, as we shall see, have many traits in common, present several textual problems. The text adopted here agrees in general with that given by Müller’s editions later than the first one, with one exception: onerata in lieu of ornata at 55.6.10. After Burman,1 the one modern editor choosing onerata (i.e. the reading that can be referred to L) seems to be Carlo Pellegrino,2 though in his commentary he gives no reason for this choice. In my opinion, though, there are multiple reasons to prefer onerata. In the first place we should call attention to the numerous texts in which the pearls worn by the matronae to enhance their beauty – such are the phalerae pelagiae of Petronius’ matrona, which clearly pick up the margarita mentioned in the preceding line – are described as a “weight”; some may even be either a poetic antecedent3 or a target of Petronius’ irony.4

* 1 2 3

4

114

A version of this chapter has appeared with the title Le due poesie in Petr. Sat. 55.6 e 93.2, “Prometheus” 35, 2009, 237-258 Burman 1743, I, 359. Pellegrino 1975, 87. One such antecedent is probably Hor. epod. 8.13-14 nec sit marita quae rotundioribus / onusta bacis ambulet, where even the word designating the pearl is the same: baca (cf. also Hor. sat. 2.3.241). The situation described in Horace’s epode is not far from Petronius’ adulterous matrona (Frassinetti 1954, 387-389 provides information on the sexual roles which may be described by the expression tollat pedes of line 11, and clearly shows that it can fit the feminine role). Such is probably Sen. ben. 7.9.4 iam enim exercitatae aures oneri ferendo sunt. In the next paragraph we find a reference to women wearing transparent dresses closely re-

Perverted Taste (Petr. 55.6; 93.2) Secondly, Petronius himself employs the same word to describe something that should be not a weight, but an ornament.5 Finally, the exchange between onerat- and ornat- is common in manuscripts,6 and both H and O, i.e. the branches of tradition that transmit ornata instead of onerata, present many more instances of “normalization”, which, though seemingly enhancing linguistic regularity, amount to as many specimens of lectio facilior, and must therefore be rejected.7 I shall only briefly point out a few more details. At the end of line 8 I accept Fraenkel’s correction (tuae) in lieu of the manuscripts’ meo.8 Tuae seems to be required by the insistent recurring of the second person in the poem.9 At line 9 Ribbeck’s correction quo margaritam caram in lieu of -a cara given by the tradition appears to be necessary. In elliptic interrogative clauses introduced by quo the accusative is normally employed.10

5 6 7

8 9

10

minding of Petronius (sericas vestes… quibus sumptis parum liquido nudam se non esse iurabit). Cf. also Sen. const. 14.1 quam oneratas aures. More parallels in Ovid: ars 3.129-130 vos quoque non caris aures onerate lapillis / quos legit in viridi decolor Indus aqua; med. fac. 21-22 induitis collo lapides oriente petitos / et quantos onus est aure tulisse duos. Habinnas’ crowns: Petr. 65.7 oneratusque aliquot coronis. Elsewhere it is clothing that is described as a weight: 32.2 oneratas veste cervices. As already remarked by Burman 1743, I, 359. At line 3 HO give auro againts aureo (L). This seemingly results in a more regular construction, since plumatus, in the sense of “embroidered” can be in a natural way determined by auro (cf. Lucan. 10.125 pars auro plumata nitet); also, the double epithet in reference to Babylonico (“carpet”) would be avoided. Indeed, in order to avoid it, Frankel corrected to Babylonicus, referred to the peacock. At line 6 HO have pietatis cultrix, against pietaticultrix (L), a normalization avoiding the hapax legomenon. At line 3 metrics itself confirms that the normalization cannot be accepted. Not so with ornata (v. 10), but here too, like in the other instances, it is surely more banal than onerata. Accepted by Marmorale 1947, 96. Petr. 55.6.2 tuo; 4 tibi… tibi; 9 tibi; 13 optas. The correction modo, proposed by Jacobs, adopted by Bücheler and some later editors (Friedlaender 1891, 136; Heseltine 1913, 98; Pellegrino 1975, 85), and more recently defended by Salanitro 1996, 300-301 n. 1 (also offering a survey of the various proposals), is based on the idea that 55.6 is a quotation from Publilius Syrus and therefore belongs to the late Roman republic, when the habit to serve storks at the table was recent. We shall see, however, that the poem cannot be Publilius’. Dell’Era 1995 proposes meae with the curious argument that the insistent recurrence of the second person should require a shift to the first person “in una posizione centrale e eccezionale”. Burman 1743, I, 358 already referred to Hor. epist. 1.5.12 quo mihi fortunam, si non conceditur uti? (cf. also Smith 1975, 149); Ov. am. 3.4.41 quo tibi formosam, si non nisi casta placebat?; Phaedr. 3.18.9 quo mi… mutam speciem, si vincor sono?; Mart. 5.35.2 quo tibi vel Nioben, Basse, vel Andromachen? 115

Chapter VI Line 14 must be understood as an ironic remark on the part of the speaker: (“unless you seek the Carchedonios ignes lapideos so that uprightness may receive light from rubies”11 – when it cannot be doubted that the craving for precious stones is a mark of moral corruption). So, no question mark must be placed at the end of line 14, as done in many popular editions12 and even in some critical ones.13 The most momentous textual problem posed by the poem at 93.2 is found at line 4. The reading unanimously transmitted by tradition14 is et pictis anas renovata pennis. According to some renovata is meaningless,15 while others refer it to the shifting colors of the duck’s feathers;16 but the most serious problem posed by this word concerns metrics, inasmuch as with renovata line 4 would not be a Phalaecean hendecasyllable, like all the other lines in the poem. Some scholars seem not to have realized this problem,17 but Burman already recorded many attempts to heal the line’s metrics.18 The traditional reading has had its champions too. Marmorale,19 who proposes a late date for the Satyrica, saw in our line the proof that Petronius had become insensible to syllabic quantity. He has been followed by Cugusi,20 who, like Marmorale, believes line 4 to be a Sapphic hendecasyllable inserted in the series of Phalaeceans. He points out a metrical inscription in which this happens 11 12

13 14

15 16

17

18 19 20

116

Rightly Connors 1998, 57: “unless you want it so that moral rectitude may gleam amid the carbuncles”. Courtney 1991, 22 regards the verse as ironical too. E.g. Ciaffi 19672, 158; Canali 1990, 90; Aragosti 1995, 260; Reverdito 1995, 82; Scarsi 1996, 72. The untenability of the question mark at the end of both line 13 and line 14 is made evident by the very translations accompanying the text in these editions (and in Ernout’s: cf. following note) E.g. Heseltine 1913, 98; Ernout 1923, 53. Cesareo-Terzaghi 1950, 44 place a question mark at the and of line 14, but not of line 13. Contrary to what appears in some editors’ critical apparatuses (e.g. Bücheler, Ernout, Cesareo-Terzaghi) the readings of the Florilegia too ( ) may be said to be based on an original anas (not on avis): cf. Hamacher 1975, 137. Bücheler 1862, 111 places line 4 after line 2, adopts avis in lieu of anas, and refers the line to the peacock (which in reality is not mentioned in this poem), perhaps being influenced by passages like Phaedr. 3.18.8 pictisque plumis gemmeam caudam explicans and the others quoted by Connors 1998, 58 n. 26. He gives up the verse transposition and adopts anas in his later editions. So, lastly, Habermehl 2006, 245. So Ernout 1923, 97: “le canard dont les plumes bariolées prennent mille teintes”. Cf. also Ciaffi 19672, 327; Canali 1990, 157; Aragosti 1995, 367; Scarsi 1996, 135. Reverdito 1955, 155 prefers no to translate renovata (“l’anitra dalle penne screziate”). E.g. Paratore 1933, II, 311: “levigatezza estrema, senza l’ombra di una licenza metrica”; “accuratezza della tecnica metrica”. Cf. also Ciaffi, Canali, Aragosti, Scarsi, and Reverdito, quoted in the preceding note. All print renovata with no comment. Burman 1743, I, 581. A more up-to-date list in Stubbe 1933, 173; an exhaustive, indepth discussion in Habermehl 2006, 245. Marmorale 1948, 292-293, who also offers a survey of proposed corrections. Cugusi 1967.

Perverted Taste (Petr. 55.6; 93.2) three times.21 Cugusi has been convincingly refuted by Vincenzo Tandoi,22 but it should be added that the traditional reading (et pictis anas renovata pennis) does not really offer a Sapphic hendecasyllable, even though it is close to that metrical pattern.23 This invalidates Cugusi’s argument at the very root. A critical survey of all the more important proposals to correct the corrupt word (no doubt renovata) can be found in Habermehl’s excellent commentary.24 Like him – and like Müller, in his editions later than the first one – I believe the most likely correction, though of course no absolute certainty can be reached, to be involuta, which was suggested by Busche.25 In my opinion the latter’s most convincing argument – aside from paleographical plausibility – is the parallel he established with 55.6.2-3, where the peacock is said to be covered with – or rather be “dressed in”: amictus – a plumage described as a carpet embroidered in gold. It should indeed not escape us that, if we accept involuta, in both poems a variegated plumage is described as a dress, with two expressions that Petronius has also used elsewhere, one shortly after the other, in reference to clothing.26 This is the first of the many common traits linking our two poems. 2. The poem recited at 55.6 by Trimalchio is in senarii, which several scholars have judged to resemble those of comedy, of what scanty remains of the mime have been preserved, and of Phaedrus’ fables, and so to mirror a rather obsolete taste, when compared with Eumolpus’ tragic trimeters the latter employs in his Troiae halosis at chapter 89.27 Nevertheless, this poem’s basic metrical correctness, as well as its connection with a strand of bona fide literary production, cannot but strike the reader who remembers the two poems previously recited by Trimalchio, especially the second one, which precedes it by only a few lines

21 22 23 24 25

26

27

He refers to CE 1504. The Sapphic hendecasyllables interrupting the series of Phalaeceans are lines 6, 10, and 23. Tandoi 1968, 77. With et pictis anas renovata pennis we have this sequence : spondee, trochee, dactyl, trochee, spondee. Cf. Cesareo-Terzaghi 1950, 82. Habermehl 2006, 245. Cf. above, note 18. Busche 1911, 455. Nisbet 1962, 229 pointed out a parallel to confirm involuta: Mart. Cap. 9.918 nam candidus nivosis / olor involutus alis. It is regarded as non-pertinent by Courtney 1991, 27 (who erroneously attributes the reference to Busche himself). Petr. 26.10 amicimur ergo diligenter; 28.6 involutus coccina gausapa. Amictus in this sense also appears at 65.3 and 97.2. Amiculum at 11.2. Involutus in the sense of covered (with clothing) also at 63.6 (involvere at 26.1; 54.4; 64.6). Cf. Stubbe 1933, 164; Barnes 1971, 45-46. Lastly Yeh 2006, 471-482, who, after a thorough analysis, comes to the conclusion that the meter of 55.6 is not really archaic, but rather archaizing (p. 478). Already Collignon 1892, 236 saw in our poem “une certaine affectation d’archaïsme”. 117

Chapter VI (55.3)28 and is extremely incorrect from the metrical point of view, even when compared with the uncultured, but basically respected metrical pattern of his first “epigram” at 34.10 This discrepancy admits of various explanations and we shall go back to it at the moment of tackling the problem of the authorship of the verse at 55.6, which Trimalchio presents not as his own, but as a quotation of Publilius Syrus. The twofold structure of our poem has been repeatedly pointed out.29 It is true that the first part is about the pleasures of the table, and the second about ornaments of the body, but we do not have two parts of equal length each comprised of eight lines. In reality the first line states the theme of the whole poem – the luxury perverting Rome –, which is then developed in the next seven verses through the mention of choice foods, illustrated through a list of birds made rare and sought after either by being exotic, or artificially altered, or caught and used as food despite the respect they would deserve.30 The second half of the poem is characterized by the shift from assertions to a series of rhetorical questions, breathlessly (and sarcastically) succeeding one another to the end of the composition. Six lines are devoted to precious stones, the final two to transparent clothing. The thread connecting these parts and unifying the poem is the theme of luxury and corruption, already implied in the first verse (marcent), then picked up in the middle, still in the part devoted to gastronomic pleasures (nequitiae, v. 8), and fully developed in the pictures of sexual depravation, and also through the ironical hint at rectitude (probitas, v. 14), which can hardly draw its splendor from gems. The opinions as to the poetic worth of this poem range from the most flattering appreciation to the most extreme scorn.31 They are only of limited interest to us. 28

29

30

31

118

Cf. e.g. Yeh 2007, 471. The interesting implications potentially connected with this correct remark are unfortunately nullified by Yeh’s startling interpretation (pp. 96-98) of Trimalchio’s two “epigrams” as part of one poem. See ch. V. Cf. e.g. Giancotti 1967, 270; Yeh 2007, 482. According to Connors 1998, 57-58 the peacock’s multicolored plumage (vv. 2-3) and the transparent dress (vv. 15-16) frame the poem. To uphold this idea she might have referred to Vopisc. Car. et Carin. et Numer. 20.5 iam quid lineas petitas Aegypto loquar? Quid Tyro et Sidone tenuitate perlucidas, plumandi difficultate pernobiles? (where these clothes are described with traits that in Petronius refer both to the dresses – lineas – and to the peacock – plumandi). See Paschoud 2001, 409. Another link connecting the two parts of the poem is the resemblance of the peacock’s feathers to the gems described in the second part (cf. Phaedr. 3.18.8 and related texts: above, note 14) Notice that four of these seven lines (vv. 5-8) are devoted to the “pious” stork at the end of the list (Salanitro 1996 demonstrates that the stork’s pietas refers primarily to its behavior toward its parents); two more (2-3) describe the peacock, which opens it. The guinea fowl and the capon, which are in the middle, are dispatched in just one verse (4). Steele 1919-1920, 286: “a poet of no little merit”; Paratore 1933, II, 171: “versi di fattura solida e robusta”; Maiuri 1945, 188: “esercizio di bravura… un fine saggio di

Perverted Taste (Petr. 55.6; 93.2) A conspicuous characteristic of our poem is the insistent recurrence, from beginning to end, of alliteration, which in many scholars’ view is meant to impart a hyperarchaic flavor.32 Other traits that appear to be affectedly sought after are the heaping of polysyllabic words (two of them long compounds),33 and the triple recurrence, in the last verses, of an exchange between noun and adjective.34 What is most interesting, however, is the peculiar slant our poem imparts to the widespread theme protesting luxury at the table35 (to which the polemic against extravagance in bodily ornament is added)36. The idea that the rare and the exotic are for this only reason preferred to the domestic and easily obtained is certainly recognizable here too, but it is not alone and undisputed, as it is in the poem we have paired with Trimalchio’s senarii from the very beginning, namely Eumolpus’ Phalaeceans at 93.2. In the first place rarity and high price seem to count more than exotic origin: peacocks and guinea fowls were certainly rare and luxurious articles, but in Petronius’ time they had been acclimated in Italy for a long time;37 and though the stork, peregrina hospita, came from afar, it was nevertheless captured in Italy to be served at the table.38 We have already hinted at the transgressive character of this fad: the pietas so connatural to the stork as to make it the symbol of this virtue so appreciated by the Romans39 did

32

33

34

35 36 37

38 39

poesia”. Following in the wake of Maiuri, Marmorale 1947, 95; Walsh 1970, 127 and n. 2; Yeh 2007, 484. At the opposite extreme, cf. Slater 1990a, 186: “some of the most deliciously awful poetry in the Satyricon”; Rimell 2002, 193 n. 34: “awful poem”. A strongly negative judgment on our poem is passed by Barnes 1971, 50 and 61 n. 6 too. Petr. 55.6.1 Martis marcent moenia; 2-3 palato… pavo pascitur / plumato; 7 titulus tepidi temporis; 8 nequitiae nidum; 10 phaleris pelagiis; 16 palam prostare; nudam in nebula. Cf. e.g. Collignon 1892, 237; Barnes 1971, 46. The hyperarchaic flavor of this trait is stressed e.g. by Bendz 1941, 54; Slater 1990a, 186; Courtney 1991, 21; Connors 1998, 57; Yeh 2007, 479-482 Petr. 55.6.6 pietaticultrix gracilipes crotalistria. Most scholars see an archaizing touch in the compounds: e.g. Sullivan 1968, 193; Smith 1975, 149; Courtney 1991, 21; a different opinion is voiced by Barnes 1971, 47. Petr. 55.6.13 ignes lapideos; 15 ventum textilem; 16 nebula linea. The artificiality of the triple figure is stressed e.g. by Bendz 1941, 54-55; Courtney 1991, 22: “this is (intentional) sowing with the whole sack”. A detailed survey will be found in Habermehl 2006, 241-243. For the polemic against jewels and transparent dresses regarded as a mark of corruption see the parallels already quoted by Burman 1743, I, 358-358; also above, notes 3-4. For the peacock cf. Cic. fam. 9.18.3; 9.20.2; Varr. r. r. 3.6; Hor. sat. 1.2.116. It was served at the table for the first time by Hortensius (Varr. r. r. 3.6.6; Plin. NH 10.45). Its eggs were very expensive (cf. Varr. r. r. 3.6.6); they are served at Trimalchio’s table (Petr. 33.4-5). For the guinea fowl cf. e.g. Hor. epod. 2.53; Varr. r. r. 3.9.1; 3.9.18; Mart. 13.73; Plin. NH 10.132; Colum. 8.2.2. The fad of serving up storks was initiated by Petronius Rufus, on whom see Hor. sat. 2.2.49-50, with Porphyrion’s commentary. Cf. Salanitro 1996, with the literature quoted and discussed. 119

Chapter VI not prevent the epicures from serving it up at the table. The most important signal, however, lies in the mention, restricted, perhaps intentionally, to just two words, of a bird not rare in itself, but artificially made pleasing to the palate: the capon (gallus spado).40 A criticism of the contemporary taste for artificiality – actually, for unnaturalness – is here in the foreground.41 This detail may be usefully associated with another poetic passage, which, like the poem at 93.2, is often paired with our verse:42 the opening part of the Bellum civile recited by Eumolpus. There, besides clear textual parallels with the poem at 93.2,43 we find a detail that helps us understand the real import of the reference to capons at 55.6: the castration of young boys to make them into objects of unnatural pleasure.44 Here the taste for unnaturalness, which is only hinted at at 55.6, is expressly emphasized: quaerit se natura nec invenit. The portion of our poem which is devoted to jewels has been associated by some with a poetic fragment of Maecenas, quoted by Isidore of Seville,45 mentioning emeralds and pearls, which also appear in the poem recited by Trimalchio. This association has no doubt been prompted by the coupling of the latter with Augustus’ powerful minister, which seems to be suggested by Trimalchio’s complete name: C. Pompeius Trimalchio Maecenatianus.46 Some scholars believe Petronius’ wealthy freedman to be the noble Etruscan knight’s provincial counterpart.47 It cannot be denied that certain details characterizing Trimalchio in the Satyrica appear to be reminiscent of Maecenas’ description we read in one

40

41 42 43

44

45

46 47

120

For the capon cf. e.g. Varr. r. r. 2.7.15; Colum. 8.2.3; Mart. 3.58.38; 13.63; Apic. 4.3.3, and see Bruneau 1985, 547-549. Ernout 1923, 52 peculiarly translates gallus spado with “chapon gaulois”. Petronius’ polemical attitude to artificiality and unnaturalness is stressed (in general) by Sandy 1969, 296-297. E.g. by Barnes 1971, 311 n. 29; Courtney 1991, 27; Yeh 2007, 86. E.g. Petr. 119.7-8 non vulgo nota placebant / gaudia, non usu plebeio trita voluptas ~ 93.2.5 plebeium sapit; 119.33-34 Siculo scarus aequore mersus / ad mensam vivus perducitur ~ 93.2.5-6 ultimis ab oris / attractus scarus; 119.36-37 iam Phasidos unda / orbata est avibus ~ 93.2.1 ales Phasiacis petita Colchis. Petr. 119.20-24 Persarum ritu male pubescentibus annis / surripuere viros exsectaque viscera ferro / in venerem fregere, atque ut fuga mobilis aevi / circumscripta mora properantes differat annos / quaerit se natura nec invenit. Maecen. fr. poet. V Avallone (Isid. orig. 19.32.6) lucentes, mea vita, nec smaragdos / beryllos mihi, Flacce, nec nitentes / nec percandida margarita quaero / nec quos Thynica lima perpolivit / anulos neque iaspios lapillos. On this fragment see Setaioli 2000, 262. Known to us from the inscription on Trimalchio’s tomb: Petr. 71. 12. The idea was suggested by Steele 1919-1920, 283-284; it was later picked up by Sullivan 1969, 137-138, and, with an express reference to Maecenas’ fragment, by Baldwin 1984. See, lastly, Byrne 2006, 99-102.

Perverted Taste (Petr. 55.6; 93.2) of Seneca’s epistles;48 but it is hardly possible to find anything in common between the crass ignorance of Petronius’ freedman and Maecenas’ refined culture.49 The scholar who has advanced farthest on this path is no doubt Shannon Byrne, who contends that Seneca’s Maecenas was meant as a veiled, but clearly unfriendly, portrait of Petronius. The latter, she goes on, was well aware of Seneca’s intention, and recycled some details of the philosopher’s description of Maecenas in Trimalchio’s parodic characterization.50 If it were so – the conclusion seems unescapable – Petronius’ Trimalchio would be a parody of Petronius himself! In reality, a close analysis of the two poems (Maecenas’ and Trimalchio’s) will show that the association is hardly fitting: Maecenas’ verse expresses the repudiation of gems, whereas Trimalchio’s associates them with moral corruption. Aside from this, we happen to know a real parody of Maecenas’ poem. It appears in a letter of Augustus partly reported by Macrobius.51 Augustus does not miss the rare neuter form margaritum which Maecenas had used in his poem. In Trimalchio’s verse, by contrast, the more common margarita is employed, although margaritum was not unknown to Petronius.52 The connection of the poem’s first part with the widespread “gastronomic” theme is much more tangible. This topic already appeared in Ennius’ Hedyphagetica,53 probably with satiric intentions. In Horace it is much developed in the second book of the Satires, but it appears elsewhere too. Horatian echoes can indeed be recognized in the verse at 55.6.54 The influence of Varro’s satiric writing has been surmised too. One of Varro’s Menippeae contained a list of tasty foods written in senarii, like Trimalchio’s poem.55 48

49 50 51

52 53

54

55

Sen. ep. 114.6 sic apparuerit (Maecenas), ut pallio velaretur caput exclusis utrimque auribus ~ Petr. 32.2 pallio enim coccineo adrasum excluserat caput. The two passages are placed side by side by Steele 1919-1920, 284; Sullivan 1968, 137. Byrne 2006, 99 correctly adds that in the same passages both Trimalchio and Maecenas are accompanied by two eunuchs. Although Byrne 2006, 100 asserts that “both are bad poets with similar tastes”. Byrne 2006, 106-107. August. ap. Macr. Sat. 2.4.12 Tiberinum margaritum, Cilniorum smaragde, iaspi Iguvinorum, berulle Porsenae. All the gems mentioned in Maecenas’ fragment reappear in this text. Cf. also Suet. Aug. 86.2. Petr. 63.3. Cf. Heraeus 1937, 113; Cavalca 2001, 107-108. Cf. above, note 35 (reference to the detailed survey in Habermehl 2006, 241-243). Cf. e.g. Hor. sat. 2.2.22-28, pointed out by Sullivan 1968, 193 n. 1; also epod. 2.53, quoted above. Some of these Horatian texts have probably influenced Petr. 93.2 too. Another Horatian poem, epod. 8, is echoed only in Petr. 55.6 (above, note 3). I am referring to the Menippean satire (Menipp. 403 Bücheler = Astbury = Gell. 6.16.1-5). The reference is made already by Collignon 1892, 285. More recently, e.g., Giancotti 1967, 273; Walsh 1970, 127; Smith 1975, 148. 121

Chapter VI 3. All scholars, almost with no exception, have emphasized the blatant contradiction between the verse recited by Trimalchio against luxury and extravagance on the on hand, and his own lifestyle on the other.56 As far as food is concerned, for example, the rich freedman censures serving peacocks at the table, but offers peacock-eggs to his guests;57 although he criticizes the matronae’s phalerae pelagiae, he possesses cursores phalerati.58 And the whole Cena is an obvious specimen of the unbridled luxury the verse at 55.6 inveighs against.59 This state of things must be kept in mind before tackling the most intricate problem posed by our poem: that of its authorship. As is well-known, Trimalchio takes the cue to produce these lines from the discussion about poets sparked by his recitation, a few lines before, of his second, metrically very defective, “epigram”.60 Trimalchio cuts short the discussion61 by asking the rhetorician Agamemnon what difference he thinks there is between Cicero and Publilius Syrus;62 but he does not leave him the time to reply and proceeds to answer his own question, by stating that in his opinion Cicero is disertior and Publilius is honestior.63 Immediately after, he produces the verse at 55.6, preceded by a rhetorical question: quid enim his melius dici potest? It is clear, then, that the poem is presented by Trimalchio as the work of Publilius Syrus. Before we tackle the problem of the credibility of Trimalchio’s ascription, it will be necessary to try to ascertain the meaning and import of the comparison 56

57 58 59

60 61

62

63

122

I limit myself to some of the more recent scholars stressing this contradiction: Giancotti 1967, 244; Sullivan 1968, 192; Sochatoff 1969-1970, 344; Walsh 1970, 127; Barnes 1971, 43-44, 48-49; Smith 1975, 148; Sandy 1976, 287; Gagliardi 1980, 76; Courtney 1991, 21; Panayotakis 1995, 86; Walsh 1996, 174; Conte 1996, 119; Connors 1998, 60; Courtney 2001, 197; Yeh 2007, 485. Cf. above, note 37, and see Cavalca 2001, 50 n. 61. Petr. 28.4. Cf. Cavalca 2001, 133. Connors 1998, 60-61 also emphasizes the contrast between the condemantion of gems at 55.6.9-14 and Fortunata and Scintilla displaying their jewels in chapter 57. She does not point out that to Habinnas a pearl is nothing but a faba vitrea (67.10), just like Trimalchio calls the emerald pretiosum vitrum in his poem (55.6.12). See below. See ch. V. I do not dwell on the detail of the palm of poetry allotted to the unknown Mopsus of Thrace (55.4). This is hardly surprising at Trimalchio’s table. The host mentions Hannibal as the conqueror of Troy (50.5), Cassandra murdering her children, and Niobe shut up in the Trojan horse by Daedalus (52.1-2). The manuscripts have Publium, corrected by Bücheler to Pub