1 ■ 1 ■ i i \ I > 2 % s^ -t %-i^... \ oJ i , — =3 '- — • >- ,^\WtUNIVERi, •'JidjpfViur •*oaiAmir:in>' wAU»aan 3»- vAUvaaii 2 ^ ^(!/0dIlV3JO>^ A^^EUNIVER% v^lOSANCEl o %a3AINfl-3V AOfCALIF0/?^> C5 c-^ \\\EUNIVER5/4 % o ?3 ^ILIBRARY( ■%JITV3J< \WEUNIVER5'/A ^ovlOSANCEL5x> >• \^ ^n\wim^ "^AajMNiiJWV •^ ^OFCAIIFO/?^ ^^OFCAIIFO; ^<5Aavaaii: ^lLIBRARY/9/^ ^IIIBRARYO^ '" ■:£■ ^ ' ■:£■ ^^WEUmVER% ^ % ^'^ f ^lOSANCEli %a3AINfl-31 oe «:, j.OFrAiiFnp,f OFfAllFO^/l^ ^ o .5.WfUNIVER% Cr =3 S :? ^lOSANCEL •5>: o Coluinliia SEnibersitg STUDIES IN CLASSICAL PHILOLOGY THE BELLUM CIVILE OF PETRONIUS COLUMBIA UNIVERSITY PRESS SALES AGENTS NEW YORK: LEMCKE & BUECHNER 30-32 West 27th Street LONDON: HENRY FROWDE Amen Corner, E.G. THE BELLUM CIVILE OF PETRONIUS EDITED WITH INTRODUCTION, COMMENTARY, AND TRANSLATION BY FLORENCE THEODORA BALDWIN, Ph.D. THE COLUMBIA UNIVERSITY PRESS 1911 All righU reserved COPTKIGHT, 1911, By the COLUMBIA UNIVERSITY PRESS. Set up and electrotyped. Published January, 1911. NotfeoooB p«aa J. S. Gushing Co. — Berwick & Smith Co. Norwood, Mass., U.S.A. « t « t «■ « « . C C C • . . . . * i I I t ■ / « c t t f. t * « t t c • € ,*< % TO MY SISTEK a; UJ -1 a. UJ a7b CONTENTS Introduction 1-95 The Author 1 The Book . ' 3 The Poem 6 The Author's Purpose 7 Theories as to the Author's Purpose (Petronius and Lucan) 10 Examination of the Theories .... 11 The Parallel Passages 13 The Prologue 21 Imitations of Other Poets .... . 22 Conclusion 24 Petronius's Knowledge of the Pharsalia . 27 General Estimate of the Poem . 32 Petronius's Epic Theory .... . 35 The Poem in Detail Obscurity . 37 II. 177-208 . 38 Caesar's Speech . 39 Points of Style . 40 Language . 45 Syntax . 63 Prosody, etc . 56 Dis, Fortuna, and Discordia 64 vii VIU Contents PAGE Reminiscences of the Bellum Civile in Mediaeval Latin 67 Editions and Translations .... . 69 Parallel Passages : Petronius and Lucan 71 Petronids, /Satirae, 1-4 89 Petronius, Satisae, 88-9 92 Bellum Civile {Satirae, 118-24) 96 Commentary 109 Translation 2.31 The Mss. of Petronius 246 Critical Notes 248 Vita 265 THE BELLUM CIVILE OF PETRONIUS INTRODUCTION In discussing any part of the work of Petronius, we are hampered at the outset by uncertainty ^^ ^^ as regards the identity of the author. The author -^t- o • i • His bahrae '■ are not mentioned by any writer prior to the second century a.d.,^ and even then, although repeatedly cited by grammarians, as Petronius, Petronius Arbiter, or Arbiter alone,^ 1 For the title, see p. 3, n. 2. 2 The first reference known is that of Terentianus Maurus, De Metris, 1. 2489, Arbiter disertus, and 2852, Petronius. For further references, see Buecheler's editions (Berlin, Ed. Maior, 1862, Ed. Minor, 1892, 1886, 1895), Collignon, Petrone en France (Paris, 1905), Introd., and Burmann, Ed. (Utrecht, 1709, 1743), Part II, pp. 254, 257, 271. ' E.g. Serv. ad Verg. Aen., Ill, 57 ; Fulgent. Mytholog., Ill, 8, p. 124 (in Mythographi Latini, Amsterdam, 1681), and the references in (2) above. In Souligcr's Ms. (see p. 246) he is called C. Petronius Arbiter Afranius. B 1 2 The Bellum Civile of Petronius he is never spoken of as a contemporary or as- signed to any period. As a result of this uncer- tainty, many theories have been advanced and vigorously contested, placing him anywhere from the reign of Augustus ^ to the fourth century a.d.- The internal evidence, however, and especially the two epic fragments, the Troiae Halosis ^ and the Bellum Civile, point decidedly to the reign of Nero.^ Beyond this there is a considerable amount of evidence, not conclusive indeed, but collectively of no little weight, which leads us to identify him with the Gaius Petronius of Nero's court,^ the elegantiae arbiter of whom Tacitus has left us so striking a portrait.'' This evidence is : first the name or designation Arbiter; second, the indications in the fragments that their author was not only familiar with the writings of mem- bers of the imperial court, but took a lively and 1 C. Beck, The Age of Petronius Arbiter (Cambridge, Mass., 1856). He admits that the work may date as late as 34 a.d. " Statilius, Apologia. 3 See p. 5. * This view is held by Teuffel, Buecheler, Ritter, Fried- lander, Boissier, Studer, Merivale, Mackail, Heitland, Momm- sen, and others. All but the first four of those mentioned accept him also as the Petronius of Tacitus. ^ Called Titus Petronius by Pliny, H. N., XXXVII, 7, 3. and Plutarch, Quo modo Adulator ab amico internoscatur, 19 (60 E.)_^,...--— """'^ , « Ann., XVI, 18 ff. Introduction 3 controversial interest in them ; third, the striking resemblance between the genius of the work and the character of the brilliant and cynical courtier who could flay his master's vices with his satire when the hand of death was already on him ; fourth, the absence of any record of another Petronius of literary pretensions or qualifica- tions.^ In the following discussion of the re- lation of the Bellum Civile to the Pharsalia of Lucan, therefore, this identity has been as- sumed, but all conclusions as to the purpose of the poem have been based upon internal evi- dence alone. The Satirae ' of Petronius Arbiter, now lost except for fragments of the fifteenth and sixteenth „, , , books, partakes of the character of a The book -» <- • ... Menippean satire m its mixture of prose ' For a list of the other known Peironii, see Burmann, Part II, p. 278. - In the Tragurian Ms. the title is : Petronii Arbitri Satyri Fragmenta ex libra XV et XVI ; in So., C. Petronii Arbitri Afranii Satyrici Liber; in B, VI, Br., F2, Mess., Petronii Arbitri Satyricon (Br. Satiricon) ; in V2, Petronii Arbitri Satyri- con Liber. The other Mss. give the author's name only. The older editors generally accepted the title Satyricon (genitive plural),. but Buocholer adopted the Latin Saturae. The more high-sounding Greek title would naturally belong to a later date than the rest of the work (see Casaubon, De Satyrica Poesi, Paris, 1605, II, p. 326). 4 The Bellum Civile of Petronius and verse, but with important differences. In the Apocolocyntosis of Seneca, the only complete specimen of this kind of work extant, the min- gling of prose and verse is much like that in the text of an old-fashioned opera. The prose tells the story and furnishes the themes, the verse embroiders them and marks the climaxes. As in the opera it furnished the material for the florid passages, so in the satire it supplied the author with the vehicle for his wildest burlesque. In these Satirae, on the other hand, some of the passages in verse are quotations from other au- thors, ^ others are poems and epigrams introduced as the compositions of some of the characters.^ Eumolpus, especially,^ is continually breaking into verse, extemporaneous or premeditated.^ And of the remaining fragments, which now appear scat- tered through the work without anything to con- nect them with the prose, it is probable that many were originally an organic part of the whole.^ In content the book is a romance, composed by a master hand, of low life and strange adventure 155, 6; 68, 4; 112, 2. 25; 23, 3 ; 34, 10 ; 133, 3 ; 134, 12. 3 See p. 7 f. 4 89 ; 93, 2; 109, 9; 119-124. = E.g. 83, 10 ; 139, 2. Introduction 5 in southern Italy. Mingled wnth its fantastic episodes are more _serious passages, in which the author makes his personages the mouthpiece of his own views. Such especially are chapters 1-4 on the rhetorical training of the day^ 88 on the decay of arts and sciences/ and -118 on poetry. All of these end with verses. The last two are assigned to the poet Eumolpus, and the poems which they introduce are the longest in the book : the Troiae Halosis - of 65 iambric senarii, and the Bellum Civile ^ of 295 hexameters. It may be noted in paseing that Eumolpus recites the Trojan piece merely a apropos of a picture at which his companion is gazing intently, but offers the other as a practical illustration of his theory of epic poetry. As the entire work, or rather all that remains 1 See p. 92. ' Teuffel-Schwabe, History of Roman Literature (tr. by G. C. W. Warr. London, 1891-1892), 305, 4, thinks that it alludes to a poem of Nero's on the same subject. E. Thomas, L'En- vers de la societe Romaine (Paris, 1902), p. 93, describes it as a parody on a juvenile Iliacon of Lucan. Its ob\'ious relation is to the Second Book of the Aeneid, which it constantly recalls. ' Or Carmen de Bello Civili. The Dresden Ms. (see p. 247) calls it Satira Petroni poete satyrici contra vicia Rom.anorum. It has also been called Carmen de luxu Romanorum, Satira de pessimis Romae morihus, Satira in qua vitia Romanorum reprehenduntur, De mutatione rei publicae Romanae. 6 The Bellum Civile of Petronius to us, has been a fruitful source of differences of opinion, so this fragment, deahng with the struggle _,, between Caesar and the Roman Sen- T. he poem ate, the most ambitious poetic effort contained in it, has become a sort of secondary storm-center. The questions which have been asked about it, and variously answered, may be summarized as follows : Was it intended to give a complete picture in little of the war, or an outline to be filled in, or an introduction to be continued ? What was the author's object in writ- ing it? Did he intend it to be taken seriously? The first question, while the least important of the three, is also the most difficult to answer with any degree of certainty. The concluding line of the poem : factum est in terris quicquid Discordia iussit might serve equally well to introduce a continua- tion of it. The fragment might also, though with less probability, be an outline, fairly complete through 244, but from there on to be expanded should the poet decide on an exhaustive treat- ment of the subject ; as exhaustive, that is, as his theory would allow. Or, finally, it might be already complete,^ describing the opening scenes 1 Complete, that is, in scope, not finish. Introduction 7 of the struggle and foreshadowing the rest. Per- haps Petronius never took the trouble to make up his mind on this point, but, if any, the last- named choice seems the most likely. The words of the introduction : tamquam si placet hie impetus, etiamsi nondum recepifultimam manum, indicate that it is merely intended for a spirited attack on the subject which, though forever im- perfect, would show what the author meant and what he might have done. Nor should it be for- gotten that, whatever Eumolpus, as a character in the story, may be supposed to contemplate, it is not likely that Petronius, the satirist, would ever have thought seriously of attempting to ex- pand his verses into a full-fledged epic, indepen- dent of his romance.^ He had made his point, and that would be sufficient. The second question is complicated at the outset by the manner in which both the Bellum The au- Civile and the Troiae Halosis are intro- thor's pur- duced. Eumolpus, into whose mouth ^°^® they are put, is a wretch whom his creator holds up alternately to ridicule and loath- ing. His public recitations are greeted with show- ers of stones, which, from long habit, he receives 1 Of course such a poem could not have been included iu the Satirae without destroying its proportions entirely. 8 The Bellum Civile of Petronius calmly enough. On this occasion his companions, with whom he has just planned a particularly daring smndle, allow him to hold forth in peace to beguile a tedious journey on foot. When poem and journey are ended, literature is dropped with- out further comment, and the intrigue goes on as before. All this, on the face of it, might be con- sidered sufficient proof that nothing assigned ^p the old poet was to be taken seriously. But it is Petronius's way to throw a dash of satire even over what he seriously means. The remarks on the decay of the fine arts and of science are in- trusted to this same Eumolpus, and the criticism of the declamatores, the justice of which is proved by abundant evidence,* is divided between the sponging and shifty Agamemnon and the cowardly profligate Encolpius. It is in the same spirit that Apuleius allows the exquisite story of Cupid and Psyche ^ to be told by the repulsive old hag in the robbers' den. Either Petronius did not consider that the incongruity involved made much differ- 1 E.g. Pers., Ill, 44 ff.; Juv., VII., 150 ff.; Quintil., II, 20, 4 ff. ; V, 12, 17 ff. ; VIII, 3, 76. For examples of the declamationes themselves, see those ascribed to Quintilian, and Seneca's Controversiae. It should be noted that part of Petronius's discussion of this subject and the whole chapter on poetry (118) are included in Saintsbury's Loci Critici (Boston, 1903). 2 Met., IV, 28 . . . VI. 24. Introduction 9 ence, or else he realized that just such incon- sistencies of intellect and character are common enough in real life. Looking without prejudice at Eumolpus's discussion, we may disagree with some of his conclusions, but we shall not find anything to ridicule. Beginning with poetry in general, he gives us an appreciation of Horace which has become justly celebrated,^ and then goes on to the epic and his own ideal of a work freed from the restraint of troublesome facts and elevated above the commonplace by an indirect method of treatment and the introduction of mythological elements : ut potius furentis animi vaticinatio apparent quam religiosae orationis sub testihus fides. He then, by way of illustration, offers the much-disputed epic fragment which one party has hailed as superior to Lucan ^ and almost equal to Vergil,^ and the other has decried as tasteless and stupid bombast. That Petronius, a mocker and a realist, should engage in the battles of artificial literature, and even essay to contribute to it himself, seems so strange, that the question at once suggests itself : Had he, in composing this poem, any special motive 1 Horatii curiosa felicitas. 2 Doiisa, Praecidanea, II, 12. 3 Anton, Ed., note to 1. 295. 10 The Bellum Civile of Petronius which can still be discovered? The almost unanimous answer has been that it was an undis- Theories as guised Satire upon Lucan.^ Wester- to the au- burg " and E. Thomas ^ add that it is pose(Petro- double-edged, ridiculing the conserva- nius and tives in the person of Eumolpus, even ^°^°) while attacking the newer school. The Troiae Halosis, whether considered as directed against Lucan or Nero,^ is included in all this. More moderate views are those of Margaritori ^ and Heitland.'' Margaritori considers the lines a mere reproduction of the epic convention of the day, to which Petronius could not have attached much importance, since he assigned them to Eumolpus,^ while Heitland finds that they were "throAvn off half in rivalry, half in imitation of Lucan." To CoUignon also the poem is "tout au plus une refonte partielle du poeme d'apres une point de vue 1 E.g. Dousa, I.e., Teuffel in Rhein. Mus., Vol. IV (1846), p. 511 ff. E. ; Thomas, op. cit., p. 93 f. Boissier, L'Op])osition sous les Cesars (Paris, 1892), V, III, adds that, in attacking Lucan, Petronius may have intended to please Nero. ^Petron und Lucan, Rhein. Mus., Vol. XXXVIII (1883), p. 92 ff. 3 Op. cit., Ch. IV, § 1. * See p. 5, n. 2. ^ Petronio Arbitro (Vercelli, 1897), pp. 49 and 54. * Introduction to Haskin's Lucan (London, 1887), (31). ^ But see p. 7 f . Introduction 11 specielle. Les deux poetes different sur la fagon de comprendre le role et le caractere du merveilleux dans Vepopee." ^ Tailhade," in his translation, omits as interpolations "d'un scholiaste hete" all verses in the work that are not closely joined to the story (Avis Premonitoire, p. xxvi). This does not seem to include the Belluni Civile, but he omits it also, although he translates chapter 118, even to tamquam si placet hie impetus, etc., and then, having merely inserted the name of the poem, continues: " Eumolpus, ayant avec sa rhapsodie, etc." De Boisjoslin, in his Preface to the book, calls the Bellum Civile an "episode de hasard, que la deraison du copiste a insere dans cette aventure." The view that Petronius was criticizing or paro- dying Lucan rests, first upon the language of Examina- chapter 118, second upon the internal tion of the evidence of the poem itself. The criti- eones cism is plain. The words of Eumolpus : non enim res gestae versibus comprehendendae sunt, quod longe melius historici faciunt, point directly to the school of which Lucan was the most dis- tinguished member, while the fact that he takes as his example belli civilis ingens opus, and that 1 Collignon, Etude sur Pitrone (Paris, 1892), Ch. V. ' ^Petrone, le Satire (Paris, 1902). 12 The Bellum Civile of Peteonius his own lines are so full of reminiscences of the Pharsalia, shows that he must have had Lucan himself particularly in mind.^ But it does not follow that his purpose was parody or travesty,^ Criticism and parody do not go well together, and it seems utterly unnatural that, having stated his objections to Lucan's method in a sober and reasonable manner, he should immediately nullify them all by offering in support of his own theory a mere burlesque, or a poem so absurd and tasteless as to discredit all that he had taken the trouble to say. Nor, indeed, is it easy to see how he could have aimed at both schools at once without inevitably missing both. The fact that his taste may have been at fault or his pro- posed plan for the ingens opus a bad one, proves nothing, if it appears that he presented it in good faith.^ 1 As Teuffel says (I.e.), Petronius probably omits Lucan's name because he was still alive. His meaning was clear enough without it. 2 Westerburg, op. cit., distinguishes the two forms in this poem, parody in its details, and travesty in the form and spirit of the whole. 3 In the argument which follows I shall, of course, be com- pelled to rely largely upon my own feeling with regard to the poem. Humor is not to be discovered by analysis, and those who seek it so are apt to miss it where it is and find it where it is not. Introduction 13 Let us turn now to the evidence afforded by the parallel passages.^ The first point that The parallel must strike any one is their number. passages jj^ ^^^q historical parts of the poem (including the portents, 122-140, which are treated as history by both authors) they are practically continuous. Those portions devoted to the super- natural element, as is natural, show only incidental reminiscences. The opening passage on Roman decadence is also full of parallels, but Caesar's descent from the Alps, 177-208, appears to be quite independent.- Returning for further light to the critics, we are told, first of all, that Petro- nius "follows Lucan step by step," ^ and this is taken as evidence of a satirical purpose. It should be remembered, however, that, as both authors are drawing from the same historical sources, this resemblance of outline is only natural. The tragic poets of Athens all used the same cycle of myths and legends, yet, with one exception,^ they seem to have been content to pursue each his own 1 See p. 71 S. ' For further discussion of this passage, see p. 38. ' Moessler Quaestionum Petronianarum Specimen Alterum el Tertium (Hirschberg, 1865), p. 9. Boissier, op. cit., V, II, p. 244. * Eurip. EL, 520-544. But the intention of criticism here is denied by Murray, in his Translation (London, 1908). Note to 11. 510-545. 14 The Bellum Civile of Petronius course without assailing that of any one else. Going farther, Westerburg ' finds the whole poem a travesty of the Pharsalia both in manner and in spirit; 1-60 is a ^'Kapuzinerpredicht." As specific cases of parody he cites fames premit advena classes, 16, where Petronius applies to tigers for the arena the words which Lucan had used of Actium : et quas premit aspera classes Leucas (I, 42 f). Julia's prophetic words : praeparat innumeras puppes Acherontis adusti portitor III, 16 f. are "very well parodied" by Fortuna : vix navita Porthmeus sufficiet simulacra virum traducere cumba ; classe opus est. 117-119. Petronius's sentit terra deos mutataque sidera pondus quaesivere suum 264 f. ridicules Lucan's address to Nero : aetheris iinniensi partem si presseris unam, sentiet axis onus, librati pondera caeli orbe tene medio. I, 56-58.^ and the Sullanus ensis personified, 1. 98, and repre- sented as drinking blood, parodies : 1 Op. cit. 2 See p. 19. Introduction 15 sic et SuUanum solito tibi lambere ferrum diirat, Magne, sitis. I, 330 f. It will be admitted, I think, that a parody should, above all things, be amusing. But compare these, or any of the parallels, with Aristophanes's paro- dies of Euripides,^ and it will at once appear how far from genuine parody they are. Nor is the poem a burlesque in tone, like the A-pocolocyntosis of Seneca. Yet, as Petronius's prose narrative shows, he had a rare talent for that artistic ex- aggeration and use of the unexpected and incon- gruous, which is the very soul of this kind of fun. Moessler,2 going more into details, finds in 116-121 a triple reflection on Lucan, III, 14-19. Fortuna is substituted for Julia as the speaker. Tisiphone is painted more luridly than Lucan's Eumenides, and laceratus ducitur orbis is meant to be at once simpler and more elegant than lassant rumpentis stamina Parcas. Again, in describing the panic at Rome, 209-244, answering to Lucan, I, 466-522, Petronius first exaggerates by giving Caesar a 1 E.g. in Alcestis, 181-182, the heroine, dying to save her husband, bids farewell to her marriage-bed : . . . I, 299-351, at Ariminum ; VII, 250-329, before Pharsalus. » I, 195-203, 225-227, 288 f. 8 292-294. See p. 84. Bed dum tela micant, non vos pietatis imago ulla nee adversa conspecti fronte parentes commoveant : vultus gladio turbate verendos. 320-322. Petronius's: victores ite furentes, ite mei comites et causam dicite ferro 168 f. merely answers to VII, 261 f.: si pro me patriam ferro flammisque pctistis, nunc pugnate truces, gladiosque exsolvite culpa. c 18 The Bellum Civile of Petronius reddenda est gratia vobis, non solus vici with Lucan, I, 340-345 ; VII, 257 f., and 264-269. Petronius's mei comites crystallizes feelings upon which Lucan enlarges again and again. Then there is the quiet : quia poena tropaeis imminet et sordes meruit victoria nostra 172 f. against the hysterical appeal : Caesareas spectate cruces, spectate catenas, et caput hoc positum Rostris effusaque membra VII, 304 f. which is immediately weakened by what follows : vestri cura movet : nam me secm-a manebit sors quaesita manu : fodientem viscera cernet me mea qui nondum victo respexerit hoste. 308-310. Lastly, there is nothing in the Petronian passage to recall the absurd : cuius non militis ensem agnoscam ? caelumque tremens cum lancea transit dicere non fallar quo sit vibrata lacerto. VII, 287-289. Had ridicule been Petronius's object, he would Introduction 19 scarcely have let pass so tempting an oppor- tunity as that. In the remaining passages it will be found that Petronius, while frequently the more ob- scure of the two, is also as a rule the more re- strained. Take, for example, the passage already quoted : sentit terra decs mutataque sidera pondus quaesivere suum. 264 f. This ascription of weight to the gods, also found repeatedly in Seneca's tragedies,^ is utterly un- poetical, and, like the operation of the law of gravity in general, brings us down to earth with a thud. But how does Petronius's use of it compare with Lucan's ? That consistent patriot applies it, not to the old noblesse of Olympus, but to the divine Nero, urging him, when the day of his apotheosis shall come, to " keep the heavens trimmed," " by sitting carefully in the middle ! ^ Certainly, Petronius's simple use of the idea is no match for this high development of it. Finally, to understand what would be necessary in order really to outdo Lucan in his most reckless mood, the account of the storm in » See on 264. ^ The expression is Heitland's, op. cit. (46), d. ' For the lines, see p. 14. 20 The Bellum Civile of Petronius which Caesar crossed from Dyrrachium to Brun- disium ^ should be considered, where the waves are held down by the clouds, and the sea would leave its bed to overwhelm the land, did not the winds, all blowing at once, protect each his own coast.^ And to appreciate what Petronius could have done, had he chosen to amuse hhnself with Lucan and his heroes, we need only turn back to some of the excursions into mythology and history made by his wonderful Trimalchio. " Rogo, inquit, Agamemnon mihi carissime, numquid duodedm aerumnas Herculis tenes, aut de Ulixe fabulam, quemadmodum illi Cyclops pollicem porcino ex- torsit f solebam haec ego puer apud Homerum legere" (48). "Scitis, inquit, quam fabulam agant?^ Diomedes et Ganymedes duo fratres fuerunt. ho- rum soror erat Helena. Agamemnon illam rapuit et Dianae cervam subiecit. ita nunc Homeros dicit, quemadmodum infer se pugnent Troiani et Paren- tini. vicit scilicet et Iphigeniam, filiam suam, 1 V, 560-677. This will also throw some light on the state- ment of E. Thomas, op. cit., p. 91, that Petronius, in imitating portions of the Pharsalia, had chosen those that we would condemn sans riserve. 2 It must not be forgotten that this is not represented as a miracle, wrought by the personal intervention of a god, as in Verg., Aen., I, 81-123, but a storm weathered — in an open boat — by a historical personage in a rationalistic epic. 5 59. A Homeric scene is about to be acted. Introduction 21 A chilli dedit uxor em. oh earn rem Aiax insanit et statim argumentum explicahit." One passage remains to be considered, the open- ing denunciation of luxury and avarice and their The pro- attendant evUs, which occupies over logue one fifth of the whole poem, and in fullness of detail far exceeds anything similar in Lucan. It has been remarked ^ that many of these details belong rather to the Empire than to the Republic, and it has accordingly been marked as a special piece of satire, aimed, not at Lucan only, but also at Seneca and other moral writers.^ It is true that most of the crying evils which it assails had been and continued to be made the object of attacks which were powerless to reform them, but it does not follow that Petronius, in adding his voice, was insincere.^ Indeed, the length and vehemence of the passage, even its overloaded obscurity, may have been due to a warmth of indignation that marred, instead of making, verses. There is nothing in the words which shows a desire to ridicule those who pro- fessed to say such things in earnest. At worst they are simply the stereotyped sermon without 1 Mocsslcr, De Petronii Poemate de Bello Civili (Breslau, 1842), § 15. ' Westerburg, op. cit., p. 93. ' Cf. the position of Juvenal. 22 The Bellum Civile of Petkonius which no serious poem of that time was com- plete.i And there is quite as much probabiHty that Petronius, with his broad if perverse out- look,^ would sometimes be stirred by such abuses, as that they would move Lucan, absorbed in his epic and its wonderful author. As to the dispro- portionate length of the passage, which occupies more space than any other single division of the poem, it may be due merely to the lack of ultima manus which would have removed superfluities and adjusted the parts to the whole. Why, then, if satire or positive criticism was not his purpose, did Petronius, after declaring Imitations against the historically conducted epic, of other follow its prophet so closely ? In seek- ^°^ ^ ing to answer this we must bear in mind that Lucan is not the only poet from whom Petronius has borrowed, but merely the chief of 1 See the notes on this passage. ^ For a touch of human sympathy, cf. Trimalchio's words (71) : '^ amid, inquit, et servi homines sunt et aeque unum lactem biberunt etiam si illos malus fatus oppressit." The man into whose mouth these words are put treats his slaves as a spoiled child his toys, but still, what he says here, ungrammatically and in a moment of drunken sentimentality, reads like the honest conviction of his creator. Cf. also the frequently repeated phrase, homo inter hom-ines, applied to the freed slave, and at the same time illuminating the other side of the great gulf. Introduction 23 many. A large number of parallel passages, from various authors, will be found scattered through the notes/ Those from the tragedies of Seneca are the most numerous, outside of Lucan, and most noteworthy. A comparison of the quotations from the works of the uncle and nephew will show, incidentally, some of the many resemblances between them, and show also that if, in his bor- rowings, Petronius generally went to Lucan for his language, the line of thought is apt to be closer to that of Seneca. This is especially true of the figures with which he embroiders his web. The points of the compass,^ the laws of physics,^ have a potent attraction for these men, to whom really poetical ideas so seldom come.* There are also the same set descriptions of nature in her angry moods, of yawning chasms and unfruitful fields. Tombs "ope their ponderous and marble jaws," and the light of day is suddenly flashed ^ Those from later writers will serve to illustrate the general prevalence of the custom. « See on 1. 2. ' E.g. the center of gravity, 264 f. ; the conflict between weight and supporting power, 85. * Contrast such a fancy as that of Vergil, where Aeneas in the Underworld sees Dido : obscuram, qualem primo qui surgere mense aut videt aut vidisse putat per nubila lunam. VI, 453 f. 24 The Bellum Civile of Petronius upon the realms of everlasting night. A conceit of which Seneca is particularly fond, introducing it, commutatis verbis atque sententiis, no less than four times, ^ is the enumeration of the grim phan- toms at the gates of Hades. In this he is follow- ing Vergil,^ and is followed in turn by Petronius,^ but Lucan, although his witch Erich tho * has much to say about the Lower World, never men- tions them. Comparisons drawn from storms at sea are found almost everywhere in Latin poetry, but there is a link between the similes of this trio made by their fondness for the word regimen. The Furies, too, are the common property of the poets of the Empire, who overwork them mercilessly. For the locus communis of reflec- tions on the fates of the members of the First Triumvirate, or of Pompey and his two sons, see the notes to 11. 61-66. To sum up, Petronius has drawn, first and chiefly, from Lucan ; secondly, but still Conclusion , "^ ' ^ ' .^ . ,, for a great amount, considermg the difference in subjects, from Seneca ; ^ thirdly, from ^H. F., 92-103, 690-696 ; Oed., 588-592, 650 f. 2 Aen., VI., 273 ff. ' Letumque Insidiaeque et lurida Mortis imago, 257. * VI, 507 ff. * Or perhaps from the same traditional literary stock from which Seneca drew. Introduction 25 Vergil/ Horace, Catullus, rarely from Lucretius,^ perhaps never directly from Ennius. In like manner, later poets borrow from him as well as from more voluminous writers.^ His epic frag- ment, while not original enough to lend much weight to the theory which it illustrates, is, when judged by contemporary standards, the reverse of absurd. For its object, it appears, as Heitland ^ says, "to have been thrown off half in rivalry, half in imitation of Lucan." Probably the versa- tile author wished to add to his satira of moods and manners a specimen of legitimate epic poetry, as then understood. Perhaps, too, piqued by Lu- can's air of possessing the entire field, he wished to show how easily such conventional work might be done. It does not follow that he thought the result great poetry, or expected any one else to think so. It is also probable that he really did > One sometimes finds in Vergil a phrase which has roused several echoes in Petronius, none of them complete. E.g. Ge., II, 510: gaudent perfusi sanguine fratrum. Cf. 75, 214. ' But in chapter 134, 12, 4 f. there is a distinct Lucretian reminiscence: mihi pontus inertes \ suinmittit fluctus. There are other echoes of Lucretius in the same poem. The resemblance here is not in sense but in sound. ' Cf. Stat. Theb., Ill, 661 : primus in orhe deos fecit timor, and Petr., Frag. 27 (Buecheler), Poetae Latini Minores, XLIII, 76 (Teub.) *0p. cit. (31). 26 The Bellum Civile of Petronius believe in keeping epic poetry within the realm of imagination — as he understood it — and distinct from history. What he failed to see was the futil- ity of attempting such treatment with a subject which was, and must remain, historical, whose per- sons could not be demigods and did not deserve to be puppets. By his manner of introducing the poem he contrived, first, to disarm criticism to a certain extent, and, second, as Tacitus says of his indomitable Petronius,^ to show that his ignavia had achieved as much as the industria of others. Finally, in so far as it is critical, its criticism is of a negative kind, incompatible with parody or travesty, which consists in writing sensibly where Lucan allowed himself to drift into folly,^ The many close resemblances to the Pharsalia would serve precisely to bring out the differences and make his readers reflect on them.^ Besides this, in handling Lucan's subject, he probably felt that, until superseded, Lucan had said the last word on it, and therefore he borrowed from him whenever convenient as frankly as Lucan himself iL.c. 2 " It reads like a fair copy written to show Lucan how to do it." Heitland, op. cit. (6). ' Collignon, Etude sur Petrone, Ch. V, gives as a reason for not considering the poem a parody of the Pharsalia: " Elle lui fait trop d'emprunts non deguises.'[ Introduction 27 from Vergil ^ and Vergil from Ennius ; nay, even more, for Petronius was not, either by nature or training, a poet, and needed outside assistance to make up the deficiency. At this point the question presents itself : How much of the Pharsalia did Petronius know?^ _ ^ . , Following the words of the Life of Petronius s knowledge Lucan attributed to the sixth-century of the grammarian Vacca,^ it has been gener- Pharsalm n i i- r. , ally believed that only the first three books of his epic were published during this author's lifetime. His suicide, commanded by Nero for his participation in Piso's conspiracy, took place in 65 a.d., Petronius's in 66, Nero's own in 68. It w^ould stand to reason that, under 1 See Heitland, op. cit. (51), for a list of Vergilian reminis- cences covering 16 pages, and Nisard, Les Pohtes latins de la decadence (3d ed., Paris, 1867), vol. 2, p. 333 ff., for a discus- sion of Lucan's debt to Vergil, Ge., I, 4G6-488. 2 Assuming him to have been C. Petronius. ' Quippe et certamine peniaeterico ado in Pompeii theatro laureis redtante Nerone fuerat coronatus; et ex tempore Orphea acriptum in experimentum ingenii ediderat et III libros quales videmus . . . (ad fin.) sua sponte coactus vita excedere venas sUn praecidit . . ., nan sine iactura utilitatis cum patriae quae tantam immature amisit indolem turn studiorum quoque. re- liqui enim VII belli civilis libri locum calumniantibu^ tamquam, mendoai nan darent. qui tametsi sub vero crimine non egent patrocinio, de iisdem did quod in Ovidii libris praescribitur potest : ' emcndaturu^ si licuisset erat.' (A list of his other writings follows.) 28 The Bellum Civile of Petronius the circumstances, Lucan's friends would not have dared to publish the remaining books until after the Emperor's death, so that Petronius could not have read them. Westerburg,^ how- ever, noticing Petronius's unmistakable imita- tions of the Seventh Book, admitted this also, attributing his knowledge of it to recitations. Outside of these four books he found but one parallel, that between VI, 817 f. : Europam, miseri, Libyamque Asiamque timete; distribuit tumulos vestris Fortuna triumphis. and Petronius's epitaph on the First Triumvirate, 61-66. This he considered accidental, as indeed, if it stood alone, it might well be.^ Heitland, however, though he does not mention this case, finds six others, making no distinction between Book VII and the rest of the group, and inclines to the belief that Petronius knew them all.^ In addition to these the following striking in- stances may be noted : maerentia tecta Caesar habet vacuasque demos. Phar. V, 30 f . of. 225 : ^Op. cit., p. 95. ^ See p. 24 and notes to 61-66. 5 He cites IX, 706 f. (14-16); X, 133 f. (21 f.); IX, 426- 430 (27-29); X, 338 f. (64); VI, 718 (95-97); VII, 125-127, (235-237). Introduction 29 rumoris sonitu maerentia tecta relinquunt. auribus incertum feralis strideat umbra. VI, 623. cf. 138 : umbrarum facies diro stridore minantur. baud multum terrae spatium restabat Eoae ut tibi nox tibi tota dies tibi curreret aether, omniaque errantes stellae Romana viderent. VII, 423-425. and X, 155-158 : quod terra quod aer ' quod pelagus Nilusque dedit, quod luxus inani ambitione furens toto quaesivit in orbe, non mandante fame. cf. 1-2 : orbem iam totum victor Romanus habebat qua mare qua terrae qua sidus currit utrumque. primaque Thessaliam Romano sanguine tinxit. VII, 473. cf. 294 : Thessalicosque sinus Romano sanguine tingue.'' spumantis caede catervas VII, 699.^ cf. 214 : perfusas sanguine turmas ' For parallels from other authors, see notes on 1-2. ' For tho reading, see Notes. ' For the Vergilian parallel, see p. 26, n. 1. 30 The Bellum Civile of Petronius and 281 : toto fluitantes orbe catervas. quae seges iiifecta surget non decolor herba? VII, 851. cf. 99 : extulit in lucem nutritas sanguine fruges. Taking the Pharsalia as a whole, and arranging the books according to the number of times that Petronius appears to have drawn upon them,i we have the following series : I, VII, X, VI, II, III, IV, IX, VIII, V. The number of parallels from I-III and from IV-X are almost equal The reason for the preponderant influence of Books I and VII is not far to seek. Petronius, even if he had known all ten books equally well, could hardly have found space for borrowings from every part of them unless he had made his 295 lines a mere cento of Lucan. The books which naturally fixed his attention were the First, containing the opening scenes of the war, to which he has devoted most of his own poem, and the Seventh, treating of Pharsalus, the climax of both the struggle and the epic. But admitting 1 For a full list of parallel passages, see p. 71 S. The reck- oning is, of course, only approximate, as there would inevitably be a difference of opinion as to how many passages were certainly, and how many only probably, conscious imitations. Introduction 31 all these evidences that Petronius was acquainted with the last seven books of the Pharsalia, how, if he died before their publication, could he have gained his knowledge? The question is easily answered. First, there were the "author's read- ings," so much in vogue at that time, where not only short poems, but passages from longer ones, not yet published, were declaimed. Heitland ^ has remarked on the disastrous effects of this custom upon Lucan's work. Even when there were books still to write, and improvements to make in those already written, Lucan's favorite passages, those which had stirred his imagination most, would probably have been perfected and read to the court circle before the ban of the im- perial displeasure silenced him.^ Another — or an additional — explanation is offered by Heit- land : "My own view of the probabilities of the case is that the words of the earlier (or Suetonian) Life, relating to Lucan's emendation of some verses just before his death,^ are the foundation of the notion that the later books are uncorrected. There was some record of the publication of three books, none of the rest. Out of these materials lOp. cit. (27). «Tac., 4nn., XV, 49. Dio, LXII, 29. Vacca. * ' Impetrato autem mortis arfntrio libero codicillos ad palrem corrigendis quibundam versibus suis exaravit. 32 The Bellum Civile of Petronius the story of the later Life ' was made up, and has no authority : though it may for all that be true." 2 Such, then, is the Bellum Civile of Petronius, and as such, although an unfinished fragment, it General claims a place in the poetic literature estimate of of its period. This period, known as e poem ^^^ Silver Age of Latin Literature, is one of decadence. In poetry it includes Lucan, Seneca, Persius, Silius Italicus, Statins, Juvenal, Martial, and some lesser names. Its characteris- tics are those which belong naturally to a genera- tion devoid of original inspiration : rhetoric at all times, overdevelopment of trivial themes, ceaseless rehandling of a small and rather cheap set of ideas, worn thin in the continual effort to polish them to new lustre, endless straining after sensational effects, sententiae, antitheses, paradoxes, a general lack of subtlety, of the ability to produce great effects by suggestion and restraint, of beauty, delicacy, and real passion ; a ceaseless shouting, quibbling, and reveling in bloodshed and physical horrors.^ This does not mean that the poetry of 1 See p. 27, n. 3. * ^Op.cit. (32). ' E.g. Lucan, II, 173-190 (the murder of the brother of Ma- rius) ; VI, 507-830 (episode of Erichtho) ; Seneca, Oed., 960- Introduction 33 the Silver Age is devoid of dignity, force, and brilliant workmanship, but its finer growths are always in danger of being choked by the weeds that spring up with them. To take oneself very seriously was perhaps the first requisite of success in this school ; to be of fastidious taste would have been a fatal handicap. In view of all this, even were there nothing else to commend his work, Petronius would deserve praise for self-restraint, for moderation in the use of meaningless details and the invention of absurdities, and for sparing us any nauseating tale of horrors. But this is not all. Into his fragment of less than 300 lines he has worked a remarkably full and suggestive narrative. Such little vignettes as 61-66, 87-88, and the prophecy of the great series of battles, 111-115, open up deep vistas to any one familiar — as was Petronius's original public — with the history of those times. There is power — the power of a keen mind to summarize, and the power of a strong imagination to visualize — in such bits as : ut bibat humanum populo plaudente cruorem. 18. 978; and the entire finaleof H. O. (1132-) and Med. (868-). Contrast Vergil's treatmentof the Deiphobus episode Aen., VI, 494 ff. D 34 The Bellum Civile of Petronius non homo pulsus erat, sed in uno victa potestas Romanumque decus, quare tarn perdita Roma ipsa sui merces erat et sine vindice praeda. 48-50. luxuriam spoliorum et censum in damna furentera. 86. the grim summing up : iam pridem nuUo perfundimus ora cruore nee mea Tisiphone sitientis perluit artus ex quo SuUanus bibit ensis et horrida tellus extuUt in lucem nutritas sanguine fruges. 96-99. and the fine picture of the unconquerable Caesar defying the elements, and lifted by sheer strength of will and purpose to a level with the gods (201- 208) . In these passages, as in the bulk of the poem, the style is terse and vigorous, with the sugges- tion of some strong motive power behind it. But there are also bits of great beauty, refreshing spots beside the road over which the poet is driv- ing his muse so relentlessly. For instance : iam Phasidos unda orbata est avibus, mutoque in litore tantum solae desertis adspirant frondibus aurae. 36-38. and : (non) verno persona cantu moUia discordi strepitu virgulta loquuntur. 72 f. With a considerable variety of subject and treat- ment, and with remarkable freedom from the ver- Introduction 35 bosity and exaggeration of its time, the poem ranks high when compared with its more preten- tious fellows. If it falls short of the brilliance of Lucan's best work, it escapes his worst bathos. If it lacks the grace and charm of Statins, it is free from the mere prettiness into which these so often degenerate. It is in every way superior to the Bellum Punicum of Silius. And all this, it must not be forgotten, is achieved by a writer whose regular metier — whoever he may have been — was as far as possible from epic poetry. As to the soundness of our author's theory, it merely represents one horn of a dilemma. Per Petronius'3 ci7)ibages deoruiuque ministeria seems a epic strange course for a narrative of well- '^^'^ known public events to pursue. But the severely historical treatment, on the other hand, makes poetry well-nigh impossible. Nisard ^ finds two ways in which the Pharsalia might have been composed : "C'etait ou de receuillir a Rome et par toute Vltalie les souvenirs nationaux sur ces dernieres guerres de la liberie, de courir en Grece, en Egypte, sur les traces de Pompee et de Cesar, d'interroger les pdtres de la Thessalie, et de composer une epopee de tous ces bruits populaires; ou bien de peindre a grands traits la corruption d'ou 1 Op. ciL, II, p. 123 f. 36 The Bellum Civile of Petronius sortirent les guerres civiles, et d'expliquer le grand changement qui rendit Cesar maitre du monde." Lucan chose neither of these ways, but it will be observed that the second is not unlike what Petronius attempted. And before one utterly condemns the plan of compelling the weary gods once more to put their shoulders to the wheel, it should be noted that what Petronius meant, as indicated by what he did, was not, like Sihus, to reduce his characters to mere pawns pushed about by divine players, but simply to have the gods set the terrific engine grinding — as who would venture to say that Caesar and the Senate of the moment could alone have wrought such a thing ? — to oversee, to prophesy, and to furnish an im- posing background for human activities, but with- out putting any constraint upon human will and reason. The gain by such a method would have been mainly negative, in the escape from the mass of details, military and political, which give the impression of religiosae orationis sub testihus fides, and the effect of the work as a whole would inevi- tably have been one of incongruity. But so is that of the Pharsalia} At best, had the author 1 It must not be forgotten that Lucan does not dispense with the supernatural entirely. Fortuna flits through his epic, except where forced to yield to the sterner abstraction, Fatum, Fama helps to carry the tidings of war to Rome (I, 469-472). Intkoduction 37 of such an epic possessed sufficient ability, he might, Hke Lucan, have made it memorable for fine passages which would compel the reader to forget for a moment the faulty design of the whole. The worst positive defect of the Bellum Civile, its frequent obscurity, is a fault shared with The poem other poets of the age — Lucan, Per- in detail. sius, Juvcnal — who wrotc under high scuri y prgssm-e of one sort or another. It doubtless arose, in the first instance, from a striv- ing after the unusual and epigrammatic, the con- stant effort to catch and hold the attention. Petronius is by no means a prime offender; but the contrast with his smooth-flowing prose makes the contortions of some of his metrical passages all the more noticeable, and perhaps these con- tortions, with the violent and jerky effect of the struggling lines, have helped the idea that the au- thor meant to make what he imitated ridiculous. But besides the general tendency to obscurity at that time there is, in this case, the additional ex- cuse of lack of experience in poetical composition, lack of ultima manus, and the problem of crowd- Palria appears to Caesar at the Rubicon (I, 185 ff.), the shade of Julia terrifies the sleeping Pompey (III, 9 ff.). a woman prophesies the course of the war (I, 673 ff.), and the witch Erichtho exercises the blackest of black arts for the benefit of Pompey's son (VI, 607 ff.). 38 The Bellum Civile of Petronius ing much material into a small compass. It is also more than likely that some of the most desperate difficulties are due to the corruption of the text, and would disappear if we could recover what Petronius really wrote. The most independent passage in the poem, 177-208, is at the same time one of the least suc- 177-208 cessful. The author's indebtedness here is not to other poets, but to Livy,^ and in his effort to adapt to his purpose and at the same time magnify to epic proportions the historian's account of Hannibal's crossing of the Alps and the Apennines — which, by the way, had lost their terrors by Caesar's time — he has shown pretty clearly that clouds and mountain- peaks were not favorable to the flowering of his imagination. Were it not for the poetical inde- pendence of the whole paragraph, the crabbed Latin and tangled meteorological phenomena might well have lent color to the burlesque theory. The parallel description by Silius Itali- cus ^ forms an interesting comparison. The work of a poet as much more tasteless as he was more ambitious, it not only equals Petronius in 1 XXI, 35-36 and 58. See the Notes. s III, 500 ff. and 547 ff. ; IV, 749 ff. See notes on 187 S. and 203. Introduction 39 exaggeration, but leaves him far behind, telhng, for instance, how the broken ice amputated the hmbs of those who fell, and then was melted by their blood. In striking contrast to this florid descriptive passage is that immediately preceding,^ contain- Caesar's ing Caesar's speech to his army. Its speech twenty lines are perhaps the finest in the poem, and in simplicity, clearness, and dignity of thought and expression are not unworthy of the man to whom they are assigned. It is marked by soldierly brevity and a strong man's modera- tion of speech. Caesar's affection for his soldiers and his absolute confidence in them are revealed without undue protestation. The rewards of vic- tory and the penalty of inaction — the possibility of failure is not considered — are brought out by single firm touches. The only thing which is not in keeping is the scornful reference to the Roman plebs (164-166). This is doubly inappro- priate, first, as it makes of Caesar a reactionary of the school of the Scipios; and second, as it throws the responsibility for the war on the peo- ple instead of on the Senatorial oligarchy, where it belonged. Petronius's object, of course, was merely to make use of the younger Scipio's sting- i 156-176. 40 The Bellum Civile of Petronius ing retort to his political enemies/ and after more than a century, during which Caesar had been canonized, anathematized, and generally misun- derstood, ihe inaccuracy is pardonable. The metrical form of this admirable bit of rhetoric is appropriately measured and dignified, lending special effectiveness to the two lines which are allowed four dactyls each dum Gallos iterum Capitolia nostra petentes. 161. representing the terrific downward sweep of the Gallic invasion ; and at reor, hand impune, nee banc sine vindice dextram vinciet ignavus. 167 f. where Caesar's calmness is broken for a moment by the thought of the vindictive injustice of his enemies. The whole ends with a line finely ex- pressive of his unshaken confidence in the out- come : inter tot fortes armatus nescio vinci.^ 176. Points of Examining our author's style more style minutely, we notice the following char- acteristics : Fondness for paradox or contrast : » See note on 1. 166. « See p. 58. Introduction 41 detritaque commoda luxu \ailneribus reparantur. inops audacia tuta est. 56 f. muneribusque meis irascor. 108. \'incendo certior exul. 162. sanguine Germane sexagintaque triumphis esse nocens coepi. 163 f. causam dicite ferro. 169. quia poena tropaeis imminet et sordes meruit \'ictoria nostra. 172 f. gaudet Roma fuga. 224. praedamque in proelia ducit.' 232. This device was, of course, especially popular with the Stoic Seneca. Used in moderation, it is a highly effective means of setting the irony of things clearly before the reader, but overdone it soon becomes tiresome, the word-play giving an exaggerated and artificial importance to the thought, instead of the thought ennobling the words. Petronius, however, has used it with dis- cretion, and in Caesar's speech it becomes a powerful means of bringing out the injustice of the position into which he has been forced. Fondness for startling and novel expressions : ' Cf. Lucan, I, 504: in helium fugitur. Examples from the Troiae Halosis (89) : ihat iuventas capta dum Troiam capil, 27. iacet aacerdos inter aras victima, 51. contraque Troaa invocat Troiae aacra, 65. 42 The Bellum Civile of Petronius belua dente ad mortes pretiosa. 15 f. fames premit advena classes. 16. (mensa) maculis mutatur villus aurum. 29. ingeniosa gula est. 33. sitientis , . . artus. 97. nimbos . . . ligatos. (i.e. ice) 187. vincta fluctus stupuere ruina.^ 191 This tiresome trick arose from the decay of true imagination and originality and the pernicious training of the schools of oratory, with their impossible controversiae and suasoriae} Petronius the novelist laughed at them,^ but Petronius the poet laid aside for the moment his splendid sense of humor and copied — though not in their most desperate forms — the foibles of his serious- minded contemporaries.* Sententiae are few, and the author has kept his own rule ^ against introducing them in digres- sions. One is woven into the address to Fortuna: 1 Examples from the Troiae Halosis; in suo voto latent. 10. peritura Troia perdidit primum deos. 53. « See p. 8, n. 1. 3 Chs. 1-4. See p. 89 ff. ^ His own words (88, 6) : vitia tantum docemus et discimus, are not without application to some of his poetry. * 118, 5. See p. 1. Introduction 43 Fors cui nulla placet nimium secura potestas, quae nova semper amas et mox possessa relinquis. 80 f. Others are appended as tags to the lines which suggest them : inops audacia tuta est. - 57. hos gloria reddit honores. 66. Colloquial expressions have sometimes been admitted in spite of the poet's ban.^ E.g. the use of accersere,^ 117, 158. esse nocens coepi.^ 164. Legal or quasi-legal expressions are : sine vindice praeda. 50. nullum sine pignore corpus. 53. causam dicite. (ironical) 169. causa peracta est. 175. On the other hand, the language has many of the characteristics which distinguish poetry from prose. E.g. : The free use of verbs and adjectives which practically personify the noun to which they belong : 1 118, 4. See p. 1. ' Lucan is also fond of this word. See notes to 158. ' Cf. Ov., Met., X, 132 : velle mori statuit. Such circumlocu- tions are plebeian. Cf. the redundant tun in vulgar German. 44 The Bellum Civile of Petronius (tellus) hostis erat. 6. ingeniosa gula est. 33. maiestas auro corrupta iacebat. 44. hoc niersam caeno Roman somnoque iacentem. 58. ira rebellat. 105. laceratus ducitur orbis. 121. (tellus) armis laeta meis. 157. non pugnavit humus. 186. consensu . . . timor. 246. terras . . . furentes. 247. lacerataque tecta rebellent. 287. Transferred epithet : corruptis miles vagus exstruit armis. 32. Effect for cause : rigido . . . flamine. 196. Zeugma : qua mare qua terrae qua sidus currit utrumque. 2, bustorum flammis et cana sparsa favilla. 77. Middle passive : tu . . . satiare ruina.^ 119. Frequent use of possessive pronouns : e.g. sua . . . arva, 12 ; mea Tisiphone, 97 ; mei comites, 169 ; Caesaris sui, 267. These, how- ever, are never superfluous, but always have some special emotional fitness. See the Notes. 1 See note on mutatur, 29. Introduction 45 Difficulties. — In addition to the obscurities already mentioned, the poem contains some lines which, in spite of the Latin words, hardly read or construe as Latin. Such are : hinc Numidae accusant, illinc nova veHera Seres atque Arabum populus sua despoliaverat arva. 11 f. hunc nive dura claudit hiems canoque ad sidera vertice toUit.^ 146 f. pepulitque meatibus auras. 178. atque hoc Romano tonitru ferit omnia sigua. 212. atque inter torto laceratam pectore vestem. 276. The difficulties in 24-27 and 229-231 are prob- ably due to the corruption of the text. In sharp contrast with the ingenti volubilitate verborum of his prose, Petronius has employed but a small vocabulary for his poem. He was, of course, obliged to dispense with many of the picturesque expressions which ^ lend color to the narrative, and does not seem to have had a sufficiently large store of voces a plehe semotae to replace them. Thus there is much repetition. Vincere appears a dozen times, rein- forced by victor and victoria, quaerere in various senses eight times, rumpere (usually the past par- ticiple) and its compounds ten times, tellus ten times, pulsus and pulsatus nine times together, 1 See Notes. 46 The Bellum Civile of Petronius orhis six times. The following words are used from three to five times each : ^ arx, hella (the plural by preference), clades, concretus, corrumpere, damnum, efferre, frangere, furere and furor, horridus, iacere, incendia, ingens, laetus, late, maerere, mergere and compounds, orhis, perdere, and perire,^ perfundere and other compounds of fundere, personare, and persona, petere, placere, pondus, potestas, rehellare, regnum, ruina, sanguineus, sanguis, sidus, solvere, strepitus, sumere, tritus, and detritus, turma, vertex, vincire, vulnera, vultus. Besides these repetitions, suh- missa is applied to the waters of the Bosporus in 242 and to Fides in 252. Fortuna and Fama are both volucer.^ The world distracted by civil strife, the garments of Discordia, and the war- swept tecta are all lacerata.* Gradiens is used first of the caged tiger and second of Discordia.^ The language of 50 : ipsa sui merces erat et sine vindice praeda. 1 This list is not meant to be exhaustive, and includes only the more important words so repeated. Others are mentioned in the Notes. 2 Especially periturus. See note to 19. 5 78 and 210. * 121, 276, 287. Cf. anserem laceratum verubus confixit, 137, 12. « 17 and 279. Introduction 47 is recalled in another connection by mercedibus emptae, 165, followed two lines below (167) by sine vindice. A particular weakness is the repe- tition of a word within a few lines but in a dif- ferent connection, and where there is nothing to gain by it, as where quaerere occurs five times in twenty-one lines.^ Thus we find eruta (mensa), 27 ; eruta . . . conchylia, 35. Moles, which stands for the unwieldy bulk of Rome in 83 and 109, is used between them, in 91, of the masses of rock excavated from the earth. Insolitae voces, 180, is followed by insolitos . . . ausus, 184.2 In the description of the panic at Rome we have maerentia tecta, 225, and maerentia pec- tora, 229. The repetition of cerno, 111 and 114, has a purpose. That of potestas, 79-80, may be intended for an antithesis: ''Power that hatest power," but the effect is not pleasing.^ A comparison of the Bellum Civile with the rest of the Satirae will show that, as far as possible, Petronius has simply transferred to the poem his prose vocabulary, shorn of its exuberances, and augmented by a few new terms. Words for which he shows a marked fondness in one part > 7, 10, 14, 24, 27. 2 Cf. ignihus insolilis, 136 ; Ch. 136, 6 : planctibus insolitis. ' For a similar habit of Lucan'e, cf. Heitland, op. cit. (47) E. 48 The Bellum Civile of Petronius of the work reappear frequently in the other. Thus perfundere, a special favorite, occurs in the prose no less than twelve times, together with other compounds of fundere. Accersere ^ appears five times ; periturus three times in the Bellum Civile and six times outside of it.^ Eruta,^ 33, 4, and vindida* 95, 7 and 136, 7, may also be noted.^ There is, besides, the author's characteristic use of male to indicate that an act or condition is unreal, ineffectual, incomplete, or unfortunate.^ In the prose : 12, 1 : fidem male ambulantem. 86, 4: male dormientis. 87, 3 : male repugnant!. 103, 6: male soporati. 132, 8, V. 3: male dabat usum. Bellum Civile: 20 : male pubescentibus annis. 30: male nobile lignum. > See p. 43. 2 Cf. p. 46, n. 2. Cf. also Fr., XXX, 13 (Buecheler) : aut premit eversam periturus navita puppem, a characteristically Petrouian line. 3 Cf. 27 and 35. See preceding paragraph. Cf. Verg., Ge., III. 498 f. : immemor herbae victor ecus fontisque avertitur. and Stat., Theh., VI. 192. This use of the word (as also of aversari. Cf. 6.Toe