STOP Early Journal Content on JSTOR, Free to Anyone in the World This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in the world by JSTOR. Known as the Early Journal Content, this set of works include research articles, news, letters, and other writings published in more than 200 of the oldest leading academic journals. The works date from the mid-seventeenth to the early twentieth centuries. We encourage people to read and share the Early Journal Content openly and to tell others that this resource exists. People may post this content online or redistribute in any way for non-commercial purposes. Read more about Early Journal Content at http://about.jstor.org/participate-jstor/individuals/early- journal-content . JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people discover, use, and build upon a wide range of content through a powerful research and teaching platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please contact support@jstor.org. VERGIL'S APPRENTICESHIP. Ill By Tennet Frank vergil's theory of his art If the arguments in the preceding papers are valid, we may safely, with Suetonius, assign to Vergil at least the Culex, the Ciris, and all of the Catalepton, and consequently reconstruct the history of the poet's artistic development from these. In the last study I attempted to collect whatever biographical material was available in them; I wish now to consider in more detail the poet's theory of his art as disclosed by his youthful work. Vergil's earliest judgment regarding his own verse is the line in which he describes the style of the Culex (48 B.C.). This (1. 35), he says, is not heroic verse : mollia sed tenui pede currere carmina, versu viribus apta suis Phoebo duce ludere gaudet. 1 In the Ciris, the prologue of which was written some five years later, he characterizes that poem, and presumably all the poetry he had thus far written, in similar terms (1. 20) : quamvis interdum ludere nobis et gracilem molli liceat pede claudere versum. 2 Phrases like tenui pede and gracilem are obviously suitable in assigning a poem to the third of the three time-honored divisions, the "plain" style; and Jackson (Harvard Studies, XXV, 117 ff.) accordingly concludes that the poet referred his early work to this type. The classification of styles into the three x a P aKT VPes, the grand (neyaXoirpeires, sublime, grande, grave, etc.), the middle (p4, medium), and the plain (iaxvov, tenue, subtile, gracile, etc.), was, of course, the orthodox procedure in the days of the Auctor ad Herennium, Varro, and Cicero. I venture to believe, however, that Vergil in using the words molle and ludere had particularly in mind the third type in the four- fold classification of styles: grand (neyaXcnrpewes), plain (iaxf^f), 1 Cf . i. 1 , gracili modulante Thalia. 2 In Catalepton ix. 61 he seems to refer to the Ciris as an approach to humilis Cyrenas. In the same poem he speaks of Messalla's eclogues as written molliler (1. 17). Classical Philology XV, July, 1920] 230 Vergil's Apprenticeship 231 polished (y\avp6v), and forceful (8eiv6v), which is given by Deme- trius, De elocutione 36, and that he would consider the y\axf>vp6v (sometimes treated as a subtype of the tenue) as more nearly satis- fying his ideals than the Stoic 'urxriv. 1 Jackson (loc. cit.) has seen that Horace's description of Vergil's early work (Serm. i. 10. 44) molle atque facetum is reminiscent of Vergil's own characterization of his work. The line was written probably some six years after the Ciris. It is, I think, not exaggerated pedantry to see in Horace's molle a direct reference to Vergil's own use of the word and to find in the technical term facetum a reference to Vergil's ludere, when we consider that 1 The problem of the "styles" is too involved to discuss here. My point of view is briefly this: Cicero in the De oratore makes actual use of only two classes, the grande and the tenue. The middle style is there considered a vague intermediate mixture of the two. Though later in the Orator (20 ff.) he identifies the three styles with the three traditional officia of the orator (docere, delectare, movere) and fills the indefinite medium with the connotations of delectare, his interest still lies with the two definite types which he divides each into two classes according as they are treated with natural simplicity or with artifice. He thus arrives at four distinct types: (la) grandis sed aspera, (16) grandis el ornata, (2c) tenuis et impolita, (2d) tenuis sed concinnior, faceta et florida (Orat. 20) . It seems to me that Demetrius and Philodemus mean practically the same thing when, without speaking of the "characters" of style, they mention four methods of composition. Demetrius has (to give them in Cicero's order): (a) Suvbv, the forceful, often closely related to the grand; (6) p.eya\owp(Tks, the grand and ornate; (c) laxvbv, the simple and plain; and (d) y\avp6v, a polished style used in unambitious subjects. Philodemus' plasmata seem to me to rest upon the same tradition ; they are (Rhet. Sud. 165. 4) : abpbv, p,iya, Urxvbv, yXaxfrvpbv. Indeed, if we recognize that in his chapter on method Dionysius (De comp. xxii) is chiefly con- cerned in the two manners of composition, the austere and the ornate, and that the "middle" manner is a more or less inconsistent concession to Theophrastian traditions, we may apply these two manners to the two chief styles (De dem. iii) and again arrive at Cicero's four divisions. The authors may be compared as follows: CICERO OBAT. DEMETRIUS PHILODEMDS HORACE "SERM." 20. i. 10. 40. grande et asperum heivbv fikya Varius and Pollio grande et ornatum p.eya\vp6v Vergilius molle atque facetum I therefore assume that the tradition of four manners of writing was in general use at Roman schools in the days of the neoteroi, and that these poets claimed to practice the simple plain style (c) in their nugae, and the related polished and pleasing, unambitious style (d) in their epyllia. Stroux, De Theophrasti virtutibus dicendi (1912), 104 ff., and Herrle, Quaest-rhetoricae (1912), 54 ff., contain many good suggestions. 232 Tenney Frank Horace is here in the midst of a literary controversy and inten- tionally choosing terms from the contemporaneous critical vocabu- lary that had well-recognized connotations. 1 Now if Vergil and Horace are using the terms with the same meaning, it may be fair to find in Horace the clue to Vergil's intention. Horace's well-known lines are (Serm. i. 10. 36—45) : Turgidus Alpinus jugulat dum Memnona, dumque Defingit Rheni luteum caput, haec ego ludo, Quae neque in aede sonent certantia judice Tarpa, Nee redeant iterum atque iterum spectanda theatris. Arguta meretrice potes Davoque Chremeta Eludente senem comis garrire libellos Unus vivorum, Fundani; Pollio regum Facta canit pede ter percusso; forte epos acer Ut nemo Varius ducit; molle atque facetum Vergilio annuerunt gaudentes rure Camenae. Horace is here obviously speaking of various literary genres, but his grouping of them seems to me to betray a consciousness of the four- fold classification of styles. The plain style is naturally represented by the comic poet Fundanius. For the grand style he has no good contemporary example since Pollio 2 and Varius, whose tragedies and epics might serve, eschewed by choice all ambitious adornment, and they consequently represent the forceful style (Seivov: forte epos acer Varius). 3 The grand style is with covert humor assigned, in the absence of a real claimant, to the turgidus Alpinus who repre- sents its exaggerated form, the yf/vxpbv. To Vergil is left, as it seems to me, the remaining of the four types, the yXacjivpov of Demetrius, analyzed by Horace into molle atque facetum. Let us then examine Demetrius' definition and description of this style. '0 ■yXa<£up6s \6yos, he says, xapwrwpbs * a ' iXapos X670S eort ("charm and geniality," op. cit. 128, Roberts' edition, p. 120). 1 Horace's phrase has frequently been discussed. See especially Bayard, Rev. de Phil. (1904), p. 213; Jackson, op. cit.; Ogle, Amer. Jour. Phil. (1916), p. 327; Knapp, Amer. Jour. Phil (1917), p. 194. 2 Pollio seems to have chosen the chaste Sophoclean style as his model, Verg. Eclogues viii. 10. Tacitus Dial. 21 calls his tragedy, as well as his prose style, durus et siccus. Puristic tendencies seem to underlie his criticism of Livy, Sallust, and Catullus. 3 Martial viii. 18. 7, fortius ore. Vergil's Apprenticeship 233 What he means by xapteJ'TWjufo — and its connotations are numerous — appears from the contextual examples and definitions. Deme- trius' instances of x6p""« include witticisms, graceful and charm- ing phrases, and passages without obvious humor, beautiful names, and what musical critics called "smooth words" (128-31, 173, 176); and in subject-matter they associated especially with themes like "the gardens of the Nymphs, marriage songs, love stories, and indeed were found in the poetry of Sappho generally" (132). The chief quality (x<*p«) of the polished style is therefore just that combination of charming grace and well-bred good humor that the "new poets" sought to embody in their more ambitious work. When they used the term facetum to characterize this in part, they knew that the inadequate word had to be filled with a new literary connotation which it had not contained in Cicero's discussion of the facetiae of prose orations. And Quintilian was aware of this differ- ence when, in explanation of Horace's phrase, he said (vi. 3. 20): Decoris hanc magis et excultae cuiusdam elegantiae appellationem puto. Molle seems to be a translation of the Greek nahanov and fiakdaicov, and Dionysius, though not recognizing a y\avpos xapaK"?P, does define a smooth diction (y\avpa) as requiring that all words 1 be AeTa Kai naXaica ical irapOevcoira. It appears, then, that in the tech- nical language of the day the qualities that were most marked in the graceful epyllia of the neoteroi were precisely those that Demetrius placed in his definition of the y\avp6v so far as he discussed poetry rather than prose, and that this term connoted what Vergil and Horace meant by molle atque facetum. Of course we do not know the date of Demetrius. Since he has been assigned to so many of the hundred or over bearers of that name already known, it may be futile to add new conjectures. But 1 " Melodious, smooth, soft as a maiden's face," ed. Roberts, p. 235. When Demetrius discusses the excess of this style, calling it kclkoI'tiXop (186), "affectation," he reminds us of nothing so much as the simpering phrases that Seneca has quoted from the works of Maecenas Epist. Mm. 114. 5. Is it not clear that Maecenas, influenced perhaps by Vergil's early work, unsuccessfully set himself the model of the y\a4>vp6s xapaKrifp ? And is not this the meaning of Agrippa's witticism on Vergil and Maecenas quoted by Donatus (Vita 44): "a Maecenate eum suppositum appel- labat novae cacozeliae repertore, non tumidae nee exilis [not the excesses of the grand or plain styles] sed ex communibus verbis atque ideo latentis" ? This criticism was probably written in the early thirties. 234 Tenney Frank it seems to me that a more improbable guess could be made than that he might be the very rhetorician, Demetrius, at whom Horace flings his abuse in lines 79 and 90 of his tenth satire. 1 Of course there is but little value in such conjectures. The only purpose in making it is to attract attention once more to the possibility of dating this interesting work in the first century B.C. The important fact is that the fourfold division of the styles (apparently no inven- tion of Demetrius') had been made public property before Horace wrote this satire; for Philodemus, as was long ago pointed out, had knowledge of it. 2 No one who has considered what the Neapolitan hortus meant to Vergil in his youth would be at all likely to assume that Vergil did not read the essays of the Gadarene on their appear- ance. Indeed, even Cicero in the passage cited by Quintilian, Ne illi sunt pedes faceti ac delicatius ingredienti molles, is apparently using quotation marks if, as seems likely, the line is sarcastic. The phrase was possibly a catchword of the neoteric apologists. It would be interesting to know to how much of Vergil's early work the term of Horace applied. Horace specifically calls atten- tion to the pastoral (gaudentes rure Camenae), but, as we have seen, both Horace and Vergil seem to class the Culex as a kind of pastoral poem. The Ciris, by Vergil's own definition, also falls into the group of the y\avpdv yivos. This poem Horace may have been 'Horace's taunts: "men moveat .... quod vellicet absentem Demetrius?" (Serm. i. 10. 78) and Demetri, teque, Tigelli, Discipularum inter jubeo plorare cathedras (ibid. 91) probably refer to ill feeling engendered by professional jealousy on the part of rhetors who had not, like Apollodorus, the entree into the circles of Augustus and Maecenas. If, however, this be the author of the well-known work De elocutione, we can well understand that he must have been more sympathetic toward the romantic schools of poetry than Horace and Apollodorus could be. For the date of Demetrius see Roberts' edition, pp. 49-64. The expression in 108, irop{ipats irXarttots, seems to refer to some architectural ornament, but even so it is not decisive. Broad expanses of crimson and red coloring occur first in the "second style" of house decoration at Rome (about 60 B.C. +, see the "house of Livia" at Rome), but they continue to be fashionable into the early Empire. It is, however, difficult to explain 179: ovSiyapTwyTrpli'eipfiTai Tin wepl y\cuj>vpas avuBkaaas, unless we place Demetrius before Dionysius. 'The plasmata of Philodemus (Rhet. Sud. 165) are very near to the styles of composition of Demetrius (Wilamowitz, Hermes [1900], p. 30, n. 4). Vergil's Apprenticeship 235 permitted to read. Indeed, the lines of the Ciris (19-20) already quoted would seem to imply that the youthful verses in general were to be considered of this genre. Certainly the definitions of the genre given by Demetrius are sufficiently broad to cover the poems Catalepton, not to mention the very graceful bit of ornate realism of the Copa. From this we may conclude that Vergil at first identified him- self quite completely with the neoteric school, that he accepted as the proper description of such work the definitions of the "polished" style propounded by critics like Demetrius, and that even after he had written the Bucolics his contemporaries thought of him as the leading representative of the polished style. He reveals, however, the fact that it was not his ultimate ideal. He very early enter- tained ambitions of writing a philosophical poem (Ciris i. 5; Georgics ii. 477), and even before Julius Caesar's death aspired to write the epic of his country (Cat. 14; Eclogues vi. 2), tasks which he con- sidered worthy of a loftier style than that employed in the Alexan- drian work he was doing. Vergil's next expression is the well-known fifth Catalepton bidding farewell to his books of rhetoric and turning for all time from a public career. This poem, of course, does not contain any creed of poetic art. It obviously expresses the hearty disgust at the dry details of rhetorical or grammatical rules of one who has caught visions of what a positivistic philosophy can promise. There are, however, some especially significant phrases which show that Vergil's teacher has forfeited his respect by too strong a faith in the old-fashioned rules now being attacked by the Atticists. Such scornful phrases as rhetorum ampullae, injlata .... verba, madens pingui .... inani, cymbalon iuventutis, indicate that this teacher had not been of Calvus' school. It is probable also that the Varro of whose book the poet had had a surfeit was the well-known encyclopedist, who, as Cicero casually says, had the bad taste to like the style of Hegesias, the father of Asianism. 1 Varro was himself an author of a book on the types of style. If that was written on a substructure of Hegesianic taste and Vergil had to study it, the tone of these lines can well be 1 See Norden, Rhein. Mus. (1893), p. 547. Varro had probably not been giving rhetorical exercises at this time (Cic. Ad fam. ix. 1). 236 Tenney Frank excused. At any rate, the lines show that, though Vergil had adopted the molle atque facetum as his forte in verse, he did not intend to practice it to excess. And as for the content of this work, his last line suggests that he had already seen the need of pruning the facetum which in his beloved Catullus had led to coarseness. " If the Muses return, as they may, they shall still be welcome, but they must come pudenter et raw." He does not, therefore, turn his back upon the Alexandrian charm and sweetness; he merely shows a determination to write with more artistic reserve. There is, also, in a brief Catalepton (No. vii) a passing reflection of the literary controversy centering about the subject of pure diction. Here Vergil pretends to acquiesce in Varius' insistence upon Latinitas by correcting the Greek word pothos of the second line to puer in the fourth. Varius, and perhaps Quintilius Varus, had already begun that movement — soon to receive the aid of Pollio, Horace, Messalla, and Octavian, the pupil of the Atticist Apollodorus — toward a stricter diction and a more classical form in Latin verse; a movement which seems to have been begun by a puristic secession within the neoteric group and to have ended very soon in a complete opposition to the neoteric ideal. 1 Vergil's epigram is in entire good humor. Varius is an excel- lent friend and Vergil by no means derides his hobby. Vergil him- self, as we see from his later work, learned some caution from these critics; but he was too intelligent to adopt an extreme theory which threatened to shackle poetry. This epigram affords an early date for the new movement so often referred to by Horace; it seems to identify that movement with Vergil's friend Varius, and it shows Vergil in the sane attitude of a man who intends to rely upon his own judgment. The second Catalepton contains another somewhat obscure refer- ence to the literary controversies of the day. Many attempts at restoring and interpreting the epigram have been made. Professors DeWitt and Fairclough have perhaps contributed most to its 1 Cf. Hendrickson, Class. Phil. (1917), p. 341, who gives a clear explanation of the relationship between the Horatian group and the neoteroi with reference to the ques- tion of diction. Birt gave the right interpretation of this poem in Jugendverse und Heimatpoesie Vergils, p. 83. Vergil's Apprenticeship 237 elucidation. 1 I can add but little to what they have said and I ven- ture that only for the sake of bringing this epigram into the account of Vergil's utterances on style. The poem seems to me to read as follows: Corinthiorum amator iste verborum, iste iste rhetor, namque quatenus totus Thucydides, Britannus, Attice febris! tau Gallicum min et sphin ut male illisit, ita omnia ista verba miscuit fratri. "That lover of archaic words, that tiresome rhetor, surely he's as much (and as little) a Thucydides as he is a real British prince! He the bane of real Attic style! It was his mixture of Greek archaisms and Celtic pronunciation, I suppose, that killed his brother." The reference is, of course, to the rhetorician Annius Cimber, the friend of Antony. There are apparently two points of attack: (1) Cimber seems to have been a man of low station (Cic. Phil. xi. 14) who tried to pass, as freedmen then frequently did, 2 as the descendant of some unfortunate barbarian prince. Since his Celtic brogue betrayed him, he chose to call himself British. Vergil in the epigram seems to imply that he had assumed the brogue {tau Gallicum) as well as the name to escape the odium attached to Eastern freedmen. (2) Cimber claimed to be an Atticist, but his ignorance of Greek — shown in his use of non-Attic forms and the adoption of Thucydides as a model for Atticistic orators — made him the laughingstock of the docti. Now the Thucydides craze at Rome had a very interesting history, and this passage affords a pertinent commentary on some passages in Cicero. In the Orator (30) Cicero drives full tilt at some unknown person who, while claim- ing to be an Atticist, has adopted Thucydides as his model. What 'DeWitt, Amer. Jour. Phil. (1912), p. 317; Fairclough, Trans. Amer. Phil. Assoc, XLVII, 43 ff. I adopt DeWitt's text with a slightly different interpretation. 2 So Hermeros in Petronius 57. Cimber's father bore a Greek cognomen, but Gallic and Germanic slaves were not infrequently given Oriental slave names. His own choice of "Cimber" was probably an early attempt to point to a loftier Western extraction. Presumably, like Antonius Gnipho, he was a Gaul who secured a Greek education. The reader will also remember Valerius Cato's claims to good birth defended in his Indignatio; cf. Hendrickson, Class. Phil., XI, 268. 238 Tenney Frank results is to Cicero a novum quoddam imperitorum et inauditum genus, as it was to Vergil. Reference to the same strange aberra- tion is found in Cicero's Brutus 287-88. Cicero indeed suggests in part (loc. cit.) how the shibboleths got mixed. In Calvus' day when the Roman Atticistic group were making their program they made liberal use of Peripatetic and Stoic theories advocating conciseness, perspicuity, and purity of grammar and diction. 1 The Stoic axiom recommending the antique simplicity that was nearer nature, since it antedated the baneful school of rhetoricians, seemed to lead back logically to Thucydides as the first important Attic writer of prose. Thus it was that some- one, following purely theoretical considerations with insufficient insight or taste, wrote the name of Thucydides on the banner of that Atticism which in general was supposed to stand for the artistic simplicity of Lysias. Since Sallust adopted Thucydides as his model in history and also aspired to fame as a writer of speeches, it is not impossible — if, indeed, Sallust had thus early made his preference known — that he was the man in question. Cicero's reference, however, seems to be aimed at some earlier doctrinaire, possibly at Vergil's victim, Cim- ber, 2 who had already entered politics in the year when the Brutus was written. Even Pollio and Messalla, who were later accused of fondness for patinated diction, seem in their youth to have been influenced by this heresy. Cicero's plain words in 46 b.c. should have corrected the error of those who did not know enough Greek to comprehend the color of Thucydides' style. Both Demetrius and Dionysius seem to take cognizance of the misunderstanding when they point out the many stately and ornate passages, not to men- tion the many lapses into obscurity and roughness, which would naturally disqualify Thucydides from being a model of Atticistic simplicity. 3 1 See Stroux, De Theophrasli virlulibus dicendi; Hendrickson, Amer. Jour. Phil., XXV, 125; XXVI, 249 ff.; Smiley, Latinitas and 'EWTjxiff/tos ; Fiske, The Plain Style in the Scipionic Circle. 2 Cimber's importance is attested by no less a person than Augustus (Suet. Aug. 86). 3 Demetrius 40, 44, 72, etc. Dionysius, Roberts' ed., pp. 178, 228; Ad Ammaeum ii. 2. Vergil's Apprenticeship 239 The point of Vergil's taunt is, then, that despite Cicero's attack Cimber still harps upon Thucydides, while pretending to be an Atticist. To the poet Cimber was not an Atticist but an "Attic plague," an allusion perhaps to the Athenian plague so gruesomely described by Thucydides. 1 As for Vergil himself, his epigram is, therefore, an attack not upon the genuine Atticists but upon the inconsistent claims of an ignorant rhetor. There are but few other references in Vergil's works to theories of style. It should, however, be noted that while the youthful Vergil, like Horace, connects himself closely with the group which was laboring for a stricter classicism — Quintilius Varus, Varius Rufus, Pollio, Valgius (who translated the Ars of Apollodorus), and, of course, Octavian, the pupil of Apollodorus — he still speaks of Cinna and Gallus, 2 the sole important survivors of the neoteric school, with more than respect, pays his regards to Parthenius and Euphorion when the new classicists had abandoned them, and honors Catullus as his master by open imitation even in the Aeneid? In a word, Vergil before the publication of the Georgics, while asso- ciating with the group that combated the program of Valerius Cato, considered himself and was considered by his associates as a representative of the neoteric school. 4 In attempting thus to place Vergil in a more or less circum- scribed niche hewed out by rhetorical analysts, I would not be understood as believing that the poet while composing set himself the task of seriously concerning himself with theories of style. He certainly did this far less than Horace. As a confirmed Epicurean he scorned slavish adherence to rhetorical rules, and the outburst 1 See Fairclough, Trans. Amer. Phil. Assoc, loc. tit. 2 Cornificius, who, to judge from Cicero's letters (especially Ad fam. xii. 17), was a very important member of the group in its latter days, may possibly be the subject of Vergil's pastoral elegy Daphnis. This poet was killed in 41 B.C. while attempting to hold Africa for the republican cause, a possible reason why obvious allusions were not incorporated in the poem. If one recalls Bion and Moschus, it is difficult to escape the conviction that some friend of the guild is concerned here, and Cornificius is the only poet whose death can with certainty be placed at this time. See Class. Rev., 1920, 49. 3 DeWitt, The Dido Episode, p. 75. * For the influence of Vergil's early habits of work on the Aeneid, see Duckett, Class. Jour. (1915), p. 333. 240 Tenney Frank against the rhetors in his youth expressed, I think, an attitude which he maintained throughout life. Attempts, for instance, to analyze, say, Sinon's speech according to the schematized laws of Hermagoras seem to me wholly misleading. Vergil certainly did not look upon rhetoric as an kiriarruiri : he was very far from being an Apollodorean. Nevertheless he was certainly in touch with the theories of style then current, as every man of culture had to be, for it was the rhetoricians and grammatici who were the literary critics of the day, who wrote the book reviews, as it were, who analyzed the poets, placed them to some extent on their proper pedestals, coined the literary phrases, and told the public what ought to be read. Horace's scoffs at critical rhetoricians like Demetrius and Hermogenes, con- signing them back to their ladies' seminaries, show that they counted for something when they spoke. And after all the poets themselves had studied with the rhetoricians, learned their systems, and memo- rized their phrases. When, therefore, the poets spoke of their own work they must use the literary phraseology of the schools, with the connotations that were orthodox. Hence, despite the freedom of composition that a poet like Vergil might display, when he sat back and analyzed the product of his creative mind he would think and speak of it in scholastic terms in so far as he could find adequate ones current. It is with this fact in view that we are justified in searching for the peculiar categories of the schools which Vergil and his contemporaries thought befitting his style of work. To be sure, literary criticism in Vergil's schooldays was very young and timid. One of its gravest shortcomings, as a reading of Dionysius and Demetrius will show, lay in its failure to deal with poetry apart from prose. 1 Perhaps the traditional interest in forensic prose on the part of Roman young men compelled the teachers even against their inclination to center their dicta about the orators. Be that as it may, it would seem that the very con- sciously scholastic group of new poets who had gathered about Valerius Cato in Cicero's day had to think out their program along l Varro apparently illustrated the three styles in Roman literature by reference principally to poets. He chose Pacuvius, Lucilius, and Terence as the representa- tives respectively of the ubera, gracilis, mediocris (Gellius vi. 14). Vergil's Apprenticeship 241 the lines of the orthodox classification of styles invented for the analysis of oratory. Very little has survived to inform us what were the main tenets of Cato's formulae by which he "made the poets" of that circle. It is generally assumed that the prose style of Calvus, who called himself an Atticist and a follower of Lysias, probably harmonized with the teachings of Cato. Calvus, as we know, tried to be direct, lucid, simple, pure, and homely in diction. He shunned the pompous and emotional in style, unusual and affected words, and all meretricious effects of metrical prose. The lucid and limpid nugae of Catullus written in the artless words of everyday Rome justify us in supposing that much of the poetry coming from Cato's pupils was inspired by the same ideals that Cicero attributes to Calvus. That, however, could hardly have been the whole story. Catullus' epyllion is neither lucid nor limpid, nor can the diction be described as pure and colloquial Latin. It abounds in poetic diction that borrows freely from the Greek vocabularies for effects in grace, smoothness, and softness of sound; its narrative is intri- cately involved, its emphasis subjective, and its color entirely emo- tional. And we know enough of other adherents of Cato, poets like Cinna and Ticidas, not to mention Calvus as author of the Io and Vergil of the Ciris, to know that Cato must have encouraged the practice of such writing in certain genres of poetry. For this style I think we may find a fairly close counterpart in the prose of Calidius, who must have said something about his program when he criticized 1 Cicero's oratory. Calidius, whose style is very generously described by Cicero (Brut. 274), was a friend of Caesar and the Atticists in general, and apparently thought of himself as a kind of Atticist but as one whose prose was more closely connected with the neoteric epyllia than with the nugae. Cicero's description of his orations is one of the most interesting in the whole fascinating book. Ita reconditas exquisitasque sententias mollis et pellucens vestiebat oratio. Nihil tarn tenerum .... nullum nisi loco positum et tamquam in vermiculato emblemate, ut ait Lucilius, structum verbum videres. Nee vero ullum aut durum aut insolens aut humile aut longius ductum; ac non propria verba rerum sed pleraque translata, etc. 1 Cic. Ad. Att. v. 19. 3; se solet anteferre (mihi). Ad. Att. vi. 8. 3. 242 Tenney Frank What better description could we have of such a style than the phrase molle atque facetum employed as Vergil and Horace used that phrase ? I have quoted this passage to show reason for supposing that, in the group of poets and orators which gathered about Valerius Cato, the so-called subtile genus was quite regularly divided into two styles, and that, just as in poetry Catullus represented natural simplicity in his lyrics, but a polished and graceful tenuosity in the epyllia, so the thin-drawn Atticism of Calvus was quite distinct from the graceful though unemotional artistry of Calidius. Surely it is in recognition of such growing theories that Cicero, while pretending in traditional fashion to hold to the threefold division of styles, actually abandons the "middle" style and works out his criticism in the Orator and the Brutus by a dichotomy of the other two. 1 Only thus could he discuss the contemporary poets and orators in con- temporary language. We need not, of course, suppose that his description of Calidius consisted of quotations: Calidius' style was apparently so striking that a use of plain Latin would be adequate for the purpose. However, Cicero knew these people intimately from Valerius Cato to Gallus, 2 and there is every reason for believ- ing that he intentionally adopted their terms, though he still refused a frank acknowledgment of their fourfold analysis. The point of this digression is that the designation of Vergil's early work not only as tenue and gracile but as molle and facetum would seem to have been a recognition of his close affiliation with the ideals of the romantic group just passing away. Soon Vergil became an intimate member of the new group gathering about the young Octavian, the tendencies of which were drawing in a new direction. It would be very interesting to know who was the guiding spirit here. The revolt was, of course, not all a matter of conscious theory. In poetry as in prose, the outburst of the Civil War changed men's outlook on life, men's interests and psychology. As Cicero's leisurely periods lost a hearing in a new 1 In a letter to Cornificius, one of the neoteric poets, Cicero uses a strange phrase that has led to much discussion: vos magnos oratores (Ad. fam. xii. 18). Did Corni- ficius claim to represent the forceful style — the Stivdv of Demetrius — as Calvus repre- sented the iaxvo" and Calidius the y\avpoi> ? The iieyaKoTtpeith was of course completely avoided by the group. 2 "Cicero and the Poetae Novi," Amer. Jour. Phil. (1919), p. 396; for Gallus see Ad fam. x. 31. 6. Vergil's Apprenticeship 243 world of realities, till he was piqued into defending his creed and practice in several pamphlets, so the romantic and to some extent affected and sentimental poets of the fifties began to cloy in the hard days of Caesar's dictatorship. A revolt was inevitable, and when it came it brought, as such things will, not a little strife. Documents like the third Eclogue and the tenth Satire show that there was battling and blood-letting. Wilamowitz 1 has said that the new classicism had no connection with the immediate past and that Apollodorus was its founder. This is doubtless an exaggera- tion, though the great rhetorician must not be forgotten. Apollo- dorus taught at Rome before 44 b.c. — how long we do not know. He must have been a man of great influence to be chosen by Caesar to teach his heir. Valgius (presumably Valgius Rufus), one of the best friends of the youthful Vergil and Horace, had translated the master's book into Latin, and finally the old teacher seems to have been a daily companion of the group about Maecenas, if I am correct in identifying him with the Heliodorus of Horace, Sermo I. 5. Although there is no gainsaying the weight of Apollodorus' influence, we must recognize the fact that the germs of the new Augustan literary growth had been cast in the soil of Rome in the generation that preceded. In the first place, many of the older men like Cicero had lived through the neoteric period without being deeply moved by it, and had in their literary works never tired of repeating the lines of Ennius and the great Greeks from whom the Ennian period drew. In fact, the classical authors had not wholly been pushed aside in the fifties, even by men like Catullus. Moreover, the influence of the neo-Atticistic prose — developed largely, it will be remembered, by men closely associated with the neoteric poets — leaned strongly toward a classical purity and simplicity. Men like Messalla and Pollio were forces working with Horace and not against him; Pollio indeed had grown up with Catullus and Calvus, while Messalla probably owed much to his devotion to Brutus. Furthermore, some of the rules most insist- ently advocated by the Augustan school, especially simplicity, lucidity, and purity of diction, were apparently a part of Valerius Cato's program in so far as it applied to lyrical poetry. These 1 Wilamowitz, Hermes (1900), pp. 1 ff. 244 Tenney Frank particular qualities are no more vital to Horace's odes than to Catullus' epigrams and songs. And leaders of Augustan criticism like Varus and Varius Rufus seem to have grown up in the Po Valley where Cato's influence was strongest. Vergil's youthful epigram to Varius does not need a reference to Apollodorus in order to be understood. The Augustan revolt undoubtedly struck at certain tendencies of the Catullian group, as we have said, and, as is natural, differences are apt at such times to be emphasized over and above similarities; but after the dust of battle has passed away we can see how the Augustans, after all, evolved their favorite doctrines out of some of the qualities that had been consciously sought for by the preceding generation of poets. That certain of their qualities, those of the y&os y\avp6v, no longer found favor was largely due to the change in the world that ushered in the Augustan age. As for Vergil, he began his schooldays as a confirmed neoteric poet. In his years of apprenticeship, when his associates began to shake off the influences of the past, Vergil refused to sell his birth- right. To be sure, he saw as clearly as they the superficial vices of the more languid neoteric poetry, but he also felt the charm and emotional appeal in the best of it. Its external tricks of style, in so far as they were false, he pruned away, but he treasured what his ear and heart told him was good poetry. From his associates he accepted the program of chastening and patient application, but he refused to go the long road with some of them to impersonal formalism. Though he maintained his independence, he was too great to be thrust aside as a negligible nonconformist like Furius Bibaculus and Cato's ape. 1 He was accordingly accepted as one of the friendly group and set down as representing a style all his own. To the last he was criticized by the formalists. Agrippa found him too freely colloquial, Horace had to defend him from the opposite charge of using unusual and foreign-sounding words, while the purists parodied his diction as smacking of the provinces. But superficial criticism probably concerned him little. He never forgot that, though he wrote for the Rome of Augustus, he had for many years been the pupil of Catullus. Johns Hopkins University 1 1 have indicated elsewhere that simius iste (Horace Serm. i. 10) may be the poet Ticidas.