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. HORACE, CATULLUS, AND TIGELLIUS By B. L. Uhman The fourth and tenth satires of Horace's first book are sources of the first importance for the history of Latin literature, for they present the theories about literary style which Horace held at a crit- ical time, not only in his own life, when he was producing his first permanent work, but also in the progress of the literature of the Augustan Age. Virgil, Varius, Pollio, Horace, and others 1 had been and still were working out the characteristic features of Augustan literature. We may look upon Horace as a spokesman, more or less official, for the whole group. 2 But, unfortunately, Horace's remarks were intended for contemporaries fully familiar with details which are lost to us. Hence the interpretation of these poems has been very uncertain. Yet we need not despair of attaining an approximate certainty on some of the matters in doubt by careful study of the shreds of evidence. 3 In this paper it is proposed to deal with the opening portion of the tenth satire, particularly with Horace's attitude toward Catullus as there displayed. It seems desirable, however, to take up first a related matter, the identity of the Tigellius Hermogenes mentioned in the tenth satire, for the two subjects throw light on each other. I. HORACE AND TIGELLIUS The scholiasts identify the Tigellius mentioned in Serm. i. 2. 3 and i. 3. 4 with the Tigellius Hermogenes mentioned elsewhere in the Satires. The identification was rejected by Dacier in his edition of Horace, but only a few scholars followed him until Kirchner 4 presented a detailed argument for Dacier's view. Most modern • Serm. i. 10. 40 ff. * Horace intimates this in Serm. i. 10. 81 ff.: "Plotius et Varius, Maecenas Ver- giliusque, Valgius, et probet haec Octavius," etc. He of course feels fairly certain of their approval. ' Contributions of especial importance have been made by Hendrickson in AJP, XXI, 121 ff., and Studies in Honor of Basil L. Gildersleeve, pp. 151 ff. 'Quaestiones Horatianae, Leipzig, 1834, cap. ii, "De utroque Tigellio." [Classical Philology X, July, 1915] 270 Horace, Catullus, and Tigellius 271 editors have followed Kirchner without question. It seems to me, however, that the identification of the scholiasts is not only possible but plausible, and that Kirchner's arguments against it are not irrefutable. It is certain that the Tigellius mentioned in i. 2. 3 and i. 3. 4 is the same man that Cicero speaks of in several letters. From Cicero and Horace we know these facts : he was a Sardinian, a fine singer, and a close friend of Julius Caesar. Cicero, Horace, and Calvus (quoted by Cicero) call him either Tigellius or Sardus Tigellius, never Hermogenes. It is therefore maintained by Kirchner that he had no such cognomen. But this argument from silence is so uncer- tain that some editors who accept the assertion that Horace speaks of two men by the name of Tigellius reject Kirchner's argument about the name and assign the cognomen Hermogenes to the Sar- dinian Tigellius. 1 In the interest of truth I may note that Kirchner overlooked a further argument in his favor. With one easily ex- plained exception the putative second Tigellius is never referred to merely by the nomen Tigellius. In this way, it might be said, Horace distinguishes between the two men. In i. 3. 129 he is called Hermogenes; in i. 4. 72, Hermogenes Tigellius; in i. 9. 25, Hermo- genes; in i. 10. 18, Hermogenes; ini. 10. 80, Hermogenes Tigellius; in i. 10. 90, Tigellius. The fact that the full name was given ten lines before would easily explain the use of Tigellius alone in the last passage. On the other hand, it is very likely in itself that the cognomen of the Sardinian Tigellius was Hermogenes. The combination of a Roman nomen and a Greek cognomen usually means that the owner is a freedman. Now Cicero tells us that Tigellius was the grandson or nephew (nepos) 2 of a certain Phamea. Judging from his foreign name and the probability that he was a rich parvenu, 3 it is altogether likely that Phamea was a freedman. 4 In that case Tigellius may well 1 Once this is granted, however, the argument for two Tigellii vanishes. For we need only put ourselves in the place of Horace's contemporaries who certainly would have understood i. 3. 129 and other passages as referring to the well-known Tigellius Hermogenes and not to some obscure hanger-on. • More likely nephew, for Phamea died in 49 B.C. (Cic. Att. ix. 9. 4) and Tigellius about 40 B.C. (Hor. Serm. i. 2. 3). • Cic. Att. ix. 9. 4; Fam. ix. 16. 8. • Tyrrell (on Fam. ix. 16. 8) assumes this without question. 272 B. L. Ullman have been of the same station, and the cognomen Hermogenes would be entirely likely, supposing that he was a Sardinian of Greek ancestry. Furthermore, Tigellius was a cantor, and his profession was looked down upon by the Romans and given over largely to Greeks. Again Cicero alludes to him in a way that suggests, even if it does not prove, that he was a freedman: "Cipius, opinor, olim 'non omnibus dormio': sic ego non omnibus, mi Galle, servio." 1 Cipius had overlooked many things that were going on in his house, but he drew the line at a slave's running off with the silver. So Cicero was willing to slave for many persons, but he drew the line at (the former slave f) Tigellius. 2 With regard to the supposed second Tigellius, we have of course no actual knowledge of such a char- acter, but from the fact that he was a singer like Tigellius it is thought that he was a freedman or adopted son of Tigellius. 3 Another argument of Kirchner's is that the Tigellius of Serm. i. 2. 3 and i. 3. 4 does not fit the description of the Tigellius Hermo- genes elsewhere alluded to. Once the suspicion was aroused that two individuals were the objects of Horace's satire, it was natural of course to look for a differentiation of the pair. Kirchner admits similarities (both were well-known singers), 4 but urges that the » Fam. vii. 24. 1 (cf. All. xiii. 49. 2). - Even Kirchner admits that Tigellius was "in gentem Tigelliam aliquando quocumque modo adscriptus," which means that he was probably a freedman. But Kirchner believes that Sardus was his cognomen, a belief which is not based on facts. Calvus and Cicero (imitating Calvus) clearly call him Sardus to make fun of his nationality, and it is therefore fair to assume that in the one other passage where he is so designated (Hor. Serm. i. 3. 3) a sneer is intended. » Horace mentions Fannius, Crispinus, Demetrius, and Pantilius in connection with Hermogenes. The last named is clearly the leader of the set, for he is mentioned by Horace more often than any of the rest; he has an imitator (simius; probably one of the above), and Fannius is his parasite (conviva). This situation harmonizes with the assumption that Hermogenes was the more or less well-known Tigellius. 1 Let me again, in the interest of truth, bolster up Kirchner's side. He admits that the two men are alike in being handsome, judging from Cicero's description of Tigellius as bellus tibicen and Horace's pulcher Hermogenes in i. 10. 18. I had noted this similarity independently of Kirchner and had come to the conclusion that it was decisive evidence that the Hermogenes of Horace was identical with Cicero's Tigellius. But a comparison of Cicero's next words (sat bonus unctor — or cantor) made me con- clude that bellus meant, not "handsome," but "good," being, after all, a diminutive of bonus. I find that the Thesaurus gives the same meaning, quoting the convincing parallel bellus architectus in Att. xiv. 3. 1, where there can be no question of personal appearance. Horace, Catullus, and Tigbllius 273 dissimilarities were greater. In this I cannot follow him. One might argue that the Tigellius of the second satire was different from the one of the third because in the former Tigellius is a man with the artistic temperament, refusing to sing when asked, and in the latter more or less of a spendthrift. Nor is Hermogenes always like himself. In i. 3. 129 and i. 9. 25 he is the typical singer, in i. 10 he is a literary critic. It has been argued further that Horace was "distinctly hostile" to Hermogenes but merely disrespectful to Tigellius. Such a distinction does not exist. The references to Hermogenes in i. 3. 129 and i. 9. 25 have even been called complimentary (but see below). One reference to Hermogenes stands out as particularly abusive (i. 10. 90), and a reason for this (if a reason be necessary) will be suggested below. Kirchner makes much of the point that Tigellius, whose wealth is indicated in Serin, i. 2 and i. 3, should be teaching for a living (i. 10. 90). But it is a gratuitous assumption that he was teaching for a living. To Kirchner it seems to be unthinkable that a man should teach except for a living. There is no end to the motives which Tigellius might have had — such as a desire to get acquainted with ladies of the upper crust. But Kirchner makes an excellent point in saying that Serm. i. 10. 18, as ordinarily understood, cannot refer to Tigellius. This point will be taken up in detail in the second part of the paper, but we may anticipate to the extent of pointing out that Kirchner's theory of two Tigellii does not solve the difficulty. Yet Kirchner's first and main argument, the one which he found in Dacier and without which the theory of two Tigellii would never have been evolved, is that Tigellius is clearly referred to as dead in i. 2 and i. 3, while Hermogenes in the other (and later) poems seems to be very much alive! If this argument can be fairly met and if the reference to Hermogenes in i. 10. 18 can be explained, it will be granted, I think, that there is no need to doubt the identification made by the scholiasts. The burden of proof, it must be remem- bered, rests on the shoulders of the skeptics. The starting-point for an explanation of Horace's references to Tigellius as still alive is in Serm. ii. 1. 39 ff. He says that he will not voluntarily (ultro) attack any living being (quemquam animantem). The natural inference is that he will attack those who attack him 274 B. L. Ullman first and those who are not living at the time. 1 The latter class might be subdivided to include characters fictitious, literary (especially from Lucilius), and Ciceronian (i.e., men whose lives fell between the time of Lucilius and Horace). It is of course possible that this program was not carried out by Horace in the first book, but it is altogether likely that in the main it was. Horace gives no hint in the second book that this is a new program ; rather he emphasizes the similarity between the first and second books: cf. ii. 1. 1 and espe- cially ii. 1. 22, which is a repetition of i. 8. 11. In the latter passage Horace makes fun of Pantolabus and Nomentanus as if they were living beings. In the second book the line is repeated in a way that shows that the names were either fictitious or Lucilian or belonged to persons who were dead when the earlier satire was written. In the same way i. 4. 92 repeats i. 2. 27. There are other passages where Horace seems to attack dead persons as if they were alive. After setting forth his program in ii. 1. 39 ff., as just noted, Horace goes on to give us a sample (vss. 47 ff.). The passage is overrun with proper names, but the allusions are all harmless (see Morris' excel- lent note). Among the names we find that of Turius. There is no reason why we should reject, though we must correct, the testi- mony of both scholiasts about him. He apparently was praetor in 76 b.c. and no doubt had died long before Horace's satire was written. In Serm. ii. 3 Horace makes Damasippus the chief speaker. This is evidently the Damasippus alluded to by Cicero in a letter (Fam. vii. 23. 2) probably written in 61 b.c. It is reasonable to suppose that he was not living when Horace wrote. In i. 4. 28 Albius is referred to as if he were alive. I have elsewhere suggested that Albius was probably dead at the time. 2 Other instances might 1 It is difficult to discover just how scholars interpret animantem. The impres- sion that one gets is that they take it in the sense of an unemphatic hominem. To me this seems impossible. The Pseudacronian scholia rightly interpret it as vivum hominem habenlem animam. In this poem Horace places himself under the banner of Lucilius, and Lucilius attacked dead men (Marx, Lucilius, i. p. xl). In fact the custom became a feature of satire (Marx compares Seneca Apoc; cf. also Juv., i. 170-71). The failure to note the significance of animantem has led Lejay (Satires, p. 613) to observe (he wrongly attributes this view to Nettleship) that Horace is insincere in his protestations, because he at once proceeds to attack Canidia, Cervius, Turius, and Scaeva. It is scarcely credible that Horace would be guilty of such inconsistency, which is suitable to a buffoon, not to a man of Horace's refinement. ! AJP, XXXIII, 160; accepted by Smith (Tibullus, 1913) in his note on Tib.i. 1. 2. Horace, Catullus, and Tigellius 275 be brought forward, but they are less certain. Parallels from other authors, as Juvenal, 1 readily suggest themselves. Yet it must not be assumed that in every instance is Tigellius referred to unequivocally as alive. In at least two instances, i. 10. 18 and i. 10. 80, there is room for doubt. Besides, in every case the allusion is such that no strikingly incongruous effect is produced. I may illustrate the point from something near at hand : I note that in this paper I constantly use the present tense in referring to Kirchner, whose pamphlet was published in 1834. One may carry out the comparison in detail, first quoting Horace and then making up a similar expression appli- cable to Kirchner: i.3.129: "Tigellius sings well"; " Kirchner argues well"; i. 4. 71-2: "I publish no books for ignorant persons like T. to handle"; "I publish no books for great scholars like K. to criti- cize"; i. 9. 25: "My singing is such that even T. would envy it"; "My argumentation is such that even K. would accept it"; i. 10. 18: "T. never read Aristophanes"; "K. never read Juvenal"; i. 10. 80: "Fannius, a follower of T., makes attacks"; "Professor X, a fol- lower of K, refutes this point"; i. 10. 90: "Good-bye to you, T."; "I bid you and your arguments adieu, K." Contrast with these passages some of the references to Tigellius' satellites, presumably alive when Horace wrote: i. 4. 14, "Crispinus challenges me to a contest." It would be absurd to say, "Kirchner challenges me to answer him." We see then that there is no good reason on chronological grounds for rejecting the identification of Tigellius and Hermogenes, and we may take the scholiasts' word for it without further question. Let us now proceed to examine Horace's references to Tigellius in order to understand his attitude toward him, at the same time gaining further evidence for our identification. The one thing that Horace emphasizes in most of his allusions is that Tigellius was a singer. The motive for this is not far to seek. Horace looked down on sing- ing, putting it in the same class as dancing and prolific verse-writing (Serm. i. 9. 23; see, e.g., Wickham's or Morris' note). Horace is "rubbing it in" 2 in his gentle but effective manner. In i. 2. 3 he » Among numerous examples in Juvenal, cf. especially i. 155, where Tigellinus, who died in 69, is mentioned as if he were living, though the satire was not written before 100; cf. Duff's edition, pp. xxxiii ff . ' The slang phrase may be pardoned because it has a classical counterpart in Horace's sale defrieuit (Serm. i. 10. 3). 276 B. L. Ullman speaks of the death of Tigellius, the singer, where the reference to his profession is entirely unnecessary. In i. 3. 1 Horace says that all singers have the same fault, and then cites Tigellius. In i. 3. 129 the Stoic is made to say, in support of the doctrine that one may be rich without having a cent, that Hermogenes is a fine singer and composer even though he does not sing. 1 The reference is anything but complimentary, for to Horace there was a vast difference between a cantor optimus and a vir optimus. It is just possible that Horace is hinting at another point at issue between himself and Hermogenes, for in the same passage Crispinus is mentioned as a Stoic, and in the fourth satire he clearly belongs to the literary circle, of Tigellius. This situation, together with the fact that it is Horace's Stoic oppo- nent who brings Tigellius forward as the typical musician, makes it plausible perhaps that Tigellius himself was a Stoic. 2 In i. 4. 72 there is no reference to singing. Tigellius is now a literary critic and perhaps a poet. Horace satirizes him by saying that his (Horace's) poems are not to be found in bookstores for the volgus and Tigellius to thumb over. The scholia rightly take this to mean that Tigellius is classed with the ignorant volgus ("imperitam turbam .... qualis est Hermogenes"), i.e., Tigellius is offered as a concrete example.* At the same time, when we consider the poem as a whole, and with it the tenth satire, it seems probable that Horace meant to imply also that Tigellius wrote for the rabble. 4 In i. 9. 25 Tigellius is again the typical singer, but, as in i. 3. 129, it is not Horace himself who cites him as such, but the character whom Horace is satirizing, or rather allowing to satirize himself. As already indicated, the reference is anything but complimentary. 5 We shall discuss i. 10. 18 1 "Quamvis tacet Hermogenes, cantor tamen atque optimus est modulator." 2 The reference in this passage to Hermogenes as a singer is further evidence that Hermogenes and Tigellius are one and the same, for by this device Horace comes back toward the end of his poem to the subject-matter of the beginning, as he is fond of doing: i. 1. 108 is especially noteworthy, and for repetition of names one should com- pare Tillius in i. 6. 24 and 107, Dama in ii. 5. 18 and 101, Albius in i. 4. 28 and 109. Tacet in i. 3. 129 (the passage under discussion) is an evident allusion to Tigellius' behavior as described in vss. 1-6. 8 This characterization is similar to that in Serm. i. 2. 1, where Tigellius is called the associate of flute-girls, drug-sellers, beggars, actresses in the mimes, and parasites. «Cf. especially i. 4. 21 ff. and below, pp. 278 f., 291 ff. s In this passage the bore enumerates three accomplishments : prolific verse-writing, dancing, and singing. Since the first is an obvious allusion to the coterie of Tigellius Horace, Catullus, and Tigellius 277 at some length in a moment, so that it is sufficient to point out that doctus cantare is said of a man standing in some close relation (called a simius) to Tigellius. Whatever cantare may mean here, it evidently was chosen to emphasize and satirize the fact that the simius, like Hermogenes, was a cantor. In i. 10. 80 the reference is only inci- dental, Fannius being mentioned as the parasite of Tigellius. There is again an allusion to the singing in i. 10. 90, where Horace tells Demetrius and Tigellius to go "wail" (plorare) among the "easy- chairs" (cathedrae) of their female pupils. 1 Demetrius is a Greek name, cathedrae were Greek chairs whose use was considered effemi- nate at this time, and iubeo plorare is an evident translation of the Greek icekevoo Khaeiv or olfm^eiv, as editors have noted, and a male- diction in place of iubeo valere. 2 The heaping up of Greek references may be significant, especially in connection with i. 10. 20 ff. (see below, p. 284). In plorare there probably is also a humorous refer- ence to Tigellius' singing, as some editors suggest. We see then that in the main it is the singing of Tigellius that is singled out for satirizing, and that there is therefore all the less reason for supposing that there was more than one Tigellius. In the tenth satire and probably elsewhere this emphasis on the singing serves the purpose of implying that Tigellius knows nothing about poetry, as the scholiast well saw. 3 To Horace poetry was an art and its practice was a profession. The best comment on Horace's attitude toward Tigellius as a literary man is the passage in Epist. ii. 1. 108 ff. : "Everybody, young and old, is crazy to write verse" is, in effect, what Horace says. " Sailors, physicians and skilled workmen have undisputed possession of their respective fields, but scribimus indocti doctique poemata passim." Tigellius is a literary dilettante, whose work appeals to the untrained volgus, but who has no knowledge of the subject. His training is narrow and superficial, for he has not even studied the masterpieces of the Old Comedy, and his pupils (Serm. i. 4) and the last is specifically attributed to Tigellius, it may be that we should infer that Tigellius was also a dancer. In that case simius of i. 10. 18 may contain an allusion to dancing, as tW^kos in Arist. Poetics 1461 b 34. 1 "Demetri, teque, Tigelli, Discipularum inter iubeo plorare cathedras." » Cf. Serm. ii. 5. 69. 'Serm. i. 2. 3: "Ideo autem dixit cantoris, quia dicebatur in poematibus suis placere voce, non carminum probitate." 278 B. L. Ullman can do nothing but cantare Calvum et Catullum. His proper place, therefore, is inter cathedras of his music pupils, for he is a good musi- cian, just as Alfenus is a good cobbler (i. 3. 130). Cobbler, stick to your last! Looking over the examples from another point of view, we see that the earlier references to Tigellius (in i. 2 and i. 3) are veiled satire, as is one of the later references in i. 9, though here the satire is more apparent. In i. 4 Horace is somewhat more outspoken, In i. 10, probably the last in date of composition of the satires of the first book, Horace allows himself three references. In the first of these he associates Tigellius with a simius, in the second with Pantilius, the cimex, and ineptus Fannius. Immediately after Tigellius' name there follows a long contrasting passage containing names of important men, and the first verse of the passage contains names only ("Plotius et Varius, Maecenas Vergiliusque"). At the end of the passage Horace attacks Tigellius again, for the first time abandoning all innuendo, by telling him to go plorare. The poem, as the book, then ends with the line, "I puer atque meo citus haec subscribe libello." The line is usually (and, I think, rightly) taken to mean that the satire is intended as an epilogue for the book (libellus). But we may still see a grain of truth in the interpretation of the scholiasts that Horace wants the slave to add these words about Tigellius to his poem because they are so appropriate. I should interpret thus: "Go slave and add these verses [i.e., the whole poem and specifically the last lines] to my book [that they may serve as an epilogue], while I still have the courage to back up my insult to Tigellius." The explanation of Horace's method of attacking Tigellius I would find in Serm. i. 3. Tigellius was a great favorite of Augustus, as he had been of Julius Caesar, and Horace did not dare at the outset to attack him openly. The most that he could do was to call him a can- tor, with a hidden sneer, for this was the truth and not denied. With the passage of time after Tigellius' death, Horace could allow himself greater freedom. It must be remembered, too, that Horace probably became a friend of Augustus through Maecenas, and several of the satires were written before Horace was introduced even to Maecenas. Augustus is referred to but once in the first book, and then only as the friend of Tigellius, as just noted. Editors have remarked on Horace, Catullus, and Tigellius 279 the omission of Augustus' name 1 at the end of the tenth satire, where Horace hopes for the favor of Maecenas, Pollio, Messalla, and their circles 2 as contrasted with the circle of Tigellius. Though Horace's words take the form of a hope, it is clear from the context that these men actually do look upon his work with favor. The name of the emperor is omitted because it is he who must decide between the Horatian and the Tigellian style of poetry. Augustus is the judge, Maecenas and the rest are the witnesses. That the outcome of the trial is a foregone conclusion is clear enough. The farewell (or rather "fare-ill") to Tigellius is the last shot in the war, and Tigellius is never mentioned again. The last line of the first book represents the triumph of an artistic ideal. In the first satire of the second book all is serene, the emperor's favor has been won, and his name is men- tioned three times. 3 At the very outset (vss. 10-11) Trebatius says to Horace in a phrase which has become famous and typical, "Aude Caesaris invicti res dicere." This of course means that the suggestion has come from a high source that Horace should turn his talents — now recognized — to other things. More specific in showing Augustus' approval of the satires is vs. 84, in which Horace ends the whole discussion between Trebatius and himself by leading the trump card: "Bona [carmina] siquis Iudice condiderit laudatus Caesare?" 4 Caesar has handed down his decision in favor of Horace, the matter is closed, and Tigellius may rest in peace. II. HORACE AND CATULLUS The aim of the second portion of our inquiry is to interpret and classify Horace's theories of literary style as shown in the tenth satire, and thus to see what the artistic ideals were that put to rout the Tigellian school. In vss. 18-19 there is a reference to Horace's attitude toward Calvus, Catullus, and Tigellius ("quos neque pulcher Hermogenes umquam legit, neque simius iste Nil praeter 1 The Octavius of vs. 82 has sometimes been supposed to be Augustus, but those editors are undoubtedly right who reject this supposition on the ground that Horace never refers to Augustus by that name. « Cf. AJP, XXXIII, 161 ff. » He is mentioned but once more in the second book (ii. 6. 56), and there the refer- ence indicates nothing about his direct relation to Horace. • Cf. Kiessling's note. 280 B. L. Ullman Calvum et doctus cantare Catullum"). Since it is obvious that a correct interpretation of this attitude is essential for an understanding of Horace's views, it has seemed that my presentation would be most effective and convincing if it led up to the interpretation of these lines. Horace's supposed attitude toward Catullus, as indicated by this passage, has been much discussed, and not without reason, for it affects our understanding of the attitude of the poets of the Augus- tan age to their immediate predecessors. Interpreting Horace's attitude as unfriendly to Catullus, scholars have long searched for the possible grounds for disagreement between Catullus and Horace. Especially has Horace's claim to the distinc- tion of having introduced lyric measures to Rome been taken as a direct slap at Catullus. Others have even been led to say that all the poets of the Augustan age were opposed to Catullus. But the weight of reason is against all these assumptions. Horace's claims as the first Roman lyricist have nothing to do with his attitude toward Catullus. The vagueness of his brief remark in the Odes (hi. 30. 13) is cleared up by his fuller statement in the Epistles (i. 19. 25 ff.). Though he here mentions both Sappho and Alcaeus, he takes pains to put forth only Alcaeus as the one whom he first introduced to Rome. Catullus wrote no Alcaics. Besides, Horace points out that his originality is in the subject-matter. Of Catullus' two Sapphic poems, one is largely a translation, the other is purely personal. In contrast to these are the more serious poems of Horace, which are the ones he no doubt has in mind as assuring his future fame. The supposed opposition between Catullus and the Augustans simply does not exist. 1 It is a product of the imagination working on Horace's allusion in the tenth satire, and perhaps encouraged unconsciously by the modern controversy (carried on especially by Munro and Conington) concerning the poetical merits of Catullus and Horace, Lucretius and Virgil. On the contrary, a deep regard was felt for the earlier poets by their successors. Horace himself was an imitator of Catullus. 2 In that case, however, we are forced to argue that there is no sneer for Catullus in Horace's words — the i Rand, Han. St., XVII, 15 ff. 2 Perhaps in this very satire sint qualiacumque (vs. 88) is an imitation of Cat. i. 8, quidquid hoc libelli, qualecumque, as editors suggest. Hobace, Catullus, and Tigellius 281 sneer is only for his imitators. The argument is put in its most ingenious form by Rand: "In satirizing a Methodist of pre- Raphaelite leanings — I hope I am not straining analogy — whose acquaintance with English poetry was limited to two of his recently sanctioned hymns, the 'Crossing of the Bar' and the 'Recessional/ we should not thereby intend disrespect to Tennyson or Kipling." But this argument from analogy does not convince. If reason tells us that there is no opposition between Catullus and Horace, intuition tells us that Horace's words, as usually understood, do convey a sneer, be it ever so mild, at Catullus. We are left on the horns of a dilemma. Reason and intuition are opposed to each other, and we cannot lightly choose either. There is only one way out of the difficulty, and that is to reinterpret the passage in such a way that reason and intuition will not clash. 1 In trying to arrive at a correct conclusion about the interpretation of vss. 18-19 we have available four kinds of evidence: certain facts concerning the relations of Calvus and Tigellius; Horace's attitude in certain literary matters in this passage and elsewhere; Catullus' attitude in the same matters; the implications of the passage itself. Let us take up the evidence in order. We know from Cicero (Fam. vii. 24. 1) that Calvus made a bitter attack on Tigellius: "eumque [Tigellium] addictum iam turn puto esse Calvi Licinii Hipponacteo praeconio": "one, moreover, who (as I take it) had been even then [i.e., when he slighted me in a marked way] knocked down as a cheap lot by the scazontic hammer of Calvus" (Tyrrell). Porphyrio (on Serm. i. 3. 4) quotes a line from Calvus' epigram : " Sardi Tigelli putidum caput venit," 2 and Cicero's words show that he is alluding to this line. Cicero continues his allusions to it by quoting the proverb "Sardos venalis alium alio nequiorem" (cf. "Sardi .... venit") and by calling Tigellius a salaco (=pididum; see below, p. 292). It is evident that Tigellius' feelings toward Calvus could not have been very friendly after such 1 Scholars frequently comment on this clash, e.g., Norden (Einleitung in d. Altertumswias., I, 496): "Auch Horaz ist von Catull in den Epoden doch starker beeinflusst als sein Seitenhieb (s. i. 10. 19) auf einen freilich sklavischen Nachahmer erwarten lasst." 2 From veneo, not venio. 282 B. L. Ullman an attack, and one wonders about the meaning of vss. 18-19. Kirch- ner saw the difficulty and made use of it for his theory of two Tigellii: he argued that Tigellius (the Sardinian) could not have lived in close friendship with a man whose custom it was to sing nothing but the poems of his (Tigellius') most bitter opponent; that therefore the Hermogenes of vs. 18 cannot be Tigellius the Sardinian. But the difficulty is not lessened in any appreciable degree by this reasoning. For the sake of argument let us suppose that there were two Tigellii. In that case the younger must have been, as is generally assumed, an adopted son or freedman of the older. Is it likely that this younger man would be so ungrateful to one to whom he owed everything as to form a close friendship with the devoted follower of Calvus? It is possible, of course, just as it is possible that the eccentric Tigellius from Sardinia may have chosen to ignore Calvus' attack and to have continued his devoted admiration for his vilifier. But neither the one explanation nor the other is likely, and we leave the matter with the feeling that there is something radically wrong in the current interpretation of vs. 19. Our next problem is to determine some points in Horace's literary creed, especially as shown in the tenth satire. For this purpose it will be necessary to interpret the opening portion of this satire. Horace's criticism of Lucilius in the fourth satire had provoked replies from the successors of Tigellius Hermogenes, and so in the tenth satire Horace reasserts his statement that Lucilius' verse is rough, calling attention at the same time to the praise which he had given Lucilius for his wit. But to make people laugh is not the sole aim of satire. 1 Brevity, geniality as well as invective, urbanitas as well as rhetorical and poetical language, are necessary. Geniality is more effective than invective, as the writers of the Greek Old Comedy well knew — but their plays have never been read by Hermogenes or his hench- man, who has learned only to cantare Catullus and Calvus. "But," says Horace's imaginary opponent of the ilk of Hermogenes, "Lucilius 1 "Non satis est risu diducere rictum Auditoris." After long and careful study of this passage I am more confident than ever of the correctness of Hendrickson's thesis (CP, VI, 129; cf. VIII, 172) that the absence of the word eatura and the resulting vagueness is due to Horace's unwillingness to use the word satura. I should say that the omission is just as striking here as any in i. 4. That Horace's words here are not meant to apply to all poetry, as some would hold, can easily be shown (why, e.g., would poetae be mentioned in vs. 12 ?) ; he is merely giving the recipe for Horatian satire. Horace, Catullus, and Tigellius 283 did a great thing in mixing Greek words with his Latin." "Late- learners," retorts Horace, "do you consider that wonderful when even Pitholeon succeeded in doing it ? " "But," the opponent replies, "the blending of the two languages, like the blending of Chian and Falernian wine, is more pleasant." Horace answers by asking whether this should apply to oratory also, and introduces Messalla's name as that of the typical orator. First of all it will be well to discuss together the meanings of several words in the passage, because a general principle of interpre- tation is involved. In vs. 21 Horace uses the words seri studiorum. Editors regularly point out that this is a translation of the Greek 6ipi.fJia.8eh, but only two apparently (Wickham and Lejay) call atten- tion to the really important point that Horace is setting the example of avoiding Greek words, for oif/i/xaSels must have had some currency in Latin, being used by Cicero (Fam. ix. 20. 2; cf. Gell. xi. 7. 3). But his translation is rather crude, 1 and was no doubt intended for humorous effect. Horace was laughing at himself and inviting his readers to laugh with him — ridentem dicere verum. Now such an undercurrent of meaning is not unique, it is a common feature of Horace's manner in the satires and must be looked for on all occa- sions — it is part of his felicitas. There are other examples in the immediate vicinity. Simius in vs. 18 has been interpreted in two ways, as referring to the ugliness of the person concerned or else to his aping of Hermogenes. There are parallels for both of these meanings, and something is to be said for both. The latter is favored by the context, the former by the contrast furnished in pulcher, applied to Hermogenes. I do not see why we should not accept both interpretations. Horace had the idea of imitation in mind in the first place and then chose pulcher to contrast with simius in its other sense. 2 Again in doctus we have a double meaning. On the surface the meaning is that the simius was taught (by Hermogenes, of course) to sing, etc. But to see the names of that inseparable pair, 3 Calvus and Catullus, separated by the word doctus, inevitably 1 Serus with the genitive is an innovation of Horace's, according to Lejay. 2 With either interpretation it is clear that the simius was closely associated with Hermogenes. ■It is very common to find them mentioned together, e.g., Prop. ii. 25. 4; 34. 87— 89; Ov. Trist. ii. 427-31, and especially Am. iii. 9. 62 ("cum Calvo, docte Catulle, tuo ")• 284 B. L. Ullman reminds one of the frequent application of the epithet dodus to these Dioscuri of Latin literature, on account of their wide learning, especially in mythology. 1 The simius is dodus, yes, in the ordinary sense, as a schoolboy is dodus, but what a contrast between his dodus, with the narrow outlook it implies (nil nisi Calvum, etc.), and that of Catullus! Finally we have a double meaning in cantare. Whatever it may mean in this line (see below), it cannot be used in the literal sense. It is not reasonable to suppose that Catullus' poems were actually sung. Yet the word was chosen to call to mind the idea of singing, with its derogatory force, as was suggested above (p. 277) ? Here Horace has been less felicitous than elsewhere in his choice of words, for the meaning of cantare is not at once clear to readers unfamiliar with certain contemporary happenings. It is our task to piece together the bits of evidence. In looking over the first forty lines of the satire as summarized above it is natural to wonder whether Horace's doctrines can be identified as those of some school of literary theory. We find a strong clue when we note that Horace objects to the use of Greek words and believes that Latin should be kept pure, citing the instance of Messalla. The doctrine and the authority quoted are significant, for Messalla, as is well known, was one of the leading Atticists of the day and purism was one of the leading tenets of the school. When, furthermore, we find Horace urging brevity (vs. 9), another cardinal principle of the Atticists, we are fairly certain that we are on the right track. As we look through the passage with this in mind, we see an astonishingly large number of allusions or possible allusions to Atticistic tenets, and we conclude that this is the unifying theme of the whole discussion. Particularly illuminating is a comparison with Cicero, the chief source for our knowledge of Roman rhetorical theory. 3 The Atticists, at whose head stood Calvus, believed in the • I find confirmation of my interpretation in Ellis, Commentary on Catullus, 2d ed., p. xxvii: "In the well-known nil praeter .... the choice of the word doclus was perhaps determined by its constant application to Catullus." * For the double meaning cf . plorare in vs. 90 (see above, p. 277) . > Lejay (intro. to Serm. i. 10, p. 258) has seen the close relation between Horace and Cicero, and has in fact suggested that Horace read Cicero, which may well be so. It is strange that Lejay did not push the matter to its logical conclusion. Again, Hendrickson, in his study of Roman Atticism, has noted traces of it in Horace and promised a discussion of it (CP, I, 104, n. 2). He must have had the tenth satire in Horace, Catullus, and Tigellius 285 exclusive use of the plain style (genus tenue), in which respect they were following the Stoic doctrine. 1 The grand style was freely used by the Asiatics, while Cicero as an eclectic asserted the ad- vantages of the plain, middle, and grand styles as occasion demanded. The first quality which Horace mentions is brevity, which is one of the cardinal points of Stoic doctrine in the fullest account that we have. 2 In Roman Atticism it was a quality that did not need special mention, being implied in other qualities. It is, however, specifically mentioned in connection with Scaevola (Cic. Br. 145), a devotee of the plain style. It is a quality which Lucilius and his successors lack (Serm. i. 4. 9 f.; i. 10. 60 f.). Then Horace says that there is need of invective (sermo tristis) occasionally (modo), but (more) often (saepe) of geniality (sermo iocosus) ; here Horace implies that Lucilius had the former but not the latter. It is clear that the Lucilians of Horace's day wanted the sermo tristis only, or at least that Horace charges them with this desire. Tristis means "biting" or "savage," and in its place acri is used in vs. 14, as ridiculum is used for iocosum.* Yet it gets this meaning only from the context: Horace is applying a general principle of style to a specific department, mind. For the transference of rhetorical theory to the field of poetry, Norden has a noteworthy article in Hermes, XL (1905), 481, "Die Composition u. Litteraturgat- tung der Horazischen Epistula ad Pisones." 1 Hendrickson, "Origin and Meaning of the Ancient Characters of Style, AJP, XXVI, 249 ft ! Diog. Laert. vii. 59. ' Hendrickson was not the first to propose this interpretation, as he thought (Gildersleeve Studies, p. 152). It was explicitly given by Schiltz in his edition of 1881, and is perhaps implied by others. It has since been adopted by Lejay. As regards the meaning of tristis , there is no sharp distinction between the meanings ' ' serious ' ' and "biting." The serious may become stern, and the stern may become biting invec- tive. One thing must be guarded against and that is the assumption that tristis and acer necessarily represent a form of wit, and that by contrast iocosum and ridiculum are humor. Horace is not dealing with humor in the narrow sense (facetiae, Cic. Or. 87), as contrasted with wit (dicacitas, ibid.), but with good-natured jest as contrasted with invective. In none of the parallels quoted for the meaning of tristis is the idea of jest necessarily present, and in some of them it is out of the question. One may compare, too, the elegiac use of (rise's as "cross." The relation of invective to wit is this: first, invective may sometimes (not always, by any means) cause laughter, which is especially true in satire, about which Horace is writing; second, it is often difficult to draw the line between wit and invective, because a given remark may contain both. So, too, Quintilian (vi. 3. 18) points out that what is salsum is not necessarily ridiculum, for the former is the broader term. Horace's point of view about the ridiculum and its proper sphere is shown in other passages, but a discussion of this must be left to another occasion, because other matters are involved. 286 B. L. Ullman satire. So it is that we can perhaps see in the phrase sermo tristis a reminiscence of Cicero's characterization of the grand style. 1 As a stylistic term it is contrasted with hilaris by Quintilian (viii. 3. 49; cf. Horace's iocosus). It sometimes has an unfavorable connotation, whereas the similar word severus has not. 2 Severus again is contrasted with comis (Cic. Or. 34). From a comparison of De off. ii. 48 with other passages (e.g., De or. iii. 1777), it appears that comitas adfabilitasque sermonis is a characterization of the plain style. Acer, too, is an epithet of the grand style which Horace has here given a special application, as is natural enough. Horace himself uses it of the grand style (poetry) in this satire: acer Varius (vs. 43; cf. i. 4. 46). Cicero applies it to the grandiloqui in Or. 99 . 3 For the meaning of vss. 14-15 ("Ridiculum acri fortius et melius magnas plerumque secat res") editors rightly compare De or. ii. 236, which Horace seems to have had in mind. Cicero tells why wit is worth while: "quod ipsum oratorem politum esse hominem significat, quod eruditum, quod urbanum, maximeque quod tristitiam ac severitatem mitigat et relaxat odiosasque res saepe, quas argumentis dilui non facile est, ioco risuque dissolvit." While Cicero is not dealing here with the tenuis orator specifically, it must be remembered that he finds the proper place for jest in the plain style. Further- more, it is noteworthy that two of the epithets here given to the orator, politum and urbanum, are put among the three epithets which the Atticists strive for in Br. 285, while the third, eruditus, is in Br. 133 associated with elegans, the third epithet which Cicero attributes to the Atticists in Br. 285. 4 1 Tristi oratione. Or. 20. Horace's use of sermo instead of oraiio may show how he adapts to satire Cicero's dicta concerning oratory. Horace likes to describe his satire as sermo, contrasted with epic poetry on the one hand and rhetorical prose on the other. 2 Severa, non tristia, Quint, xii. 10. 80. 8 The Auctor ad Herennium speaks of amplificatio in connection with the genus grande (iv. 8. 11). His words show that he is thinking of his threefold division of the mollitudo vocis (iii. 13. 23) into sermo, contentio, and amplificatio. In general, sermo belongs to the plain style, the other two to the grand style (comparison with Cicero indicates the same thing). Contentio is described as oratio acris, while one side of amplificatio is described as oraiio quae in iracundiam inducit, and more fully later (24) as oraiio quae aliquod peccatum amplificans auditorem ad iracundiam adducit. This is exactly the idea that we get of Lucilius from passages like Juv. i. 165, and Hor. Serm. i. 4. 3 f. and ii. 1. 68. • Eruditus is twice applied by Cicero to Brutus, very probably an Atticist (Or. 40, 174). Hobace, Catullus, and Tigellius 287 The terms iocosum and ridiculum can also be illustrated in an interesting way from the field of rhetoric. The Auctor ad Herennium and Cicero find a place for jest in the plain style only, because they both present the view that all styles should be used as occasion de- mands, and they apparently hold that the grand style is not suitable for jest. Both warn against certain kinds of jest: "Iocatio [cf. Horace's iocosum] est oratio quae ex aliqua re risum pudentem et liberalem potest comparare. " : Cicero (Or. 88-89) has a much fuller treatment of the ridiculum, as he calls it (cf. Horace). Of particular interest in connection with Horace is the statement that the tenuis orator will not use wit that is pert (pelulans) lest it become impudent (improbum), and especially that such a speaker "vitabit insanabilis contumelias, tantummodo adversarios figet," etc. In other words, there is a point where a joke ceases to be a joke and becomes brutality and invective (acre, triste). Cicero indicates the incompatibility of invective and the plain style (of the philosopher) in Or. 64: "nihil iratum habet, nihil invidum, nihil atrox," etc. The last word is particularly noteworthy. 2 Horace's next point is that there is need occasionally of the rhe- torical and poetical, (more) often of the urbane. 3 It is implied that Lucilius and his followers had the first quality but not the second. The description of the urbanus reminds the editors of the Greek elpuv. Yet this comparison has obscured rather than illumined the passage, in that it seems to have tended to block further reflection on the meaning of Horace's words. The dpuveia must be a part of something broader. It is obvious that urbanus is contrasted with rhetoris atque poetae, just as iocoso (cf. ridiculum) is contrasted with tristi (cf. acri). But what is the point of the contrast? Scholars do not express an opinion on this point, probably because they imagine that urbanus means the "man of wit," contrasted with the 1 Ad Her. Hi. 13. 23. Iocatio is classed under sermo, for which see above, p. 286, n. 3. • De off. i. 104 describes the two kinds of jest, but not from the point of view of style. • "Sermone opus est modo tristi, saepe iocoso, Defendente vicem modo rhetoris atque poetae, Interdum urbani, parcentis viribus atque Extenuantis eas consulto." The form of expression shows that interdum is a variation for saepe, and the context shows that Horace thinks that the second quality in each pair (i.e., iocoso and urbani) is the more important, and that the inclusion of the first is largely a concession to the opposition. 288 B. L. Ullman (seriousness of the) rhetorician and poet, thus amplifying what they take to be the thought of vs. 11 (tristi .... iocoso). 1 But urbanitas is a very broad term, and wit (of a certain kind) is but one of its special applications. Though urbanitas as applied to satire would of course involve wit, it is not that side of it that Horace is stressing, but rather its general significance. For he goes to the trouble of defining the urbanus as parcentis viribus, etc. The definition is an excellent one, for urbanitas may perhaps be best rendered in English as "restraint." 2 But how does this contrast with rhetoris atque poetae? Is it that wit (as urbanitas is usually understood) is re- strained, whereas seriousness (rhetoric and poetry) is unrestrained ? Surely not. Or is it that urbane, restrained wit contrasts with unre- strained wit ? Epic poetry and rhetoric can hardly be called wit of any kind. The contrast must be between restraint and lack of restraint — and wit does not specifically enter into the matter at all. The only satisfactory explanation seems to be from the standpoint of stylistic doctrine: urbanitas is one of the qualities of the plain style and one of the ideals of the Atticists. 8 As for rhetoric, it was itself almost synonymous with the grand style, since the plain style was that of the Stoic grammarians, who opposed all rhetoric. 4 It is 1 Hendrickson (AJP, XXV, 135, n. 1) compares the line with Aristotle's flexible, changing mean. The comparison to the mean is of no significance, for any quality that Horace defends is likely to be one of moderation. The idea of flexibility Hendrick- son seems to see in a contrast between parcentis and extenuantis, a contrast which obviously does not exist. Augeo, not parco, would be necessary. 'Mueller well comments: "Ein Haupttheil der urbanitas ist das Masshalten, die Riicksicht auf das nequid nimis." For the various meanings of urbanitas, cf. O. Lutsch, "Die Urbanitas nach Cicero," Festgabe fUr W. Crecelius, Elberfeld, 1881, pp. 80 ff. • If the Latin equivalent for Hellenism is Latinitas, as Hendrickson (CP, I, 100) says, the Latin equivalent for Atticism might be said to be urbanitas (from urbs, i.e., Roma). At any rate, urbanitas sums up better than any other one word the Roman Atticistio ideal, clpaveta. is too narrow a term, being but one application of urbanitas. Cicero says that dpwvda. belongs properly to the plain style (sermo, not contentio, De or. iii. 203). ' Hendrickson, AJP, XXVI, 290. The discussion centers around the opposition between sophistry-rhetoric-grand style on the one hand and dialectic (philosophy)- grammar-plain style on the other. Hendrickson's comment on Diogenes' definition of Hellenism (ibid., p. 258), the first quality of the plain style, serves admirably as a comment on Horace's words: " . . . . Conversational idiom (as opposed to the poetical and elaborated style of conventional rhetoric)." Diogenes' Hellenism, then, corre- sponds to Horace's urbanitas, in so far as the latter applies to diction. Horace's words are of course meant to have a wider application. Horace, Catullus, and Tigellius 289 hardly necessary to dwell on the close relation between rhetoric and poetry (i.e., epic and tragic poetry, which are in the grand style). 1 The very words that Horace uses in describing the urbanus show that his thoughts were centered on the theories of style. Parcentis is paralleled by Cicero's frequent use of various words of the same stem in discussing the plain style (parous, Or. 81, parcius, parcum, 83, parsimoniae, 84). Vis and vires are often applied to the grand style; of oratory, e.g., Or. 76, 97; of poetry, e.g., Hor. Serm. i. 4. 46; ii. 1. 12; Epist. ii. 1. 259. Especially noteworthy is the contrast of vis and elegantia in Br. 89; Galba (an Asiatic) had vis, Laelius had elegantia, the "watchword of the Atticists." 2 Extenuantis is an evident allusion to the genus tenue? cf. Carm. iii. 3. 72, magna modis tenuare parvis, where Horace turns away from the grand style of epic, and especially i. 6. 9, tenues grandia, where the genus tenue and the genus grande are obviously and strikingly juxtaposed, though scholars have not noted the fact. 4 For the description of the urbane Atticist as a man of restraint it is interesting to compare Martial (viii. 70), who calls Nerva the Tibullus of his time. The epithet is explained by vs. 2, vires cohibet pudor, for Tibullus was an Atticist and restraint was his chief quality. 6 Just as Martial's vires cohibet helps to elucidate Horace's parcentis viribus, so in turn another passage of Horace (Epist. ii. 1. 258-59) elucidates Martial's pudor: "nee meus audet Rem temptare pudor quam vires ferre recusent." Horace declines to enter the field of epic poetry, with its grand style. 1 Cf. especially Theophrastus Ap. Ammon. in At. de Interpr. Com., p. 65. 31 (cited by Hendrickson, op. tit., p. 255) and Norden, Antike Kunstprosa, I, 30 ft., 73 ff. » Hendrickson, op. tit., p. 264. ' Called figura extenuata in Auct. ad Her. iv. 8. 11. 4 In the Pseudacronian scholia on this line there is a note extenuat vires suas, which is more intelligent than it seems, for Horace is made to interpret himself by a quotation of Serm. i. 10. 14. Norden (Einleitung in d. Alt., I, 505) is right in seeing an allusion to the genua tenue in two other passages of Horace, though he would have done better to quote the ones mentioned above: "Gedichtchen des laxvov y4ms — spiritum Graiae tenuem Camenae (Carm. ii. 16. 38), tenui deducta poemata filo (Epist. ii. 1. 225) : das ist der auch von Augustus approbierte attizistische Klassizismus auf die Poesie ilbertragen." Since the above was written, Jackson has independently pointed out the allusions to the genus tenue in Carm. i. 6. 9 and iii. 3. 72 (Harvard Studies, XXV, 134). »Cf. Burger, "Beitrage z. Elegantia Tibulls," in Leo, Xdpires, p. 393; AJP, XXXIII, 167, note. 290 B. L. Ullman Attention has already been called to the circumstance that Horace's attack on the use of Greek words in Latin was good Atti- cistic doctrine. We may pause for a word about exsudet. Horace charges his opponent with dragging in a Greek word when he is at a loss for a Latin word, while Messalla perspires (exsudet) to find the proper Latin idiom. Sudo is used in a similar sense in AP, 240, a passage in which Horace gives his theory of poetry in words taken from Cicero's description of the genus tenue of the Atticists. 1 It is passages such as these which show why Horace tried to avoid the terms satura and schedium for his own work. 2 In vs. 36 Horace turns from the subject by saying that he leaves epic poetry (with its grand style, as Morris rightly interprets) to Alpinus, while he himself ludit (in the plain style). Alpinus is de- scribed as turgidus, obviously because of his style; cf. Epist. ii. 3. 27: professus grandia turget. Similarly tumidus and inflatus are epithets of the Asiatics in Quintilian xii. 10. 16-17. Turget et inflata est is applied to an exaggerated grand style in Auct. ad Her. iv. 10. 15. In Semi. ii. 5. 41 the same Alpinus seems to be alluded to under the name Furius, 3 and he is described as pingui tentus omaso. Wickham was tempted to explain turgidus as "stuffed out" on the analogy of this phrase. But the situation is reversed: the phrase is eluci- dated by turgidus. It is an allusion to Furius' turgid style, which Horace illustrates by a quotation, but he chooses to play on the meaning of the rhetorical term. Since Furius was a Gaul, Horace makes mention of omasum, "tripe," then as now a favorite Gallic dish. Pinguis itself is an epithet for the grand style, like crassus. But most striking as illustrating Horace's words is Cicero's descrip- tion of the Asiatic style as "opimum quoddam et tamquam adipatae [fatty] dictionis genus" (Or. 25).* In vs. 65 limatior is a stylistic 1 Horace says: "Ex noto fictum carmen sequar, ut sibi quivis Speret idem, sudet multum frustraque laboret Ausus idem." Cf. Cic. Or. 76: " [Orator tenuis] summissus est et humilis, consuetudinem imitans, ab indisertis re plus quam opinione differens. Itaque eum qui audiunt, quamvis ipsi infantes sint, tamen illo modo confidunt se posse dicere; nam orationis subtilitas imitabilis ilia quidem videtur esse existimanti, sed nihil est experienti minus." 2 CP, VIII, 189. « My interpretations of Horace's epithets of Furius have a bearing on the identi- fication of this writer with Furius Bibaculus. < Cf., too, Verg. Cat. 5. 4, natio madens pingui (of rhetoricians). Hoeace, Catullus, and Tigellius 291 term which, as Lejay remarks, Horace took from Cicero. But, it should also be remarked, Cicero regularly uses it of the plain style of the Atticists (cf. Or. 20 and Sandys' note). Further illustrations of all these matters might be adduced and other words and phrases in this satire might be pointed out as pos- sibly suggestive of the dispute between Asianism and Atticism, 1 but it has, I think, been conclusively shown in the foregoing that the satire must be interpreted from the standpoint of this dispute. I lay especial stress on the fact that a satisfactory antithesis between the urbane and the poetic-rhetorical (vss. 12-13) can be discovered in no other way. The use in this satire of so large a number of terms which seem to refer to this dispute is also of moment. Horace is attacking the extreme Asiatic view. He himself is not an extreme Atticist, though inclining more to that side than the other (vss. 11- 14). His temperamental inclination toward aurea mediocritas modi- fied the extreme views of the Atticistic school to which he belonged. The opposition was represented by Tigellius and his circle, who, asserted that they were following the tradition of Lucilius. 2 We can see then the significance of contrasting Tigellius and his coterie of Asiatics in vss. 78-80, 90-91, with Pollio the Atticist and with the circles of Maecenas and Messalla 3 (vss. 81-87), the tendencies of which were decidedly Atticistic at the time this satire was written. 4 The Asiatic tendency of Tigellius is perhaps hinted at in the epithet putidum caput, applied to him by Calvus, for putidus is 1 Ineptus, applied to Fannius, recalls the use of this word for a quality opposed to the Attic ideal (cf. Br. 284, and especially Or. 29, with Sandys' note). Crispinus, another follower of Tigellius, is called ineptus in Serm. i. 3. 138. The mention of Laberius in vs. 6 recalls this writer's impure diction (Gell. xvi. 7). Cassi rapido ferventius amni Ingenium (vs. 62) reminds one of Cicero's description of the Asiatic style of his youth (Br. 316) : Quasi extra ripas diffluentis. Horace's ferventius is paral- leled by deferveral oratio in the same passage of Cicero. In writing this paper, I pur- posely left out of the discussion the phrase molle atque facetum in vs. 44 because it seemed to me to need a special paper. This need has since been excellently met by Jackson in Harvard Studies, XXV, 117 ff. He correctly interprets the phrase as an allusion to the plain style, and his whole paper harmonizes perfectly with the present discussion. 1 Horace implies (vs. 65) that there is some ground for believing that Lucilius' tendencies were Atticistic compared with other writers of his time, but that Lucilius was more or less of an Asiatic compared with the higher standards of Horace's day. • I have argued elsewhere (AJP, XXXIII, 161 ff.) that the men mentioned in vss. 80-87 belonged to one or the other of these circles. ' Cf., e.g., Hendrickson, CP, I, 104. 292 B. L. Ullman commonly applied to the Asiatics in the sense of "affected" (e.g., Or. 27 and Br. 284, where it is joined with ineptus, for which see above, p. 291, n. 1). Cicero in the letter (Fam. vii. 24) in which he alludes to Calvus' epigram seems to take putidum in this sense, for he calls Tigellius a salaco, which is its proper Greek translation. 1 Now Calvus was the head and front of the Atticists, and his close friend Catullus was in entire accord with him. 2 Is it possible that the followers of Tigellius, 3 Asiatics all of them, would imitate or praise or sing or recite the poems of the Atticists Calvus and Catullus ? We leave this matter with a feeling even stronger than before that the current interpretation of vs. 19 is wrong. Now for the third point. It is not necessary to make a detailed examination of Catullus' relation to Atticism, but it is of importance to find out his attitude toward those qualities of style which Horace discusses in the immediate vicinity of our famous vs. 19. First, there is the matter of Greek words. Here we may quote Ellis (Com. on Cat., 2d ed., p. xxxi): "The Latin of Catullus will bear comparison with that of Lucretius in its purity : Lucilius, whom Catullus some- times copies, had spoilt his satires by a barbaric admixture of Greek; M. Varro was repeating Lucilius' error in his own time: with these two warnings before him Catullus wisely introduced Greek sparingly, and preferably such words as had become or were becoming natural- ized." Ellis follows this comment with a list of the Greek words in Catullus. Horace's custom is exactly the same. Kiessling (p. xxi) says that Horace's vocabulary is confined almost exclusively to the simplest expressions of the cultured conversational idiom without a strict avoidance of such foreign words of Greek or Gallic origin as were necessary in daily life. Horace uses a fair number of Greek words in his satires. In the matter of the ridiculum and acre, Catullus' position would not be essentially different from that of Horace. It is true, of course, that there is much of the acre in Catullus, and scholars have thought 1 Hesychius defines is contrasted with the etptov in Ar. Eth. Eud. 1234a and elsewhere (cf. Ribbeck in Rh. Mm., XXXI, 388), and dpaneia has much in common with Atticism (see above). 2 Cf., e.g., CP, I, 103. 8 It makes absolutely no difference whether there were one or two Tigellii. Horace, Catullus, and Tigellius 293 of it in interpreting Horace's reference to Catullus. But there is also a great deal of the ridiculum, as everyone knows, and, besides, Horace concedes that there is plenty of room for the acre (witness his own Epodes). As for urbanity and brevity, the other Atticistic qualities which Horace urges in the tenth satire, Catullus' poems not only speak for themselves, but set forth their author's position in so many words. In the twenty-second poem, addressed to Varus, 1 he good- naturedly attacks Suffenus, a bad poet, of whom he says: "Homo est venustus et dicax et urbanus, Idemque longe plurimos facit versus." Idem means "notwithstanding this," as Merrill says, and indicates an "unexpected contrast," or, better, an impossible combination. Catullus of course means that Suffenus is anything but venustus et urbanus, as he pretends to be. 2 The idea of the two lines is exactly that of Horace's tenth satire. Urbanity (the goal of the Atticists) is bound to result in brevity. 3 Both qualities are insisted upon by Horace as by Catullus. Catullus goes on to say that Suffenus must have written ten or more thousand verses. 4 This recalls Horace's words about Lucilius in i. 4. 9 f. and i. 10. 60 f. Still closer to Horace is Catullus' language in poem 95, where Cinna's epic, carefully worked over and polished for nine long years, 5 is con- 1 Editors agree that this is Quintilius Varus, spoken of with great respect by Horace in AP 438 as a literary critic of refined taste. Since it is evident from Catul- lus' poem that he and Varus are agreed on the very matter of good taste, it would be logical to argue that Catullus and Horace are agreed in this matter. 2 Similar to this is Horace's "fuerit Lucilius comis et urbanus" (.Serm. i. 10. 64). The Lucilians maintained that their idol was comis et urbanus, and Horace grants it only for the sake of argument. A close parallel for idem is Mart. i. 9. 1 : " Bellus homo et magnus vir idem." ' Cf . Quintilian on Horace's Satires (x. 1. 94): "tersior ac purus magis [quam Lucilius]." • "Puto ego esse illi milia aut decern aut plura." Editors have difficulty with this (Baehrens even emends) because the double aut is regularly used where the alter- natives presented are mutually exclusive, and such they believe not to be the case here. But the editors take this wrongly, I think; the alternatives are exclusive. Catullus is moek-serious throughout: "If I'm not mistaken [puto], he has written — let me think — it's either ten thousand or — more." We should expect eleven thousand or some other exact number. The word order favors this interpretation, as does the position of plura at the end of the line, the favorite place for a irapi. irpocrfioKlav. I should punctuate with a dash before plura. Similar cases of pretended exactness are 5. 7 ff. and 7. 3 ff. 6 Horace takes the same attitude as Catullus on this point. In AP 388 he advises writers to hold back their work till the ninth year. Editors of Horace quote the Catullian passage, and it is likely that Horace had it in mind. 294 B. L. Ullman trasted with Hortensius' facile output of five hundred thousand verses a year. 1 Everyone knows that Hortensius was to Asianism what Calvus was to Atticism. Volusius is attacked in the same poem for the same quality of voluminousness as Hortensius. In poem 36 his Annals are described as pleni ruris, i.e., lacking in the urbanity of the Atticists (cf. 22. 9-10, 14). Rhetorical and poetical (i.e., epic) qualities, the importance of which Horace minimizes, are con- spicuous for their absence in Catullus' shorter poems, which alone enter in the discussion. It is clear then that Catullus and Horace are entirely in harmony on the points which Horace discusses in his tenth satire, and that Catullus, like Horace, attacks those who hold different views. The evidence that I have presented cannot be ignored in the inter- pretation of vs. 19. No doubt Catullus and Horace were not in exact agreement in all their views about life, but we must interpret Horace's attitude toward Catullus as indicated in the tenth satire from the standpoint of the subject-matter of the satire. Horace is defending a theory of style which is essentially Atticistic and attacking one which may be called Asiatic. 2 Calvus and Catullus are leading Atticists. This it is that Horace must have in mind concerning them, not that they were veurepoi, — imitators of the Alexandrian writers, to whom Horace is wrongly 3 supposed to be opposed — or hostile to Caesar, 4 or anything else. Grant, as everyone must, that the tenth satire deals with matters of style, and that Horace and Catullus are one in their attitude in favor of the plain style, it might still con- ceivably be said that Horace attacks the Tigellians because they 1 A verse of Catullus' poem is lost and either anno or die must be supplied. s I do not of course insist on these terms. In the nature of things they are flexible terms, like radical and conservative, but they are convenient for purposes of classi- fication. For the interpretation of vs. 19 it is sufficient to show that Catullus and Horace are entirely in harmony on the points which Horace discusses. » Cf . Lejay, p. 258: "La minutie D'Horace rappelle souvent celle des Cantores Euphorionis Horace avait le m6me souci de perfection que ces patients et curieux ciseleurs d'epyllies." Again, p. 254: "Limitation des Alexandrins n'est pas tout a fait bannie de ses odes." It seems to be generally assumed that all Alexandrians were alike in every respect. Nothing could be less true. As far as style is concerned, Calvus, a vedrepos, was an Atticist, while Hortensius, also probably a veiirepos, was an Asiatic. Nor does Alexandrianism always mean antagonism toward all earlier writers. Catullus foreshadows the eclecticism of the Augustan age by his interest in Sappho. < Scholars forget or ignore that Calvus and Catullus were reconciled to Caesar (Suet. Jul. 73) and that Tigellius was a great favorite of Caesar and Augustus. Horace, Catullus, and Tigellius 295 out-Calvus Calvus in their tenuity. But this is evidently not so; he attacks them and their idol Lucilius for their grandiloquent qualities — verbosity, savage invective, obscurity, impurity of diction, inur- bane rhetoric, exuberance, turgidity, ineptness. They are therefore no imitators of Calvus and Catullus, the great masters of the plain style. Let us revert again to Horace's words. Ridiculum is better than acre, says Horace, and the writers of the Old Comedy knew this well. But Hermogenes does not know this and therefore he and his fol- lowers are devoted to the acre alone ("nil praeter Calvum et doctus cantare Catullum"). If we accept the current interpretation of cantare, it is evident that we should expect, not the names of Calvus and Catullus, but that of Lucilius. The preceding discussion has made it clear that Calvus and Catullus can hardly be classified with Lucilius. The passage itself shows that the current interpretation is wrong. The difficulty is centered on the word cantare. We have already rejected the literal meaning ,as unlikely, if not impossible. Other meanings that have been suggested are "recite," "praise," and "imitate," the last named apparently being the favorite. If Horace had one of these meanings in mind, it makes little difference which, for, in the light of the foregoing study, the sense must be ironical: "the simius who has been taught to imitate (?!) Calvus and Catul- lus," implying that he would imitate (or recite, etc.) anything in the world rather than Calvus and Catullus. This interpretation is perhaps favored by the fact just noted that, instead of the names of Calvus and Catullus, the reader naturally expects that of Lucilius. Of the various meanings suggested for cantare, "imitate" is the most natural, but is unparalleled except perhaps by the phrase cantores Euphorionis (Cic. Tusc. iii. 45), the meaning of which is very uncer- tain and is often explained by reference to the supposed meaning of cantare in Horace. "Praise" can be paralleled, being merely a variation of the notion of "tell about," but the sense is not very satisfactory. "Recite" is on the whole the most likely. 1 But it is 1 The Thesaurus puts the Horatian example under alicuius carmina cantare, but this must be meant loosely to include the idea of "reciting," for the only other passage cited is Juv. xi. 180, where the meaning is clearly "recite." 296 B. L. Ullman possible to solve the problem without recourse to irony. Cantare may mean to "tell about" not only in a good sense but also in a bad sense. For the meaning "tell about" as applied to persons the Thesaurus quotes two examples of the unfavorable connotation as against one of the favorable (Cic. Q. fr. ii. 11. 1). One of the two is from the satires of Horace themselves, ii. 1. 46: "Qui me com- morit Flebit, et insignis tota cantabitur urbe." The parallel is the more convincing because this satire is similar to the tenth in dealing with Horace's relation to satirical writing, and as one reads he naturally thinks of the tenth satire: "I will satirize whoever attacks me — as the simius satirized Calvus and Catullus." We may per- haps preserve the tone of the tenth satire by paraphrasing thus: "The simius who was taught by his teacher Hermogenes to sing but one song, 'Calvus and Catullus, Calvus and Catullus.'" While it is scarcely possible to summarize this paper, containing as it does a great deal of detailed interpretation, the following are the chief points brought out: It is unnecessary to assume that there was more than one Tigellius. Horace's references to him are always more or less satirical, his opposition being due to differing ideals of literary style. Horace's tendencies are in the main Atticistic; Tigellius has the faults of Asianism. Calvus and Tigellius were personal and literary enemies. Catullus and Horace are in entire harmony on the literary matters with which the tenth satire deals. Hence Horace's reference to Calvus and Catullus must be interpreted in such a way as to bring out the opposition of Catullus, Calvus, and Horace to Tigellius. University op Pittsburgh