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. A PHASE OF THE DEVELOPMENT OF PROSE WRITING AMONG THE ROMANS 1 By Charles Knapp In his Etymologiae i. 38. 2, Isidorus, Bishop of Seville, speaks as follows of the beginnings of Latin prose writing: Praeterea tam apud Graecos quam apud Latinos longe antiquiorem curam fuisse carminum quam prosae. Omnia enim prius versibus conde- bantur, prosae autem studium sero viguit .... apud Romanos .... Appius Claudius Caecus adversus Pyrrhum solutam orationem primus exercuit. Centuries earlier Cicero had expressed a similar sentiment (Brutus 61). Having named and characterized, so far as was pos- sible, various early orators of Rome (52-60), he says of Cato Censor (61), "Nee vero habeo quemquam antiquiorem, cuius quidem scripta proferenda putem, nisi quem Appi Caeci oratio haec ipsa de Pyrrho et nonnullae mortuorum laudationes forte delectant." Cer- tainly if oratory is fairly tested by the standard of effectiveness, the speech of Appius Claudius Caecus deserved its reputation. For references to it see Cicero Cato Maior 16; Valerius Maximus viii. 13. 5. It was still read in Seneca's time (Seneca Epp. 114. 13). Of the laudationes funebres Cicero speaks in detail in Brutus 62. He condemns them vigorously, as corrupters of history (cf. Livy viii. 40. 4-5) ; this part of the Brutus is, however, really a digression, since it has nothing to do with matters of style. Of the style of the laudationes Cicero says nothing in the Brutus. In De oratore ii. 44-45, and ii. 341, he speaks of their oratorical inferiority. 'This paper is based mainly on my own independent studies. I am glad to acknowledge, however, that those studies were at first guided by a paper entitled "A Short Account of the Development of Classical Style in Latin Prose," which Henry Nettleship printed on pages 39-64 of his book, Passages for Translation into Latin Prose (London, 1887). This paper, afterward included in Nettleship's Lectures and Essays: Second Series (Oxford, 1895), appeared eleven years before the great work of E. Norden, Die Antike Kunstprosa (Leipzig, 1898: 2d ed., 1909). My references to Norden are to the first edition (the second edition, I believe, is practically a reprint of the first). References in this paper to Nettleship are to the version in his Lectures and Essays: Second Series. [CLASSICAL Philology XIII, April, 1918] 138 Pbose Writing among the Romans 139 That the speech of Appius Claudius seemed rough and uncouth to the educated ears of Cicero's day is clear enough from Cicero's words quoted above; note especially nisi .... forte. Compare also Seneca Epp. 114. 13; Tacitus Dialogus 18. In the field of oratory, then, the Romans had nothing of real consequence to offer prior to the time of Cato Censor. 1 In the other great field of prose writing, history, they had nothing at all to offer, in the days before Cato Censor, that could be described as literature. 2 Indeed, in the judgment of Cicero, the Romans had nothing down to his own day of consequence in this field. In De legibus i, after the preliminaries have been disposed of, Cicero and Atticus talk most interestingly on this theme (5-9). Atticus declares that nothing can be ieiunius than the Annates Maximi. He condemns as exiles Fabius Pictor, Cato Censor, qui tibi ( = Ciceroni) semper in ore est (see below, p. 143), Piso, Fannius, Venonius, Gellius, Clodius, Asellio, Macer. Only one writer, Caelius Antipater, gets a word of praise, but even this is much qualified. The passage continues thus: Atticus: Abest enim historia litteris nostris, ut et ipse intellego et ex te persaepe audio. Potes autem tu profecto satis facere in ea, quippe cum sit opus, ut tibi quidem videri solet, unum hoc oratorium maxime. Quam ob rem adgredere, quaesumus, et sume ad hanc rem tempus, quae est a nostris hominibus adhuc ignorata aut relicta Cicero: Intellego equidem a me istum laborem iam diu postulari, Attice. 3 We may now sum up. In oratory Cicero rated Cato Censor higher than he did any of Cato's predecessors (Brutus 62). In history, however, Cicero grants no specific credit to Cato. Yet for us, Cato is the father of Latin prose writing. He is the first writer of Latin prose of whom we have today a complete or nearly complete work; he is the first writer of Latin prose of whom we have considerable remains. 1 Cf. Brutus 61, 299, and see below, pp. 141-44. ' It is well known that the first Romans who wrote in prose on Roman history wrote in Greek. 8 Cf. also De oratore ii. 52-59, especially 53. Cicero's point in 54, that the earlier Roman historians successfully combined brevity with clearness, is well taken. It is also of great importance. The brevity of early Latin prose is a very different brevity from that of Tacitus (see below, pp. 152-54). 140 Charles Knapp Let us begin our study of Cato's style with an examination of a passage from the De agri cultura. I cite, as a fair example, chapter v (Teubner text, by Keil, 1895): Haec erunt vilici officia. Disciplina bona utatur. Feriae serventur. Alieno manum abstineat, sua servet diligenter. Litibus familia super- sedeat: siquis quid deliquerit, pro noxa bono modo vindicet. Familiae male ne sit, ne algeat, ne esuriat: opere bene exerceat, facilius malo et alieno prohibebit. Vilicus si nolet male facere, non faciet. Si passus erit, dominus impune ne sinat esse. Pro beneficio gratiam referat, ut aliis recte facere libeat. Vilicus ne sit ambulator, sobrius siet semper, ad cenam nequo eat. Familiam exerceat, consideret quae dominus imperaverit fiant. Ne plus censeat sapere se quam dominum. Amicos domini, eos habeat sibi amicos. Cui iussus siet, auscultet Iniussu domini credat nemini: quod dominus crediderit, exigat. Satui semen, cibaria, far, vinum, oleum mutuum dederit nemini. Duas aut tres familias habeat, unde utenda roget et quibus det, praeterea nemini. Rationem cum domino crebro putet. Operarium, mercennarium, politorem diutius eundem ne habeat die Parasitum nequem habeat Segetem ne defrudet, nam id infelix est. Opus rusticum omne curet uti sciat facere, et id faciat saepe, dum ne lassus fiat: si fecerit, scibit in mente familiae quid sit, et illi animo aequiore facient. Si hoc faciet, minus libebit ambulare et valebit rectius et dormibit libentius. Primus cubitu surgat, postremus cubitum eat. Prius villam videat clausa uti siet, et uti suo quisque loco cubet et uti iumenta pabulum habeant. Straight to the point, certainly, Cato goes in this passage; force- ful, certainly, his injunctions are. Brevity, too, the passage has, a brevity on the whole not inconsistent with clearness. 1 But grace and charm the passage has not, though some effort has evidently been made to attain effective word-order. There is no variety. One soon wearies of the staccato effect of the string of injunctions, set forth in the same form, as co-ordinate sentences, or in paratactic arrangement of what are, logically, parts of one sentence. There is little or no evidence of skill in the handling of particles, the skill so characteristic of the mature style of Cicero and of Livy. 2 1 See above, p. 139, note 3. The sentence Vilicus si nolet .... non faciet is, however, decidedly obscure if, as is probable, the subject of non faciet is familia. Probably, too, familiam is to be supplied with facere. 8 Examples of the lack of skill are the three ne-clauses in a row, Familiae male . . . . ne esuriat; the paratactic arrangement in opere bene .... prohibebit; the absence of the copulative conjunction in such expressions as Familiam exerceat, con- sideret . . . . ; the awkward anacoluthon, Amicos domini, eos habeat sibi amicos (one thinks of Plautus here). Compare ipsa verba compone et quasi coagmenta, said by Cicero, Brutus 68, of certain improvements thatf might be made in Cato's style (see below, p. 143). For Cicero's skill in the use of particles see the fine discussion by Pro- fessor Showerman, CJ, VIII, 184. Prose Writing among the Romans 141 I give now a passage from one of Cato's speeches: 1 Scio solere plerisque hominibus rebus secundis atque prolixis atque prosperis animum excellere, atque superbiam atque ferociam augescere atque crescere. Quo mihi nunc magnae curae est, quod haec res tarn secunde processit, ne quid in consulendo advorsi eveniat, quod nostras secundas res confutet, neve haec laetitia nimis luxuriose eveniat. Advorsae res edomant, et docent quid opus siet facto, secundae res laetitia trans- vorsum trudere solent a recte consulendo atque intellegendo. Quo maiore opere dico suadeoque, uti haec res aliquot dies proferatur dum ex tanto gaudio in potestatem nostram redeamus Atque ego quidem arbitror, Rhodiensis Persen noluisse nos ita depugnare, uti depugnatum est, neque regem Persen vinci. Sed non Rhodienses modo id noluere, sed multos populos atque multas nationes idem noluisse arbitror. Atque haud scio an partim eorum fuerint, qui non nostrae contumeliae causa id noluerint evenire, sed enim id metuere, ne, si nemo esset homo quern vereremur, quidquid luberet faceremus. Ne sub solo imperio nostro in servitute nostra essent, libertatis suae causa in ea sententia fuisse arbitror. Atque Rhodienses tamen Persen publice numquam adiuvere. Cogitate, quanto nos inter nos privatim cautius facimus. Nam unus quisque nostrum, si quis advorsus rem suam quid fieri arbitratur, summa vi contra nititur ne advorsus earn fiat: quod illi tamen perpessi. Finally I cite a passage from the Origines: 2 Dii immortales tribuno militum fortunam ex virtute eius dedere. Nam ita evenit: cum saucius multifariam ibi factus esset, tamen volnus capiti nullum evenit, eumque inter mortuos defetigatum volneribus atque quod sanguen eis defluxerat cognovere. Eum sustulere, isque convaluit, saepeque postilla operam rei publicae fortem atque strenuam perhibuit, illoque facto, 1 This paragraph, from Cato's famous speech against declaring war on the Rhodians, is preserved by Gellius vi. 3. 14-16. Duff, A Literary History of Rome, p. 257, describes the passage of which it is a part as the "best, as it is the best-known example of Cato's style " Of its style Duff says (ibid.): "This .... is luminous and forcible, but it lacks form and comeliness. Synonyms are piled on each other to secure emphasis. There is a want of variety both in expression and sound. There is no studied rhythm, and no horror of a jingle at the end of clauses. In respect to art, we are still a long way from Cicero." * See Gellius iii. 7. 19. In § 1 of this chapter Gellius says: "Pulchrum, dii boni, facinus Graecarumque facundiarum magniloquentia condignum M. Cato libris Originum de Q. Caedicio tribuno militum scriptum reliquit." Then in 2 he says, "Id profecto est ad hanc ferme sententiam." Then follows, in 3-17, a vivid account of the exploit. In 18 Gellius says: "Sed quod illi tribuno, duci militum quadringentorum, divinitus in eo proelio usus venit, non iam nostris, sed ipsius Catonis verbis subiecimus." Then comes 19, cited in full above. Gellius' words in 2 and 18 ought to imply that for the form of the narrative in 3-17 he is himself largely responsible. It is likely, though, that he here reproduced much of Cato: one word he definitely marks (6) as cited from Cato (verruca =locus editus asperque). In ii. 29 Gellius professes to be giving his own prose version of a metrical passage in Ennius' Saturae. But metrical tags and archaisms abound in Gellius' version (see my discussion of this point in AJP, XXXII, 28-31) ; we see that in reality, in ii. 29 at least, Gellius is using, in large part, Ennius' words and phrases. 142 Chables Knapp quod illos milites subduxit, exercitum ceterum servavit. Sed idem bene- factum quo in loco ponas nimium interest. Leonides Laco, qui simile apud Thermopylas fecit, propter eius virtutes omnis Graecia gloriam atque gratiam praecipuam claritudinis inclutissumae decoravere monumentis: signis, statuis, elogiis, historiis, aliisque rebus gratissimum id factum habuere. At tribuno militum parva laus pro factis relicta, qui idem fecerat atque rem servaverat. Of the speech Pro Rhodiensibus, Gellius (vi. 3. 53) said: "Ea .... omnia distinctius numerosiusque fortassean dici potuerint, fortius atque vividius potuisse dici non videntur." Nettleship (p. 97) characterizes the style as clear, forcible, and luminous, but as lacking harmony, beauty, and rhythm. There are few connecting particles; those employed are of the simplest sort. The order of the words is at times entirely without art, as in the sentence "secundae res ... . trudere solent a recte consulendo atque intellegendo." This word-order, I may note, is that of plebeian Latin in general. There are jingles, at the ends of the clauses, like the jingles of Plautus and Ennius. Finally, words more or less kindred in sense are heaped up. One may well see in this phenomenon a movement toward the copia of Cicero (see below, p. 146). Clearness, however, marks both passages: there is no mistaking Cato's meaning (see above, p. 139, note 3, p. 140, note 1). The brevity of early Latin prose writing is seldom, I think, marred by obscurity. There is no evidence of striving to attain the periodic sentence. It is time now to return to Cicero's account of Cato. Of Cato as a historian he says little in the Brutus, naturally, since that work is a treatise de claris oratoribus. The one sentence he does write in this connection is important (66) : " lam vero Origines eius quem florem aut quod lumen eloquentiae non habent!" This high praise must, beyond doubt, be discounted in the light of what is said below (pp. 140-41) of Cicero's comments on Cato's oratory, and in view of Atticus' discussion, in De legibus i. 5-9, of Roman historical writing down to Cicero's own day (see above, p. 139). Of Cato's oratory Cicero has something to say in each of his three great treatises on oratory. In De oratore i. 171 he makes Crassus say of him: "Nonne .... eloquentia tanta fuit, quantam ilia tempora atque ilia aetas in hac civitate ferre maximam potuit et iuris civilis omnium peritissimus ? " Prose Writing among the Romans 143 In the Brutus (63-71), Cicero compares Cato to Lysias. In the very number of their speeches they are alike; their speeches too are alike, since they are acuti, elegantes, faceti, breves. Still, though Lysias has certos sui studiosos, Cato is not read at all. And yet as an orator Quis illo gravior in laudando, acerbior in vituperando, in sententiis argutior, in docendo edisserendoque subtilior? Refertae sunt orationes amplius centum quinquaginta . . . . et verbis et rebus inlustribus. Licet ex his elegant ea quae notatione et laude digna sint: omnes oratoriae virtutes in eis reperientur. lam vero Origines eius quem florem aut quod lumen eloquentiae non habent! Amatores huic desunt, sicuti . . . . et Philisto Syracusio et ipsi Thucydidi. Nam ut horum concisis sententiis, interdum etiam non satis apertis cum brevitate turn nimio acumine, officit Theo- pompus elatione atque altitudine orationis suae . . . . , sic Catonis lumini- bus obstruxit haec posteriorum quasi exaggerata altius oratio. Sed ea in nostris inscitia est, quod hi ipsi, qui in Graecis antiquitate delectantur eaque subtilitate, quam Atticam appellant, hanc in Catone ne noverunt quidem. Hyperidae volunt esse et Lysiae. Laudo: sed cur nolunt Catones ? Attico genere dicendi se gaudere dicunt. Sapienter id quidem — atque utinam imitarentur nee ossa solum, sed etiam sanguinem! Gratum est tamen, quod volunt: cur igitur Lysias et Hyperides amatur, cum penitus ignoretur Cato? 1 To us this praise of Cato is astonishing. 2 If Cicero is right, why have we so little now of Cato's speeches? Why have We no 1 In De re publico, ii. 1 Cicero makes the younger Africanus speak warmly of Cato's oratory: "erat in homine .... modus in dicendo et gravitate mixtus lepos et summum vel discendi studium vel docendi et orationi vita congruens." In Tusc. disp. i. 3 Cicero is much less ardent: "Attamen oratorem celeriter complexi sumus, nee eum primo eruditum, aptum tamen ad dicendum, post autem eruditum. Nam Galbam, Africanum, Laelium doctos fuisse traditum est, studiosum autem eum qui eis aetate anteibat, Catonem. . . . ." 8 Norden (pp. 164-69) thinks highly of Cato as one of the most original of Latin writers. The style of the De agri cuttura he characterizes as "ganz roh," naturally, because Cato is talking to the farmer for practical purposes, and so he speaks in the tone which the farmer, when he came to market, heard ringing in the laws. The style of Cato's history, says Norden, is much more developed, "kurz, derb, kraftvoll." To his speeches, continues Norden, Cato owed much of his fame in later years. In these we see clearly the results of Greek influence. ". . . . jedesmal trifft er den Nagel auf den Kopf; .... er sagt meist mit einem Wort mehr als Cicero mit einem Satz . . . ." Norden sees also in the fragments of Cato's orations striving for periodicity and certain graces. I may note that above (p. 142) I hazarded the suggestion that Cato sought to work toward the copia of Cicero. If there is any merit in that suggestion, it is in sharp opposition to Norden's overenthusiastic praise of Cato's brevity. Norden omits, be it noted, entirely most of the passages in which Cicero writes about Cato; most extraordinary is his failure to mention the long passages in the Brutus which deal with Cato. 144 Chables Knapp speech of his complete? Our judgment finds confirmation in what Atticus says in the Brutus 292 ff.: Equidem in quibusdam risum vix tenebam cum Attico Lysiae Catonem nostrum comparabas .... singularem virum .... sed oratorem? sed etiam Lysiae similem? .... Bella ironia, si iocaremur; .... orationes autem eius, ut illis temporibus, valde laudo, significant enim quandam formam ingeni, sed admodum impolitam et plane rudem. Origines vero cum omnibus oratoris laudibus refertas diceres et Catonem cum Philisto et Thucydide comparares, Brutone te id censebas an mihi probaturum ? It will appear presently that in these words Atticus gives the opinion current in his day with respect to the earlier Roman orators. 1 Cicero meets Atticus' attack by insisting that he is perfectly serious (297 ff.), but that he has not time to discuss the theme now. He assures him that Cato possessed every excellence that an orator should have, "nisi eorum pigmentorum quae inventa nondum erant florem et colorem." In Orator 150ff., while he is discussing hiatus, Cicero makes clear his opinion that orationes Mae ipsae horridulae Catonis were stricter in this matter, as were the early Roman orators in general, than the Greeks had been. It will be noticed that, in his first discussion of Cato, in the De oratore, Cicero speaks of Cato with no great warmth; his last utterance, that in the Orator, has even less warmth. Why, then, the extraordinary fervor of his remarks in the Brutus about Cato? The Brutus and the Orator were both published in 45. Cicero's career as an orator was to all intents and purposes over. Even if we forget the fact that Caesar was then supreme, so that no orator had proper opportunity for the display of his talents, we must remember that, as a senex, Cicero was feeling the burden of years (see the Cato Maior 1-2) and was facing the thought that before long he must leave the future of Roman oratory to others. With these feelings he fights, twice in this one year, the battle for what he regards as the one hope for Roman oratory. 2 1 Cf. Quintilian ii. 5. 21 ; Tacitus Dialogus 18. « Cf. Norden, pp. 21&-21. Prose Writing among the Romans 145 The Brutus is a verbal picture gallery of the heroes of Roman oratory. The Orator is a pen picture of the ideal orator. As he wrote these works, Cicero was keenly aware that the principles which had guided and determined his own oratory were being vigorously challenged. In both works Cicero was seeking to win Brutus to his side in the great struggle between the Attic and the Asiatic schools of Roman oratory. 1 In Orator 110 he clearly indicates that he thinks of Brutus as his successor. In Orator 76-101, Cicero insists that the ideal orator must show complete mastery of the three genera dicendi. 2 The perfect orator, however, will display his genius best by harmoniously combining the three (100-101). Such a happy combination he had himself, he insists, achieved. If, however, we read 109 in close connection with 107-8, in which he had cited, apologetically, examples of his own iuvenilis redundantia, we see clearly that Cicero had been severely criticized for his tendency toward the Asiatic style (compare again Hendrickson, AJP, XXVI, 273). In the Brutus and the Orator, Cicero sought to wean Brutus from his leaning to an (overstrict) Atticism, and to induce him to accept as his own the oratorical creed of Cicero (see below, pp. 146-48). The distinction between the Asiatic style and the Attic style is too well known to need extended discussion here; for the Asiatic style see, e.g., Brutus 325-27 (an account of the style of Hortensius); Norden, pp. 131-52, 186ff., 218-22, 251-70. For the Attic style see Brutus 283-91; Norden, pp. 149-52, 219-21, 251-70. Since Cicero was assailing, mainly, the Atticists, we must quote here part of Brutus 284-91: Turn Brutus Atticum se, inquit, Calvus noster dici oratorem volebat: inde erat ista exilitas, quam ille de industria consequebatur. Dicebat, inquam, ita, sed et ipse errabat et alios etiam errare cogebat Sin autem ieiunitatem et siccitatem et inopiam, dummodo sit polita, dum urbana, dum elegans, in Attico genere ponit, hoc recte dumtaxat; sed quia sunt in Atticis alia aliis meliora, videat ne ignoret et gradus et dissimili- tudines et vim et varietatem Atticorum. 1 Cf. Hendrickson, AJP, XXVI (1905), 272-73; Teuffel, §119. 4. 2 See Hendrickson, ibid., pp. 249-90; C. N. Jackson, Harvard Studies, XXV (1914), 117-37; M. S. Dimsdale, A History of Latin Literature, pp. 175-77. The three styles, Asiatic, Attic, and Rhodian, are well described by Quintilian xii. 10. 16-26. 146 Charles Knapp The Attic style, then, made but a sparing use of verbal ornament, tropes, antitheses, and other rhetorical devices; it was restrained, dignified, severe, with a leaning toward (excessive) brevity rather than to copia. The Asiatic style was the antipodes of all this; it was exuberant, florid, given much to tropes and rhetorical display. Each style had its excellences. Each had its characteristic danger. The Attic style was likely to become too brief, bald, obscure; the Asiatic tended to become florid and bombastic. What style did Cicero himself adopt ? In the Brutus 304-19 he describes in detail the training by which he had sought to fit himself to win distinction as an orator. In 310 occurs a very significant Commentabar declamitans — sic enim nunc locuntur — .... idquefacie- bam multum etiam Latine, sed Gra'ece saepius, vel quod Graeca oratio plum ornamenta suppeditans consuetudinem similiter dicendi Latine adferebat, vel quod a Graecis summis doctoribus, nisi Graece dicerem, neque corrigi possem neque doceri. In 313ff. he explains why, after delivering his speech for S. Roscius, in 80, he went abroad, to remain away two years. Having studied at Athens and in Asia under various philosophers and rhetoricians he put himself a second time under Apollonius Molo, a man (316) .... in notandis animadvertendisque vitiis et in instituendo docendoque prudentissimum. Is dedit operam, si modo id eonsequi potuit, ut nimis redundantis nos et supra fluentis iuvenili quadam dicendi impunitate et licentia reprimeret et quasi extra ripas diffluentis coerceret. Ita recepi me biennio post non modo exercitatior, sed prope mutatus, nam et contentio nimia vocis resederat et quasi deferverat oratio With the last sentence or two we may compare the passage referred to above (p. 144) from the Orator, 107-8, in which Cicero gives examples of his iuvenilis redundantia. Plainly, by natural inclination Cicero was an Asiatic in style. Recall his incessant references to copia as an indispensable element of good oratory. Copia, copiosus, copiose, when applied to style, are to him terms of highest praise. 1 1 See copia, copiosus in Merguet's lexicons to Cicero. Cicero demanded copia of the philosopher too; cf. e.g., Tusc. i. 7. On Cicero's copia see Professor Shower- man, CJ, VIII, 182-86. Prose Writing among the Romans 147 We see now why Cicero praised Cato.so highly in Brutus 63 ff., and in particular why he compared Cato with Lysias. This passage and Brutus 283-91, 292-95 show that Cicero is seeking to discomfort the Roman Atticists by suggesting that they do not know Attic oratory when they see it or hear it. If plainness, he argues, is of the essence of the Attic style, Cato is the most Attic of Roman Atticists; why, then, has he no following ? In practice, Cicero remained, all his life, to some extent an Asiatic. Yet, from his return to Rome in 78, he was in theory not an Asiatic at all, but the representative of the truest Atticism. 1 For our purposes only two characteristics of Cicero's style need be considered. Of one, his copia, something has been said; more will be said here at once, and again below, in the account of Tacitus (pp. 152-53). Just now I wish to dwell on one statement in Cicero's account of his own training — the sentence Commentabar declamitans .... doceri, cited above (p. 146). When I put this beside his account of the Atticists of his day, and when I think of his many allusions to copia as a most desirable element of style, I cannot escape the con- clusion that to Cicero, early in his life, came the thought that Latin prose style could be perfected only by a departure from the brevity of the earlier times, and by the substitution for that brevity of a fuller, richer, more rotund, as well as more rhythmical, style. Latin suffered all too readily from the defects of its qualities. Brevity and plainness tend to become overbrief and overplain and to end in obscurity, in baldness, in a monotonous succession of short sentences, too staccato in their effect and so wearisome, and in utter absence of the finer effects, alike in logic, syntax, and rhythm. That we are not indulging here in mere a priori speculations will be clear enough to one who recalls the discussion above (pp. 139-44) of Cato's style, 1 See here J. E. Sandys' edition of the Orator, xliii-xlv; Norden, pp. 216-21. Note especially Tacitus Dialogue 18: "satis constat ne Ciceroni quidem obtrectatores defuisse, quibus inflatus et tumens nee satis pressus sed supra modum exsultans et superfluens et parum Atticus videretur." Tacitus goes on to say that from letters sent by Calvus and Brutus to Cicero it appears that, to Calvus, Cicero seemed tamquam solutus et enervis; to Brutus he appeared tamquam fractus atque elumbis. According to Quintilian xii. 10. 12-14 some dared to criticize Cicero as tumidiorem et Asianum et redundantem. In my day, he continues, some think him ieiunus atque aridus; his personal enemies criticized him for his nimii fiores et ingenii affluentia. "Falsum utrumque, sed tamen ilia [here = " the latter view"] mentiendi propior occasio." 148 Charles Knapp both in historical writing and oratory, who has not forgotten Cicero's references to the exilitas and tenuitas of Calvus (Brutus 64, 284 ff.) and other Atticists, who thinks of Sallust, and who has a clear con- ception of the style of Tacitus. Cicero, then, saw one great defect of Latin prose style — its wrongly developed brevity — and set himself to eradicate this defect from his own style. It is not at all strange that in working out his design, the achievement of copia, he fell into the very error he charged against his predecessors, overindulgence in a quality in itself good, and that sometimes, especially in his earliest efforts, he himself displayed the defects of his qualities by being guilty at times of iuvenilis redundantia. Nor is it strange that, always, whether he realized it or not, he had a tenderness for the Asiatic style. The other great characteristic of Cicero's style is the periodic structure. Here it will be enough to refer to Nettleship, pp. 105-9. If, now, we try to recall the names of the great writers of Latin prose, apart from Caesar, Cicero, and Livy, we shall think of Sallust, of the two Senecas, of Quintilian, of Pliny the Elder and P lin y the Younger, of Suetonius, and of Tacitus. Of these, Sallust and Tacitus will mean most to us at present. As we call to mind the style of these various authors, we shall be struck by at least two facts: (1) they all differ markedly from Cicero and Livy, and in much the same way, in that they are less copiosus, terser, in a word more like the prose writers that preceded Caesar and Cicero: 1 (2) in them, the periodic structure, libration, and the like are far less con- spicuous. Sallust and Tacitus in particular stand out in sharp distinction to Cicero and Livy. Sallust's writings lie, apparently, between 44 and 36 b.c. They were all published before the first part of Livy's history appeared. In a word, in all probability but a short time after the publication of the Brutus and the Orator, we have prose writings in Latin widely different in style from Cicero's, in a style, too, which shows markedly the influence of the study of Cato Censor. Latin prose writing of the empire culminates, surely, in the developed style of Tacitus. That style shows the results of a profound study of Thucydides, of Cato, and of Sallust — all apostles, in one form or 1 To this one characteristic — brevity — I must confine the rest of this paper. Pbose Writing among the Romans 149 another, of brevity, all far removed from the copia of Cicero, all in harmony with the earlier brevity of Roman prose writing rather than with the development represented by Caesar, Cicero, and Livy. 1 How are we to account for the triumph of the earlier style, in spite of the splendid results achieved by Cicero ? Nettleship (pp. 110-11) found the answer in part in the loss of freedom, as the republic gave way to the empire. Oratory could no longer be free-spoken and sincere, in the forum, or in the courts, or in the senate. "And as the sphere of oratory became narrower, the cultivation of style became nicer and more minute." Another cause on which Nettleship (pp. lllff.) lays great stress is a change in the character of Roman education, especially that of the future orator. Especially harmful was the declamatio, which dealt with fictitious themes. Compare Seneca (Contr. 9, Praef. p. 241, Bursian=Kiess- ling, p. 391): qui declamationem parat, scribit non ut vincat sed ut placeat. Omnia itaque lenocinia conquirit: argumentationes, quia molestae sunt et minimum habent floris, relinquit : sententiis, explicationibus audientis deliniri contentus est. Cupit enim se approbare, non causam. Sequitur autem hoc usque in forum declamatores vitium, ut necessaria deserant dum speciosa sectantur. See also Petronius 1. In attaching weight to these factors, Nettleship is beyond ques- tion right. But another factor and a very important one he leaves entirely out of his account — the development of the archaizing spirit, a spirit which reached fullest development in the second Chris- tian century, the days of Hadrian, Fronto, Gellius, and Apuleius. The history of this archaizing development I shall now trace, briefly. 2 To the conservatism of the Romans allusion has often been made. That conservatism shows itself as markedly in the field of education 1 Convenient discussions of the style of Sallust may be found in Norden, pp. 200- 204; Teuffel, §206.5-9; Dimsdale, pp. 221-22 (a brief, but very good account). In the matter of brevity, Sallust's style is essentially an instance of "reversion to type," a resurrection, in far finer form, to be sure, of the brevity of Cato. On Sallust's archa- isms see the dissertation of P. Schulze, De Archaismis Sallustianis (Halle, 1871). For an ancient reference to those archaisms see Suetonius Augustus 86. Gellius, a pro- fessed archaist, had a high admiration for Sallust. See below, pp. 150-51. * I discussed this matter in a paper entitled "Archaism in Aulus Gellius," pub- lished in Classical Studies in Honour of Henry Drisler, pp. 126-71 (New York, 1894). See especially pp. 126-41. See also Norden, pp. 258-63, 348 ff. 150 Chables Knapp and literature as it does elsewhere. The Laws of the Twelve Tables formed part of the curriculum of Roman schools in Cicero's time; plagosus Orbilius helped Horace to study the carmina hivi Andronici. Cicero was criticized by those who favored earlier writers: compare Tacitus Dialogus 22: "Ad Ciceronem venio, cui eadem pugna cum aequalibus suis fuit quae mihi vobiscum est. Illi enim antiquos mirabantur, ipse suorum eloquentiam anteponebat." We have clear echoes of this conflict in Orator 169-71. In 168, Cicero begins his discussion of the oratorical arrangement of words according to the rules of artistic rhythm. The value of such arrangement is to him self-evident: his ears tell him this. He continues thus: "Quiddico meas aures ? contiones saepe exclamare vidi, cum apte verba cecidis- sent. Id enim expectant aures, ut verbis colligetur sententia. Non erat hoc apud antiquos." 1 Clearly there were standpatters, in matters oratorical. Of Asinius Pollio, Tacitus Dialogus 21, says, "Asinius quoque, quamquam propioribus temporibus natus sit, videtur mihi inter Menenios et Appios studuisse. Pacuvium certe et Accium non solum tragoediis sed etiam orationibus suis expressit: adeo durus et siccus est." Due allowance must, of course, be made for the bias of the speaker: yet Quintilian x. 1. 113 says of Pollio, "a nitore et iucunditate Ciceronis ita longe abest ut videri possit saeculo prior." Horace, in Epistles ii. 1, defends the new school of Latin poetry, the school to which Vergil, Varius, and Horace himself belonged, and assails those who decried all contemporary poets in their admiration, real or professed, of the ancients. The whole Epistle is addressed to this defense. But mark especially vss. 18-88. Compare also Persius i. 76-78, with Conington's admirable transla- tion; Martial viii. 69; v. 10; xi. 90, with the comments of the editors. Tacitus, Dialogus 18 makes a speaker say, " . . . . vitio .... malignitatis humanae vetera semper in laude, praesentia in fastidio esse. Num dubitamus inventos qui prae Catone Appium Caecum magis mirarentur?" Seneca (Epistles 114. 13) declares that "Multi ex alieno saeculo verba petunt, duodecim tabulas locuntur; Gracchus illis et Crassus et Curio nimis culti et recentes sunt; ad Appium usque et ad Coruncanium redeunt " 'This sentence gives an objection to Cicero's preceding statement: "rhythm was unknown among the ancients." Antiquos covers all Roman orators prior to Cicero himself. Pbose Writing among the Romans 151 I add a final array of passages which all tell the same story. Quintilian x. 1. 43 ... . quidam solos veteres legendos putant neque in ullis aliis esse naturalem eloquentiam et robur viris dignum arbitran- tur Suetonius Augustus 86 Cacozelos et antiquarios ut diverso genere vitiosos pari fastidio < Augustus >sprevit, exagitabatque nonnumquam. .... Sed nee Tiberio parcit et exoletas interdum et reconditas voces aucupanti. M. quidem Antonium ut insanum increpat, quasi ea scribentem, quae mirentur potius homines quam intellegant .... addit haec: Tuque dubitas, Cimberne Annius an Veranius Flaccus imitandi sint tibi, ita ut verbis quae Crispus Sallustius excerpsit ex Originibus Catonis utaris ? Tacitus Dialogus 23: Neminem nominabo, genus hominum significasse contentus; sed vobis .... versantur ante oculos isti qui Lucilium pro Horatio et Lucretium pro Vergilio legunt, quibus eloquentia Aufidi Bassi aut Servilii Noniani ex comparatione Sisennae aut Varronis sordet, qui rhetorum nostrorum commentarios fastidiunt et oderunt, Calvi mirantur. Quos more prisco apud iudicem fabulantis non auditores secuntur, non populus audit, vix denique litigator perpetitur. Quintilian x. 1. 93 Lucilius quosdam ita deditos sibi adhuc habet amatores ut eum non eiusdem modo operis auctoribus sed omnibus poetis praeferre non dubitent. Evidently there was in Cicero's day, and continually from that time on, eager strife between the advocates of the old and the sup- porters of the new. That Sallust threw in his lot with the advocates of the old is clear enough from Suetonius Augustus 86, cited above. Writer after writer has called attention to the archaisms in Sallust. Yet, on the whole, throughout the first century of the empire, the "modern" tendency had the upper hand, thanks, in large part, to the influence of Quintilian, to whom Cicero was the prince of Roman orators and stylists. 1 It has been held by many scholars that both Tacitus and Pliny the Younger were pupils of Quintilian. At any rate, the style of Pliny was much more like that of Quintilian than like that of Tacitus; at times, at least, Pliny imitated Cicero (see Ep. i. 2). We come now to Tacitus. The style of Tacitus shows the effects of four distinct influences: (1) that of Cicero, (2) that of Sallust (here we may set the influence, direct or indirect, of Thucydides, 1 See Dimadale, pp. 445-46; G. A. Simcox, A History of Latin Literature, p. 171; Norden, p. 269; W. Peterson, edition of Quintilian, Liber Decimus, xxxix-xli (Oxford, 1891). 152 Chablbs Knapp direct, as the result of study of Thucydides himself, indirect, as the result of the influence of Thucydides on Sallust), (3) that of Vergil, (4) that of the silver age in general. 1 The wide divergence between the style of the Dialogue, Tacitus' earliest work, and that of the Annates, his latest work, is a common- place. In the Annates every word is charged with almost more than its proper share of meaning, and the mind of the reader is kept always on the strain. In the Dialogus the style is easy, full (copiosus), fluent, and continuous. There is a regular, well-balanced, periodic structure, which involves at times a copiousness that borders on redundancy. Clearly, in his youth 2 Tacitus was an ardent admirer and close student of Cicero. In all probability Tacitus had modeled his early efforts at the bar, as nearly as possible after such an interval, on the oratory of Cicero and his contemporaries. Peterson, in his edition of the Dialogus (pp. xlvi-liii), points out at length illustrations of the effects of Cicero's style on that of Tacitus. The part of this exposition which most concerns us here is that which deals with a certain fulness (copia) that marks the style of the Dialogus. Such expressions as the following, consisting of pairs of words of kindred meaning, have a distinctly Ciceronian ring: animi et ingenii, clamore plausuque, divitiae et opes, jama et laus, vis et facuUas, memoria et recordatio, modestia et pudore, operae curaegue. We may next, with 'On the style of Tacitus cf. A. Gudeman, edition of the Dialogus, "Prolegom- ena," pp. xliv-xlviii, li-lvii, ciii-cxix (Boston, 1894), and in the revised German edi- tion (Leipzig, 1914), pp. 20-29, 99-111; H. Furneaux, edition of the Annates, I, 29-62 (Oxford, 1884) ; W. Peterson, edition of the Dialogus, pp. xliii-lxii (Oxford, 1893) ; Norden, pp. 321-43. s Norden, however (pp. 322-24), holds that stylistic differences between writings of an author afford no ground for the assumption that those writings belong to different periods in the author's life; at one and the same time, he maintains, writers practiced widely different styles. He then (pp. 324-25) argues that the Dialogus waB written after 91a.d. He concludes (p. 326) that the Dialogus may not be regarded "als Dokument fur die allmahliche Entwicklung der taciteischen Diktion." The Dialogus, he insists, is rather a literary tour-de-force in the Ciceronian manner. The develop- ment of Tacitus, both as historian and as orator, begins, he continues, with the Germania and the Agricola; "von da ab ist es ein Weg, der ununterbrochen auf- warts fuhrt, seine Signatur ist der immer starker werdende Streben nach dem Unge- wohnlichen, hervorgerufen durch seine immer mehr sich auspragende Subjektivit&t." Gudeman argues, in the second (German) version of his edition of the Dialogue, 29-55, against the view that the Dialogus belongs so late in Tacitus' life. For his direct answer to Norden see pp. 31-33. I side with Gudeman. So, I see, does Professor F. G. Moore, CP, XIII, 108. Pbose Writing among the Romans 153 Peterson, compare Tacitus directly with Cicero, setting D. 1 memoria et recordations beside Lael. 103; D. 2 industriae et laboris beside Brutus 237, Ad jam. xiii. 10. 3; D. 5 metum et terrorem, A. 32. 2 metus ac terror, beside Verr. iv. 41; D. 7 tueri et defendere, G. 14. 2 ilium defendere, tueri, beside De orat. i. 172, Ad jam. xiii. 64. 1; D. 32 eloquentia circumcisa et amputata, beside De jin. i. 44 amputata cir- cumcisaque inanitate omni et error e, Acad. ii. 138 circumcidit et amputat muUitudinem, De orat. i. 65 circumcidat atque amputet (see Wilkins, ad loc.). Such, then, was the style of Tacitus in his earlier days. One may think of him as then full of hope and confidence with respect to the future of Roman oratory, and as seeing in a return to the style of Cicero the best promise of the fulfilment of that hope. While that belief was yet strong within him, he wrote the Dialogus, and employed in that work a style which, in its copia and its use of the period, was sharply at variance with the style prevalent at that time. One would give a deal for a record of the struggle that went oh in Tacitus' mind against his growing conviction that the world would have none of a style modeled closely on that of Cicero. That the struggle was a long one is proved by the fact that he did not at once and completely lay aside the full, rotund, copious style, of the Dialogus. Indeed, he never wholly laid aside his earlier manner. Compare, for example, such expressions as nova et recentia iura, H. iv. 65; abiectum et sordidum, Ann. xiii. 46; nemora et lucos, G. 9, 10, 45; vis ac potestas, H. ii. 39, hi. 11, G. 42; incensum ac flagrantem animum, A. 4; quies et otium, A. 6, 21, 42; extremo ac novissimo iactu, G. 24; maneat duretque, G. 33; formam ac figuram animi, A. 46, etc. I think, then, of Tacitus as struggling long and hard against the conviction that his day and generation demanded brevity and "point," and I imagine a moment at which, giving up the fight and yielding to the tendency of his times, he resolved, so to say, to beat his fellows at their own game, to outbrevity them in brevity. At all events, his final style represents Latin brevity at its very best, and, at times, as it seems to me, at its very worst. 1 As one reads the Germania, for example, he finds Tacitus' brevity over and over 1 Cf. my note on Tacitus Agricola 31. 5 in PAPA, XXXIII, xlix-li. 154 Chables Rnapp obscure, so that he lays down the book with a sigh of regret that it is impossible, in many places, to determine, by a study of the Latin itself, just what is meant by the words of the man whose testimony, understood or not understood, is of such prime importance to the student of Teutonic origins. We have now, in a very imperfect way, traced the development of Latin prose writing, in a single phase, through a period of four centuries or more. We have seen that writing marked at the close of the period under review, as at its beginning, by one and the same characteristic — brevity, full of energy and power, epigrammatic, striking sharp blows upon the ear and the mind, over and over saying things in unforgettable formulas, and yet on the other hand tending to display the defects of its qualities, in baldness, in the impression conveyed of undue strain and struggle, and, worst of all weaknesses of style, in obscurity. 1 When we examine Latin style in this way, we get a fresh sense of the wide departure Cicero made in his style from that of his predecessors, and we get an added appreciation of the greatness of Cicero's contribution to the literary and intellectual glory of his countrymen. Our discussion has, I hope, enabled us to understand why Cicero had but one direct successor, Livy, and has taught us how, within the limits of a century and a half, the style of Sallust rather than that of Cicero won supremacy in the field of Latin prose writing. Barnard College Columbia University 1 This, be it remembered, is true of the final rather than of the initial period (cf. p. 139, n. 3).