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. THE HISTORICAL INFINITIVE III. IMITATION AND DECLINE By J. J. Schlicheb The middle of the last century b.c. was a time of rapid and important changes in Roman literary standards and practice, which incidentally had a serious effect on the fortunes of the historical infinitive. The construction had been intimately connected with the sermo urbanus, 1 and the influences now at work were tending toward the disintegration of this native idiom. They were princi- pally two: the increasing importance of outsiders in Rome, with its effect on the language, and, secondly, the adoption of Alexandrian models by practically a whole generation of poets. Of the two external foes of good Latin (urbanitas), namely, rusticitas and peregrinitas, 2 the former was at this time the less dangerous. Some people fell into it, according to Cicero, 3 when they tried to be old fashioned in their speech. But he makes it clear that the two things are not the same. 4 With the influence of foreigners upon polite speech it was differ- ent. This class of the population was constantly increasing, both in numbers and in importance, and the problem became a serious one, it appears, when the influx began from Northern Italy and Gaul, reinforced as it was by a number of talented young literary men whom it brought to Rome. Cicero, who belonged to the old school and had grown up in the traditions of the sermo urbanus, makes reference several times in his later writings to a state of things which caused him no little annoyance. He felt that the purity of » Cf. Class. Phil, IX. 279. 2 Quint, vi. 3. 17 and 107; De or. i. 50. 218. 3 De or. iii. 42: "Rustica vox et agrestis quosdam delectat, quo magis anti- quitatem, si ita sonet, eorum sermo retinere videatur; et . . . . L. Cotta .... illud quod loquitur, priscum visum iri putat, si plane fuerit rusticanum." 4 De or. iii. 45 : " Equidem cum audio socrum meam Laeliam — facilius enim mulieres incorruptam antiquitatem conservant, quod multorum sermonis expertes, ea tenent semper quae prima didicerunt; sed earn sic audio, ut Plautum mihi aut Naevium videor audire. Sono ipso vocis ita recto et simplici est, ut nihil ostentationis aut imitationis afferre videatur; ex quo sic locutum esse eius patrem iudico, sic maiores; non aspere ut ille, quern dixi, non vaste, non rustice, non hiulce, sed presse et aequaliter et leniter." Cf. also De or. Hi. 39 and 46, and Brut. 259. [Classical Philology X, January, 1915] 54 The Historical Infinitive 55 the language was being corrupted by these foreigners. 1 And it is especially noteworthy that he regularly looks for the cause of the corruption among the Gauls. 2 This prejudice was evidently but a part of a more general feeling against the newcomers who were suddenly playing such an important part in Rome. 3 It seems to have survived for a considerable time, and it is reasonable to suppose that when Asinius Pollio suspects quandam Patavinitatem in Livy, his criticism may have been prompted largely by the old feeling against the Transpadanes which he had imbibed in his younger years when he was training with the Atticists 4 and the arrival of a number of bright young poets from that province had suddenly upset literary values at Rome. The protest of the Atticists and their insistence upon the purity of the language cannot well be thought of as entirely independent of that corruption of the language which led Cicero himself to say (Brut. 258): "expurgandus est sermo et adhibenda tamquam obrussa ratio quae mutari non potest." It is certainly strange that in spite of the prominence of the Cisalpine Gauls in the literary life of the period, and while they constituted the very backbone of the Alexandrian movement, and several of them were even close friends of the Atticists, so far as we know none of the Atticists themselves were Gauls. 5 1 Brut. 258: "Sed omnes turn fere, qui nee extra urbem hanc vixerant nee eos aliqua barbaries domestica infuscaverat, recte loquebantur. Sed hanc certe rem deteriorem vetustas fecit et Romae et in Graecia. Confluxerunt enim et Athenas et in hanc urbem multi inquinate loquentes ex diversis locis." Similarly in a letter to Paetus (Fam. ix. 15. 2). 2 "Cum in Galliam veneris, audies tu quidem etiam verba quaedam non trita Romae" (Brut. 171). "T. Tinea Placentinus" (Brut. 172). "Bracatae et trans- alpinae nationes" (Fam. ix. 15. 2). Perhaps the repeated reference to the bad Latin of Caecilius (Att. vii. 3. 10; Brut. 258) may be partly due to the same feeling. 3 Suet. Jul. 76: "Eadem licentia spreto patrio move .... civitate donates, et quosdam e semibarbaris Gallorum, recepit in curiam." Ibid. 80: "Peregrinis in senatum allectis, libellus propositus est: Bonum factum; ne quis senatori novo curiam monstrare velit! et ilia vulgo canebantur: Gallos Caesar in triumphum ducit, idem in curiam. Galli bracas deposuerunt, latum clavum sumpserunt." * He was much like them in the characteristics of his style (Quint, x. 1. 113), and also in his narrow standards, which led him to find fault with the language of other writers (references in Teuffel, Bdm. Lit., 221, 6). 6 Nepos is conjecturally, but not very plausibly, connected with the school. Indeed, the individuals in the movement are, with a few exceptions, much in doubt. 56 J. J. SCHLICHEK But while Atticism died out very soon as an active propaganda, 1 its sister-movement, the imitation of Alexandrian models in poetry, had important and far-reaching consequences. Cicero was opposed to this, as he was to the radicalism of the Atticists. 2 But the ridicule of staid old men may have served, if anything, merely to confirm the younger generation in its ways. The drift of things favored the latter. The next generation was to be one of poetry, and some of those who began as Alexandrians lived to be its chief representa- tives, carrying with them into their mature work, if not the artificial standards of their early period, at least its devotion to highly finished workmanship. 3 To be sure, the champions of the new order did not have things all their own way. They had to fight a hard battle with the admirers of the older poets, as we see from Horace. 4 If they won in the end, we can readily believe that it was largely owing to their own sanity, which abhorred the artificiality and one-sidedness of the Cantores Euphorionis, no less than it abhorred the crudeness of Plautus and Lucilius. In the Aeneid we may even see a sort of final truce which, while maintaining all the real advantage that the new school had won, yet accepted willingly and even gratefully the contributions of an earlier time. It is this fact, perhaps, quite as much as its subject-matter and its inherent excellence, which caused Virgil to be recognized at once as the poet par excellence. If we turn now to the fortunes of the historical infinitive during this period of transition, we find that it fared badly on the whole. 1 Cic. Tusc. ii. 1. 3: "Qui iam conticuerunt paene ab ipso foro irrisus." Simi- larly, Brut. 289. For Cicero's relations to them, cf. these two passages at greater length; also Brut. 68. 283 and Tacitus Dial. 18. It is quite possible that the move- ment was less formidable than it appears to us owing to Cicero's onslaughts. Har- necker (M.J. f. Phil. u. Paed., CXXV, 604 ff.) places the school's greatest activity in 51-50 B.C., when Cicero was absent in Cilicia. After the death of Calvus (47 B.C.) and the publication of Cicero's Brutus in the next year, their defeat seems to have been assured. 8 Tusc. iii. 19. 45: "O poetam egregium! Quamquam ab his cantoribus Euphor- ionis contemnitur." Other passages in Schanz, Rirm. Lit., I, chap. xcvi. The earliest reference to the veiirepoi, according to Harnecker (Philol., XLI, 466), is Att. vii. 2. 1. {50 B.C.). The feeling could hardly have existed when Catullus wrote his forty- ninth poem. 3 On this subject see Rand, Harv. Stud., XVII, 15 ff. * Sat. i. 4. 8-13; i. 10 passim; Epp. ii. 1. 64 ff.; ii. 3. 270-74. The Historical Infinitive 57 To be sure, an independent adherent of the earlier language, like Sallust, whose method was not merely restrictive like that of the Atticists, but creative as well, and who developed that older lan- guage vigorously along its own lines, might continue to use the con- struction with a zeal which was never surpassed. But the Atticists themselves were lukewarm toward it. It is not found in the letters of Brutus, and Caesar, who may perhaps be classed with the school, at least in his practice and sympathies, 1 used the historical infinitive rather sparingly, when we consider his opportunities. From Caelius we have two examples of the construction (Fam. viii. 4. 3. and Quint. 9. 3. 58), but his connection with the Atticists is very doubtful. 2 The painful correctness of their style, and their exaltation of regu- larity and restraint, would naturally make them unsympathetic toward an exuberant expression of unrestrained and unregulated feeling or desire, like the historical infinitive. But if the Atticists were apparently averse to the construction, the provincials do not appear to have used it at all, except as an acquired form of expression. Nepos does not have it, nor does Catullus, and Virgil adopted it only by degrees, having none in the Eclogues, only two passages in the Georgics, and thirty in the Aeneid. Livy also, though he did not begin to write until the situation had greatly changed, does not treat the historical infini- tive as if it were native to him. 3 In trying to discover why Livy used the construction so much more in his first decade than he did later, 4 we should bear in mind that the composition of these early books coincides in time with that of the Aeneid, and the use of the historical infinitive may well, in both cases, have been due to the high wave of patriotic revival which swept over Roman literature during the decade following the battle of Actium. Nothing would be more natural than to see good in the days of old, now that the battle was won, both in literature and in politics, and to show this in part by adopting a somewhat archaic 1 Cf. Hendrickson, Class. Phil, I, 97 ff. ' Even if we accept Harnecker's interpretation of Brut. 273 (2V. J. f. Phil. u. Paed., CXXIX, 45-48), the actual style of Caelius is not Attic. See Becher, Vber d. Sprachgebr. d. Cael., pp. 6 ff.; also F. Burg, De M. Caeli Rvfi genere dicendi, p. 9. * Cf. Class. Phil., IX, 384-86. « Cf. Class. Phil, IX, 383. 58 J. J. SCHLICHEB tone, such as we have in the Aeneid, and also in the early books of Livy. 1 And the historical infinitive, which had been conspicuous in early Latin, naturally profited by this, the same as other archaic features did. That Livy should later on lose interest in the con- struction is also natural on the assumption that he used it partly because it was in vogue when he began to write, and partly because it had always been used in historical writing, upon which Sallust had recently impressed it more indelibly than ever. As for the Alexandrians, their influence was highly unfavorable to the construction, for the same reason, no doubt, as in the case of the Atticists, because it presupposes a habit of mind which was diametrically opposed to their scrupulous and painstaking methods. The success of their movement emphasized and made irrecon- cilable the difference between their own high finish and the compara- tive crudeness of their predecessors, who henceforth occupied a plane in the estimation of literary critics not unlike that which rusticitas had occupied in the days of Cicero. Even within their own works they made a distinction between high and low — the Odes of Horace, on the one hand, and the Sermones on the other. The latter might admit colloquial forms of speech and the old- fashioned historical infinitive, but the former, continuing a Greek rather than a Roman tradition, could not do so without losing caste. So we do not find the construction in Catullus, not even in his epic, where we might reasonably expect it, nor in any of the other short epics found in the Appendix Vergiliana. Ovid does not have a single sure case of it, even in his narrative works, where the opportunity for its use was excellent, and it is absent from the writers of elegy. All this, especially the conscious depreciation of early Latin from the standpoint of literary excellence, had given the construction a very bad standing with those who aspired to be up with the times. As a consequence it disappeared even from such forms of compo- 1 R. Jonas (Uber d. Gebr. d. verba freq. u. intens. bei Livius, and in a similar article on the same verbs in early prose) shows that the verba frequentativa and intensiva, which are especially common in early Latin and in imitators of it, like Sallust, are also common in the beginning of Livy's work, but fall off rapidly as the work pro- gresses: first decade, 575 occurrences; third, 408; fourth, 292; fifth (half), 110. The reminiscences of Ennius and other early Latin authors are also found chiefly in the first decade. Cf. Stacey, Archiv. f. lat. Lex., X, 22-33. This agrees exactly with what we found to be Livy's treatment of the historical infinitive. The Historical Infinitive 59 sition as had formerly employed it, e.g., from satire and the drama (Persius, Juvenal, Seneca), and from prose narrative and history (Valerius Maximus and Velleius). Some part of this change may have been due to the growing influence upon style of the instruction in the schools. The very conditions of instruction make it necessary to lay stress upon regu- larity of speech, and an odd construction like the historical infinitive, which belonged chiefly to the time of the grandfathers and had been frowned upon by two generations of poets, would hardly appeal to practical instructors as deserving of much consideration, especially when many of them, and of their pupils, as well, were from the provinces, straining desperately to avoid anything strange in their speech. But entirely apart from the fact that the teachers were seldom native Romans or even Italians, and that their pupils no longer had a chance to acquire the genuine sermo urbanus from the observation of living models in public life, there were changes in the conditions of public speaking which in themselves must have been distinctly hostile to the use of our construction. The discussions in the Dialogus of Tacitus make it clear that the leisurely, rambling speeches of republican times were no longer possible. 1 A pleader now had to economize his time and stick to his subject, 2 and there was little room for the broad, almost conversational treatment of a narratio, for example, which would permit of much use of a con- struction like the historical infinitive. Even in the case of Cicero himself, we must remember that about three-fourths of all the instances of its use in the speeches are found in those against Verres, particularly in those of the Actio Secunda, which were never actu- ally delivered in court, and which could hardly have been delivered in the form in which they were published, even in that day. The imitation of Virgil by the epic writers of the first century, and the return to earlier prose models under the influence of Quin- tilian, and, later on, under Fronto, brought the historical infinitive back into use to a degree. But it is easy to see that it was from now on an artificial and bookish thing, and no longer what it was before 1 Dial. 19. 3; 20. 1; 22. 4 and 7; 23. 3. 2 Ibid. 19. 6: 20. 2. 60 J. J. SCHLICHEK Sallust, an integral and living part of the common speech. The historians and epic poets still used it as part of the recognized armory of expression in their respective fields, just as the later Greek epic copies the language of Homer, or the preachers of the gospel that of the Bible. But outside of these two fields, and a slight use in the letter and the novel, it was practically extinct after the time of Augustus. THE POETRY OP THE EMPIRE Of the epic poets of the first century a.d., Lucan, who is the most original and the least influenced by Virgil, also makes least use of the historical infinitive. Of his four passages, three are found near the beginning of the first book of the Pharsalia. And two of these (i. 129-35 and 146-49) are set characterizations, composed of infini- tives co-ordinate with descriptive adjectives, such as are found in Terence and Sallust, for instance, but not in Virgil. It is in Valerius Flaccus that we meet for the first time the real cause which sustained the failing construction in epic poetry. His general dependence on Virgil is well known, 1 and so far as the his- torical infinitive is concerned the relation between the two poets is certainly striking enough. It appears most conspicuously in the frequent use of the same infinitive at the same point in the hexameter. Argonautica Aeneid 1. 3 trepidare, ii. 47. trepidare, ii. 685; vi. 491; ix. 538. 2. B consurgere, ii. 478. resurgere, ii. 520. 6 insurgere, iv. 274. consurgere, x. 299. 3. mugire, ii. 498. mugire, viii. 215. 4. hi tendere contra, vi. 362. illi tendere contra, ix. 377. 5. 3 petere et, v. 127. petere ac, ix. 790. 6. 2 3 audiri, v. 169. exaudiri, vi. 557; vii. 15. 7. 5 educere, n. 371. diducere, iv. 275. adducere, vi. 133. 5 pars ducere, i. 423. 1 The extent of it is roughly apparent from the fact that in Grueneberg's dis- cussion of the subject (De Vol. Flacco imitatore, Berlin, 1893), fifty pages are devoted to passages in which he shows the influence of Virgil, as against twenty pages for Ovid, Horace, Tibullus, Propertius, Seneca, and Lucan combined. The Histokical Infinitive 61 Phrases closely akin are also found in the same location: 8 8 tm 8. turn rumpere questus, iv. 42. hinc spargere voces, u. 98. 8 8 spernere voces, vii. 237. pars tollere vocem, vi. 492. 6 S 9. manibus diducere caestus, iv. umeris abscindere vestem, v. 685. 275. There are, besides this, two passages in Valerius which have a dis- tribution of metrically similar forms of the infinitive so like that in a passage of the Aeneid that it can hardly be due to accident. 10. Aen. i. 423-24: pars ducere muros molirigue arcem et manibus svbvolvere saxa. Arg. ii. 519-20: intremere inde inlidique vadis pronae atque resurgere puppes. Arg. vii. 613: armarique phalanx totisque insurgere campis. To this list should be added spargere, an unusual word, employed as a historical infinitive, in the Aeneid (ii. 98) and three times in Valerius (iii. 257, 603; vi. 133), but not in any other author. 1 When we recall that there are only thirty passages in the Aeneid which contain historical infinitives, this list assumes considerable proportions, and it appears clear that while Valerius was widely indebted to Virgil, he was especially influenced by him in the use of the historical infinitive. But the same is true also of Statius, 2 where we find the same varieties of relationship to Virgil, and in much the same proportion. With one exception the passages are all in the Thebaid. Statius Aeneid B 8 1. spumare, vn. 67. spumare, viii. 689. 2. videri, v. 40, 135; vi. 818; vii. videri, xii. 216. 315. 3. adsurgere, iv. 580. consurgere, x. 299. e e 4. relinqui, x. 151. relinqui, viii. 216. 12 12 5. nee celerare gradum, viii. 158. sed celerare fugam, ix. 378. 6. quaerere, Silvae v. 2. 42. quaerere, ii. 99. 1 In Virgil the object is vocem, in Valerius iii. 603 voces. The compound dis- pergere is found as a historical infinitive in Claudian In Eutropium ii. 85. 2 For his indebtedness to Virgil in general, see A. Deipser, De Statio Vergilii et Ovidii imitatore (Diss. Argentor, 5. 91). 62 J. J. SCHLICHBB Similar phrases: B 5 7. expromere voces, u. 101. spargere voces, ii. 98. 6 tollere vocem, vi. 492. Similar distribution of infinitives: 8. Illi ad fata rapi atque illi iam occumbere leto, Theb. v. 693. Turn sic adfari et curas his demere dictis, Aen. ii. 775. 9. Pars clipeis munire ratem, pars aequora fundo, Theb. v. 382. Ingenti trepidare metu; pars vertere terga, Aen. vi. 491. 10. .... pars .... emunire toros alteque inferre tapetas, pars teretes levare manu ac disponere mensas, Theb. i. 517-19. pars ducere muros molirique arcem et manibus subvolvere saxa, pars optare locum tecto et concludere sulco, Aen. i. 423-25. 11. Devesei et multum patrio pinguescere luctu, Theb. i. 604. Audiri et longe pastorum rumpere somnos, Theb. iv. 714. molirique arcem et manibus subvolvere saxa, Aen. i. 424. Of these, Nos. 3 and 11 are also found in Valerius (Nos. 2 and 10). Besides, Statius has in common with Valerius also: solvere, Theb. xii. 10. solvere, Arg. viii. 220. 6 6 rumpere somnos, Theb. iv. 714. rumpere questus, Arg. iv. 42. Silius, while using the historical infinitive somewhat more rarely, nevertheless shows a similar dependence on Virgil. Punica Aeneid 5 B 1. nudo non credere ferro, xiii. 198. brevibus se credere saltu, x. 289. B pedibus se credere Celtae, iv. 300. B 5 2. confidere rebus, xiv. 90. et fidere nocti, ix. 378. 3 3 t 3. trepidare metu, xv. 578. trepidare metu, ii. 685; vi. 491. 4. mere implacabile, xvii. 252. ruere et, xi. 142. S B 5. consurgere, xi. 109. consurgere, x. 298. 6 B f 6. traducere noctem, ix. 4. ducere muros, l. 423. Nos. 5 and 6 are both found in Valerius (Nos. 2 and 7), No. 5 in Statius (No. 3). Besides, Silius has in common with Statius: respicere, Pun. vi. 335. respicere, Theb. ii. 308. and with Valerius : vertere nunc, Pun. i. 242. vertere tunc, Arg. vii. 6. The Historical Infinitive 63 Ostentare, increpitare, velle, instate are also found as historical infinitives in both Silius and Valerius, though not at the same place in the verse. They also have the following similar distribution of infinitives: Inde ferae silvis sterni; turn serpere labes, Pun. xiv. 596. Turn lacrimis, turn voce sequi, turn rumpere questus, Arg. iv. 42. Bellatrix tunc gleba quati, pariterque creari, Arg. vii. 612. Of the co-ordinating words used by Virgil in connection with the historical infinitive, we find pars .... pars in Statius, as well as the more prosaic pars .... alii and hi ... . alii, which are not used by Virgil, nor by any of the other poets. Virgil's hinc .... hinc is found in Silius. Nee (nihil) solum .... sed is found in all the first-century epic poets except Lucan. The same is true of turn and hinc, used singly. In addition to the foregoing expressions, we find the following which Virgil did not use with the historical infinitive: Illi .... Mi, nee .... nee, vix .... vix in Statius, hi ... . hi, turn .... turn in Valerius, modo .... modo, hie . . . Me in Silius, and nunc .... nunc in all three. Their more extensive use of such expressions, compared with Virgil, is perhaps due to the rhetorical character of their style. They went beyond their master in still another direction. This was in the use of intensive modifiers which are at the same time richly descriptive. Modifiers of the purely intensive kind, such as omnes, magis etiam, ilico, adeo, and the like, had been a very early feature of the historical infinitive, and are common in Plautus and Terence. 1 Virgil had in addition to these used modifiers also that are at once intensive and descriptive. 2 These are perhaps more widely characteristic of Valerius, Statius, and Silius than anything else. They help the construction to produce that loud and stren- uous effect which is so peculiar to these poets. 3 On the whole we 1 Cf. Class. Phil., IX, 281. 2 Cf. Class. Phil., IX, 381. 3 The examples are so numerous that a few illustrations from each of the authors must suffice. Lucan: "reparare novas vires, longos iungere fines, cervix lassata quati, obliqua labare crura." Valerius: "monstriferi mugire sinus, maestis reposcere votis, longa silentia rumpere." Statius: " atra sede tegi, fixos superum ad penetralia currus vellere, elatis tremefacta adsurgere vittis, trepidas spumare per herbas, dira labes nocturno squalida passu inlabi." Silius: "proterere exanimos artus, telo instare sequaci, nudo non credere ferro, pondera regni posse pati." 64 J. J. SCHLICHKR may say that it was as an expression of awe, horror, despair, and human helplessness that the construction had its chief interest for these poets, just as for Virgil. 1 It must be said for them, on the other hand, that they observe moderation in the length of the series. Virgil, Statius, and Silius, in their epics, did not go beyond five infinitives, and Valerius did not go beyond four. While Virgil had used the construction once in a relative clause, the others, except Statius, who has it after quam- quam and wide, used it only in main clauses. A further restraint, as compared with prose, is found in the matter of alliterative and rhyming groups of infinitives, both of which are very rare in poetry. 2 After the authors just mentioned more than two centuries pass before we meet the historical infinitive again in poetry. Little is preserved to us during this period, and to judge from what we have it is unlikely that the construction was used, to any great extent, at least, in what has perished. When it appears once more, it is in a Christian poet — Juvencus. He is followed by others, also Chris- tians, through the fourth and fifth centuries, who use it one or more times apiece, but none of them frequently — Prudentius, Ausonius, > Virgil Aen. ii. 132, 169 (cf. Georg. i. 200); Hi. 141; iv. 422; vi. 491-92, 557 (=vii. 15); ix. 377-78, 538-39; xii. 216-17; Valerius Arg. i. 213, 608; ii. 47, 478, 498,519-20; iii. 95, 268, 601; iv.42; v. 169; vii. 6-7, 612-13; Statius Theb. i. 601-4, 621-23; ii. 544; iv. 579-80; v. 693; x. 150-51; Silius Pun. xiv. 88-90, 596-97; xv. 578-79; xvii. 17, 251-52. 8 Terence and Afranius have one alliterative pair each, and the same is true of Virgil, Valerius Flaccus, and Claudian. Some of the prose authors are equally con- servative; so Sallust, who has only 4 pairs, although he uses the construction so frequently. Livy has 23 pairs, Tacitus 15, Cicero and Curtius 4 each, the Bellum Africanum 3. Livy, Tacitus, and Cicero have, further, each a series of three infini- tives beginning with the same letter. Rhyme is most common in infinitives of the first conjugation, which constitute about one-third of all verbs used in the construction. That this was at times sought, and at times avoided, seems clear from the fact that the Orations against Verres have twelve such rhyming pairs, while the rest of Cicero's works appear to have none. Groups of three rhyming infinitives are also found in Afranius, Claudius Quadrig., Bel. African., Sallust, Tacitus, and Livy. The last two authors have series of four, and Livy even has one of five rhyming infinitives of the first conjugation without interruption by other infinitives (xxix. 15. 11, 12). In this he is followed by his epitomator Floras, who also has one series of five, interrupted by one deponent of the first conjugation (ii. 13. 82). In Sallust and Tacitus many long rhyme series are interrupted by infinitives of other conjugations. Of the poets, Plautus and Terence have each one rhyming pair of infinitives of the first conjugation, as have also Valerius Flaccus, Prudentius, and Claudian. The other poets have none, except Silius Italicus, who has five pairs, in this respect approaching the practice of prose. The Historical Infinitive 65 Claudius Marius Victor, Paulinus Petricordiae, Alcimus Avitus, Sidonius Apollinaris, Cyprianus poeta, and Paulinus Nolanus — nearly all of them Gauls. The indebtedness of the early Christian writers of hexameter verse to Virgil is well known, and the historical infinitive is prac- tically confined to the hexameter. 1 Twice, at least, this indebted- ness is of the cento type — trepidare metu (Cyprian Num. 447) from Aen. ii. (585 and vi. 491 (also Silius Italicus xv. 578), and a half- line in Alcimus Avitus (iii. 315), retro sublapsa referri, which is taken bodily from Aen. ii. 169 (=Georg. i. 200). But, generally speaking, Virgil's influence is no longer so distinctly traceable in them as it was in the poets of the first century a.d. It is evident rather from the fact that the historical infinitive is generally found in those authors and compositions which show the influence of Virgil and classical models in other respects also. The use of the construction by the Christian authors above mentioned is quite limited, the nine authors having between them only a total of 29 passages containing the historical infinitive. And those among them who use it with any frequency at all show a marked tendency to use it mechanically. This is especially apparent in the hexameter, where the infinitive gravitates toward the fifth foot. Thus, of the nine historical infinitives which Prudentius has in the hexameter, seven are found in this foot; in Paulinus Petri- cordiae, eight out of fourteen; in Juvencus, five out of six. Authors who use it only occasionally tend more and more to the use of single infinitives instead of series, and the assumption is easy that they employed them chiefly as a metrical convenience, as, for example, in the fifth foot of the hexameter. Claudian is the only poet of this period who uses the construction with freedom (19 passages with 41 infinitives), and with something like its old effect. He does not exhibit the same extensive use of Virgilian methods as the poets of the first century, and yet there are some unmistakable examples in which he shows his relation both to Virgil and to them. 1 The exceptions are two passages in Prudentius (Peristeph. iii. 21-22, xiii. 31-32), which are written, however, in other forms of dactylic meter (tetrameter catalectic and greater archilochian) , and the passage in Ausonius (Com. Prof. Burdig. iii. 11-12), which is in the elegiac meter. 66 J. J. SCHLICHER s 1. extollere (Claud. De raptu Pros. ii. 321). tollere (Virg. Aen. vi. 492). » 2. mugire (Claud. De raptu Pros. ii. 152). mugire (Virg. Aen. viii. 215; Val. Arg. ii. 498). 3. videri (Claud. In Eutrop. i. 346). videri (Virg. Aen. xii. 216; Stat. Theb. v. 40, 135; vi. 818; vii. 315). 4. ire (Claud. In Eutrop. i. 257). Ire (Stat. Theb. ix. 110). 5. moliri (Claud. De cons. SHI. i. 42). moliri (Sil. Pun. xiii. 198). A somewhat similar relation between Claudian and the earlier poets is found also in the distribution of the infinitives within the verse. Thus Virgil Aen. ii. 98; v. 686; vi. 491 are like Claudian De raptu Pros. ii. 152; In Eutrop. ii. 85; Epithal. Hon. 239; Val. Fl. iv. 42 is like Claud. De sext. cons. Hon. 32; Sil. xiv. 595 is like Claud. In Ruf. i. 178; Virg. Aen. ii. 775, vi. 557, and Sil. xiii. 198 are like Claud. De raptu Pros. iii. 211 and De cons. Stil. i. 42. The co-ordinating tunc .... tunc and pars .... alii occur in Claudian, and the historical infinitive is used once in a continu- ing cum-cl&use (Epithal. de nupt. Hon. 238-40). Aside from a certain shifting of the vocabulary due to the difference in time, Claudian's use of the construction is altogether classical. In the vocabulary of the construction he is much more independent than the poets of the first century — only 7 of his 41 historical infinitives being words which are found thus used in the epic poetry before him. On the other hand, of the Christian poets several remain quite strictly within the earlier limits. In Corippus four out of five infinitives are old, in Juvencus three out of five, in Alcimus Avitus three out of four, in Paulinus Nolanus three out of five. Paulinus Petricordiae is the most original in this respect. Prudentius shows wide acquaintance with classical authors, combined with originality. Neither does Claudian show any of the stiffness of the con- temporary Christian poets. Only 14 of his 41 infinitives are in the fifth foot, a smaller proportion than in Virgil himself. Claudian is further quite free from the rhetorical excesses which characterize the construction in the epics of the first century. The Historical Infinitive 67 The last use of the historical infinitive to any extent worth mentioning, in poetry, is made, apparently, by Corippus in the sixth century, and his employment of it, as well as his comparatively correct versification, is due to his very close dependence upon classi- cal models, especially Virgil. We find another occurrence in the same century in Fortunatus, if the text is correct, and one in Bede at the beginning of the eighth. After that the construction seems to be no longer used, even in the hexameter. The epics of the Carolingian period do not appear to have it, even to the slight extent of the Christian poets above mentioned. THE PROSE OF THE EMPIRE The very general use which the epic poets of the first century made of the historical infinitive left its mark to an extent also upon the writers of prose. Pliny, Curtius, and Tacitus, for example, use it in the description of impressive natural phenomena, 1 which, as we have seen, was common in Virgil and his successors, as it had been in tragedy, but is not found in prose after the Silver Age. Of the various features of the construction which had been gradually introduced in the course of its history, several lasted to the very end. Among these are the use of the passive voice, the series composed of infinitives co-ordinate with indicatives, and the extended list of subordinate clauses, modifying the infinitive clause, which we find first in Sallust and Livy. On the other hand the practice of using the historical infinitive itself in subordinate clauses reaches a rather sudden end with Tacitus. Only two cases are found in prose after that (Hegesippus iii. 3. 4, in a continuing relative clause, and Lactant. De mort. persecut. 17. 6, after "postpositive cum"). Co-ordinating and contrasting words like alius .... alius, nunc .... nunc, etc., also maintain themselves pretty well, although they are found more in some authors (Curt., Justin, Pliny, Apul., Sidon., Dictys) than in others. The common strengthen- ing words which had been associated with the construction since the earliest period (omnes vehementer, etc.) are found very fre- quently and generally in the prose authors who use it after the Augustan Age. 1 Curtius iv. 3. 17; ix. 9. 10; Pliny vi. 20. 11-12; vii. 27. 8; ix. 33. 4; Tacitus An. i. 70; ii. 23; iv. 49. 68 J. J. SCHLICHEB On the other hand, a number of features that had been common at one time or another disappear, in whole or part. This is especially true of several which had become very prominent in Livy. The infinitive dicere, which he used so frequently (over 40 times), and which had been used freely by Cicero as well, occurs only once more in our construction after Livy's time (Heges. iv. 26. 2). Negare does not occur again except in two passages in Sulpicius Severus (Chron. ii. 8. 5; Vita S. Mart. 17. 2). Esse is not found after Livy, except once in Petronius (62. 5) and once in Tacitus (A gr. 19), and videri is found only once more in prose (Justin xviii. 3. 11). Of these words only the last occurs in poetry, where the example of Virgil (Aen. xii. 216) was followed by Statius (four times), by Clau- dian (once), and by Paulinus Nolanus (once), every one of these cases being in the sixth foot of the hexameter. Generally speaking, however, and aside from the inevitable changes of vocabulary wrought by time, the words most commonly used in the later authors are still as a rule the same which had been employed in the classical period. For example, out of thirty words which occur three or more times in prose after Tacitus, there are only three which had not been used as historical infinitives before. 1 This suggests that the loss was greater than the gain. There was, in fact, no expansion. The types which we find after Tacitus are still practically the same which had existed from the beginning. Some of the newer ones introduced by Sallust and adopted by others maintained a foothold also. For instance, descriptions of battles and other military operations continued, after Sallust's fashion, to be done by historical infinitives. The appositional series 2 is still found occasionally. On the other hand, the careful preparation which Sallust so often made to give strange infinitives a standing by supplying the proper emotional tone is found to any extent worth mentioning only in the earlier writers of the empire. And of the subtle manner in which the historical infinitives of Tacitus were 1 They are prohibere, commovere, and fateri. The other 27, in the order of fre- quency, are: habere, orare, hortari, mirari, circumire, audire, agere, polliceri, dbnuere, credere, dare, extollere, monere, terrere, caedere, gratulari, instare, incusare, meditari, obtruncare, opperiri, perstare, praebere, precari, resistere, sedare, urgere. 2 Cf. Class. Phil., IX, 389. The Historical Infinitive 69 staged, there is, of course, hardly a trace after his time. On the whole, we may say that while the construction is still employed correctly enough, there is a general decline in the discrimination and effectiveness with which it is used. Some of the writers of the empire have certain special types of the historical infinitive which they prefer, quite to the exclusion of the rest. Thus Curtius and Floras have a preference for those expressing strong assertion or attitude, and for those expressing pursuit of a goal. Sulpicius Severus confines himself pretty strictly to the former of these, and has far more of that sort than of all the others together. In Aurelius Victor, and especially the Epitome, an even greater predominance is enjoyed by the characterizing type which, usually composed of co-ordinate infinitives and descriptive adjectives, had grown out of the old type expressing habit or incli- nation. The series in this latter type is sometimes carried to an absurd length, as in Fronto and Sidonius. The construction as it is used by Fronto and Apuleius differs from that found in the other late authors, in the many unusual words which they employ and in their preference for types which other prose writers of this period do not use so much, particularly, in Apuleius, those expressing emotion and distraction or excitement. Their unusual words are in line with their general vocabulary, and it is possible that in preferring those types of the historical infinitive which had gone somewhat out of date, they may have felt that they were doing honor to the writers of the ante-classical period. A relatively large proportion of very long series is especially evident in some of the prose writers of the empire. Pliny has two, of seven and nine infinitives respectively (vii. 27. 8 and ix. 33. 4), Fronto (Naber, p. 207) has one of eighteen, Apuleius one of twelve (Met. viii. 7), Sidonius one of nine (i. 7. 8) and one of thirteen (vii. 2.4), Hegesippus one of twelve (iii. 3. 4) and several of eight, Lactantius has one of ten (Inst. epit. 54. 7), the Epitome of Aurelius Victor one of eight (48). There are some individual peculiarities besides this, such as the frequent use of two infinitives in Dictys, the absence of the series in Petronius, the infrequent occurrence or absence of the single infinitive in Apuleius, Hegesippus, and Sidonius. The fond- ness of several of the late writers for long series may reasonably be 70 J. J. SCHLICHEB set down to the credit of Sallust, whose influence upon the prose of this period was very great. In a number of the later authors the historical infinitive is rather unevenly distributed over their works. All but two of the passages in Florus are found in i. 5 and ii. 13. All but two in the Metamor- phoses of Apuleius are in the last part of Book ii, the first part of Book iii, and the first part of Book x. The five passages in Julius Capitolinus are all in the Lives of the Maximins and the Gordians. In Aurelius Victor's Caesares they are confined to the first 14 chap- ters, the remaining 28 having none. In the Epitome they are found similarly from the beginning to chap. 14, and from chap. 41 to the end, these latter chapters covering a period not treated in the Caesares. In Sulpicius Severus' Chronica they are confined to the second half of the first book, and the beginning and end of the second. In Sidonius they appear in two letters of the first book and one of the seventh. The two cases in Dares are found in chaps. 36 and 37. In the other prose authors and in the epic poets, except Lucan, they are more evenly distributed. Three of Lucan's four passages are contained within less than forty lines near the beginning of his first book (11. 133-70). His virtual discontinuance of the con- struction after that suggests that he felt it to be foreign to him, even though it had with Virgil become a part of the epic style. No doubt a similar condition existed in the case of some of the prose authors of whom we have just spoken. They are known to have followed earlier authors as their models, or to have been strongly influenced by them. Florus' work, for example, is an epitome of Livy, Apuleius is an archaizer, Aurelius Victor and Sulpicius Sev- erus are under the influence of Sallust, and Sidonius, of Pliny the Younger. The authors who influenced them had used the historical infinitive, and its uneven use by the later authors seems like a sporadic following of the practice of their models, while their real inclinations and the practice of their own times, or, possibly, other literary influences, were in the opposite direction. The so-called Epitome of Aurelius Victor is an interesting and instructive case in point. It is found to be made up of three distinct parts: 1 chaps. 1-11, following closely the corresponding chapters of Aurelius i Wolfflin, Rhein. Mus., XXIX, 285 ff. The Historical Infinitive 71 Victor's Caesares; chaps. 12-39, in which the Caesares are not thus used; and chaps. 40-48, in which the style again changes, and which extend the account beyond the Caesares. The historical infinitive is used in the first section and extends even to chap. 14, in the second, just as it does in the Caesares. Then there are no instances of it till we reach chap. 41. But from this point on to the end they become quite frequent. It is a remarkable fact that the revival of the historical infinitive in the fourth and early fifth centuries, so far as prose is concerned, is found most prominently — indeed, almost exclusively — in the writers and works which show strongly the influence of Sallust. These are Sulpicius Severus, Aurelius Victor, Dictys, and the so- called Hegesippus. Those who employ the simple epitome or chronicle style (Eutropius, Festus, etc.) do not have the construc- tion, Florus and Justin, who reproduce features of the style of their originals, being the exceptions which prove the rule. In the Chronica of Sulpicius Severus, where the narrative is fuller in parts, and more slender in others, the historical infinitive is found only in the former, and the same relation between the construction and the fulness of treatment is seen also in the Epitome mentioned above, and in Florus, five of whose nine passages are found in the unusually full account of the war between Caesar and Pompey (ii. 13). The biographers of the Historia Augusta, with one exception, Julius Capitolinus, do not use the construction, in this perhaps following the practice established by Suetonius. Summing up briefly the later history of the historical infinitive, we find a revival of interest in it, owing to the influence of Virgil and Cicero, in the second half of the first century a.d., and to the influence of the early Latin authors, in the second century. The early Christian writers, both in prose and in verse, use it quite generally, but at the same time sparingly. 1 There is another revival of interest in the construction during the fourth century, when authors like Claudian, Sulpicius Severus, Hegesippus, and Dictys again use it with some of the old-time freedom, though evidently under the influence of earlier authors, especially Sallust in prose and 1 Moat of their prose writings, being expository or argumentative, are not well suited to its use. 72 J. J. SCHLICHER Virgil in poetry. The most important historian of the period, Ammianus, has but slight traces of the construction, and the same is true of Orosius. Its last use in prose is found apparently in Sidonius, in the second half of the fifth century. 1 It maintained itself somewhat better in poetry, where the firm structure of the hexameter furnished it support and shelter. With a few exceptions the authors and works mentioned below have been carefully searched by the writer himself. Some omissions will no doubt be found, but the lists are substantially complete, sufficiently so for deter- mining the extent to which the construction was used in late Latin. In addi- tion to the authors and works mentioned, which have the construction, a considerable number were examined in which it was not found. Among these are the minor pagan poetry, especially that having a narrative element, between the beginning of the second and the fifth century, the Querolus, Sedulius, Orientius, Dracontius, Paulinus of Pella, Ennodius, Prosper, Orator, various minor Christian poems found in Petschenig's edition and elsewhere, Aldhelm (De laudibus virginum), and about half of the longer narrative pieces in Dummler and Traube's Poetae Latini Aevi Carolini. Narrative works in prose which do not have the construction are: the writers of the Historia Augusta (except Julius Capitolinus), Festus, Eutro- pius, Jordanes, Julius Exsuperantius, the Excerpta Valesiana, the Historia Apollonii, Eugippius (Vita S. Severini), Victor Vitensis (Hist, persecut. Vandal.), the Origo gentis humanae and De viris illustribus, Bedes' Ecclesi- astical History, Paulus Diaconus (De gestib. Langobardorum) , Isidore of Seville (De regib. Gothorum). Further, the letters of Cyprian, Symmachus, and Ennodius. POETRY OF THE EMPIRE Lucan: i. 132-35, 147-48, 167-70; iv. 624-25. Valerius Flaccus: i. 213, 608-9; ii. 47, 371-72, 457, 478, 498-99, 519-20, 526; iii. 95, 257-58, 268, 601-3; iv. 42, 274-75, 494-95, 500, 690; v. 127, 169, 267, 650; vi. 133, 153-54, 362-63; vii. 6-7, 64, 237, 612-13, 625; viii. 176, 220-21. Statius: Theb. i. 413, 518-19, 602-4, 622-23; ii. 100-101, 308, 544, 592-93; iii. 571-72, 582, 587; iv. 428, 580-81, 714, 732; v. 40, 135, 382-83, 490, 513, 693; vi. 818; vii. 67, 315, 442; viii. 158; ix. 110, 588-90; x. 150- 51; xi. 474-75; xii. 9-10; Achil. ii. 150-51; Silvae ii. 1. 122-23; v. 2. 41-46; v. 3. 146-47. 1 Unless we count two cases — abire and Iransire — in Gregory of Tours, Hist. Franc. But the frequent confusion of i and ie in the MSS make them rather doubt- ful. The variants abiere and transiere also occur in these passages. The Historical Infinitive 73 Silius Italicus: i. 241^2, 248-51, 262-64; iv. Ill, 300; v. 101-4; vi. 335, 462; vii. 369-70; viii. 18, 577; ix. 4-7; xi. 109; xiii. 197-98; xiv. 89-90, 596-97; xv. 578-79, 719-24; xvi. 340; xvii. 251-52. Ilias Latina 799. Juvencus: i. 55-56; ii. 351, 566; iv. 446, 614. Prudentius: Apotheosis 455-59; Hamart. 429; Peristeph. iii. 21-22; xi. 56-58; xiii. 31-32. Ausonius: Com. Prof. Burdig. iii. 11-12. Claudian: De raptu Pros. ii. 152-53, 320-21; iii. 210-12, 215-16; De cons. Stil. 41-43; Laus Serenae 90. 222; De sext cons. Hon. 32, 237, 564; Paneg. Manl. Theod. 20-21; Epithal. Hon. Aug. 238-39; In Rufin. 178-79; In Eutrop. i. 257-58, 346-48; ii. 45-46, 84-87, 191, 432-33. Cyprianus poeta: Num. 447. Claudius Marius Victor: Aleth. ii. 494 ff., iii. 111-14. Sidonius Apollinaris: Carm. ii. 156. Paulinus Petricordiae: Vita Mart. ii. 540; iii. 112, 326; iv. 281-84, 362; v. 251-52, 289-90. Alcimus Avitus: Poemat. iii. 315, 320; v. 70, 662. Paulinus Nolanus: xv. 175-76; xviii. 389-90, 396. Corippus: Johan. i. 33; iv. 28; In laud. lust. iv. 18-19, 89. Fortunatus: Vita S. Mart. iv. 47. Bede: Vita S. Cuthb. xliv. PROSE OF THE EMPIKE Petronius: 52. 4; 62.5; 62.8. Curtius: iv. 3. 17; 4. 9; 9. 20; 10. 30; 10. 33; 12.21; 13. 16; 15. 19; v. 4. 13; 10. 11: 12. 7; vi. 2. 20; 7. 8; 11. 14; viii. 2. 5; 3. 2; 3. 7; 4. 9; ix. 9. 10. Pliny: Epist. ii. 20. 7; vi. 20. 11-12; 20. 15; viii. 27. 8; ix. 13. 7; 33. 4; Paneg. 22. 3. Florus: i. 39. 2; ii. 5. 4; 5. 9; 13. 9; 13. 17; 13. 39; 13. 42; 13. 82; 15. 2-3. Justin: v. 2. 14; 7. 4; ix. 8. 12-14; xi. 6. 8-9; 9. 8; xii. 6. 8; 8. 12; 8. 14; 11. 7; xiv. 2. 8; xvi. 4. 18; xviii. 3. 11; xxi. 5. 4-8; xxx. 1. 5; xxxii. 2. 8; 3. 2; xxxvi. 4. 2; xxxix. 3. 7; xli. 3. 4; xiii. 4. 12; xliii. 4. 11. Fronto (Naber): Pp. 128, 207, 215, 237, 238. Apuleius: Metam. ii. 19, 26, 27, 29; iii. 1, 10; viii. 7; ix. 30; x. 6, 10; Flor. 3. 16; 16. 65; 19. 95. Granius Licinianus (Flemisch) : Pp. 4, 5. Minucius Felix: iv. 1. Arnobius: iv. 33 (2); v. 5 Lactantius: Instit. epit. 54. 7; De mort. persecut. 7. 4; 17. 6; 19. 4; 47. 2. 74 J. J. SCHLICHER Augustine: Confess, vi. 1. Itinerarium Alexandri, chaps. 45, 69. Julius Capitolinus: Vita Maxim. 3. 6; 6. 2; 22. 6; Vita Gord. 13. 4; 31. 7. Ammianus Marcellinus: xxix. 2. 13; 3. 7. Aurelius Victor: Caesar. 3. 14; 3. 19; 6. 1; 14. 6; Ejrit. 2. 6; 9. 15; 13.5; 14.3; 14.7; 41.14; 41.21; 42.16; 45.6; 46.3; 47.4; 48.9; 48. 13; 48. 17-18. Hegesippus: De bello Judaico: i. 15. 3; 31. 3; 37. 2; 40. 2; 40. 4; 40. 12; ii. 15. 6; 15. 7; iii. 2 (2); 3. 4; iv. 11; 13. 1; 23.1; 25. 2; 26. 2; v. 13; 24. 4; 31; 33. 2; 38; 42. 5; 50. Dictys: i. 9, 19 (2), 21; ii. 1 (2), 5, 6, 13 (?), 16, 23 (2), 24, 25, 30, 34, 38, 41, 42, 43, 46, 48 (2), 49, 50, 51 (2); iii. 1 (2), 3, 6, 9, 16, 20, 22; iv. 1, 2, 3, 4 (2), 5, 11, 12, 13, 16, 17, 20, 22; v. 1, 6, 7 (3), 8, 10 (3), 12, 14 (2), 15 (2), 16(3); vi. 15(2). Orosius: iii. 4. 2. Sulpicius Severus: Chron. i. 22. 5; 25. 4; 26. 4; 31. 5; 32. 5; 33. 7 35. 8; 38. 8; 43. 4; 43. 7; 43. 8; 50. 5; ii. 4. 5; 7. 4; 8. 5; 11. 3; 44. 2 45. 4; 47. 5; Vita S. Martini 2. 8; 3. 2; 13. 2; 13. 6; 13. 9; 16. 5; 17. 2 23.4; Dial.i. 18.3; ii. 3. 8; 11.4. Dares: 36, 37. SidoniusApollinaris: i. 7. 3; 7.8; 11.2; 11.7-8; 11.16; vii. 2. 4r-5. Gregory of Tours: Hist. Franc, i. 15 (?); vii. 35 (?). Tebrb Haute, Ind.