XII.—The Virgilian Authorship of the Helen Episode, BY PROFESSOR F. W. SHIPLEY WASHINGTON UNIVERSITY It requires no little courage to reopen the question of the Virgilian authorship of this passage, in the face of Norden's sweeping statement 1 that efforts to prove the authenticity of the passage have been based upon ignorance of the author's style, language, metre, and technique of composition, and especially in the face of the pained prophecy with which he closes his indictment to the effect that such "Rettungsversuche" will continue to be the stamping ground for the dilettante. But, in parliamentary law, it is at least permissible for the supporter of a measure to make a motion for its reconsideration. From the time of the appearance in 1903 of Heinze's book on Virgil's epic technique 2 until a few months ago, I myself held, with Thilo,3 Leo, Heinze, and Norden, the orthodox belief that the passage was an interpolation, notwithstanding Fairclough's 5 able defense of it in Classical Philology for 1906. My reasons for so regarding it were based, like those of Norden, upon peculiarities of language and versification, in addition to the artistic considerations adduced by his three compatriots. 2 One does not lightly abandon a long cherished belief. My reasons for reconsidering my own conclusions in regard to the passage are based upon the very peculiarities of versification which caused me, as well as Norden, to regard the passage as non-Virgilian, and the work of an interpolator. My present position in regard to the matter was forced upon me by a recent 1 Aeneis, Buch VI2, pp. 261-262, also Anhang x1, 3, p. 454. 2 Vergils Epische Technik, 1903, pp. 45 f. 3 Thilo, p. xxxii of his edition. 4 Leo, Plaut, Forsch.2 p. 39, n. 3. 5 "The Helen Episode in Vergil's Aeneid," Class. Phil. 1 (1906), 221. study of the verse technique of the Aeneid, which was undertaken originally as the outgrowth of a study of the half lines, and had no reference either directly or indirectly to the passage in question. As a result, I am satisfied that the apparent metrical peculiarities not only do not prove un-Virgilian authorship, but, on the contrary, are strong evidence in favor of the hypothesis that the passage was originally drafted by Virgil in its present form, but was omitted by Varius and Tucca, his literary executors, in accordance with the poet's known wishes in the matter. The passage is as follows: 567 570 575 580 585 Iamque adeo super unus eram, cum limina Vestae Ulcisci patriam et sceleratas sumere poenas. 'Scilicet haec Spartam incolumis patriasque Mycenas Coniugiumque domumque patres natosque videbit, Iliadum turba et Phrygiis comitata ministris? Occiderit ferro Priamus? Troia arserit igni? These lines do not occur in the older and better MSS.,7 and 7 Ribbeck in his critical note records the readings of the MSS. as follows: om M Py abc m cum ceteris Bernensibus, 4; habent Dorvillianus Gothani duo cum paucis aliis apud Heynium recentissimis. 8 are not mentioned by any ancient authority except Servius, who is probably the source from which the later MSS. derived them. They were clearly not in the text which Servius used, as he gives no annotations. He makes four references to them in the introduction and in his commentary. In the introduction he states that these lines (which he quotes in full), like the first four lines of the Aeneid, were removed by Tucca and Varius, Virgil's literary executors, and in the commentary he gives two reasons: (1) that it was not chivalrous for a brave man to wish to slay a woman, and (2) that the passage is inconsistent with the story told by Deiphobus in book VI. So much for the statements of Servius. If we take them at their face value, we might assume that, in view of the intimate terms existing between the poet and his literary executors, Varius and Tucca had omitted the passage because of the poet's own dissatisfaction with it for the reasons stated, and because he had not before his death written the passage which was to take its place. In any event, if the passage is omitted, there is an obvious lacuna, and as the second was one of the three books read to Augustus in fairly complete form, it may be justly assumed that the editors had stricken out either these lines, or others which stood in the original draft, because the poet was dissatisfied with them, and that the discarded lines 8 Servius, ed. Thilo and Hagen. (a) p. 2, 1. 12: Augustus vero, ne tantum opus periret, Tuccam et Varium hac lege iussit emendere, ut superflua demerent, nihil adderent tamen: unde et semiplenos eius invenimus versiculos, ut hic cursus fuit, et aliquos detractos; nam ab armis non coepit, sed sic (here he cites the four lines beginning with Ille ego) et in secundo hos versus esse detractos (citing the lines of the passage in question). (b) ad Aen. II, 592: Dextraque prehensum: ea corporis parte qua Helenae ictum minabatur, quae in templo Vestae stabat ornata. ut enim in primo diximus, aliquos hinc versus constat esse sublatos, nec immerito. nam et turpe est viro forti contra feminam irasci, et contrarium est Helenam in domo Priami fuisse illi rei, quae in sexto <495> dicitur, quia in domo est inventa Deiphobi, postquam ex summa arce vocaverat Graecos. Hinc autem versos esse sublatos, Veneris verba declarant dicentis <601> 'non tibi Tyndaridis facies invisa Lacaenae.' (c) ad 11, 595: quia furor est virum fortem ruere in mulieris interitum. (d) Servius Daniel. ad 11, 566: post hunc versum hi versus fuerunt qui a Tucca et Vario obliti sunt. had something to do with Helen is to be inferred from line 601. We are not here concerned with the question as to whether Virgil was justly dissatisfied with the passage, or the editors justified in expunging it, but with the other question, raised by modern scholars, as to whether Virgil was really the author of the lines cited by Servius or whether it was some later interpolator who, according to the opinion of Thilo, Leo, Heinze, and Norden, sought to fill in an existing lacuna. For lack of time, I shall merely summarize the evidence upon which the interpolation theory is based. The passage has been subjected to a closer scrutiny than almost any other in classical literature. In addition to the lack of chivalry, and the conflict with the story of Deiphobus in book vi, which were mentioned by Servius, the process of combing with a very fine comb has developed the following points which have been used as evidence against Virgilian authorship: the expressions "limina Vestae servantem," "sceleratas poenas," "ultricis flammae," "merentis poenas"; that 579 is imitated from XI, 269; that the soliloquy beginning with 577 is the only one of its kind in books II and III where Aeneas is telling his own story; that the mention of Paris in 601 shows that Venus did not have Helen alone in mind; that the words of Venus which follow point not to the thought of killing Helen, but to a purpose on the part of Aeneas to end his life in battle or even by his own hand. Most of these arguments have been answered by Noack, Fairclough, and others, with sufficient success to warrant for the interpolation theory a verdict of not proven. 9 But the arguments against Virgilian authorship, which seemed to me to be most conclusive, were the metrical considerations adduced by Norden, and it is with these in particular that I propose to deal in the present paper. In Anhang XI, 3 of his second edition of Aeneid vi, Norden has pointed out that while synaloepha of a spondaic word between the second and the third foot is not common in Virgil, occurring on the average but once in 89 lines, this form of synaloepha occurs in the Helen episode three times in fifteen lines. • Rh. Mus. XLVIII, 420 f. 573, Praemetuens, Troiae et patriae communis Erinys Norden has here hit upon what seemed to me to be an insurmountable objection to the authenticity of the passage. The last two lines certainly require a sense pause after turba and flammae, and with the traditional method of eliding, which places the caesura after et, the pause cannot function in the place in which it is demanded by the sense. Though not mentioned by Norden, who confined his observations to words of spondaic scansion, the following lines would present a similar difficulty: 572, Et Danaum poenam et deserti coniugis iras since in each of them the sense demands a pause before the et, while, with the traditional method of reading the lines, the caesura falls after the et. The piling up, within the compass of twenty-two lines, of six examples of an awkward difficulty, in which the sense is apparently sacrificed to the mere mechanics of verse making, seemed to me to be the strongest evidence against the theory that the passage was the work of a great artist like Virgil. There can be no doubt in regard to the question, if our conventional method of pronouncing et in full and slurring the preceding syllable is correct, and it matters little whether the preceding word is a spondee, as Norden has noted, or a trochee or a longer word ending in a short vowel or m. The effect would be wooden and monotonous in the extreme. But did Virgil actually read such lines as these in the way in which they are usually read? I am now convinced that he did not. In a paper presented to this association at the last meeting,1o and published in the Transactions, I have discussed the question of elisions at 3m, that is to say, at the position 10 "Hiatus, Elision, Caesura, in Vergil's Hexameter," T. A. P. A. LV, 137– |