No CrossRef data available.
Article contents
Virgil, Aeneid 2.349–50
Published online by Cambridge University Press: 11 February 2009
Extract
A textual problem in Virgil, Aen. 2.349–50 has puzzled scholars since antiquity and still divides editors and commentators today. Aeneas is exhorting his comrades to join him in the final battle for Troy, but the variants audendi and audentem leave it uncertain whether he says, ‘si vobis audendi extrema cupido/ certa sequi, quae sit rebus fortuna videtis’, or ‘si vobis audentem extrema cupido/ certa sequi, quae sit rebus fortuna videtis’. The variant audendi has been discussed and defended in several commentaries and articles, whereas audentem, though often the choice of editors, has yet to receive a detailed defence. In this note, I will demonstrate that audentem is the correct reading and that the false reading audendi came about through a common scribal error.
- Type
- Research Article
- Information
- Copyright
- Copyright © The Classical Association 1987
References
1 Some of the editors and critics who have chosen audendi: Ladewig, T.(Berlin, 1865); F. A. Hirtzel (Oxford, 1900); P. Lejay (Paris, 1919); R. Sabbadini (Rome, 1930); L. Castiglioni (Paravia, 1945); R. G. Austin (Oxford, 1964); R. D. Williams (London, 1972); M. Geymonat (Paravia, 1973). Among those who have chosen audentem:J. L. de la Cerda (Cologne, 1664); C. G. Heyne and G. P. E. Wagner (Leipzig, 1832); J. Conington (London, 1876); O. Ribbeck (Leipzig, 1894); T. E. Page (London, 1894); H. R. Fairclough (London, 1934); H. Goelzer and A. Bellesort (Paris, 1956); R. A. B. Mynors (Oxford, 1969); J. Perret (Paris, 1981).Google Scholar
2 See Ladewig, op. cit. n. 1; E. Baehrens, ‘ Emendationes Vergilianae’, Neue Jahrbiicher fiir Philologie undPadagogik131 (1885), 395–6; P. Deuticke, ‘Vergil: 1884 bis 1888’, Jahresberichte des Philologischen Vereins zu Berlin14 (1889), 330; Lejay, op. cit. n. 1; W. Aly, ‘Zwei Interpretationen aus dem zweiten Buch der Aeneis’, Philologische Wochenscrift48 (1928), 878–9; and Austin, op. cit. n. 1.Google Scholar
3 Conington, op. cit. n. 1, in stating that he believes no construction is possible with audendi, comes closest to defending audentem, but this is the extent of his remarks.
4 In the fourth or fifth century manuscript P, a hole in the parchment has destroyed the final syllable, leaving auden.The space is large enough for either two or three letters; cf. H. R. Fairclough, ‘Observations on Sabbadini's Variorum Edition of Virgil’, TAPA 63 (1932), 212.
5 Zetzel, J. E. G., Latin Textual Criticism in Antiquity(Salem, N.H., 1981), 132. It is worth noting that Tiberius Claudius Donatus read audendi, and the text quoted at Isid. Etym.1.37.20 shows audendi, although Isidore's paraphrase (which is identical to Servius) presumes audentem.Google Scholar
6 Austin (op. cit. n. 1), for example, in defending audendi, relies heavily on its ‘obvious authority’ in the manuscripts. Baehrens, Deuticke, and Aly (op. cit. n. 2) all argue that the reading audentemoriginated as Servius ‘conjecture’ and can therefore be discounted.
7 Although no commentator argues this in so many words, C. Hardie (review of R. G. Austin,Virgil Aeneid II, JRS54 (1964), 244) believed that it underlay Austin's preference for audendi.
8 Deuticke (op. cit. n. 2) changed sequito sedet;Baehrens (op. cit. n. 2) endorsed this emendation and also suggested niforsi.Sabbadini (op. cit. n. 1) changed certa sequitocertast qui, which was adopted by Castiglioni and Geymonat (op. cit. n. 1). For an argument against this emendationsee Funaioli, G., Sludi di Lelteratura Antica(Bologna, 1948), ii(l).237–40.Google Scholar
9 Cf. Williams R. D., Virgil, Aeneid Book III(Oxford, 1962), 142, on 3.405.Google Scholar
10 Examples of elision of long -iwith etare not uncommon in Virgil; the -iin sequiis elided with etin Aen.7.300. Aly (op. cit. n. 2) suggested a similar interpretation to Lejay's, taking both audendi extremaand sequi certaas complements to cupidobut cited no parallels of this unusual construction.
11 Dirais the most common adjective with cupido(e.g. G.1.37, Aen.6.373, 9.185), also tanta (Aen.7.263), immensa (Aen.6.823), insana(Aen.9.760).
12 Cicero, for example, has 587 instances of the genitive gerundive but only 24 of the genitive gerund with accusative object. Caesar has 280 of the former but 7 of the latter. Both Caesar and Cicero use it mainly to avoid the awkward repetition of the -orumending or for particular emphasis. Cf. J. B. Hofmann and A. Szantyr, Lateinische Syntax und Stilistik(Munich, 1965), sect. C, 372–3;Kühner, R. and C. Stegmann, Ausfiirhliche Grammatik der Lateinische Sprache(Darmstadt, 1976), ii(l).735–6.Google Scholar
13 In Curtius Rufus, Books 5–10, there are 22 instances of the genitive gerund with accusative object versus 9 instances of the genitive gerundive construction; cf. Hofmann and Szantyr, op. cit. n. 12.
14 The use of the infinitive with a noun expressing desire is Virgilian (e.g. G.1.21, Aen.2.10,3.298–9,6.133–4). Patterned after the infinitive complement to verbs of desiring, the construction was employed as a metrically convenient alternative to the gerund form; cf. Austin, op. cit. n. 1, on 2.10.
15 In support of this explanation are Servius remarks on the ‘obscurity’ caused by the ‘confusion’ of the hyperbaton in these lines: ‘obscuritatem autem facit hoc loco et synchysis, id est hyperbati longa confusio, et falsa lectio.’
16 For conclusive argument that longamis the correct reading, cf. the commentaries on Book I by R. S. Conway (Cambridge, 1935), 118, and Austin, R. G. (Oxford, 1971), 211. See also Heyne-Wagner's (op. cit. n. 1) application of the principle of lectio difficilior to these lines.Google Scholar
* I wish to thank my colleagues Catherine Mardikes and Dorothy Graddy for their useful criticism. I am also grateful for the helpful comments of an anonymous referee.