1. ‘Jederman wird gestehen, dass hier [Goethe is discussing a painting] nicht an Allegorie zu denken sei. Es ist nach unserem Ausdruck ein Symbol’, Goethe, Werke (Zürich, 1948–1954), XIII. 868.

2. ‘Es ist die Sache, ohne die Sache zu sein, und doch die Sache’, Ibid.

3. Wilkinson, L. P., The Georgics of Virgil (Cambridge, 1969), p. 165, describes this as a ‘remarkable symbolic vision of epic’.Google Scholar

4. ‘Eine Geschichte der allegorischen Erklärung [of Greek Literature] ist dringend erwünscht’, says Pohlenz, M., Die Stoa ², II (Göttingen, 1955), 55.Google Scholar Cf. Tate, J., CQ 28 (1934), 105 ff.CrossRefGoogle Scholar, and Pépin, J., Mythe et allégorie (Paris, 1958), pp. 85–214.Google Scholar

5. The Augustan revival of early Roman religion made it difficult for the Romans to acquiesce in a literal interpretation of outworn religious practices; cf. Latte, K., Römiscbe Religiongeschichte (München, 1960), p. 297.Google Scholar

6. Iliad 2. 308–19. It is not easy to define exactly what an omen is (Reiss, E., Pauly-Wissowa, RE xviii. i (1939), 352 ff.).Google Scholar I have in mind the kind of sign frequently found in the Homeric poems, where the symbolism is usually easily understood (cf. Stockinger, H., Die Vorzeichen im homerischen Epos (St. Ottilien, 1959), p. 165).Google Scholar In popular beliefs it is often impossible to understand why an omen has a particular meaning. See also Grassmann-Fischer, B., Die Prodigien in Vergils Aeneis (München, 1966), p. 118 and bibliography.Google Scholar

7. On the allegorizing tradition applied in Alexandria to the Hebrew scriptures, cf. Chadwick, H. in the Cambridge History of Later Greek and Early Medieval Philosophy, ed. Armstrong, A. H. (Cambridge, 1967), pp. 137 ff.Google Scholar, and Pépin, , op. cit., pp. 215 ff.Google Scholar; Clement of Alexandria extended the practice to the exegesis of the New Testament (Chadwick, , p. 180Google Scholar; Pépin, , pp. 265 ff.).Google Scholar

8. Comparetti, D., Virgilio nel Medio Evo, rev. ed. by Pasquali, G. (Firenze, 1943), vol. I, pp. 71 ff. and pp. 128Google Scholar ff.; Robertson, F., PVS 6 (1966–1967), 34–45Google Scholar; Coleiro, E., PVS 13 (1973–1974), 42–53.Google Scholar

9. I do not think he actually uses the word ‘symbole’ or its derivatives. But the idea may be seen, e.g. in ‘on peut dire que l'idée des guerres puniques, l'idée d'Annibal, plane sur la composition de l'Enéide’ (Saint-Beuve, C. -A., Étude sur Virgile (Paris, 1857), p. 183)Google Scholar and in ‘Énée, a dit énergiquement Gibbon, contient en lui le germe de tous ses descendants’ (ibid., p. 86).

10. e.g. ‘we may … recognise some symbolical meaning’ in Georg. 1. 24–42. (Sellar, W. Y., The Roman Poets of the Augustan Age: Virgil ³ (Oxford, 1908), p. 224).Google Scholar

11. ‘We have come to equate “symbolist” with modern and to see the symbol as the sole instrument of the modern poet’, Weinberg, B., The limits of symbolism: Studies of five modern French poets (Chicago, 1966), p. 6.Google Scholar

12. The first edition was translated into English by G. Seligson under the title The Art of Vergil: image and symbol in the Aeneid (Ann Arbor, 1962).

13. e.g. La Penna, A., DArch 1 (1967), 220–44Google Scholar, and Eichholz, D. E., G & R 15 (1968), 105–12.Google Scholar

14. Putnam, M. C. J., HSPh 66 (1962), 233Google Scholar, inclines to this view, and R. D. Williams in his edition of Aeneid I–VI (London, 1972), ad loc. is prepared to consider it.

15. C.f.e.g. Heinze, R., op. cit., pp. 291 ff.Google Scholar, on the one hand, with Boyancé, P., La Religion de Virgile (Paris, 1963), pp. 17 ff.Google Scholar, on the other; personally I incline to the caution of Bailey, C., Religion in Virgil (Oxford, 1935), pp. 312 ff.Google Scholar

16. A comprehensive survey is a desideratum; cf. however, Maehler, H., Die Auffassung des Dichterberufs im frühen Griechentum bis zur Zeit Pindars (Göttingen, 1963)Google Scholar, and Kambylis, A., Die Dichterweibe und ihre Symbolik (Heidelberg, 1965).Google Scholar

17. Cf. Bowra, C. M., Pindar (Oxford, 1964), pp. 3 and 230.Google Scholar

18. Cf. Wimmel, W., Kallimachos in Rom (Wiesbaden, 1960), pp. 103 ff. and 222 ff.Google Scholar

19. Cf. especially Propertius 4. 6. 1–10, and Eisenhut, W., Hermes 84 (1956), 121–8.Google Scholar

20. For an attempt to elucidate this symbolism, cf. Griffiths, C., PVS 9 (1969–1970), 1–19a.Google Scholar

21. Verse 658. So too the suicides for love are placed in a wood of myrtle (sacred to Venus) at Aen. 6. 443 It is hardly more than a coincidence that the bay and the myrtle are found mentioned together at Ecl. 2. 54 (‘et uos, o lauri, capiam, et te, proxuma myrte,/sic positae quoniam suauis miscetis odores’). As Conington notes, the two shrubs are constantly mentioned together. But cf. Putnam, M. C. J., Virgil's Pastoral Art (Princeton, 1970), p. 106Google Scholar, for a contrary view.

22. Cf. Fraenkel, E., Horace (Oxford, 1957), p. 282Google Scholar, and references there given: Snell, B., Die Entdeckung des Geistes ³ (Hamburg, 1955), p. 397Google Scholar; and Hölscher, T., Griechiscbe Historienbilder des 5. und 4. Jahrbunderts v. Chr. (Würzburg, 1973), p. 71, n. 332.Google Scholar

23. Aen. 6. 27; cf. Enk, P. J.. Mn, Ser. IV, 11 (1958), 322–30Google Scholar, and Knight, W. F. J., Vergil: Epic and Anthropology, (London, 1967), pp. 188 ff.Google Scholar

24. Knight, , Roman Vergil ² (Harmondsworth, 1966), pp. 206 ff.Google Scholar These views were developed with a wealth of uncritical learning by Cruttwell, R. W. in Vergil's Mind at Work (Oxford, 1946).Google Scholar

25. Putnam, , The Poetry of the Aeneid (Cambridge, Mass., 1965)—contrast pp. 40–1Google Scholar with 20, 27, and 67; cf. also Knox, B. M. W., AJPb 71 (1950), 380 and 395Google Scholar. See Schlunk, R. R., The Homeric Scholia and the Aeneid (Ann Arbor, 1974), pp. 38 ff.Google Scholar, for an attempt to show that the important part played by snakes in Aeneid 2 is a result of Virgil's knowledge of the view that was held by certain critics (cf. Schol. bT ad Iliad. 2. 316 (Erbse)) to the effect that the snake in the omen of Iliad 2. 303 ff., symbolized by its winding method of progression and by coiling round its prey the indirect route (Troy-Tenedos-Troy) by which the Greeks travelled to capture Troy. He thinks Virgil followed the scholiasts' sources in taking this symbol to signify ‘the eventual sack of Troy, and especially … the devious means by which the Achaeans contrived to capture it’, and made it a dominant theme in Aeneid 2 for this reason. I do not find his argument convincing, but cannot discuss the problem here.

26. West, D.. JRS 59 (1969), 42.Google ScholarWilliams, R. D. (ad loc. in his edition of Aeneid I–VI (London, 1972))Google Scholar, is more cautious: ‘Perhaps too we may connect the renewed snake (novus) with the renewal of Achilles in Neoptolemus ….’

27. How much Virgil's simile in the Georgics owes to reminiscence of Nicander is explained by Cazzaniga, I., SIFC 32 (1960), 26–8, and by 1Google Scholar. Gualandri, , Acme 23 (1970), 149–51.Google Scholar

28. Aen. 4. 579–80. Cf. Bradley, D. R., CPh 53 (1958), 234–6Google Scholar, and Quinn, K., Virgil's Aeneid (London, 1968), p. 57, n. 1.Google Scholar

29. ‘They turn the prows of their ships to face the sea (they must be ready for departure—who knows what may happen in this strange land?)’, Quinn, , op. cit., p. 161.Google Scholar

30. There is a full discussion by E. Norden on Aen. 6. 136. For more recent views, cf.Brooks, R. A., AJPh 74 (1953), 260–80.Google Scholar

31. Cf. Knauer, G. N., Die Aeneis und Homer (Göttingen, 1964), pp. 354 ff.Google Scholar

32. Aen. 1. 648 ff. Cf. Austin on 650: ‘ornatus … Helenae. an ominous association: Virgil makes Aeneas seem extraordinarily insensitive, and the sinister character of the gift is further underlined in inconcessos hymenaeos (651).’

33. Cf. n. 9 supra, and R. G. Austin's edition of Aeneid J (Oxford, 1971), pp. ix ff., and references there given.

34. Cf. n. 9 supra, and also e.g. Pease, A. S., op. cit., pp. 23 ff and 47 ff.Google Scholar; Glover, T. R., Virgil ⁷ (London, 1942), p. 166Google Scholar; and Binder, G., Aeneas und Augustus (Meisenheim am Glan, 1971).Google Scholar

35. A useful list may be found in Knight, G & R 13 (1944), 10–14.

36. Aen. 2. 557; cf. Servius ad loc: ‘Pompei tangit historiam. Quod autem dicit “litore”, illud, ut supra (506) diximus, respicit quod in Pacuuii tragoedia continetur …’, and ad Aen. 2. 506: ‘alii dicunt quod a Pyrrho in domo quidem sua captus est, sed ad tumulum Achillis tractus occisusque est iuxta Sigeum promunturium …’

37. In his note on Aen. 2. 557; he gives no reason why Virgil should have alluded to the fate of Pompey.

38. Cf. n. 36 supra, and the discussions by van der Kolf, M. C., Pauly-Wissowa, RE xxii (1954), 1888 ff.Google Scholar, and by Stabryla, S., Latin Tragedy in Virgil's poetry (Wroclaw, 1970), pp. 46–8Google Scholar (Polska Akad. N. - Odd. w Krakowie: Prace Kom. Filol. Kias., Nr. 10).

39. A condensed version of an earlier draft of this paper was read as a discussion paper at the Triennial Joint Meeting of the Classical Societies at Oxford on 26 July 1975, and I am grateful for the criticism offered by a number of scholars on that occasion. They will, I trust, appreciate that it has not been practicable to take account of their views here. The preparation of the paper was greatly facilitated by the generous hospitality of the Institute of Classical Studies of the University of London which I enjoyed during May and June of 1975, and by the opportunity of discussing aspects of my views with scholars I met there (especially with Dr. N. Horsfall and Dr. A. Johnston).