¹ See Chr. G. Heyne and G. P. E. Wagner (Leipzig and London, 1832⁴ [= Hildesheim, 1968]), ad loc, vol. II, pp. 693f. According to Heyne, the facta impia are Dido's, but they would refer to her plan to kill Aeneas, as she says at 4.592–4: ‘exhorruerat ilia modo ac damnaverat v. 595 atrocitatem consilii de persequendo et perdendo hospite; silet, et mox: quasi vero nunc demum nefas ac scelus exhorrescendum mihi sit! turn debebam exhorrescere, cum, defuncti mariti memoria animo expulsa, in thalami ac regni societatem perfidum hospitem recipiebam’. This is clearly impossible. Another explanation which Heyne takes into account is to refer nunc tefacta impia tangunt to Aeneas’ perfidy, and turn decuit, cum sceptra dabas ‘ad ipsius Didonis fidem marito datam et nunc laesam’. It is a very unlikely combination, pace C. Buscaroli, Il libro di Didone (Rome/Milan, 1932), p. 414.

² Cf. e.g. T. Ladewig, C. Schaper, and P. Deuticke (Dublin and Zürich, 1912¹³), ad loc: ‘Freilich hatte sich Dido damals noch nicht über Aeneas zu beklagen, aber er gehört zum volke des Laomedon (s. 542) und Dido hatte darum Anlass genug zur Vorsicht und zum Mistrauen.’

³ Gruen, P., ‘Facta impia and Dido's soliloquy (Aeneid 4.590–629)’, CB 56 (1980), 65–9Google Scholar, see p. 67 (emphasis in original).

⁴ Monti, R. C., The Dido Episode and the Aeneid (Leiden, 1981), pp. 62–8Google Scholar, see esp. p. 64. Facta impia are correctly referred to Aeneas also by Farron, S., ‘Plus Aeneas in Aeneid 4.393–6’, in C., Deroux (ed.), Studies in Latin Literature and History, VI (Bruxelles, 1992), pp. 260–76Google Scholar, at 271; and A. Traina, Enc. Virg. S.V. ‘pietas’ (excellent), p. 94.

⁵ Cf. Pease, ad loc.: ‘aiunt: half-incredulous of what she has heard and making no allusion to Aeneas’ having told her this (2.717–23; 2.747–8)’ Williams, ad loc.: ‘the sarcastic use of the vague aiunt to introduce the traditional qualities of Aeneas (guardian of his country's gods and rescuer of his father) gives a cold sense of distance as she ignores the fact that he had told her things himself (2.707, 717)’

⁶ The first reference to Aeneas’ betrayal is to be found in Menecrates of Xanthos (FGrHist 769 F 3 = Dion. Hal. 1.48.3), who explicitly maintains that Aeneas, because of Alexander's feelings of hostility towards him, handed Troy over to the Greeks. For this reason the Greeks allowed Aeneas to save his family, and he became one of them. This hostile account of Aeneas’ escape, which is based on Iliadic references to the hostility between Aeneas and the Priamidae (13.461, 20.178–86), is elaborated as anti-Roman propaganda at the time of the conflict between Rome and Pyrrhus (Gabba, E., ’Sulla valorizzazione politica della leggenda delle origini troiane di Roma fra III e II secolo a.c.’, in M., Sordi (ed.), I canali della propaganda nel mondo antico [Milan, 1976], pp. 84–101Google Scholar, at 92 = Aspetti culturali dell‘ imperialismo romano [Florence, 1993], pp. 101f, persuasively dates Menecrates between the third and second century B.c), i.e. in a moment when Roman propaganda was trying to emphasize and put to political use the mythological foundation of Rome by Aeneas. Pro-Roman writers will react to this account by displacing the charges of treason on Antnor (cf. Braccesi, L., La leggenda di Antenore da Troia a Padova [Padua, 1984], pp. 123–46). Sisenna (fr. 1 E) and Alexander of Ephesus (ap. Origo gentis Romanae 9.2) contrast Antenor's betrayal with Aeneas’ pietas. A Lutatius (Quintus Lutatius Catulus according to Gabba, pp. 93f. = 102f; Lutatius Daphnis according to Braccesi, pp. 127f.) also maintains that Aeneas was the traitor: at vero Lutatius non modo Antenorem, sedetiam ipsum Aeneam proditorem patriam fuisse tradit (ap. Origo gentis Romanae 9.2). Horace, Carm. saec. 37–44, rather explicitly denies this tradition:… Per ardentem sinefraude Troiam / castus Aeneaspatriae superstes/liberum munivit iter (41–3; Serv. Aen. 1.242 cites ardentem sinefraude Troium, and adds: nemo… excusat nisi rem plenam suspicionis; on the problems posed by this passage, see recentlyGoogle Scholar Braccesi, L., Grecità difrontiera [Padua, 1994], pp. 147–62Google Scholar; and Lapini, W., RCCM 38 [1996], 156–69).Google Scholar The motif of Aeneas as a traitor will subsequently be found in Dares Phrygius (37–44) and Dictys Cretensis (4.4,4.22, 5.12, 5.16E). On the legend of Aeneas‘ escape from Troy, and especially on his betrayal, there are several works; in addition to Gabba and Braccesi, quoted above, see esp. Ussani, V. jr , ‘Enea traditore’,SIFC 22 (1947), 108–23 = Id.Google Scholar, Memoria Classica (Rome, 1996), pp. 153–67; Galinsky, G. K., Aeneas, Sicily and Rome (Princeton, 1969), pp. 3–61, esp. 46–51.Google Scholar

⁷ The expression facta impia is used by Catullus (23.10 and esp. 30.4), as well as by Lucretius (1.83: the sacrifice of Iphigenia is an instance of the scelerosa atque impia facta caused by religion). It is intriguing to see in Aeneas’ acta impia an echo of the scelerosa atque impia facta which at Lucr. 1.83 mark the beginning of the Trojan war: the war begins with the impia facta of the Greek leader Agamemnon, and ends with the facta impia of the Trojan traitor, Aeneas.

⁸ All these accounts are already too embarrassing for Virgil, or, better still, for Aeneas: in Aen. 2 the Greeks do not let Aeneas leave thanks to his pietas, or—as we read in Livy 1.1.1–3—thanks to an old relationship of hospitality and because he and Antenor had argued that Helen should be returned to the Greeks. InAen. 2 Aeneas leaves against his will, and only because of divine orders.

⁹ Austin (Oxford, 1971), ad loc, p. 172.

¹⁰ Cf. Horsfall, N., ‘Dido in the light of history’, PVS 13 (1973–4), 1–13Google Scholar, at 4 = S.J., Harrison (ed.), Oxford Readings in Vergil's Aeneid (Oxford, 1990), pp. 127–44Google Scholar, at 132.

¹¹ Serv. ad loc.:‘it nigrum campis agmen’…hemistichium Ennii de elephantis dictum [Enn. Ann 502 Sk.], Quo ante Accius usus est de Indis. See O. Skutsch (Oxford, 1985), ad loc.: ‘Elephants were used by Pyrrhus, Hannibal and Antiochus, and the placing of the fragment is therefore uncertain. Hannibal’s approach to Italy through the Plains of Gaul or his movements in the Po valley before he lost many of the beasts in crossing the Apennines seem likely settings.’ I think that if we reflect on the context of the Ennian quotation in Vergil, we may add further evidence to the hypothesis that Ennius’ elephants belong to Hannibal. Any reader cannot help noticing the hurnour of the transference of the same phrase from enormous elephants to small ants. In Vergil's text the ants stand for the Trojan army fleeing from Troy, an action fated to provoke Dido's anger and her subsequent curse, in which she will unequivocally refer to Hannibal's revenge: exoriare aliquis nostris ex ossibus ultor, l quiface Dardanios ferroque sequare colonos (4.625f.). I think that the ‘hurnour’ of the transference would be even more effective (and much more bitter) if Ennian elephants were exactly Hannibal's elephants.

¹² See Lyne, R. O. A. M., Words and the Poet (Oxford, 1989), pp. 63–99.Google Scholar

¹³ Conington, ad loc., p. 291.

¹⁴ See Grant, J. N., ‘Dido Melissa’, Phoenix 23 (1969), 380–91CrossRefGoogle Scholar, esp. 384–6; and Putnam, M. C. J., Virgil's Aeneid. Interpretation and Influence (Chapel Hill and London, 1995), pp. 66–8.Google Scholar The imagery of the sacking army looks ahead to the Trojans’ descendants: cf. Nisbet, R. G. M., ‘Aeneas Imperator: Roman generalship in an epic context’ PVS 18 (1978–80), 50–61Google Scholar, at 53 = Harrison (n. 10), pp. 378–89, at 382: ‘There could be no better description of the legalized destructiveness of the Roman army with its requisitioning of corn (frumentum imperatum), organized supply-columns, and discipline on the march’. Interesting remarks on the simile at 4.402–7 and its gloomy tone are also in Briggs, W.W., Narrative and Similes from the Georgics in the Aeneid (Leiden, 1980), pp. 53–5.Google Scholar

¹⁵ Lyne, R. O. A. M., Further Voices in Vergil's Aeneid (Oxford, 1987), p. 210.Google Scholar The right interpretation of the paintings in Juno's temple is given by Horsfall (n. 10), pp. 7f. = 136f.; see also, for a different perspective, Fowler, D. P., ‘Narrate and describe: the problem of ekphrasis’, JRS 81 (1991), 25–35, at 31–3Google Scholar; Barchiesi, A., ‘Rappresentazioni del dolore e interpretazione nell; Eneide’, AuA 40 (1994), 109–24, at 114–24.Google Scholar

¹⁶ Conington, ad loc., p. 57

¹⁷ References to Aeneas’ betrayal in Virgilian scholia can be found esp. in Serv. Aen. 1.242, 1.488, 1.647,2.17,2.35,2.298. An allusion to the betrayal at 1.488 is detected by Ahl, F., ‘Homer, Vergil, and complex narrative structures in Latin epic: an essay’, ICS 14 (1989), 1–31, at 26–9Google Scholar (curiously, Ahl does not mention Servius). In nunc te facta impia tangunt I cannot help the temptation to hear an echo of 1.462 sunt lacrimae rerum et mentem mortalia tangunt (only in these two places does tangunt appear in all of Virgil): Aeneas was wrong, then, when he believed that the Carthaginians were compassionately ‘touched’ by the fate of Troy; it is only Dido who will be ‘touched’ by them compassionately, but she will also be forced to change her mind and will align herself with the painters of the temple decoration, who were ‘touched’ by Aeneas’ plight only in a ‘negative’ sense.

¹⁸ We might even doubt that Aeneas actually misinteprets the meaning of the temple decoration. In his account in Book 2 he will take pains, however implicitly, to explain to Dido the Possible origin of the rumour of a ‘Greek Aeneas’. In the fateful last night of Troy he and some fellow soldiers happened to put on the armours of some slaughtered Greek soldiers (2.387–401), and they subsequently found themselves ‘mixed up’ with the Greeks, immixti Danais (2.396). See the interesting remarks by Ahl (n. 17), pp. 26–9.

¹⁹ Turnus appears to be aware of this account, since he calls Aeneas desertorem Asiae (12.15), ‘a reference to the legend that Aeneas had acted in a cowardly fashion in leaving Troy’ (Williams, ad loc.).

²⁰ Even in this account of the story there are details that should have warned Dido on how Aeneas dealt with the women who were an ‘obstacle’ to the accomplishment of his mission (or, from another point of view, of his interests). Ovid's Dido picks up Aen 4.597–9 and focuses on the fate of Creusa, whom Aeneas left behind (cf. Her. 7.79–86): his behaviour can thus be considered one of the facta impia. Moreover, Dido's inclusive remark at Her. 7.81 (omnia mentiris) could also conceal a reference to the more serious charge of Proditio Troiae. Ovid alludes to alternative accounts about Aeneas departure from Troy and to his betrayal at Met. 13.623–7: cf. Casali, S., ‘Altre voci nell’ Eneide di Ovidio’, MD 35 (1995), 59–76, at 59–62.Google Scholar The echo of Aen. 2.790f. lacrimantem et multa uolentem / dicere deseruit (cf. Georg. 4.500f: an intertext which is very mournful, but surely also very ‘effective’ from Aeneas’ point of view, because it establishes a connection between the moving story of Orpheus and Eurydice on the one hand, and Aeneas and Creusa on the other) at 4.390f. multa metu cunctantem et multa uolentem / dicere seems to point out the connection between Creusa's and Dido's fate. That deseruit is referred to Creusa by Aeneas as he talks to Dido can only sound in retrospect as bitterly ironic to the deserted queen of Carthage.

²¹ Virgil achieves a similar effect when he makes Juno refer to Aeneas as the aggressor at 10.66 hostem regi se inferre Latino and 10.78 arua aliena iugo premere atque auertere praedas: ‘Aeneas is guilty of neither of these charges in the Aeneid, having been granted land and gifts by Latinus (cf. 7.259ff.), though in other accounts of Aeneas’ arrival in Italy the Trojans do seize land and plunder Latium’ (S. J. Harrison [Oxford, 1991], ad loc.; cf. Cato, Orig. 1 fr. 10 Jordan, Livy 1.1.5, Dion. Hal. 1.57.1). Also in this case the reference to a covered up alternative tradition prompts the reader to reflect on the tendentious process of ideological selection inherent in a propagandist poem. Juno also refers to a formal engagement between Turnus and Lavinia at 10.79 gremiis abducere pactas (cf. Amata at 7.363–66, Allecto at 7.423–24), a version coincident with that of Livy 1.2.1, but apparently untrue in the Aeneid: ‘pactas is tendentious… note the careful language of the poet at 7.54–8’ (Harrison ad loc). Nevertheless, the language of Virgil at 7.54–8 is carefully elusive, and suggests that the propagandist poet may be no less tendentious than the envious goddess.

²² Homeric specialists are less chary than their Virgilian counterparts, and do not hesitate to question the truth of Odysseus‘ account: cf. recently Richardson, S., ‘Truth in the tales of the Odyssey’, Mnemosyne 49 (1996), 393–402CrossRefGoogle Scholar (with bibliography: esp. p. 396, n. 5): ‘We must accept that, because Homer does not himself cover this period in Odysseus’ life, we cannot know for certain what is fictional fact and what is the hero's fabrication’ (p. 396). The insinuation that Odysseus' tale is a fabrication is already known in ancient times: cf. e.g. Dio Chrys. 11.34; Lycophr. Alex. 764; Lucian. V.H. 1.3; Juv. 15.13–26 (with E. Courtney [London, 1980] on 15.16). It might be useful to recall that Aeneas’ speech at Dido's court echoes that of a well-known liar to the Phaeacians.

²³ It is perhaps worth noticing a fact that, in the light of the interpretation I have outlined, might be significant: these two lines, if we consider Aeneas’ own narrative in Aen. 2, sound at least exaggerated: ‘… Aeneas declares that he carried his father out of Troy “per … mille sequentiatela” (6.110), whereas the narrative does not show that a single shot was aimed at them (2.721–44). Henselmanns, V. (Die Widersprüche in Vergils Aeneis [Aschaffenburg, 1913])Google Scholar considers this to be an insoluble contradiction (p. 127); yet Aeneas is trying to persuade the Sibyl and therefore stresses his love for his father’ (Highet, G., The Speeches in Vergil's Aeneid [Princeton, 1972], p. 289).Google Scholar