1. E.g., Bandera, C., ‘Sacrificial Levels in Virgil’s Aeneid ’, Arethusa 14 (1981), 217–39Google Scholar; Farron, S., ‘Aeneas’ Human Sacrifice’, AClass 28 (1985), 21–33Google Scholar; Nicoll, W.S.M., ‘The Sacrifice of Palinurus’, CQ 38 (1988), 459–72CrossRefGoogle Scholar; O’Hara, J.J., Death and the Optimistic Prophecy in Vergil’s Aeneid (Princeton 1990), 19–35Google Scholar, 106–10; Hardie, P.R., The Epic Successors of Virgil: A Study in the Dynamics of a Tradition (Cambridge 1993), 19–23Google Scholar, 25–29, 32–36, 40–42, 51–53; Smith, R.M., ‘Deception and Sacrifice in Aen. 2.1–249’, AJP 120 (1999), 502–23Google Scholar.

2. Hardie (n.l above), 19–56.

3. So, e.g., Liebeschuetz, J.H.W.G., Continuity and Change in Roman Religion (Oxford 1979), 173 Google Scholar: ‘Compared with Virgil, or Livy, Silius Italicus seems to be less interested in the rich variety of traditional religious and political institutions of Rome.’ Similarly, Feeney, D.C., The Gods in Epic: Poets and Critics of the Classical Tradition (Oxford 1991), 311f.Google Scholar, who notes ‘Silius’ failure to evoke the potency of Rome’s antique cult and ritual’, and adds that ‘even in comparison with Livy, let alone Vergil, it is striking how feeble an impression one derives from reading the Punica of any grandeur, power, or vitality in connection with religious matters.’ He would later revise this opinion, however: ‘I should have seen that he had understood very clearly Virgil’s procedure, and wanted to create a Virgilian, rather than a Livian mode of representing ritual, even if it was historically Roman ritual’ (Feeney, D.C., Literature and Religion in Rome: Cultures, Contexts, and Beliefs [Cambridge 1998], 142 Google Scholar n.20).

4. For Virgil, see Johnson, W.R., Darkness Visible: A Study of Vergil’s Aeneid (Berkeley 1976), 117–19CrossRefGoogle Scholar; O’Hara (n.l above), 82–85; Hardie (n.1 above), 28f.; Leigh, M., ‘Hopelessly Devoted to You: Traces of the Decii in Virgil’s Aeneid ’, PVS 21 (1993), 89–110Google Scholar. Also, see Pascal, C.B., ‘The Dubious Devotion of Turnus’, TAPA 120 (1990), 251–68Google Scholar, at 251 n.3, where he cites earlier literature, though he denies Virgil’s use of the deuotio. For Lucan, see Hardie (n.l above), 30f., 53f; Leigh, M., Lucan: Spectacle and Engagement (Oxford 1997), 118 Google Scholar n.4 (for earlier literature) and 128–43. For Statius, see Vessey, D.W.T.C., ‘Menoeceus in the Thebaid of Statius’, CP 66 (1971), 236–43Google Scholar, at 239f.; Heinrich, A., ‘ Longa retro series: Sacrifice and Repetition in Statius’ Menoeceus Episode’, Arethusa 32 (1999), 165–95CrossRefGoogle Scholar, at 180–90.

5. Cf. Virg. Aen. 1.712, 11.442, 12.234; Luc. BC 2.308, 3.311, 4.272, 533, 695, 8.91, 112, 10.176; Stat. Theb. 3.233, 10.794, 11.371.

6. Heinrich (n.4 above), 181. The Romans commonly used the word deuotio to refer to the practice Heinrich describes here, which they identified closely with the example of the Decii Mures. Versnel, H.S., ‘Two Types of Roman Devotio ’, Mnemosyne 29 (1976), 365–410CrossRefGoogle Scholar, calls this deuotio a deuotio ducis to distinguish it from another kind of deuotio, which he calls a deuotio hostium, in which ‘the object of the vow…is not the general himself, but the enemy territory with all movable and immovable property, including the civil and military population’ (366). In this paper I am only concerned with the former type, i.e. the deuotio ducis.

7. My text is Delz, J. (ed.), Silii Italici Punica (Stuttgart 1987 Google Scholar).

8. For this idea, see Hommel, H., ‘Per aspera ad astro ’, WJA 4 (1949/1950), 157–65Google Scholar, and for Silius’ use of it here in the Punica, see von Albrecht, M., Silius Italicus: Freiheit und Gebundenheit rbmischer Epik (Amsterdam 1964), 17f.Google Scholar; Niemann, K.-H., Die Darstellung der römischen Niederlagen in den Punica des Silius Italicus (diss. Bonn 1975), 26f.Google Scholar; Feeney (n.3 above), 305f.; Ripoll, F., La morale héroïque dans les épopées latines d#x2019;époque flavienne: tradition et innovation (Louvain and Paris 1998), 324f Google Scholar. Also, cf. Pun. 4.603: explorant aduersa uiros (‘adversity tests men’).

9. It should be noted, however, that, unlike Paulus and Marcellus, Fabius does not die in battle or as a result of wounds sustained in battle and that Silius does not even mention his death in the epic. For other ways of accounting for why Paulus, Fabius and Marcellus are separated from Scipio in the prophecy, see Lorenz, G., Vergleichende Interpretationen zu Silius Italicus und Statius (diss. Kiel 1968), 34 Google Scholar; Schubert, W., Jupiter in den Epen der Flavierzeit (Frankfurt am Main 1984), 62f.Google Scholar; Ahl, F., Davis, M.A. and Pomeroy, A., ‘Silius Italicus’, ANRW 232A (1986), 2492–2561Google Scholar, at 2504; Marks, R., From Republic to Empire: Scipio Africanus in the Punica of Silius Italicus (Frankfurt am Main 2005), 96–99Google Scholar, 215–17.

10. I should make it clear that when Silius attributes the foundation of Roman regnum to the sufferings of Paulus, Fabius and Marcellus, he is not suggesting that the victories of Scipio, though separated from them in the prophecy, as we have seen, have no role to play in shaping Rome’s future, including her empire. The important distinction here is that Paulus, Fabius and Marcellus are credited with establishing Rome’s empire in the wider sense of the term, namely, her world empire, whereas Scipio, while building on that foundation, differently, contributes to the emergence of her empire in a narrower, political sense, namely, the Principate or, more specifically, the Principate of the Flavians. The ways in which Silius’ Scipio paves the way for Rome’s Principate are extensively discussed by Marks (n.9 above), who also shows how Scipio’s place in Roman history is so defined through Jupiter’s prophecy here in Book 3 (Marks [n.9 above], 211–17).

11. Niemann (n.8 above), 26f. also reads 5.674–76 in connection with 3.588 as well as with 3.584f. (iamque tibi ueniet tempus, quo maxima rerum/nobilior sit Roma malis, ‘and there will soon come a time for you when Rome, more noble because of her misfortunes, will be the head of an empire’), but, differently, focuses on how these passages give shape to the notion that Rome’s defeats and wounds in the war will win her fame; so also Czypicka, T., ‘Funzionalità del dialogo tra Venere e Giove nel libro III delle Puniche di Silio Italico’, Eos 75 (1987), 87–93Google Scholar, at 89; Fucecchi, M., ‘Empieta e titanismo nella rappresentazione siliana di Annibale’, Orpheus 11 (1990), 21–42Google Scholar, at 25 n.9. For the connection between uulnera and fame, cf. also 9.350f. (adora uulnera laudes/perpetuas paritura tibi, ‘adore wounds that will win you everlasting praise’) with Niemann (n.8 above), 28. Note that as Hannibal admires the courage and fortitude of the Romans slain on the battlefield of Trasimene (5.669–73), he points out their uulnera: quae uulnera cernis,/quas mortes! (‘look, what wounds, what deaths!’, 5.669f.).

12. On compensatory exchange and reciprocity with regard to sacrifice, see van der Leeuw, G., ‘Die Do-ut-des-Formel in der Opfertheorie’, ARW 20 (1920/1921), 241–53Google Scholar; Hubert, H. and Mauss, M., Sacrifice: Its Nature and Function, tr. W.D. Halls (London 1964 Google Scholar); Mauss, M., The Gift: Forms and Functions of Exchange in Archaic Societies, tr. I. Cunnison (London 1969), 12–15Google Scholar; van Baal, J., ‘Offering, Sacrifice and Gift’, Numen 23 (1976), 161–78CrossRefGoogle Scholar; Versnel, H.S., ‘Self-sacrifice, Compensation and the Anonymous Gods’, in J. Rudhardt and O. Reverdin (eds.), Le Sacrifice dans l’antiquité (Geneva 1981), 135–85Google Scholar, at 163–85; Grottanelli, C., ‘Do ut des?’, Scienze dell’antichità: storia, archeologia, antropologia 3–4 (1989/1990), 45–54Google Scholar; Burkert, W., Creation of the Sacred: Tracks of Biology in Early Religions (Cambridge 1996), 129–55Google Scholar, esp. 149–52.

13. On substitution in sacrifice, see Girard, R., Violence and the Sacred, tr. P. Gregory (Baltimore 1977), 3–10Google Scholar, 68–88, 250–73; Bandera (n.1 above), 223–26; Versnel (n.12 above), 159f., 163–69; Hardie (n.1 above), 3–10, 27–35,49–56.

14. That a crisis, posing a threat to the stability and order of a community, is resolved through sacrifice is consistent with Girard’s understanding of ‘sacrificial crisis’, but for him the term entails much more than that, including the escalation of violence within the community, the failure of sacrificial institutions to check it, the effacement of differences between members of the community, among others; for more, see Girard (n.l3 above), 39–67. While none of these aspects is evident in the crisis that Jupiter describes in his speech, the god’s solution, the war, does bring about such a Girardian crisis insofar as it precipitates violence between Rome and Carthage in such a way that differences between the two are effaced, as we shall see on p. 133 below. For the notion that violence effaces differences, see Girard (n.13 above), 47–66, 70–79; Bandera (n.1 above), 230–34; Hardie (n.1 above), 22–26. For words of caution on applying Girard to reading Roman epic, however, see Hardie (n.l above), 21 n.5 and Smith (n.1 above), 507f. For a less cumbersome definition of ‘sacrificial crisis’, see Versnel (n.12 above), 143, which I quote on p. 135 below.

15. Silius lends programmatic force to Catus’ death not only by making it the first for Rome (cf. Livy 21.47.1), but by emphasising the primacy of the Ticinus battle itself. The tossing of the spear toward the enemy underlines this point as it is an act that often constituted an informal declaration of war and one which the Romans formalised into a ritual performed by the fetiales; see Bayet, J., ‘Le rite de fécial et le cornouiller magique’, MEFRA 52 (1935), 29–76CrossRefGoogle Scholar, at 51–53, and Watson, A., International Law in Archaic Rome: War and Religion (Baltimore 1993), 1–9Google Scholar, 20–30. Spaltenstein, F., Commentaire des Punica de Silius Italicus, livres 1 à 8 (Geneva 1986 Google Scholar), ad 4.134, points out that the siege of Saguntum similarly begins with Hannibal throwing a spear that strikes and kills the Saguntine Caicus (1.305); an echo of that event would be apt here since Catus’ death marks the beginning of Rome’s self-destructive phase in the war, just as the death of Caicus marks the beginning of Saguntum’s conflict with Hannibal, which ends with the city’s own self-destruction, namely, the internecine slaughter and suicides of her citizens (2.526–680).

16. The self-destructive tendencies of some of Rome’s leaders in Books 4–10 have been recognised by others (e.g., Ahl, Davis and Pomeroy [n.9 above], 2520–23; Marks [n.9 above], 16–20), but a thorough examination of the extent to which Silius departs from the historical record to convey this impression has not been made. This is necessary, however, in order to establish that Silius’ emphasis on self-destruction is his own.

17. For the rescue, see Polyb. 10.3.4–6, Livy 21.46.7, Val. Max. 5.4.2, Sen. Ben. 3.33.1, and Plin. NH 16.14.

18. So Venini, P., ‘Tecnica allusiva in Silio Italico’, RIL 106 (1972), 532–42Google Scholar, at 536; Niemann (n.8 above), 38; Spaltenstein (n.15 above), ad 4.515. Also, Spaltenstein (n.15 above), ad 4.510, notes that Silius fails to mention the success the Romans had against the Carthaginians immediately before the Trebia battle (Livy 21.52.9–11).

19. According to Appian (Hann. 7) and Nepos (Hann. 4.2) Scipio took part in this battle, but Polybius (3.70.10) and Livy (21.53.9) suggest that he did not because of the wound he received at the Ticinus. Livy, though, implies elsewhere (21.15.4, 57.2) that Scipio fought in the battle; on this, see Nicol, J., The Historical and Geographical Sources Used by Silius Italicus (Oxford 1936), 30f.Google Scholar; Niemann (n.8 above), 40, 94; Nesselrath, H.-G., ‘Zu den Quellen des Silius Italicus’, Hermes 114 (1986), 203–30Google Scholar, at 21 If.; Spaltenstein (n.15 above), ad 4.622. Historical details such as the flooding of the Trebia (Livy 21.54.9) and rainy conditions during the battle (Livy 21.56.3, 56.6) may have prompted Silius to invent Scipio’s battle with the river god (so Nicol, 30f.; Spaltenstein [n.15 above], ad 4.638), but the episode is principally inspired by Achilles’ river-battle in Il. 21.211–382: so Nicol, 30f.; von Albrecht (n.8 above), 148f.; Romano, D., Poesia e Scienza. Silio Italico: uomo, poeta, artista, attraverso una moderna interpretazionefilologica e psicologica (Naples 1969), 75f.Google Scholar; Juhnke, H., Homerisches in römischer Epik flavischer Zeit: Untersuchungen zu Szenennachbildungen und Strukturentsprechungen in Statius’ Thebais und Achilleis und in Silius’ Punica (Munich 1972), 13–24Google Scholar, 201; Niemann (n.8 above), 94–99; Spaltenstein (n.15 above), ad 4.638, 649, 662, 666, 667, 671; Feeney (n.3 above), 308–10.

20. Also cf. App. Hann. 38 with Lucarini, CM., ‘Le fonti storiche di Silio Italico’, Athenaeum 92 (2004), 103–26Google Scholar, at 113. The simile comparing Flaminius to an incompetent helmsman in 4.713–17 may be inspired, though, by Polybius, who also compares him to a ship’s pilot (3.81.11); so Nicol (n.19 above), 35; Spaltenstein (n.15 above), ad 4.713. The detail of Flaminius’ being at the head of his army (5.28f.) is only found in Polybius, however; so, Klotz, A., ‘Die Stellung des Silius Italicus unter den Quellen zur Geschichte des zweiten punischen Krieges’, RhM 82 (1933), 1–34Google Scholar, at 31f.; Nicol (n.19 above), 33f.; Nesselrath (n.19 above), 226; cf. Spaltenstein (n.15 above), ad5.29.

21. Unlike Livy, however, Silius makes the will of the gods a point of contention in Flaminius’ deliberations with his advisers and thus foregrounds the consul’s impious arrogance as a contributing factor to his demise. The priest Corvinus, who does not appear in any other source, is the prominent adviser in the episode. The priest advises him to avoid battle not only because it would be better to wait until Servilius’ army arrives (5.97–100)—advice which is also given in Livy (22.3.8–10)—but because the gods have expressed their disfavour through ill omens (5.82–93; cf. 5.102–04), a warning that is not mentioned in Livy even though omens are there reported (22.3.11–13; cf. 22.1.8–13). As for Flaminius’ disregard of the will of the gods, this is referred to by Livy before the consultation (22.3.4f.), but in Silius shows itself when the consul mocks his advisers for their superstitiousness (5.117–20, 125–27).

22. And note that in 5.392f. he rushes at the enemy. Flaminius’ bravery and daring in the battle, though, come across in Livy’s account too (22.5.1, 6.2): Venini (n.18 above), 534 n.4.

23. For Silius’ departures from his sources here, see pp.l39f. below.

24. Minucius is as reckless and hasty in Silius at these points as he is in Livy: cf., e.g., 7.495f. with Livy 22.27 (Spaltenstein [n.15 above], ad 7.495), 7.523f. with Livy 22.28.2–4 (Spaltenstein [n.15 above], ad 7.523), and 7.525–29 with Livy 22.28.9.

25. For this omission, see Klotz (n.20 above), 7; Nicol (n.19 above), 73–76; Nesselrath (n.19 above), 217. While it may be, as Klotz proposes, that Silius is following a source other than Livy—Appian (Harm. 13) also speaks only of the second battle between Minucius and Hannibal—he is, nevertheless, doing so by choice, for he could not have been unaware of Livy’s account.

26. Silius, in fact, identifies Minucius with Flaminius through a simile in which he compares Hannibal, waiting for Minucius to attack, to a fisherman (7.500–03); this recalls the fisherman simile applied to Hannibal when he lay in wait for Flaminius at Trasimene (5.47–52); so Spaltenstein (n.15 above), ad 5.47 and ad 7.500. Fabius, on the other hand, contrasts himself with Flaminius in a speech to his troops (7.229–31; cf. Livy 22.12.5). For more on Fabius contrasted with Flaminius, see Ahl, Davis and Pomeroy (n.9 above), 2520, 2523. For Fabius and Minucius contrasted, see von Albrecht (n.8 above), 70f.

27. So Danesi Marioni, G., ‘Un esempio della tecnica compositiva di Silio Italico’, in Munus amicitiae: Scritti in memoria di Alessandro Ronconi I (Florence 1986), 43–55Google Scholar, at 43–48, and Spaltenstein (n.15 above), ad 8.265.

28. For Varro as a new Flaminius, see 8.218, 309f. (cf. Livy 22.39.6; so Spaltenstein [n.15 above], ad 8.310), 9.52–55, 421 f. Also, note that Hannibal waits for Varro at Cannae (8.350f.) much as he waited for Flaminius at Trasimene (5.1–3, 47–52) and that the dispute between Varro and Paulus before the Cannae battle (9.23–65) forms a pendant to that between Flaminius and Corvinus before the battle at Trasimene (5.77–129); so Niemann (n.8 above), 172. Varro’s comparison to a captain of a ship-wrecked vessel (10.608–12) is perhaps meant to recall Flaminius’ helmsman simile (4.713–17) too; so, Ahl, Davis and Pomeroy (n.9 above), 2529; Matier, K.O., ‘A Role Reversal in Silius (10.605–639)’, Akroterion 38 (1993), 32–38Google Scholar, at 37.

29. So Niemann (n.8 above), 172; Fucecchi, M., ‘La vigilia di Canne nei Punica e un contributo alio studio dei rapporti fra Silio Italico e Lucano’, in P. Esposito and L. Nicastri (eds.), Interpretare Lucano: Miscellanea di Studi (Naples 1999), 305–42Google Scholar, at 324–27.

30. The exchange with Lentulus (10.260–92), in which Paulus, though gravely wounded, refuses to give up the battlefield, has much in common with Livy 22.49.6–13, however; so Venini (n.18 above), 534 n.4 and 536 n.10.

31. For haste as a typically Hannibalic quality in the epic, especially in Books 1–10, see Kissel, W., Das Geschichtsbild des Silius Italicus (Frankfurt am Main 1979), 103–15Google Scholar; Vessey, D.W.T.C., ‘The Dupe of Destiny: Hannibal in Silius, Punica Ill’, CJ 77 (1982), 320–35Google Scholar; Ahl, Davis and Pomeroy (n.9 above), 2511–19; Laudizi, G., Silio Italico: Il passato tra mito e restaurazione etica (Galatina 1989), 96–123Google Scholar; Matier, K.O., ‘Hannibal and the real hero of the Punica ’, AClass 32 (1989), 3–17Google Scholar; Fucecchi, M., ‘Il declino di Annibale nei Punica ’, Maia 42 (1990), 151–66Google Scholar; Marks (n.9 above), 15–31, 46f. Another point of contact between Scipio and Hannibal is Achilles: Scipio’s battle with the Trebia is modelled after Achilles’ with the Scamander (see n.19 above), and Hannibal is compared to Achilles at 7.120–22; also, Hannibal recalls Scipio at the Trebia when he is said to have clogged the Eridanus with slaughter (11.137f.). Ahl, Davis and Pomeroy (n.9 above), 2521, point out that Juno chooses Flaminius to hasten Rome’s destruction (4.708–10) and that this makes him comparable to Hannibal, who is an agent of the goddess too (cf. 4.729–38). Also, compare Flaminius’ and Hannibal’s charges into battle at Trasimene: 5.392f., 607f. In Rome Varro’s restless plea to march to Cannae (mora sola triumpholparuum iter est, ‘a short journey is the only thing delaying triumph’, 8.273f.) echoes Hannibal’s prior impatience to engage Fabius in battle (uincendi sola uidetur,/quod nondum steterint acies, mora, ‘the only thing delaying victory seems to be that the battle-lines have not yet been drawn up’, 7.99f.). The similarity between Varro and Hannibal in 9.1–8 is also noted by Hardie, P.R., ‘Tales of Unity and Division in Imperial Latin Epic’, in J.H. Molyneux (ed.), Literary Responses to Civil Discord (Nottingham 1993), 57–71Google Scholar, at 67.

32. All three instances are noted by Jr.McGuire, D.T., Acts of Silence: Civil War, Tyranny, and Suicide in the Flavian Epics (Hildesheim 1997), 133f Google Scholar. Note, though, that the thought expressed by Paulus in 8.332f. is drawn from Livy 22.39.18; so Spaltenstein (n.15 above), ad 8.332.

33. Scipio’s attempted suicide is not reported in any source; Polybius (10.3.4–6) and Livy (21.46.7) do not even say that he was dismayed by the grave condition of his father while Valerius Maximus (5.4.2) and Seneca (Ben. 3.33.1) go so far as to suggest that he responded heroically and decisively to the crisis. For Silius’ account of Flaminius’ final moments, which differs in some significant respects from the historical record, see p. 139 below. On Varro’s thoughts on suicide, Ahl, Davis and Pomeroy (n.9 above), 2534, point out that he is excussus mentem (9.644) and thus recalls Flaminius, who at Trasimene is excussus…mentem (5.540). The Saguntines, during the slaughter and suicide at the end of Book 2, are said to have excussae mentes (2.592) too. Note also that the death of Satricus’ son, Mancinus, whose armour he dons, was due to his reckless haste; in a skirmish before the battle of Cannae he eagerly rushed out to be the first to face the enemy and was struck down (9.12–14).

34. Livy’s report of two Romans who were held prisoner by the Carthaginians and fled to the Roman camp (22.42.11) may have initially prompted Silius to invent the episode, but scholars agree that it has much in common with stories connected with Rome’s civil wars; for evidence and discussion, see Klotz, A., ‘Silius Italicus’, RE 2. Reihe 5. Halbband (1927), 79–91Google Scholar, at 86; Nicol (n.19 above), 12f.; Romano (n.19 above), 90; Niemann (n.8 above), 174–77; Spaltenstein, F., Commentaire des Punica de Silius Italicus, livres 9 a 17 (Geneva 1990 Google Scholar), ad 9.66; Mezzanotte, A., ‘Echi del mondo contemporaneo in Silio Italico’, RIL 129 (1995), 357–88Google Scholar, at 362f.; McGuire (n.32 above), 134f.; Fucecchi (n.29 above), 315–22. Hardie (n.31 above), 68, also points out that the distinction between Roman and Carthaginian is blurred in the episode when Solymus mistakes Satricus for an enemy, a confusion that the father puts as follows: iaceres in me cum feruidus has-tam/Poenus eram (‘when you threw in anger your spear at me, I was Carthaginian’, 9.129f.).

35. The omens that precede the battle of Cannae (8.622–55) recall in several respects those that precede the battle of Pharsalus (BC 7.151–213): Niemann (n.8 above), 170; Fucecchi (n.29 above), 323f.; cf. also Livy 22.36.7–9 with Niemann (n.8 above), 165–69. Silius’ prophetic soldier (8.656–76) is modelled after Lucan’s prophetic bacchant (BC 1.673–95): Mendell, C., ‘Silius the Reactionary’, PhQ 3 (1924), 92–106Google Scholar, at 105; Schönberger, O., ‘Zum Weltbild der drei Epikern nach Lucan’, Helikon 5 (1965), 123–45Google Scholar, at 137; Niemann (n.8 above), 169f.; Fucecchi (n.29 above), 324; cf. Spaltenstein (n.15 above), ad 8.656. The dispute between Paulus and Varro (9.23–65) echoes that between Cicero and Pompey before Pharsalus (BC 7.62–127): Fucecchi (n.29 above), 327–29 (see also n.28 above). The gigantomachy simile at Cannae (9.304–09) is inspired by Lucan’s at Pharsalus (BC 7.143–50): von Albrecht (n.8 above), 153; Niemann (n.8 above), 189 n.2. The death of Paulus in Book 10 also occasions several allusions to Lucan’s Pompey; for details, see Marpicati, P., ‘Silio “delatore” di Pompeo (Pun. 5,328 ss.; 10,305 ss.)’, MD 43 (1999), 191–202Google Scholar. For more on Lucan’s influence on the Punica, see, among others, Häussler, R., Das historische Epos von Lucan bis Silius und seine Theorie: Studien zum historischen Epos der Antike (Heidelberg 1978), 161–77Google Scholar; Brouwers, J.H., ‘Zur Lucan-Imitation bei Silius Italicus’, in J. den Boeft and A.H.M. Kessels (eds.), Actus: Studies in Honour of H.L.W. Nelson (Utrecht 1982), 73–87Google Scholar; Ahl, Davis and Pomeroy (n.9 above), 2501–04; Fucecchi (n.31 above), 157–66; McGuire (n.32 above), 82–85, 211–19; Marks (n.9 above), 273–76.

36. Jr.McGuire, D.T., ‘History Compressed: The Roman Names of Silius’ Cannae Episode’, Latomus 54 (1995), 110–18Google Scholar; Mezzanotte (n.34 above), 383–85; McGuire (n.32 above), 61–63, 85, 136–44. There are, moreover, signs of a developing rift between the people and senate. When the people grow restless with Fabius’ tactics and support a more aggressive approach, Silius says that Juno incited the senate to split Fabius’ command with Minucius (7.504–16), a detail that suggests that the senate was initially at odds with the people on the matter. Varro, then, appears to exploit this rift when he sets the people against the senate and gains election to the consulship (8.246–57); he even promises to put an end to the rule of the senate (8.274f.). On the historical basis for this division between people and senate, see Danesi Marioni (n.27 above), 46; Fucecchi (n.29 above), 325 n.49; Pomeroy, A.J., ‘Silius’ Rome: The Rewriting of Vergil’s Vision’, Ramus 29 (2000), 149–68CrossRefGoogle Scholar, at 162. For Varro’s election and its possible relation to future civil strife, see McGuire (n.32 above), 126f.; Dominik, W., ‘Hannibal at the Gates: Programmatising Rome and Romanitas in Silius Italicus’ Punica 1 and 2’, in A.J. Boyle and W.J. Dominik (eds.), Flavian Rome: Culture, Image, Text (Leiden 2003), 469–97Google Scholar, at 492f.

37. Versnel (n.12 above). For more on the practice, see Versnel (n.6 above); Janssen, L.F., ‘Some Unexplored Aspects of Devotio Deciana ’, Mnemosyne 24 (1981), 357–81CrossRefGoogle Scholar; Barton, A.C., The Sorrows of the Ancient Romans: The Gladiator and the Monster (Princeton 1993), 40–46Google Scholar; Feldherr, A., Spectacle and Society in Livy’s History (Berkeley and Los Angeles 1998), 85–92Google Scholar.

38. Versnel (n.12 above), 143–45.

39. Versnel (n.12 above), 159f. See also the discussions on substitution cited in n.13 above. With specific reference to the Punica, see Kissel (n.31 above), 62–67, and Hardie (n.1 above), 8–10 and 50f.

40. Although Fabius is hardly self-destructive, he is said to confront the enemy by himself (7.6–8), expresses his wish to do so to his troops (7.251f.), and is often identified as a ‘one’ (7.1, 739f., 743, 745, 8.2f., 8–11; cf. 6.637–40, 7.62–64). The identification of Fabius as such has a long tradition; see, e.g., Enn. Ann. 363 Sk and Virg. Aen. 6.845f. with Hardie (n.1 above), 4–10. The ‘one’ on the Carthaginian side is, of course, Hannibal: 1.36f., 5.151–53, 16.17–19, 17.149–51, 197f., 399f. (with Scipio), 512–16. For more on these leaders as ‘ones’ in the Punica, see Marks (n.9 above), 78–81.

41. Versnel (n.12 above), 143. For more on the ‘situation’ of a deuotio, see Versnel (n.12 above), 140–43.

42. Neither of these responses to crisis is found in our historical sources. Appian and Nepos, who, unlike Polybius and Livy, report that Scipio fought at the Trebia (see n.19 above), do not indicate that he reacted to some crisis during the battle as Silius here suggests. As for Paulus, our sources report that he generally responded to the needs of his troops throughout the battle at Cannae and that he even realised that the Romans were losing, which he expresses to Lentulus, but they never combine these two details as Silius does; see n.48 below.

43. Livy says that during the battle at Trasimene Flaminius went around shoring up the formation of his ranks and encouraging his men to stand their ground and fight (22.5.If), but does not refer to these deeds in connection with the Romans’ flight (cf. 22.6.6f.) nor speak of Flaminius trying to check that flight at all. Furthermore, Flaminius’ words of encouragement, as reported in Livy 22.5.2, have little in common with his speech in 5.633–43 (cf., instead, 5.107–29, which recalls Livy 22.3.7), and there is nothing comparable to his vaunt that he will stand and face the enemy alone.

44. Paulus encourages his men to stand their ground too (10.6–10), but does not do so when they are in flight. As for Decius’ second question, compare Flaminius’ to his men in 5.633f.: quid deinde, quid, orojrestat, io, profugis? (‘What then, what, I ask, remains for fugitives?’). On Paulus and Flaminius compared at these points, see Niemann (n.8 above), 219.

45. While the danger in which Scipio eventually found himself at the Ticinus is widely reported (see n. 17 above), neither Polybius nor Livy reports the details that Silius gives here relating to the crisis of the Romans’ flight and the consul’s response to it. Appian (Hann. 7), though, says that after his men fled, Scipio followed them and tried to encourage them: (‘Scipio, following them and exhorting them, had been wounded and was almost killed’).

46. For the significance of charging on horseback, see Versnel (n.12 above), 152–56.

47. For the charge into the enemy’s midst as a standard feature of the deuotio, see Versnel (n.12 above), 145–48; Leigh (n.4 above 1993), 91–96; Leigh (n.4 above 1997), 130–34; Heinrich (n.4 above), 182–85.

48. In none of these cases, except Paulus’, does the detail in question appear to be historical. No source reports that Flaminius rushed into the thick of the enemy as he does here, but Livy does say that he told his men per medias acies ferro uiam fieri (‘that a path be made with the sword through the middle of the lines’, 22.5.2) and that the fighting around him was quite fierce (22.6. i; cf. 22.5.5); instead, it is Ducarius, pursuing him, who per confertissimam hostium turbam impetum facit (‘makes an attack through the thickest crowd of the enemy’, 22.6.4). As for Paulus’ charges, Polybius is more explicit about this than Livy: (‘he rode into the middle of the whole line and at once was engaged’, Polyb. 3.116.3; cf. 3.116.9). Livy does report that the fighting conditions at Cannae were crowded (e.g., 22.47.2f., 48.4) and that at one point some troops were pushed into the middle of the battle (per praeceps pauore fugientium agmen in mediam primum aciem inlati, ‘borne through the ranks of those fleeing headlong from fear and right into the midst of the line’, 22.47.6). Cf. also Scipio the younger (4.459–71), Hannibal (praeceps init in medios , ‘he goes headlong into their midst’, 5.607), Fabius (perrumpit anhelumldictator cum caede globum , ‘with killing the dictator breaks through the panting mass’, 7.705f.), and Scaevola ( in medios…/…uasto conixus turbine fertur, ‘carried on a vast swirl, he is borne into their midst’, 9.378f.).

49. The deuotus is, by definition, willing to die, a uoluntarius; so Versnel (n.12 above), 145–48. For the distinction in Virgil’s Aeneid, see Leigh (n.4 above 1993), 92f., who gives several examples of ‘death rushes’, in which a warrior is willing or ready to die, and differentiates this from the standard rush at the enemy, in which a warrior’s willingness or readiness to die is not at issue or simply goes unmentioned; for examples of the latter in the Aeneid, see Leigh (n.4 above 1993), 106 n.14, and for instances in other epics, see Luc. BC 7.497, 590; Val. Fl. Arg. 1.438L, 6.238; Stat. Theb. 7.280, 8.466f, 9.727f., 10.164, 714. For ‘death rushes’ in other epics, see Luc. BC 2.523–24 (with Leigh [n.4 above 1997], 133–43 and 152–54) and Stat. Theb. 7.698–700 (Amphiaraus).

50. Note that Minucius, while not among those who rush at the enemy listed above, is intent on dying: nee mora disiecto Minuci uecordia uallo/perdendi simul et pereundi ardebat amore (‘without delay, after the wall had been knocked down, Minucius in his madness was burning with desire of killing and being killed at the same time’, 7.523f.). For non-leaders who are willing or ready to die, see Scaevola (9.372f.), Marius (9.402), and Crista and sons (10.104f., 136f.).

51. A point particularly stressed by Janssen (n.37 above), 380f.

52. Wezel, E.J., De C. Silii Italici cum fontibus turn exemplis (diss. Leipzig 1873), 73f.Google Scholar, and Venini (n.18 above), 535 n.9, both note that Silius and Livy differ on the detail of Flaminius’ killer (or killers), though Wezel, in addition, points out the similarilty between Silius and Polybius.

53. For the connection between Flaminius and Decius, note also the phrase obruitur telis (‘he is overwhelmed by missiles’, 5.656), and compare, of the Decii, obrutus telis (Livy 8.9.12), telis obrutus (Val. Max. 5.6.5), and obrutum (Livy 10.29.19). I do not want to rule out the possibility, however, that Silius may have altered the historical record in this way to identify Flaminius more closely with Paulus, who is likewise overcome by many, and thus to foreshadow the latter’s demise at Cannae.

54. Compare also the buried state in which Decius’ son is found: obrutum superstratis Gallorum cumulis (‘overwhelmed by heaps of Gauls spread on top of him’, Livy 10.29.19). Spaltenstein (n.15 above), ad 5.658, suggests that Silius might have been thinking of Polybius 3.84.10, where it is said that some Romans were trapped in the waters of Lake Trasimene and in despair committed suicide there. Even if this were the case, Polybius does not connect those suicides with Flaminius’ death in his account.

55. Note, though, that Livy uses the phrase obruere telis (22.49.12) whereas Silius does not here; see n.53 above.

56. Versnel (n.12 above), 150f.

57. In Flaminius’ case, though, Silius may have drawn this detail from Livy, who has Ducarius identify him as a sacrificial victim too: iam ego hanc uictimam manibus peremptorum foede ciuium dabo (‘now I will give this victim to the spirits of citizens dreadfully slain’, 22.6.3). For other instances where Romans and their allies are regarded as sacrifices, see 4.548, 5.574–76 (with 4.735), 7.513f., 8.253–57, and 8.333–35 (with Hardie [n.1 above], 52 n.76 on concido). Also, Hannibal is twice said to ‘send down’ (demittere) his victims, a word that is used, as Versnel (n.12 above, 154–56) notes, to denote the passage of a deuotus, once slain, to the underworld: 4.340f. (Metabus), 10.136f. (Crista and sons). Silius less often uses the language of sacrifice in Books 4–10 to identify the Romans as the sacrificers and the Carthaginians as the sacrificed. One episode alone, Scipio’s rescue of his father at the Ticinus, accounts for three such uses of sacrificial language (primitiae, 4.428; piacula, 465; mactare, 465); Scipio is there modelled after Achilles and Aeneas when they seek out sacrificial victims in revenge for the deaths of Patroclus and Pallas respectively; cf. Pun. 4.459–65 with Il. 21.26–28 and Aen. 10.513–20. Otherwise, mactare is twice used of Scipio the elder slaying his enemies (4.232, 604), and at Trasimene the Romans sacrifice their enemies as ‘offerings to their own soon-to-be manes’ (manibus ipsa suis praesumpta piacula, 5.213).

58. For the practice, see Huss, W., Geschichte der Karthager (Munich 1985), 531–40Google Scholar, and Brown, S.S., Late Carthaginian Child Sacrifice and Sacrificial Monuments in their Mediterranean Context (Sheffield 1991), 21–26Google Scholar. For sources for this practice on which Silius may have drawn, see Woodruff, L.B., ‘Reminiscences of Ennius in Silius Italicus’, University of Michigan Studies 4 (1910), 355–424Google Scholar, at 383–85, Nicol (n.19 above), 87f.; R.T. Bruère, ‘Silius Italicus, Punka 3.62–162 and 4.763–822’, CP 47 (1952), 219–27, at 219f. and 222; Romano (n.19 above), 89f; Spaltenstein (n.15 above), ad 4.765. While there is no evidence of child-sacrifices performed during the Second Punic War, it may be inferred, nevertheless, that they were being performed because, as Woodruff points out (385), Curtius says that the practice did not cease until Carthage was destroyed (4.3.23).

59. For the use of sacrificial language in these verses, see Spaltenstein (n.15 above), ad 4.827, although he does not read them in connection with the deuotio.

60. Rome’s enemies did not always ‘play along’: Pyrrhus undermined the deuotio of Decius the grandson at Ausculum in 279 BCE by ordering his men not to kill him (Dio 10.43 [= Zon. 8.5]; contra, Cic. De Fin. 2.61, Tusc. Disp. 1.89). The turn that Silius sets up here, that Hannibal by sacrificing the Romans is contributing to his own demise, may also have been prepared for in the oath scene in 1.81–139, in which Picard, G.C., ‘Le Tophet de Carthage dans Silius Italicus’, in J.P. Boucher (ed.), Mélanges de philosophie, de littérature et d’histoire ancienne offerts à Pierre Boyancé (Rome 1974), 569–77Google Scholar, Raabe, H., Plurima mortis imago: Vergleichende Interpretationen zur Bildersprache Vergils (Munich 1974), 160 Google Scholar n.62, and Martin, M., ‘Carthage vue de Rome ou “le rivage des Syrtes” chez Silius Italicus’, Eidolon 28 (1986), 59–77Google Scholar, at 71–73 (contra, Küppers, J., Tantarum causas irarum: Untersuchungen zur einleitenden Bücherdyade der Punica des Silius Italicus [Berlin and New York 1986], 80 CrossRefGoogle Scholar n.307) detect allusions to the ceremony of child-sacrifice in the tophet. If so, the identification of the boy Hannibal as a victim, and at the very moment he vows eternal enmity to Rome, is heavily ironic. And because he is not sacrificed (see 1.119–22), perhaps we are to suppose, much as we might in Book 4 where his son’s sacrifice is avoided, that one day Carthage will have to produce a victim (perhaps Hannibal or Carthage herself?) to compensate for this victim withheld. On this prospect, consider Hanno’s warning: sic propria luat hoc poena nee misceat urbis/fata suis (‘so let him pay for this with his own punishment and let him not mix the fates of his city with his own’, 2.301f.). Also, Hardie (n.1 above), 64 and n.4, notes that Hannibal unwittingly repeats Dido’s words (1.114f.) when she imagines herself hunting down the Romans after her death (Aen. 4.384–86). The allusion cleverly implies that Hannibal’s curse against Rome will be unfulfilled by the time of his death and may even suggest that he and Dido will be alike in death; she, after all, will go on to sacrifice herself by suicide (cf. Hardie [n.1 above], 29 and 41–43), and he will one day commit suicide himself (2.705–07; 13.890–93).

61. Versnel (n.12 above), 139f., 161f.

62. Cf. also Livy 10.29.3–5.

63. I leave out of the following discussion those gods whose behaviour toward Rome and Carthage never changes significantly, such as Juno and Venus, since we cannot measure the extent to which Rome’s actions influence their behaviour and thus cannot determine whether those actions conform to a ‘devotional’ pattern. For that reason I focus, instead, on those gods, namely Jupiter and the ‘anonymous gods’, whose behaviour does change and in such a way as to affect the fortunes of the two sides in the war. On ‘anonymous gods’ and their relation to the deuotio, see Versnel (n. 12 above), 171–79.

64. The attention Silius gives to the will of the gods at these points is not found in Livy; Paulus never refers to it in his conversation with Fabius (Livy 22.39.1–40.3 ∼ Pun. 8.298–348) or in his dispute with Varro (Livy 22.44.5–7 ∼ Pun. 9.23–65).

65. I have not taken into account two moments where Jupiter shows favour toward the Romans, viz. when he appoints Mars to help the young Scipio rescue his father at the Ticinus (4.417–59) and when he has Pallas removed from the battlefield of Cannae (9.470–85), because in neither case does his intervention change the outcome of the battle. For the same reason I have not addressed those instances where the ira deum is directed at individuals; see 4.229, 5.475, 588, and 9.424–26 (cf. 9.650f., 10.639).

66. A notable difference between the deuotiones of the Decii and those of Flaminius and Paulus in the Punica, however, is that the former occur during battles and change the tide of those battles to the Romans’ advantage whereas the beneficial effects of the latter are only perceptible after the battles in which they occur. This is because Silius puts Flaminius’ and Paulus’ deaths at the end of the battles at Trasimene and Cannae, respectively, or, at least, at the end of his narratives of them, which gives the same impression. This move invites us to see their deaths as climactic moments, which is surely one of the poet’s intentions, but it also necessitates that the effects of their deaths be felt subsequent to those battles. For the deuotio’s role in securing victory, see Val. Max. 5.6.5: ex cuius uulneribus et sanguine insperata uictoria emersit (‘from whose wounds and blood emerged an unexpected victory’).

67. On Silius and philosophy, see Billerbeck, M., ‘Aspects of Stoicism in Flavian Epic’, Papers of the Liverpool Latin Seminar 5 (1985), 341–56Google Scholar; Colish, M.L., The Stoic Tradition from Antiquity to the Early Middle Ages, I: Stoicism in Classical Latin Literature (Leiden 1985), 281–89Google Scholar; Billerbeck, M., ‘Stoizismus in der römischen Epik neronischer und flavischer Zeit’, ANRW 2.32.5 (1986), 3116–51Google Scholar, at 3134–43; Matier, K.O., ‘Stoic Philosophy in Silius Italicus’, Akroterion 35 (1990), 68–72Google Scholar; Marks (n.9 above), 148–61, with earlier literature given in 149 nn.97 and 98.

68. I am adapting here Bandera (n.1 above), 223, who distinguishes between two ‘sacrificial levels’ in the Aeneid, ‘the traditional sacrifice’ and ‘sacrifice in general, conceived as a law of history, governing Aeneas’ journey from beginning to end…’