Hostname: page-component-78c5997874-xbtfd Total loading time: 0 Render date: 2024-11-02T20:16:48.016Z Has data issue: false hasContentIssue false

A STYLISH EXIT: MARCUS TERENTIUS’ SWANSONG (TACITUS, ANNALS 6.8), CURTIUS RUFUS AND VIRGIL

Published online by Cambridge University Press:  10 June 2021

Rhiannon Ash*
Affiliation:
Merton College, University of Oxford

Abstract

Within the narrative for a.d. 32, Tacitus recreates a spirited speech delivered before the Senate by the eques Marcus Terentius (Ann. 6.8), defending himself retrospectively for having been a ‘friend’ of Sejanus. This speech, the only extended speech in oratio recta to feature in Annals Book 6, is historiographically rich and suggestive.

This article first analyses the speech as a compelling piece of oratory in its own right. It then explores the provocative mirroring of another important speech in Curtius Rufus (7.1.19–40). This is where the general Amyntas, defending himself before Alexander the Great against charges of participation in an alleged conspiracy, refuses to deny his friendship with the conspirator Philotas (now dead). Scholars have rightly acknowledged the significant intertextuality of these two speeches in Curtius Rufus and Tacitus. Yet the interest in this mirroring between Amyntas and Terentius has overshadowed another important intertext. This article demonstrates how Tacitus also engages with a programmatic moment from the opening of Virgil's Aeneid when Aeolus is cajoled by Juno to unleash a devastating storm. Terentius wittily casts Tiberius as a powerful divinity whose whims had to be obeyed and himself as a helpless Aeolus doing his will.

This article demonstrates that the two passages from Virgil and Curtius Rufus underpinning Terentius’ speech work together powerfully, challenging Tacitus’ readers to reflect on the difficulties of ‘speaking to power’ and on the compromises involved for men like Terentius in negotiating the complex political realities of the imperial system.

Type
Research Article
Copyright
Copyright © The Author(s), 2021. Published by Cambridge University Press on behalf of The Classical Association

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

Footnotes

I thank the anonymous reviewer for CQ whose suggestions were invaluable in improving this article and Bruce Gibson whose energetic and altruistic service as an editor of CQ has been exemplary. I am also tremendously grateful to the audiences at Christina Kuhn's Ancient History seminar in Oxford on the Julio-Claudians (including Alison Cooley and Panayiotis Christoforou, who each independently pointed me in the direction of the Cypriot oath of allegiance to Tiberius [n. 38]) and at the splendid conference in Martina Franca in Italy in April 2019 honouring John Marincola which was capably organized by Giusy Monti and Scarlett Kingsley. Finally, I dedicate this piece to the memory of my old friend Dave Mankin of Cornell University, whose untimely death coincided with the final stages of preparing this article: his teaching, scholarship and loyal friendship have inspired many people over the years. He is much missed.

References

1 PIR 2 8.18 no. 64. Martin, R.H., Tacitus Annals V and VI (Warminster, 2001), 121–4CrossRefGoogle Scholar and Woodman, A.J., The Annals of Tacitus Books 5 and 6 (Cambridge, 2017), 121–6Google Scholar discuss the speech. Dio covers the same episode at 58.19.1–5 and includes a very much shorter version of this same speech (also in oratio recta). Martin (this note), 121 is confident that Dio's speech ‘clearly derives from Tacitus’. Woodman (this note), 121 agrees: ‘the consensus seems to be that on this occasion Dio used T.’.

2 Sejanus’ sharp, sudden peripeteia provoked reactions in the literary tradition (Sen. Tranq. 11.11; Juv. 10.56–114 on the futility of striving for political pre-eminence).

3 See, for example, Miller, N.P., ‘Dramatic speech in Tacitus’, AJPh 85 (1964), 279–96Google Scholar; Adams, J.N., ‘The vocabulary of the speeches in Tacitus’ historical works’, BICS 20 (1973), 124–44Google Scholar; Miller, N.P., ‘Dramatic speech in the Roman historians’, G&R 22 (1975), 4557Google Scholar; Aubrion, E., Rhétorique et histoire chez Tacite (Metz, 1985), 491678Google Scholar; Keitel, E., ‘Homeric antecedents to the cohortatio in the ancient historians’, CW 80 (1986–1987), 153–72Google Scholar; Dangel, J., ‘Les discours chez Tacite: rhétorique et imitation créatrice’, Ktema 14 (1989), 291300CrossRefGoogle Scholar; Hansen, M.H., ‘The battle exhortation in ancient historiography: fact or fiction?’, Historia 42 (1993), 161–80Google Scholar; Devillers, O., L'art de la persuasion dans les Annales de Tacite (Brussels, 1995), 195261Google Scholar; Brock, R., ‘Versions, “inversions” and evasions: classical historiography and the “published” speech’, PLLS 8 (1995), 209–24Google Scholar; Laird, A., Powers of Expression, Expressions of Power: Speech Presentation in Latin Literature (Oxford, 1999), 116–52Google Scholar; D.S. Levene, ‘Speeches in the Histories’, in A.J. Woodman (ed.), The Cambridge Companion to Tacitus (Cambridge, 2009), 212–24; R. Mayer, ‘Oratory in Tacitus’ Annals’, in D. Berry and A. Erskine (edd.), Form and Function in Roman Oratory (Cambridge, 2010), 281–93 and D. Pausch (ed.), Stimmen der Geschichte: Funktionen von Reden in der antiken Historiographie (Berlin and New York, 2010).

4 Martin (n. 1), 121.

5 Cf. Tacitus’ reflections (Agr. 42.4) about productive collaboration vs futile resistance (often culminating in an ambitiosa mors). See further Turpin, W., ‘Tacitus, Stoic exempla, and the praecipuum munus Annalium’, ClAnt 27 (2008), 359404Google Scholar, especially 396–7.

6 On doublespeak, see Bartsch, S., Actors in the Audience: Theatricality and Doublespeak from Nero to Hadrian (Cambridge, MA and London, 1994)CrossRefGoogle Scholar. Ahl, F.M., ‘The art of safe criticism in Greece and Rome’, AJPh 105 (1984), 174208Google Scholar is an important discussion. See too Haynes, H., ‘Tacitus’ dangerous word’, ClAnt 23 (2004), 3361Google Scholar; Gallia, A., ‘Potentes and potentia in Tacitus’ Dialogus de oratoribus’, TAPhA 139 (2009), 169206Google Scholar, Strunk, T., ‘Offending the powerful: Tacitus’ Dialogus de oratoribus and safe criticism’, Mnemosyne 63 (2010), 241–67CrossRefGoogle Scholar; and Dressler, A., ‘Poetics of conspiracy and hermeneutics of suspicion in Tacitus’ Dialogus de oratoribus’, ClAnt 32 (2013), 134Google Scholar. On speaking truth to power in Tacitus, see O'Gorman, E., Tacitus' History of Politically Effective Speech: Truth to Power (London, 2020)CrossRefGoogle Scholar.

7 For similar disingenuous ‘apologies’ for monotony, see Tac. Ann. 4.33.3 and 16.16.1. At the opening of Ann. 6.8.1 the introductory nam (translated by A.J. Woodman, Tacitus The Annals [Indianapolis, 2004] as ‘for example’) implies that this is one of several notable examples involving pericula et poenas.

8 Martin (n. 1), 121. Only 72 words of this now fragmentary speech survive (cf. 245 words of Terentius’ speech).

9 The text is from Woodman (n. 1).

10 Translated by Woodman (n. 7), 169–70.

11 S.J. Harrison, Vergil Aeneid 10 (Oxford, 1991), 149–50 cites Enn. Ann. 233 Skutsch and Sen. Ep. 94.28. See too Plin. Ep. 6.16.11. The earlier speech at Tac. Ann. 5.6.2 also introduces the concept of fortune (uersa est fortuna) in relation to Sejanus (as Woodman [n. 1], 68 notes, quite possibly alluding in a subtle way to Sejanus’ personal statue of Fortuna turning its back on him before his fall: Cass. Dio 58.7.2–3).

12 The assonance of fortuna and fortasse, elegantly captured in Woodman's translation (n. 7), reprises effective usage in Cicero (Pis. 71; Sull. 73), as Woodman (n. 1) observes.

13 Woodman (n. 1), 122.

14 There is a striking movement from the first-person singular to the first-person plural in Terentius’ speech: fortunae meae, fatebor, me, essem, adeptus eram, uideram, adsumo, defendam (Ann. 6.8.1–3) as opposed to fuimus, colebamus, nostrum, nobis, spectamus, uenerabamur, nos (Ann. 6.8.3–6). This shift elegantly backs up Terentius’ basic point about collective guilt. It also mirrors the opening of the Agricola where first-person singular (at nunc narraturo mihi uitam defuncti hominis, Agr. 1.4) shifts to first-person plurals (dedimus profecto grande patientiae documentum … ita nos quid in seruitute … perdidissemus … in nostra potestate, Agr. 2.3), painfully showcasing collective guilt, even if Tacitus is taking more blame than he should: ‘The plural verb refers to T. and his contemporaries, especially his fellow-senators, but it disguises the fact that, when the philosophers were banished, T. was almost certainly absent from Rome and could not be held to share in responsibility’ (A.J. Woodman with C.S. Kraus, Tacitus Agricola [Cambridge, 2014], 80–1).

15 Dyck, A.R., Cicero Catilinarians (Cambridge, 2008), 68Google Scholar comments: ‘Rhetorical theory recognized the example … as a means of persuasion … introduced from outside the case (Quint., Inst. 5.11.1).’

16 Cicero's disregard for his own life while confronting Catiline's conspiracy features, for example, at Cat. 3.28. Compare the difficulty of finding qui auderent se et salutem suam in discrimen offerre pro statu ciuitatis et pro communi libertate, ‘men who would dare to expose themselves and their very lives to danger for the stability of our constitution and for the general liberty’ (Sest. 1). As noted by Woodman with Kraus (n. 14), 90, Cicero saw his return from exile after the conspiracy as the ‘start of a second life’ (Att. 4.1.8) and a ‘birthday’ (Att. 3.20.1).

17 Whether Sejanus really engaged in conspiracy remains uncertain: ‘… victims and an informer do not prove treason’ (R. Syme, Tacitus [Oxford, 1958], 406).

18 Cf. the lament tot hominum milibus unum iam reliquum diem, ‘that for so many thousands one day was now left’ (Ann. 1.65.7). The juxtaposition meo unius is also emphatic (Woodman [n. 1] compares Livy 28.28.12).

19 Woodman (n. 1), 125: ‘Terentius’ 16 years are a.d. 15–30 inclusive i.e. excluding the part-years of Tib.'s accession and Sejanus’ execution. Velleius’ 16 years are likely to be a.d. 14–29 inclusive, unless he projected himself counting twelve-month periods backwards from some point within a.d. 30.’ Velleius’ history was published in a.d. 30 before Sejanus’ fall.

20 See Woodman with Kraus (n. 14), 88–91 on this passage.

21 There is tension in the notion that, although Tacitus in the Agricola is one of the superstites, it is unclear whether Terentius himself will survive the current danger.

22 Damon, C., Tacitus Histories Book I (Cambridge, 2003), 244Google Scholar notes that Tacitus’ version of the scene, ‘a “reconciliation comedy” (Heubner ad loc.), is less favourable to Otho than Plutarch's’.

23 Cf. Polyb. 38.4.1–2 criticizing the timid man fearful of free speech who is not a sincere friend, and the bad citizen who omits the truth because he fears a hostile reaction. Terentius is the polar opposite.

24 Plass, P., Wit and the Writing of History: The Rhetoric of Historiography in Imperial Rome (Madison, WI, 1988), 47Google Scholar.

25 Lausberg, H., Handbook of Literary Rhetoric: A Foundation for Literary Study (Leiden, 1998), §64Google Scholar (including the point that ‘the opposing party represents a causa turpis’). See too Woodman (n. 1), 122: ‘Honourableness was intrinsic to friendship (Cic. Part. or. 88, honestati, quae maxime spectatur in amicitiis), and much of Terentius’ defence will rest on the implicit assumption that his friendship with Sejanus was honourable, as Quintilian recommends (7.4.4).’

26 Lausberg (n. 25), §197: ‘In response to the prosecutor's accusation of Fecisti, the defendant thus charges the prosecutor with Fecisti—the same particular (identical) crimen’; A.M. Riggsby, Crime and Community in Ciceronian Rome (Austin, 1999), 38, 58; Dyck, A.R., Cicero Pro Roscio Amerino (Cambridge, 2010), 76, 147–50Google Scholar.

27constantia refers not so much to the speech as to the character or views of the speaker’ (Woodman [n. 1], 127). In Stoic thought, constantia was a virtue.

28 The formal charge against Terentius was presumably maiestas.

29 Star, C., ‘Commanding constantia in Senecan tragedy’, TAPhA 136 (2006), 207–44Google Scholar, especially 211–16, usefully discusses this letter.

30 Martin (n. 1), 121.

31 Dio's short version of the speech is as follows: ‘ὥστ’ εἰ μὲν ἐκεῖνος ὀρθῶς’ εἶπεν ‘ἐποίει τοιούτῳ φίλῳ χρώμενος, οὐδὲ ἐγώ τι ἠδίκηκα· εἰ δ’ ὁ αὐτοκράτωρ ὁ πάντα ἀκριβῶς εἰδὼς ἐπλανήθη, τί θαυμαστὸν εἰ καὶ ἐγώ οἱ συνεξηπατήθην; καὶ γάρ τοι προσήκει ἡμῖν πάντας τοὺς ὑπ’ αὐτοῦ τιμωμένους ἀγαπᾶν, μὴ πολυπραγμονοῦντας ὁποῖοί τινές εἰσιν, ἀλλ’ ἕνα ὅρον τῆς φιλίας σφῶν ποιουμένους τὸ τῷ αὐτοκράτορι αὐτοὺς ἀρέσκειν’, ‘“Consequently”, he said, “if the emperor did right in having such a friend, I, too, have done no wrong; and if he, who has accurate knowledge of everything, erred, what wonder is it that I shared in his deception? For surely it is our duty to cherish all whom he honours, without concerning ourselves overmuch about the kind of men they are but making our friendship for them depend on just one thing—the fact that they please the emperor”’ (Cass. Dio 58.19.3–4).

32 Martin (n. 1), 121; Woodman (n. 1), 121; Wiedemann, T., ‘Über das Zeitalter des geschichtsschreibers Curtius Rufus’, Philologus 30 (1870), 241–64CrossRefGoogle Scholar, at 244; Devine, A.M., ‘The Parthi, the tyranny of Tiberius and the date of Q. Curtius Rufus’, Phoenix 33 (1979), 142–59CrossRefGoogle Scholar; J.E. Atkinson, A Commentary on Q. Curtius Rufus’ Historiae Alexandri Magni Books 3 and 4 (Amsterdam and Uithoorn, 1980), 37; Bosworth, A.B., ‘Mountain and molehill? Cornelius Tacitus and Quintus Curtius’, CQ 54 (2004), 551–67CrossRefGoogle Scholar. Bosworth's piece is a significant treatment of Tacitus and Curtius in general (and refers to further scholarship on the topic).

33 Philotas’ conspiracy has itself generated conflicting interpretations of what really lay behind it. See Badian, E., ‘The death of Parmenio’, TAPhA 91 (1960), 324–38Google Scholar. Heckel, W., ‘The conspiracy against Philotas’, Phoenix 31 (1977), 921CrossRefGoogle Scholar. In addition to Curtius Rufus’ account, we also have Diod. Sic. 17.79, Strabo 15.2.10, Plut. Alex. 48–9 with J.R. Hamilton, Plutarch Alexander. A Commentary (Oxford, 1969), 132–8, Arr. Anab. 3.26–27.4 with the useful appendix of Brunt, P.A., Arrian Anabasis Alexandri Books I–IV (Cambridge, MA, 1976), 517–21Google Scholar, and Just. Epit. 12.5.1–3.

34 Alternatively, he is killed by a barrage of javelins hurled by the Macedonians (Arr. Anab. 3.26.3). The pattern of one unsuccessful defence speech (Philotas, Curt. 6.10) followed by another successful one (Amyntas, Curt. 7.1.19–40) is exactly the same as the succession in Tacitus (Ann. 5.6.2–3, 6.8).

35 Arr. Anab. 3.27.2 says that Amyntas defended himself vigorously before the Macedonians, but he does not give details of the speech.

36 The text is from E. Hedicke's Teubner (Leipzig, 1908).

37 Translated by J. Yardley and W. Heckel, Quintus Curtius Rufus The History of Alexander (Harmondsworth, 1984).

38 Cf. Livy 35.50.2 eosdem … hostes et amicos quos populus Romanus, 37.1.5, 38.8.10, 38.11.3. See further L. de Libero, ‘“VT EOSDEM QVOS POPVLVS ROMANVS AMICOS ATQVE HOSTES HABEANT”: die Freund-Feind-Klausel in den Beziehungen Roms zu griechischen und italischen Staate’, Historia 46 (1997), 270–305.

39 ‘We ourselves and our children swear to harken unto and to obey alike by land and sea, to regard with loyalty and to worship Tiberius Caesar Augustus, son of Augustus, with all his house, [(lines 15–17) καὶ v. | τὸν αὐτὸν ἐκείνοις φίλον τε καὶ ἐχθρὸν v. | ἕξειν v.] ‘to have the same friends and the same enemies as they …’: see T.B. Mitford, ‘A Cypriot oath of allegiance to Tiberius’, JRS 50 (1960), 75–9. I am grateful to Panayiotis Christoforou and Alison Cooley, both of whom independently mentioned this inscription to me as a relevant intertext. See too J.E. Atkinson, A Commentary on Q. Curtius Rufus’ Historiae Alexandri Magni Books 5 to 7.2 (Amsterdam, 1994), 253–4 and Gonzalez, J., ‘The first oath pro salute Augusti found in Baetica’, ZPE 72 (1988), 113–27Google Scholar on an oath to Augustus and the male members of his family dateable to 5 b.c.

40 Bosworth (n. 32), 565.

41 See R.H. Martin and A.J. Woodman, Tacitus Annals Book IV (Cambridge, 1989), 84–7.

42 Bosworth (n. 32), 565.

43 Woodman (n. 1), 123.

44 Martin (n. 1), 122; Woodman (n. 1), 122. Walter, F., Studien zu Tacitus und Curtius (Munich, 1887)Google Scholar assembles parallels between the two authors. S.P. Oakley, ‘Style and language’, in A.J. Woodman (ed.), The Cambridge Companion to Tacitus (Cambridge, 2009), 195–211, at 196 observes: ‘The similarity exhibited by Sallust, Livy, Quintus Curtius Rufus (in his History of Alexander the Great) and Tacitus in their choice of vocabulary allows the generalisation that Latin historical style was marked by frequent employment of archaisms.’

45 See Joseph. AJ 18.6.6 on Antonia's letter to Tiberius on Capri, which was then followed up by Tiberius’ uerbosa et grandis epistula to the Senate (Juv. 10.71; Cass. Dio 58.9–10). Atkinson (n. 39), 251 comments on Olympias’ ‘proclivity for accusatory letters’ (citing Arr. Anab. 7.12.5–7, Diod. Sic. 32.1, 114.3, 118.1, Just. Epit. 12.14.3, Plut. Alex. 39.7, Mor. 180D and 332F).

46 Atkinson (n. 39), 24 and 251 gives the credit to Curtius Rufus (whose history he thinks was published under Claudius), suggesting that Curtius Rufus drew directly on his personal knowledge of Terentius’ actual speech in the Senate in shaping his version of the Amyntas story. That interpretation rests a great deal of weight on the assumption that Tacitus’ version is an accurate record of what happened in a.d. 32, and it does not allow for the possibility that Tacitus was drawing on Curtius Rufus.

47 See V. Pagán, Conspiracy Narratives in Roman History (Austin, 2004), 87–90 for the tendency for conspiracy narratives to share common elements and themes to cope with ‘epistemological gaps’. C. Damon, ‘Déjà vu or déjà lu? History as intertext’, PLLS 14 (2010), 375–88 usefully reminds us that transhistorical connections between chronologically disparate episodes need not always be mediated through texts.

48 This is a potentially huge topic, but important surveys of the debate include: W.W. Tarn, Alexander the Great, 2 vols. (Cambridge, 1948), 2.111–14; D. Korzeniewski, Die Zeit des Quintus Curtius Rufus (Diss., Cologne, 1959), 4–50; Milns, R.D., ‘The date of Curtius Rufus and the Historiae Alexandri’, Latomus 25 (1966), 490507Google Scholar; Badian, E., ‘Alexander the Great, 1948–67’, CW 65 (1971), 3783Google Scholar, at 47–8; Devine (n. 32), 142–4, 148; H. Boedefeld, Untersuchungen zur Datierung der Alexandergeschichte des Q. Curtius Rufus (Diss., Düsseldorf, 1982); Hamilton, J.R., ‘The date of Quintus Curtius Rufus’, Historia 37 (1988), 445–56Google Scholar, at 445–7; J. Fugmann, ‘Zum Problem der Datierung der Historiae Alexandri Magni des Curtius Rufus’, Hermes 123 (1995), 233–43, at 233–4; J.E. Atkinson, ‘Q. Curtius Rufus’ Historiae Alexandri Magni’, ANRW 2.34.4 (1998), 3447–83, at 3451–5; E. Baynham, Alexander the Great: The Unique History of Quintus Curtius (Ann Arbor, 1998), 201–20; J.E. Atkinson and J.C. Yardley, Curtius Rufus: Histories of Alexander the Great, Book 10 (Oxford, 2009), 2–14; and T. Power, ‘Suetonius and the date of Curtius Rufus’, Hermes 141 (2013), 117–20. Naturally, people have asked whether the Curtius Rufus mentioned at Tac. Ann. 11.20–1 is our historian or an ancestor. See S.J.V. Malloch, The Annals of Tacitus Book 11 (Cambridge, 2013), 304–5 for a helpful summary of the scholarship.

49 Devine (n. 32), 152.

50 Martin (n. 1), 121.

51 On window references, see Thomas, R.F., ‘Virgil's Georgics and the art of reference’, HSPh 90 (1986), 171–98Google Scholar.

52 Atkinson (n. 39), 37. Aufidius Bassus (FRHist no. 78) wrote a Bellum Germanicum and a history of his own times. Pliny the Elder's history began a fine Aufidi Bassi (Plin. Ep. 3.5.6), though there are different views about the endpoint. His history ‘might have reached Sejanus’ death and become available for the elder Seneca to hear soon afterwards’ (B.M. Levick in FRHist 1.520). If it was published soon after Tiberius’ death, this ‘would make it likely to be the first hostile account of his reign and influential in establishing the tradition’ (B.M. Levick in FRHist 1.521). That scenario would certainly cohere with Aufidius Bassus composing a defiant speech from Terentius—but so too would later dates of publication, and the notion that Aufidius Bassus was a source for Tacitus’ version of Terentius’ speech must remain speculation.

53 There are some complex questions about Sejanus’ marital status in a.d. 31. Tacitus’ Tiberius memorably wrote to Sejanus (a.d. 25) responding to his request for permission to marry Livi(ll)a (Ann. 4.39–40). As a result, Sejanus non iam de matrimonio, sed altius metuens tacita suspicionum, uulgi rumorem, ingruentem inuidiam deprecatur, ‘no longer talked about marriage, but, with a deeper dread, protested against the silent suspicions, the public rumours, the encroaching resentment’ (Ann. 4.41.1). Yet at Tac. Ann. 5.6.2 the unnamed speaker claims that Sejanus in a.d. 31 was Tiberius’ gener. Had Sejanus therefore married Livi(ll)a? Or another woman? Zonaras (Cass. Dio 58.3.9) suggests that Sejanus had married Julia, that is, Livi(ll)a's daughter. The Fasti Ostienses suggest that Sejanus’ wife committed suicide eight days after Sejanus’ death in October a.d. 31, but unfortunately the wife's name is missing: some restore Apicata (Sejanus’ ex-wife). Bellemore, J., ‘The wife of Sejanus’, ZPE 109 (1995), 255–66Google Scholar suggests (at 262) that Sejanus and Livi(ll)a swiftly celebrated a clandestine marriage which was revealed to him by Antonia, Livi(ll)a's mother. Bellemore ([this note], 266) proposes that the missing name from the Fasti Ostienses indicates Livi(ll)a. If Terentius in a.d. 32 is alluding to these events, then calling Sejanus gener is another bold move.

54 Woodman (n. 1), 124, citing the helpful note of R.G.M. Nisbet and M. Hubbard, A Commentary on Horace: Odes Book I (Oxford, 1970), 131 on Carm. 1.10.9.

55 As Woodman (n. 1), 124 notes, Tacitus’ choice of the verb colebamus ‘looks all the way back to Sejanus’ formal introduction at the start of Book 4 (2.3 “colique … effigies eius”)’ and ‘constitutes a reminder that he too received worship: he and Tib. had been the presiding twin deities, their statues on either side of the ara amicitiae (4.74.2 and n.)’. Suetonius notes that Tiberius had rewarded the legionaries in Syria because they were the only soldiers who refused to cultivate Sejanus’ imago amongst their standards (quod solae nullam Seiani imaginem inter signa coluissent, Tib. 48.2; cf. imagines aureas coli passim, Tib. 65.1).

56 Martin (n. 1), 123 and Woodman (n. 1), 125 observe that this phrase features at Plin. Pan. 9.5, 83.7.

57 Tacitus’ echoing of Virgil has prompted much interest. As well as the indices of the various commentaries on Tacitus, see, for example, Baxter, R.T.S., ‘Virgil's influence on Tacitus Histories 3’, CPh 66 (1971), 93107Google Scholar; Baxter, R.T.S., ‘Virgil's influence on Tacitus in Annals 1 and 2’, CPh 67 (1972), 246–69Google Scholar; Miller, N.P., ‘Virgil and Tacitus again’, PVS 18 (1986), 87106Google Scholar; Putnam, M.C.J., ‘Virgil and Tacitus Annals 1.10’, CQ 39 (1989), 563–4CrossRefGoogle Scholar; E. Henry, ‘Virgilian elements in Tacitus’ historical imagination’, ANRW 33.4 (1991), 2987–3005; Pagán, V., ‘Beyond Teutoburg: transgression and transformation in Tacitus Annales 1.61–2’, CPh 94 (1999), 302–20Google Scholar, especially 305–6; M. Nickbakht, ‘Fighting for liberty, embracing slavery: Tacitus, Annals 1.7.1’, MH 63 (2006), 39–43; T. Joseph, ‘The metamorphoses of tanta moles: Ovid, Met. 15.765 and Tacitus, Ann. 1.11.1’, Vergilius 54 (2008), 24–36; Keitel, E., ‘The Virgilian reminiscences at Tacitus Histories 3.84.4’, CQ 58 (2008), 705–8CrossRefGoogle Scholar; T.A. Joseph, Tacitus the Epic Successor: Virgil, Lucan, and the Narrative of Civil War in the Histories (Leiden, 2012). See too A. Foucher, Historia proxima poetis. L'influence de la poésie épique sur le style des historiens latins de Salluste à Ammien Marcellin (Collection Latomus 255) (Brussels, 2000) and the essays in D.S. Levene and D.P. Nelis (edd.), Clio and the Poets: Augustan Poetry and the Traditions of Ancient Historiography (Leiden, 2002) and in A.J. Woodman and J.F. Miller (edd.), Latin Historiography and Poetry in the Early Empire: Generic Interactions (Leiden, 2010).

58 Virgil's Aeolus is a very different character from Homer's version (Od. 10.1–55), where Aeolus is happily married with a large family (6 sons and 6 daughters who have married one another).

59 R.G. Austin, P. Vergili Maronis Aeneidos Liber Primus (Oxford, 1971), 50 comments on Aeolus ‘using the ceremonial style of hymns, with anaphora’.

60 Rutherford, R., ‘Herodotean ironies’, Histos 12 (2018), 148Google Scholar traces forms of irony on a spectrum ranging from those clearly indicated by the narrator to those demanding more by way of interpretation and supplement from the reader. Terentius’ play with Virgil is closer to the second than the first, but the irony enhances the situational drama of his defence-speech. Wittily collapsing the distance between human and divine spheres is a trick used by Ovid in Met. 1.175–6: hic locus est quem, si uerbis audacia detur | haud timeam magni dixisse Palatia caeli, ‘this is the place which, if boldness be granted to my words, I would not fear to call the Palatine district of heaven’. Fuhrer, T., ‘Der Götterhymnus als Prahlrede—zum Spiel mit einer literarischen Form in Ovids Metamorphosen’, Hermes 127 (1999), 356–67Google Scholar explores how Ovid regularly exploits the comic potential of divinities engaging in self-aretalogy in the Metamorphoses.

61 capesso in the sense of carrying out orders (OLD 8c; TLL 3.311.15–31) is first attested in Plautus (Trin. 299). N. Horsfall, Virgil Aeneid 11: A Commentary (Leiden, 2003), 210 observes that capesso, the desiderative form of capio, appears eight times in the Aeneid and that it is Ennian in flavour, appearing in Ilia's dream (Ann. 40–2 Skutsch: … errare uidebar | tardaque uestigare et quaerere te neque posse | corde capessere, ‘… I seemed to wander and to track with slow foot and seek you, but I was unable to embrace you [?]’, translated by S.M. Goldberg and G. Manuwald, Fragmentary Republican Latin: Ennius [Cambridge, MA, 2018], 141). O. Skutsch, The Annals of Q. Ennius (Oxford, 1985), 199 discusses the ‘very difficult phrase’ corde capessere. Vine, B., ‘corde capessere (Ennius, Ann. 42 Sk)’, Glotta 67 (1989), 123–6Google Scholar suggests emending it to corda capessere.

62 Tacitus has capesso once in the Agricola, eight times in the Histories and thirty-two times in the Annals. See Martin, R.H. and Woodman, A.J., The Annals of Tacitus Book 3 (Cambridge, 1996), 263Google Scholar for capesso used of taking up magistracies or other specific duties (TLL 3.311.31–47; ‘almost a technical political term [OLD 8]’, as Woodman with Kraus [n. 14], 108 observe), first at Livy 6.34.4 (with magistratus), to which this more generalizing usage is related.

63 Aeolus’ use of labor is obsequiously hyperbolic: the goddess settling on a desire hardly involves much work.

64 It may be suggestive that Cicero in a letter to Appius Claudius Pulcher from 50 b.c. praises his urbanitas which (Cicero says) the Stoics very correctly call a virtue (Fam. 3.7.5). Terentius’ witty allusion to Virgil in potentially deadly circumstances sees him displaying a Stoic virtue. See further E.S. Ramage, Vrbanitas: Ancient Sophistication and Refinement (Norman, OK, 1973), 161–2.

65 The fact that Juno, the goddess of marriage, has bribed Aeolus with the beautiful nymph Deiopea also seems expressive in light of Tiberius preventing some prospective marriages, including Sejanus and Livi(ll)a (Ann. 4.39–40) and Agrippina the Elder (and anyone) (Ann. 4.53), and arranging others (e.g. Agrippina the Younger and Cn. Domitius Ahenobarbus, Ann. 4.75; her sisters Julia Drusilla and Cassius Longinus, Julia Livilla and Marcus Vinicius, Ann. 6.15.1; and Julia [Drusus’ daughter] and Rubellius Blandus, Ann. 6.27.1).

66 Ihm, M., C. Suetoni Tranquilli opera. Vol. I: De Vita Caesarum libri VIII (Leipzig, 1901)Google Scholar.

67 Champlin, E.J., ‘The odyssey of Tiberius Caesar’, C&M 64 (2014), 199246Google Scholar, on 228. Io is the daughter of Inachus and (?) the nymph Melia, daughter of Oceanus. That family background might provide a watery connection, but it seems tenuous.

68 Houston, G.W., ‘Tiberius on Capri’, G&R 32 (1985), 179–96, at 179Google Scholar.

69 Kaster, R.A., C. Suetoni Tranquilli de Vita Caesarum libros VIII et de Grammaticis et Rhetoribus librum (Oxford, 2016)CrossRefGoogle Scholar, accepting the arguments of Champlin (n. 67), 225–30.

70 Heinsius's reading is cited in Ihm's apparatus criticus and discussed in Ihm, M., sogennante, ‘DieVilla Iouis” des Tiberius auf Capri und andere Suetoniana’, Hermes 36 (1901), 287304Google Scholar, especially 287–91. On Daniel Heinsius (1580–1655), Professor of History at the University of Leiden, see Somos, M., Secularisation and the Leiden Circle (Leiden, 2011)CrossRefGoogle Scholar.

71 Champlin (n. 67), 227.

72 This inference does not depend on emending the corrupt portion of the text at Tac. Ann. 4.67.3 (tum Tiberius duodecim uillarumnominibus et molibusinsederat) and reading numinibus for nominibus. Instead, it rests on the detail that there are twelve villas—and the number twelve has obvious associations with the Olympian divinities. Martin and Woodman (n. 41), 244–5 discuss the textual problems and elect to mark the text as corrupt. Woodman, A.J., The Annals of Tacitus Book 4 (Cambridge, 2018), 308–9Google Scholar comments further on the passage (still marked as corrupt), though he is cautious about whether twelve substantial villas could be contained on the small mountainous island of Capri.

73 Again, there is cause here to regret the large lacuna of almost three whole years in Tacitus’ narrative of Tiberius’ principate, with the story only resuming (Ann. 5.6–11) towards the end of a.d. 31 and the aftermath of Sejanus’ execution. On the lacuna, see Woodman (n. 1), 3–9 (including an illuminating and helpful facsimile of the problematic point in the manuscript) and 67.

74 Not all alleged Virgilian echoes in Tacitus find acceptance. A.J. Woodman, ‘Introduction’, in A.J. Woodman (ed.), The Cambridge Companion to Tacitus (Cambridge, 2009), 1–14 (especially 6–7) has an illuminating discussion.

75 The anonymous reviewer for CQ asks a pertinent question about whether there are significant clusters of Virgilian resonances elsewhere in Tacitus Annals Book 6. Woodman (n. 1) points to a number of verbal echoes, particularly in the Parthian excursus at Ann. 6.31–7 (Ann. 6.25.2 uirilibus curis ~ Aen. 9.311 curamque uirilem; Ann. 6.32.1 tardari metu ~ Aen. 11.21 segnisue metu sententia tardet; Ann. 6.35.1 horridam … aciem ~ Aen. 10.408 horrida … acies; Ann. 6.35.2 uulnus adegit ~ Aen. 10.850 uulnus adactum; Ann. 6.35.2 clamore telis equis ~ Aen. 11.609–10 clamore … equos … tela; Ann. 6.50.3 instaurari epulas ~ Aen. 7.146 and Aen. 8.283 instaurant epulas).

76 Elliott, J., ‘The epic vantage-point: Roman historiographical allusion reconsidered’, Histos 9 (2015), 277–311, at 287Google Scholar.