1. The line recalls or is recalled by Medea 1026f.: so Boyle (2014), 387, ‘Is there an ironic echo of Hercules’ apotheotic language?’ If Medea were the earlier play the irony would work differently: Hercules will not think of child-killing for another thirty lines.

2. Lawall (1983), 14.

3. Fitch (1987), 28-33.

4. See Mazzoli (1996), 28, on carmina sublimis… Lucreti (Ov. Am. 1.15.23), and on ancient foundations for an early modern sublimity Porter (2007), Hardie (2009), 67-228, and Day (2013), 30-71.

5. Hardie (2009), 184.

6. So Oliensis (1998), 132f, Santirocco (1986), 112 and 108, on the retreat from grandeur at the end of a lyric as characteristically Horatian.

7. See with great complexity Davis (1991), 133-37, on Carm. 4.2.1-32.

8. See Santirocco (1986), 142-45, on the importance of reading these poems in succession.

9. On the sublimity of Carm. 3.25 see Schiesaro (2009), 61-64.

10. On other collisions of genre (and gender) in the person of Pallas see Theodorakopoulos (1997), 157 and 164, and Reed (2007), 21-23.

11. Schiesaro (2003), 52-55; quotation on p.55.

12. On the Euripidean and more generally tragic colouring of the Virgilian passage see Horsfall (2000), 222-30, and Panoussi (2009), 124-32. On the Virgilian influence on Hercules Furens see Buckley (2013), 214-16, and Trinacty (2014), 130-38; on Thyestes Schiesaro (2003), 30-36.

13. Hardie (2012), 109,

14. On poetics of nefas deriving from the passions of the Aeneid see Ganiban (2007), 33-37. Contrast in this volume Cowan's distinction between lofty, noble epic and its ‘evil twin’, tragedy. Whatever their differences both epic and tragedy ‘soar aloft in the stratosphere of the generic hierarchy’. See further Cowan on the transgressive grandeur of tragedy more generally in the satiric imagination.

15. Fas is the emphatic first word of the third and fourth stanzas (Hor. Carm. 2.19.9 and 13). On the poem see Davis (1991), 107-11.

16. See further Davis (1991), 111-14, and Gale (2000), 192. Compare also Horace's aspirations for himself, quodsi me lyricis uatibus inseris, / sublimi feriam sidera uertice (‘But if you rank me among lyric bards, I shall soar aloft and strike the stars with my head’, Carm. 1.1.35f.) with those for Caesar: aeternum meditans decus / stellis inserere (‘planning to set amid the stars [Caesar's] immortal glory’, Carm. 3.25.5f.).

17. Longinus’ date is unknown, but the lament for lost sublimity under lost political liberty (Longin. 44) has been the basis of an argument for dating it to the first century CE. See further Segal (1959) and, though he himself concludes otherwise, Heath (1999).

18. Innes (2006), 308.

19. Cf. Hardie (2013), 135: ‘For the Flavians, at least, the freedom associated with the sublime is in part the freedom of being able to go where their sublime predecessors have not gone…to break free of the shackles of intertextuality, and in particular of Virgil.’ For Seneca see Agapitos (1998), 241-50, on Thyestes and the creation of ‘a poetic discourse radically deviating from convention’ (242).

20. Porter (2007), 174.

21. Davis (1991), 101. See also Nisbet and Rudd (2004), 7, on Hor. Carm. 3.1.1 (‘But here the religious metaphor refers primarily to content rather than literary style’), and Fitch (2006-2007), 38, on Carm. 1.3 (‘While Horace lyricus might have had doubts about the boldness of composing epic, it would have been preposterous to align Vergil with violators of cosmic order as a means of conveying disagreement over genre’).

22. See LSJ s.v. ὑπερφυής II.

23. On the Romanizing of Euripides’ Phaethon (via Ovid's Phaethon narrative in Met. 2) see Fitch (1987), 158-60. On chariot imagery and aspirational poetics see Littlewood (2004), 120-26.

24. See Sampson in this volume. Both Juno and the chorus view Hercules as dangerously disruptive, but Juno's critique is complicated by the fact that her conception of cosmic order, as threatened by Hercules, is itself violent and disorderly. Her ‘natural order of things’ has nothing in common with the simple life preferred by the chorus.

25. See Fitch (1987), 179, and Billerbeck (1999), 268. Cf. Beringer (1997), 97: ‘The absence of violent consequences from Horace's depictions of transgression of natural limits may be attributable to the lighter vein of solo lyric poetry.’

26. See Littlewood (2004), 107 and 113, on HF 128, 944-47, 1142, and Davis (1993), 133-35, on Phaethon as a paradigmatic overreacher in this and other Senecan tragedies.

27. Trinacty (2014), 8. See also Buckley (2013), 213, and cf. Allendorf (2013), 140, on a ‘multiplicity of voices and perspectives in Senecan drama’ as characteristic of its poetic texture.

28. Tarrant (1985), 104, and Jakobi (1988), 155f., on ripae uacent (Sen. HF 108) and ripae Spercheides ardent… ostia septem / puluerulenta uacant (Ov. Met. 2.250, 255f.). I note in support of these claims for Ovidian influence that Met. 2.250 begins aestuat Alpheos: the Alpheus is the dried-up river in Seneca's landscape. Fury's speech in Seneca (Thy. 107-21) follows the first part of Ovid's narrative (Met. 2.210-34) with the same sequence of elements: water retreating into the earth, the loss of the trees and their fruit, the snow melted from the high mountains, the figure of the uncertain charioteer (Phaethon in Ovid, Titan in Seneca).

29. See Lowrie (1997), 77-93, on Horatian lyric's self-imposed limits and its opposition to epic and tragedy, and in this volume Cowan on generic antithesis at Sen. Ag. 328-32: ‘The chorus… urges the poet-god to come in humble Callimachean or Horatian lyric guise rather than with the ruinous grandeur of tragedy or epic.’

30. On bella as a generic marker of epic see Hunter (2006), 117, on Virg. Ecl. 6 and Hor. AP 73f.

31. On the military metaphors see Fitch (1987), 164.

32. Fitch (1987), 163. Fitch (2002-2004), 59.

33. Fitch (2002-2004), 59.

34. On this poem see Farrell (1991), 333f., and J. Pucci (1992).

35. On the importance of Virg. Ecl. 6 as a pattern cf. Harrison (2007), 188, on the technique of Hor. Carm. 3.3, arguing that lyric's excursion into and final retreat from epic themes here ‘is reminiscent of the end of Vergil's sixth and tenth Eclogue’.

36. Putnam (1986), 202.

37. Davis (1991), 184-86, On Roman elegy's debt to Greek tragedy see Conte (1986), 121.

38. See Davis (1991), 188, on atris ignibus and ignibus atris at Virg. Aen. 4.384 and 11.186 respectively and the poetic alternative Horace sets before ‘the composer of grave epic’. On the dawn and the respite it offers cf. Hershkowitz (1998), 72, on Euripides’ Hercules Furens, ‘Sunrise imagery… figures strongly in metaphorical approaches to the onset of reason after a period of madness.’

39. Fitch (1987), 160.

40. On the paradox see Gale (2000), 171-79.

41. On the morning salutatio at proud thresholds and other points of correspondence see Fitch (1987), 160, and Davis (1993), 127f.

42. But see in this volume Cowan on satiric moralizing present in De Rerum Natura and Georgics 2 and Hercules directly and indirectly ‘satirized’ in this chorus.

43. See also Harrison (2013), 368, on Odes 2 drawing on Lucretian inspiration, but accommodating it to ‘the more moderate persona of Horatian lyric’.

44. See further Gale (2000), 44f.

45. Harrison (2007), 142.

46. Gale, (2000), 9-11, Volk (2002), 142-45. Thomas (1988), 253, is in a minority in finding the language ‘redolent of Lucretius’ but not interpreting the passage as referring to the poet.

47. Billerbeck (1999), 484.

48. See Hardie (1986), 209-11, on Lucr. 1.62-79 and 5.110-21. For repression see e.g. oppressa graui sub religione (‘crushed beneath the weight of religion’, Lucr. 1.63) and quem… nec minitanti murmure compressit caelum (‘whom… heaven with menacing roar could not quell’, 1.68f.).

49. id autem dicitur a limine superiore, quia supra nos est, discussed by Mazzoli (1996), 26. In a scientific sense the ‘sublime’ (sublimia) is the region bounded by heaven above and earth below (Sen. Nat. 2.1.1).

50. See Hardie (2009), 105f. (quotation at 106), tracing the word from Ennius fr. 210 Sk. to Lucretius 1.11 to Virgil, Aen. 7.613 and beyond. On the proem of Annales 7 as a pattern for the proem of Aeneid 7 see further Goldschmidt (2013), 131-39. On the metapoetics of the Virgilian passage, mad genius bursting forth to inspire the poet, see Fowler (2000), 185f. Compare also the metapoetic readings of Schiesaro (2003), 226-28, and Boyle (2011), lxxxiii-lxxxv, on the necromancy of Oedipus: reseranda tellus (‘we must unbar the earth’, Sen. Oed. 395).

51. If Epicurus’ victory in Lucretius itself evoked Alexander, Seneca's double recollection is natural. See further Buchheit (1971), 307-11.

52. On Lucretius see Hardie (1986), 193-200, Gale (1994), 104-10. On the Lucretian spear-cast recalling a Roman ritual declaration of war see Hardie (2009), 175f., and compare Longin. 35.3 τῇ θεωρίᾳ καὶ διανοίᾳ τῆς ἀνθρωπίνης ἐπιβολῆς οὐδ᾽ ὁ σύμπας κόσμος ἀρκεῖ (‘not even the entire cosmos is sufficient for the contemplation and intellection of the thrusting human mind’).

53. Gale (1994), 45.

54. Conte (1994), 5-7.

55. quem neque fama deum nec fulmina nec minitanti / murmure compressit caelum (‘for neither fables of the gods could quell him, nor thunderbolts, nor heaven with menacing roar’, Lucr. 1.68f.).

56. See Gale (1994), 192. On analogy between the material universe and the (inter)textual world in Lucretius cf. Fowler (2000), 148.

57. See n.19 above on (inter)textual liberation; cf. Trinacty (2014), 164, on Sen. Med. 372 and the peruius orbis as suggesting ‘the inherent freedom of Seneca's literary travels’.

58. On the paradox see Fowler (1995), 265, and Mazzoli (1996), 31.

59. Cf. Boyle (2014), 226 on Sen. Med. 369-74 and the severe prohibitions in Roman law on removing ‘sacred’ boundary stones (termini).

60. Fitch (1987), 30.

61. See also Billerbeck (1999), 546, invoking Lucr. 4.961-1036 to discuss the violence of Hercules’ delusions at Sen. HF 1082-87.

62. Tib. 1.10, Virg. Ecl. 4, and Georg. 2 feature prominently in the commentaries of Fitch (1987) and Billerbeck (1999).

63. So Conte (1994), 22-25.

64. Scholarly consensus rather than textual certainty establishes the addressee as Serenus. See Smith (2014), 147.

65. Cf. Gunderson (2015), 85, on the sublime spectatorship presented to Marcia as a perspective to overcome her grief.

66. Cf. Day (2013), 225, on Luc. 8.662-872 and ‘(t)he presence of the one construction within the other, the combination of collapse and grandeur’ and the ‘eventual recognition that, paradoxically, the grandeur of Pompey's grave cannot be separated from its ruinous state.’

67. Williams (2012), 255f.

68. Boyle (1994), 227.

69. So Hardie (2009), 154-57.

70. Lovatt (2013), 103. See further Lovatt (2013), 99-104, on ‘Philosophy and the Divine Gaze’, and 337-41, on the epic limitations of Aeneas, unable to meet Turnus’ eyes at the end of the Aeneid.

71. See Barchiesi (2005), 263 ad loc. making the connection with Longin. 3.3. On literary ambition as a perilous flight and Ovid's Phaethon falling short of epic status see Morgan (2003), 75-79. Cf. also Brownlee (1984), 141f., on ‘the Commedia's programmatic representation of Dante as a “corrected” Phaeton’, one ‘who at every stage of his journey has been led by a guide (and who has been well guided)’ (141).

72. Gildenhard (2004), 32-38

73. Mazzoli (1996), 27-29. Note also the emphasis on spectacle in the passage of De Otio with its Lucretian flight of the mind (Dial. 8.5, discussed p.12 above). Natura… spectatores nos tantis rerum spectaculis genuit, perditura fructum sui, si tam magna… solitudini ostenderet (‘Nature created us as spectators of so great a performance of events, and would waste the fruit of her labour, if she displayed such great works to solitude’, Sen. Dial. 8.5.3f.). Longinus (35.2) advances the same argument and makes this vision the inspiration of divine poetry. Cf. Sen. Med. 992-94, derat hoc unum mihi, / spectator ipse. nil adhuc facti reor: / quidquid sine isto fecimus sceleris perit (‘This was the one thing I was missing, this spectator. I think I have done nothing yet: whatever crime I did without him was wasted’). On Medea as Nature see Littlewood (2004), 169-71.

74. et nitor / percussit oculos lucis ignotae nouus… compressit oculos et diem inuisum expulit / aciemque retro flexit (‘and the strange brightness of the unknown light struck his eyes… he shut his eyes tight to expel the hated light, turned his gaze backwards’, HF 813f., 824f.).

75. Cf. Gunderson (2015), 138, ‘Hercules’ madness contains a vision of the unwriting of his own story. His labors… are tasks that become undone even as they undo Hercules himself.’

76. Lovatt (2013), 317.

77. See further Trinacty (2014), 133-35, on Juno conceiving the tragedy, Hercules Furens, in Virgilian terms, and Schiesaro (2003), 186, on events which repeat for the spectators ‘a masterplot they already know’. On Virgil's epic successors see Hardie (1993). On tragedy's monsters as emblematic of its poetic excess see Cowan in this volume; Staley (2010), 96f.; and Nussbaum (1994), 458-71, on the image of the snake in Medea and in particular ‘(t)he snake-drawn chariot… Seneca's ironic replacement for the Phaedrus’ stirring and beautiful image’ (459).

78. Gunderson (2015), 127-47; quotation at p.131.

79. Cf. Hardie (2009), 163, on Dido engulfed by tragedy: ‘Dido's inability to practise emotional detachment on her citadel mark the beginning and the end of the tragedy of Dido, in whom a number of scholars have seen the figure of a would-be Epicurean who fails to live up to the master's teachings.’

80. Cf. Hardie (2009), 157f., on the determination of Virgil's Georgics not to look beyond mundane limits, at what lies beneath or beyond the earth.

81. On Juno's anger in the Aeneid as resistance to closure see Hershkowitz (1998), 95-124, and Schiesaro (2003), 67-69, on its reception in Senecan tragedy. On the defiance of limits in Senecan tragedy see Curley (1986), 161-68.

82. ‘Farming is a heroic activity, a kind of constructive warfare’ (Low [1985], 8) and at greater length Gale (2000), 243-69.

83. See Harrison (2007), 147-49, on the laudes Italiae and the Georgics’ ‘collusion with politics’ (147) at Georg. 2.165-76. For different blurring of the boundary between didactic and martial epic cf. Barrenechea (2010) on ‘didactic aggression’ and Lucan's Caesar redefining (in conspicuous bad faith) his imperialist uirtus as intellectual enquiry.

84. Gale (2000), 194.

85. Gale (2000), 190-92.

86. On anthropomorphism see Gale (2000), 263f., and on the sublime simile see Mynors (1990), 219 ad loc.: ‘The first line [237] is reused, with significant changes, in a simile, A. 7.528-30 illustrating the growth of civil commotion.’

87. Volk (2002), 126.

88. On such Bacchic poetics see Schiesaro (2009), esp, 66 and 70. The thyrsus resembles a spear—see Fitch (1987), 241, on HF 474.

90. Mynors (1990), 226.

89. Segal (1983), 235f.

91. See e.g. Davis (1991), 87, on Hor. Carm. 2.13: ‘Virgilian echoes [of Georgics 4.481-4]… help to magnify the aura of Lesbian poetry by affiliating its effects with those of the all-puissant Orpheus’, and Lovatt (2007), 148f.

92. See Harrison (2013), 381-83, on ‘a serious and tragic episode from Vergil treated in a lighter way in Horace's lighter genre’ (in Carm. 2.9 and 2.13). Laments for lost love are of course the subject of elegy, which would suggest a different generic placement of Orpheus, but cf. above on tragic/elegiac Philomela.

93. See Littlewood (2004), 160-68, and Trinacty (2014), 154-58.

94. Morgan (1999), 207.

95. Cf. Casali (2011), 89-92, on reminiscence of the Georgics as an element of Lucan's anti-Aeneid.

96. Segal (1966), 321-25, and note especially (322) the opposition of styles which are then fused, complex and unresolved in the narrative of the Aeneid. Gale (1997), 189-91, argues for a similarly unresolved georgic tension in the epic and its eponymous hero.

97. Harrison (2005), 165f. On repetition, corrective or traumatic, in the Aeneid see Quint (1993), 50-53.

98. On Hercules’ victory as a kind of ‘harrowing of Hell’ see Hardie (1986), 115. On Hercules and Orpheus, see McNelis (2007), 28f.: ‘[T]he Virgilian hymn to Hercules (Aen. 8.285-302) recalls the song celebrating Apollo… that is sung by Orpheus in Apollonius’ Argonautica (2.698-713).’

99. Putnam (!995), 256f. and 291f. See also Hardie (1997), 319f., reading the destabilization of Roman panegyric through the influence of Attic tragedy, and in this particular case Aeneid 8 by Sophocles’ Trachiniae.

100. Gildenhard (2004), 38.

101. For reminiscences of Virgil's Hercules in Seneca see Billerbeck (1999), 409f. (on Aen. 8.194f. and HF 601 and 606) and 490 (on Aen. 268-79 and HF 893-939). More generally, see Buckley (2013), 216: ‘Seneca's Hercules does not merely re-frame the master-text of the Augustan age…’ and Trinacty (2014), 136f.: ‘This play can be seen as an expansion of Hercules’ character from Aen. 8’ (n.43, 137).

102. On tree-felling as violent intervention in a literary tradition see Hinds (1998), 11-13. On the landscape imagery of Bacchic inspiration and Hor. Carm. 3.25 see Schiesaro (2009), 61-64.

103. I am grateful to the editors and to the anonymous referees of Ramus for their comments and corrections.