1. Derrida (1981), 287.

2. I would like to thank the editors for their thoughtful, engaged, and productive comments.

3. See also Owen (1968) on the problem of finding Seneca's originality amidst his traditional aspects. The secret, he says, can be seen in Moral Letter 84 and the story of the way bees gather pollen in order to make honey. Trinacty (2014), 13-16, argues similarly. See also Henderson (2004), 46-48, on this same letter.

4. One will note, though, that this essay does not offer a specifically rhetorical exploration of Seneca's tragedies. See instead something like Wray (2009).

5. A single example: geminatio is a stylistic feature. See Cic., De or. 3.206 (which is itself repeated verbatim at Quint., Inst. 9.1.33). Compare Inst. 9.3.28f. and 9.3.41-45. Next, the same thing, but differently: geminatio is a (criminal) feature of Seneca's style and emblematic of the epigrammatic (artifice-ial) quality of Seneca's dramaturgical productions. See Quint., Inst. 8.5.18: facit quasdam sententias sola geminatio, qualis est Senecae in eo scripto quod Nero ad senatum misit occisa matre, cum se periclitatum uideri uellet: ‘saluum me esse adhuc nec credo nec gaudeo’ (‘A number of sound bites are yielded by doubling alone. Take for example Seneca's epigram in that letter Nero sent to the senate in the wake of his mother's murder where the emperor wanted to give the impression that he himself had been in danger: “Neither do I believe nor do I rejoice that I am safe”’). See also Boyle (2011) on iam iam at Oed. 28: ‘emotional duplication (geminatio), often at the beginning of a line of verse, is a feature of Seneca's tragic style: e.g. …’ Boyle then gives nine examples of iam iam as well as numerous examples of still more doubles such as nunc nunc, me me, and ite ite.

6. Littlewood (2004), 6: ‘As any commentary shows, Senecan tragedy is stitched together from lines of Virgil and Ovid: this is its primary material.’ Stronger still is Schiesaro (2003), 223: ‘Seneca's tragedies stem from a continuous, even obsessive confrontation with their [tragic] models.’ Note that Littlewood's primary source of repetitions is not quite the same as Schiesaro's. Trinacty (2014) offers a book-length examination of Seneca's intertextual relationship to his antecedents.

7. Tarrant (1976) on Ag. 15: ‘Seneca's descriptions of hell invariably include a catalogue of the famous sinners.’ Tarrant (1976) on Ag. 181: ‘The ode on mutability is a hall-mark of Senecan tragedy.’ He lists examples from five different plays.

8. See Trinacty in this volume for a discussion of Seneca's intratextuality and its relationship to intertextuality. See especially his observation on the manner in which Virgil's epic monologue turns into dramatic dialogue at the end of Seneca's Agamemnon. Meanwhile Sampson's essay on misdirection points out that repetition can be more important for the differences it enables. Raising expectations and then subverting them can also be part of the game. Littlewood makes a convergent point as he closes his essay in this volume: repetition can be but a prelude to upstaging.

9. See Boyle (1997), 34: the two plays repeat one another abundantly at a structural level and yet have an ‘ideological distinctness’ and occupy ‘quite different worlds’.

10. Boyle (1997) argues this at length in his fifth chapter, ‘The Palimpsestic Code’.

11. Tarrant (1985), 86: ‘The scene has no obvious model in extant Greek or Roman tragedy.’ The moment indirectly affirms Trinacty (2014), 4f.: Seneca's innovations arise out of rhetorically inflected returns to the literary past.

12. Derrida (1981), 291.

13. reducem expetito litori impressit pedem (‘He planted his returning step upon a longed-for shore’, Ag. 401). And feet themselves return to furnish key images in the exchange between Agamemnon and Clytemnestra (810-950). See also Ginsberg in this volume for a discussion of Polynices’ (multiply) Oedipal repetition, return, and error in the Phoenissae.

14. But see as well Garelli-François (1998), 18, on Eurybates as emblematic of Seneca's innovative relationship to the motif of the messenger.

15. This verbal repetition is itself repeated later both when the Chorus uses iuuat twice in rapid succession (Ag. 664 and 666) and then again when Cassandra raves (Ag 750, 751). And, finally, Cassandra herself repeats this same repetition of iuuat in the penultimate line of the play (iam, iam iuuat uixisse post Troiam, iuuat [‘Now, now I am glad to have lived beyond Troy, yes, glad’], Ag. 1010). The variation lies in taking only one line to do what had formerly taken two.

16. (Eurybates) iuuat uidere nuda Troiae litora, / iuuat relicti sola Sigei loca (‘We delight to look upon the denuded shores of Troy. We delight to see the desolate landscape of an abandoned Sigeum’, Ag. 435f.).

17. Ag. 446: bella narrat. This is itself a retelling of Virgil's arma canere.

18. Hectoris fortis minas / currusque et empto redditum corpus rogo, / sparsum cruore regis Herceum Iouem (‘The threats of brave Hector, / his chariot, his body restored and consigned to the pyre at a price, / Hercean Jove splattered with a king's blood’, Ag. 446-48).

19. Dolphins themselves are ‘programmatic’ within the poetic tradition, and they evoke dithyrambic choral dance. See Csapo (2003). And this dancing is itself associated with the motif of rebirth. See Kowalzig and Wilson (2013), 11.

20. tunc qui iacente reciprocus ludit salo / tumidumque pando transilit dorso mare / Tyrrhenus omni piscis exultat freto / agitatque gyros (‘In a calm sea this way and that they frolic / and with broad back leap through the swelling wave: / dolphins spring forth everywhere across the main. / They set spirals spinning…’, Ag. 449-52). Tarrant (1976) on reciprocus: ‘An extremely rare word in poetry, and far from common in prose.’ ‘Ludic’ dolphins also appear in Livius Andronicus’ Aegisthus fr. 2, tum autem lasciuum Nerei simum pecus / ludens ad cantum classem lustratur (‘Then the sportive snub-nosed flock of Nereus / frolicked in time to the music and circled the fleet’).

21. et comes lateri adnatat, / anteire naues laetus et rursus sequi; / nunc prima tangens rostra lasciuit chorus, / millesimam nunc ambit et lustrat ratem (‘They swim alongside as our companions. / They delight to outstrip the fleet and then again to follow it. / Now touching the fleet's first prow they frolic as a chorus, / now they circle and review the thousandth ship’, Ag. 452-55).

22. tum murmur graue, / maiora minitans, collibus summis cadit / tractuque longo litus ac petrae gemunt; / agitata uentis unda uenturis tumet (‘Then a heavy low noise: / more is menaced as it tumbles down from the highest hilltops. / The shore's long expanse and the crags groan. / Waves swell, set in motion by the winds to come’, Ag. 466-69).

23. uento resistit aestus et uentus retro / aestum reuoluit (‘Wave's swell struggles against the wind and back again the wind / wheels the wave’, Ag. 488f.).

24. Ag. 497f.: and compare redditam reuomit (‘regurgitates the restored’) in line 500.

25. Art set against storms is Ovidian: see Tarrant (1976), ad loc.

26. Tarrant (1976) compares Hom. Od. 5.306ff. and Virg. Aen. 1.94ff. to Seneca's 512ff. Pypłacz (2010), 54: Seneca will exaggerate each of the elements of his models.

27. See Ag. 561: feruetque semper fluctus alterna uice (‘Ever seethes the swell moving now here, now there’). Compare Ag. 564: Isthmon, arto qui recuruatus solo… (‘Isthmus, which bent back with a narrow band of soil…’).

28. The translation is from Tarrant (1976).

29. pars uehitur huius prima, pars scopulo sedet; / hanc alia retro spatia relegentem ferit / et fracta frangit (‘This ship sees the foremost part of itself sail off while part sticks to a rock. That ship strikes another which reverses and retraces its course: / broken it breaks’, Ag. 573-75).

30. On this collocation, see also Trinacty and Sampson in the first article of this volume.

31. Virg. Aen. 2.160. See Tarrant (1976), 283.

32. sed ecce gemino sole praefulget dies / geminumque duplices Argos attollit domus (‘But behold the day resplendent with a twin sun. / And a twin Argos raises aloft domains double’, Ag. 728f.).

33. et solem geminum et duplices se ostendere Thebas (‘[who saw] a twin sun and Thebes display itself doubled’, Virg. Aen. 4.470). The scene describes lovesick Dido who is repeating the experience of Pentheus (ueluti demens Pentheus 4.469). And Virgil has obviously asked us to note that he has himself reproduced and modified Euripides’ Bacchae 918f.: καὶ μὴν ὁρᾶν μοι δύο μὲν ἡλίους δοκῶ, / δισσὰς δὲ Θήβας καὶ πόλισμ’ ἑπτάστομον (‘Indeed I think I can see two suns / and Thebes with its seven mouths doubled’).

34. Compare Boyle (1997), 112: ‘Senecan tragedy contains its own autobiography: past and future.’

35. See Trinacty (2014), 199-214, for a close reading of the manifold details.

36. spectate, miseri: fata se uertunt retro (Ag. 758). Fata se uertunt retro receives its own treatment in the fifth chapter of Schiesaro (2003), a discussion of dramatic time in Senecan tragedy: ‘By creating a framework in which certain sections of the play appear to revolve back to a point in time that has already been treated, and by substituting iteration for linearity, these tragedies make repetition an essential modality of tragic representation’, 186). Compare Motto and Clark (1985) on the play's ‘backwardness’.

37. Tarrant (1976) on Ag. 759ff.: ‘The Furies whose approach is imagined by Cassandra have little direct connection with the avenging spirits of Aeschylus’ Choephoroe and Eumenides and Euripides’ Orestes, and are based instead on the picturesque descriptions of earlier Latin poets, in particular that of Virgil's Allecto.’

38. I hope that I will be forgiven for side-stepping the debate over whether or not these plays were ever staged. I believe that a notional performance works just as well as an actual one for the points that I am making.

39. This motif of back-and-forth is also replayed not just internally to the house of Atreus but also externally for the Trojans who are the doubles of the Argives. Cassandra's vision of Agamemnon's death is celebrated with exclamation, ‘Conquered, we conquer!’ (uicimus uicti Phryges, Ag. 868).

40. On the irony of the moment, see Shelton (1983), 173. See also Trinacty in this volume for more on iam iam.

41. Ag. 1111.

42. On the chorus itself, see Aricò (1996) and Shelton (1983), 170f.

43. See Ferri (2003), 53.

44. Derrida (1981), 316.

45. See Ginsberg in this volume.

46. Different-but-same: ‘Which creature walks on four legs in the morning, two legs in the afternoon, and three legs in the evening?’

47. carmen poposci (‘I demanded the carmen’, Oed. 98). Boyle (2011) on line 98: carmen is likewise a common word for verse prophecy. See Hyginus and Ovid. And compare Sophocles as well. See also the discussion of Medea’s carmen in Trinacty and Sampson in the first article. In case we missed it, carmen is repeated a few lines later at Oed. 102: ac triste carmen alitis solui ferae (‘I unraveled the baleful riddle of the wild winged beast’). Cf. Boyle (2011) on line 102: triste carmen also means ‘tragic song’.

48. See Curley (1982), 100f., on the universal generalizability that one ought to recognize when pondering the Sphinx-and-Oedipus.

49. Move on to to Braund's essay in this volume and observe the love of anaphora on Garnier's part. And note especially Garnier's repetition of being-and-nothingness as this Oedipal child of Seneca tragicus writes up the story of the house of Laius: C'est.. / Ce n'est pas… / Ce n'est pas… Meanwhile the line ends sound the same even as they attempt to draw distinctions: horreur vs. erreur: error is not horror, one hopes. And, disturbingly, we hear and feel a connection between pense and imprudence: thought is not folly, one prays.

50. I am using ‘fate’ rather loosely. It is possible to read the play with a much more technical relationship to the question of fate. See Davis (1991) who intervenes in a well-established area of inquiry.

51. On the irrelevance of moral culpability within the play see Müller (1953), 459f. On Senecan nefas see the third part of Dupont (1995). On Oedipus’ guilty conscience at the opening of the play see Boyle (1997), 92f.

52. inter ruinas urbis et semper nouis / deflenda lacrimis funera ac populi struem / incolumis asto - scilicet Phoebi reus (‘Amidst a city's ruins and ever renewed / tears shed over corpses in need of lament and a people become a confused heap / I stand here unharmed: assuredly Apollo's culprit’, Oed. 31-34). Note Oedipus’ sense that he is somehow exceptional and standing apart from the iterative process (albeit with tears in his eyes). Oedipus is very keen on his exceptional status. Boyle (2011) on Phoebi reus: the genitive is pointedly ambiguous, ‘defendant before Phoebus’ and/or ‘guilty of Phoebus/fate’.

53. Oed. 131f. See Caviglia (1996), 90-92, on the role of the chorus at this juncture.

54. Boyle (2011) on lines 166-70: ‘Again Seneca's cue is Virgilian—this time Virgil's famous description of Charon at Aen. 6.302-05… Seneca's cumba (166), senio (168), crudo (168), conto (169), turbam (170), uectare (170) are verbal debts to this passage. Cf. also HF 764-68.’ We can also note that the underworld of the HF repeats the repetitions of the Oedipus in more ways than one. See the notes on Oed. 180f.

55. Oed. 180f.

56. Boyle (2011) on lines 180f.: ‘Seneca seems fond of this kind of polyptoton—morte peor… mortis locus, “place of death worse than death”, HF 706; peior… bello timor… belli…, “fear or war worse than war”, Thy. 572… On the “novelty” of plague, see Thuc. 2.47.3, Lucr. 6.1125. For the conceit, cf. Ov. Tr. 1.2.51 and Sen. Tro. 783: o morte dira tristius leti genus, “O form of death more grim than dire death”.’

57. Derrida (1988), 20: ‘Effects of signature are the most common thing in the world. But the condition of possibility of those effects is simultaneously, once again, the condition of their impossibility, of the impossibility of their rigorous purity.’

58. ambigua soli noscere Oedipodae datur (‘To Oedipus alone is the understanding of riddles given’, Oed. 216). This is how he asks for the contents of the oracle: ‘Make it a riddle, so I alone can figure it out.’

59. Derrida (1987), 354f.: ‘[T]he Umweg is not a derivative type of path or step. It is not a passing determination, a narrower or stricter definition of the passage, it is the passage. (The) Weg (is) Umweg from the first step of the step [pas du pas].’

60. The inverted message is my message: a maxim much repeated by Lacan. See, for example, how he opens the Écrits by saying it yet again (Lacan [2006b], 3f.). New readers will be hearing this for the first time, if, of course, they are able to hear it at all.

61. Boyle (2011), 186, on lines 291-402: ‘This extraordinary scene is one of Seneca's invention and owes little to the Oedipus-Tiresias scene of Sophocles… [W]e are given one of the most surprising scenes in extant ancient drama, involving a Roman sacrifice and a ritual inspection of animal entrails.’ But, as Dupont (1995), 202, notes, what one sees is unintelligible within the traditions of Roman sacrifice. A multiple break, then. See also DeBrohun in this volume.

62. Boyle (2011) on line 350.

63. See also Poe (1983), 146, on the emblematic quality of the line.

64. Boyle (2011) on mutatus ordo est in 366: ‘a reprise of Lucretius’ description of the arrangement of atoms.’ Lucretius: wrong philosophical sect; wrong genre; wrong context. And yet perfectly fitted to present purposes.

65. Boyle (2011) on acta retro cuncta in 367: ‘this focus on “reverse” picks up on Apollo's reuolutus (238) and anticipates Oedipus’ retro reuersas (870).’

66. Boyle (2011) on these lines: ‘[N]ote the complex, alliterative patterning of the original… The alliteration is reinforced in the Latin by quadruple homoeoteleuton.’

67. On the Sphinx as truly ‘his own’: see Mastronarde (1970), 300-05.

68. See also Cowan's meditations on Senecan ‘moreness’ in this volume.

69. See Derrida (1981), 322.

70. dehisce, tellus, tuque tenebrarum potens, / in Tartara ima, rector umbrarum, rape / retro reuersas generis ac stirpis uices (‘Gape, earth, and you, master of the shadows, / to deepest hell, lord of the shades, snatch me, / a perverse, inverted interchange of birth and blood’, Oed. 868-70). Boyle (2011) on these lines: ‘Oedipus’ initial request is that he be plunged into hell … There is also a suggestion of a “return to the womb”: see below. The explosion of alliteration and assonance in the lines enhance their declamatory power.’ See Schiesaro (2003), 196, on the manner in which dehisce tellus itself appears thrice in Seneca's plays.

71. illa quae leges ratas / natura in uno uertit Oedipoda, nouos / commenta partus, supplicis eadem meis / nouetur (‘That Nature who warps her fixed laws in Oedipus' case alone, a nature who contrived unparalleled [nouos] births, let that same nature be revolutionized [nouetur] for my punishment’, Oed. 942-45). Boyle (2011) on nouos: ‘Seneca seems fond of this sarcastic use of nouus.’ He then lists six parallels from a variety of plays.

72. uiolenta Fata et horridus Morbi tremor, / Maciesque et atra Pestis et rabidus Dolor, / mecum ite, mecum. ducibus his uti libet (Oed. 1059-61).

73. Contrast the Stoicizing interpretation of libet at Henry and Walker (1983), 136.

74. Compare Schiesaro (2003), 189, on non-closure in Seneca's Thyestes: ‘To deny closure means that everything will happen again and again, that regression will know no end.’

75. Derrida (1981), 338.

76. Crébillon (1812), 172: ‘A plan so fell, / if it is not worthy of Atreus, it is worthy of Thyestes.’

77. Poe (1884), 416.

78. Lacan (2006a), 9. But there one sees un destin (‘a fate’) instead of un dessein (‘a plan’).

79. Derrida (1987), 494.

80. Polysemy and the myth of authorial intentions: Derrida (1981), 350. Conversely the heterogeneity of writing really does offer a way forward (Derrida [1981], 356).

81. Contrast the theodicy of orthodox Stoic repetition at Long and Sedley (1987), 52: the best happens over and over again. On the one hand, Stoic readings such as Staley (2010) continue to yield productive accounts of Seneca. But, on the other, the Stoic elements within the plays have always also been pointedly warped. See Gunderson (2015), or, much more briefly, Lefèvre (1981).

82. A logic that has an analogue in Bloom's parricidal ‘anxiety of influence’ (Bloom [1997]). And I am resuming while modifying a line ‘first’ opened up at Schiesaro (2003), 223: ‘The pervasive characteristic of Seneca's tragedies is their belatedness: they represent an anachronistic return to the past, a frustrated desire for lost forms mediated by an overwhelming and oppressive intertextual memory.’ See also Schiesaro (2003), 224: ‘If all post-Ovidian literature is programmatically self-conscious to a very high degree, Seneca's own narcissism takes the form of a sustained critique of authorial responsibility as it is showcased in the author's staged counterparts—a group of obsessed, determined criminals.’

83. Klossowski (1997), 67 [original emphasis].

84. Klossowski (1997), 67 [original emphasis].

85. Seneca's Moral Letter 79 offers a portrait of the advantages of coming last to the literary banquet. See Trinacty (2014), 10-13: the letter's very language for figuring innovation amid tradition is itself a reworking of that tradition. And see also the preface to the present volume on the possibilities for writing in the first century CE.

86. Klossowski (1997), 69. I suspect that not everyone will recognize ‘his’ or ‘her’ Nietzsche in this formulation, but I wish to stick with it nevertheless as a means of reading Seneca.

87. Many will have noticed that I am repeating but altering Hinds’ account of tendentious repetition within Latin verse. See the fourth chapter of Hinds (1998).

88. Freud (2001a), 245: The unheimlich is secretly familiar (heimlich-heimish).

89. Freud (2001a), 241: ‘[T]his uncanny is in reality nothing new or alien, but something which is familiar and old-established in the mind and which has become alienated from it only through the process of repression.’

90. Freud (2001a), 222, and compare Lacan (2006a), 34.

91. See Henderson (1983) on the programmatically violent sound-effects when Seneca turns the amplifier up to 11 in the Medea. A productive poetic feedback-explosion occurs.

92. This topic has a rich Lacanian bibliography. See, for example, Žižek (2006), 17-18. And all of this can be traced back to Freud's Beyond the Pleasure Principle (Freud [2001b]).

93. Freud (2001a), 238: ‘[W]hatever reminds us of this inner “compulsion to repeat” is perceived as uncanny.’

94. Deleuze (1994), 15.

95. See also the final two chapters of Gunderson (2015).

96. Derrida (1987), 492.