Skip to main content Accessibility help


  • Christopher V. Trinacty (a1)


Conclusions matter in Senecan prose and poetry. At the conclusion of his epistles, Seneca often includes an unexpected quote or alters his subject-matter in a surprising manner—a technique that Fowler has helpfully classified as an example of ‘Romantic Irony’ in the vein of Heine or selected Horatian odes. His dialogues display a similar penchant for such endings, e.g. the post-mortem speech of Cremutius Cordus to his daughter, Marcia, in the finale of the Consolatio ad Marciam (Dial. 6.26.2-7). Seneca's tragedies likewise conclude in a beguiling fashion, ‘Part of the dramatic force of the Senecan ending is its avoidance of any note of easy resolution; it serves rather to sharpen and/or problematize the central issues of the particular play.’ As a way to further encourage the reader to question or recognize major themes of the play, Seneca's conclusions feature an intertext that casts these themes in a different light or elicits metapoetic commentary. These intertexts stress ideas and language important to the particular play, especially those found in the prologue, in order to create a type of ring-composition to the tragedy as a whole. This paper investigates these intertexts and indicates not only how they operate on an inter/intratextual level, but also why Seneca would think of the texts that he does at the conclusion of his tragedies. Seneca looks back to some of his major literary influences at the conclusions of his plays (Ovid, Horace, and Virgil unsurprisingly; Seneca the Elder perhaps more surprisingly), which reveals that these moments are diagnostic of his intertextual method in general. The larger situational or generic context of the sources shade the words uttered by Senecan protagonists, but Seneca stresses the tragic impact of such intertextual echoes again and again; Seneca tragicus surely is a pessimistic reader of the Augustan tradition. The reiteration of similar language and imagery throughout the play ‘primes’ the reader to recognize the intertext at the play's conclusion—thus intratextual repetitions signpost the intertextual reference. Seneca wants these references to be noticed; he promotes a retrospective reading technique in which these intertexts recast language and themes found earlier in the play, now vis-à-vis the literary and rhetorical source material. In creating such dense verbal connections, he encourages further contemplation of the major motifs of the tragedies and inherently endorses the position of his plays as ‘open’ texts that beg for further supplementation by further reading and rereading, again and again.



Hide All

1. Fowler (2000), 273: ‘At the end of the work the writer reveals that he is after all just like us, with all our failings, and thus removes the resistance we feel to his preaching without undermining it’ (citing Ep. 56.14f.). The organization of the Epistulae into books is significant and the endings of each book would help the reader understand the intended closure (cf. Cancik [1967], 138-51).

2. Boyle (2011), ad 1061.

3. Putnam (1995) has proven that Seneca was sensitive to the ending of the Aeneid and that he incorporates notable allusions to it in each of his tragedies. The intertexts I have culled appear within the final 20 lines of their tragedy. Phoenissae does not appear to feature one and this may be further indication of its incompleteness.

4. As Mastronarde (1970), 312, points out à propos the intratextual connections in Seneca's Oedipus: ‘Groups of words with their associated moods and imagery recur with shifts of meaning which reinforce and illuminate other uses of the same word, mood, or image. Some interconnections are fairly definite, but the independent life assumed by individual words in their interplay may add vague and ominous suggestiveness to a seemingly straightforward passage’ (my emphasis). This study adds intertextual connections into the hopper.

5. See Fowler (2000), 243: ‘it seems to do more justice to our intuitions to see the tension between “open” and “closed” as one ever present in the literary work. All works leave things undone as well as done; all great works have that paradox at the core of their greatness.’ Fowler later notes the particular complexity of the end of Euripides’ Heracles including similar intra- and intertextual components, ‘the closure of Heracles may serve as a paradigm of how many elements may be involved in dramatic closure, and how important they may be for the interpretation of the play’ (277). Seneca's Hercules Furens reasserts such complexity.

6. See Boyle (2011), ad loc., and Coffey and Mayer (1990), 195, where a line from the Phaedra rivals this as ‘arguably the worst line in Senecan drama’. Both Cowan and Boyle in this volume remark on the unsettling nature of this line.

7. Boyle (2011), ad loc., cf. Fitch (2004), 154, for parallels of the sexual sense of the line. Boyle (2011), ad 14, also notes the intratextual repetition and Seneca's penchant for this (incidere + in) construction.

8. The only other appearance of a word with fall- as the root is at 119 (fallacis…Parthis).

9. Sampson in the second essay of this volume stresses the novelty inherent in Seneca's emphasis on the problems of kingship at the start of Oedipus.

10. Note that his earlier worries (timeo, 15) about parricide (perimatur, 16) are fulfilled (timui, 1044; peremi, 1045) at the conclusion of the play, further knotting together these two sections.

11. His accidental regnum (14) is ironically recalled by rege (1049).

12. While the Theban books of the Metamorphoses have often been seen as quasi-Oedipal stories, in his Oedipus Seneca offers a reading of the Metamorphoses as a whole in which characters such as Myrrha and Callisto (in addition to Thebans such as Narcissus and Actaeon) are to be seen through Oedipus—an emphatically tragic Oedipus.

13. See O'Bryhim (1990) for more on the pollution of Callisto.

14. See Davis (1983), 11: ‘Within the universe of the Met. “hunting” (uenatio) has a more or less consistent, though extremely complex, set of associations with “loving” (amor).’ Seneca has recourse to such imagery elsewhere in this play from the Bacchae hunting Pentheus (Oed. 432-44) to Actaeon's fate (Oed. 751-63).

15. See DeBrohun in this volume with Boyle (2011), lxxxvi-ii: ‘only in Oedipus does the audience witness a dithyrambic ode to tragedy's god. This play most especially points to itself as verbal, transformed, and transformative construct of theatrical imagination, thought, and passion, and as “rabid” presentation of a theatricalized world.’

16. Cf. Mastronarde (1970), 312, on the Bacchus ode, but pertinent to the play as a whole: ‘The Theban past, its metamorphoses, and its miraculous and tragic encounters with forests, springs, and animals appear wholly uniform with the present; the future may well be the same.’

17. Cf. Staley (2010), 116: ‘The Phaedra makes that sentiment and the metaphor behind it literal when it creates a tragedy out of Hippolytus’ disiecta membra.’ Most (1992) is the locus classicus on the hermeneutics of dismemberment in Silver Latin poetry.

18. Cf. Cic. de Orat. 3.190: tamen ipsa membra sunt numeris uincienda (‘although these sections should adhere to rhythmical unity’) and Ovid's description of Horace as numerosus (‘of many metres’) at Tr. 4.10.49.

19. Of the 11 appearances of forms of fingere in Senecan tragedy, seven are in this play (196, 203, 372, 496, 915, 1194, 1265). Cf. Trinacty (2014), 45f., for more on fingo in Phaedra. Note the use of pono in Theseus’ directions (1257, 1260, 1268).

20. See Gransden (1976), ad loc., for more on his punishment. The mention of Mettus appears at Virg. Aen. 8.642-45 and there are linguistic echoes between Sen. Pha. 1103 (acutis asperi uepres rubis, ‘undergrowth rough with sharp thorn-bushes’), and Virg. Aen. 8.865 (sparsi rorabant sanguine uepres , ‘the thorn-bushes were bedewed with spattered blood’).

21. If the opening of the play encourages the reader to think of a triumph (longo…triumpho, 79f.), so the ending may recall a funeral procession – the triumph and the funeral cortège were two common ‘Roman’ structuring devices (cf. Fowler [2000], 298).

22. In addition to these echoes, the entire passage looks back to Virbius’ narrative in Metamorphoses 15, in which Hippolytus is resurrected as an Italian god. The polyvalence of pars and corpus, which likewise can denote parts of a literary collection, could be activated by Theseus’ claim, cf. Ov. Tr. 2.535: nec legitur pars ulla magis de corpore toto (‘Not any part from his whole corpus is read more’). Note how Theseus employs agnoscere to indicate his ‘reading’ of the remains (laeui lateris agnosco notas, 1260): see Bartsch (2006), 260-62, and Boyle (1997), 112f., for the metapoetics of this reassembly.

23. Cf. Segal (1986b), 218: ‘The father's ultimate inability to fashion or form a complete image of his son's body is also that absence of closure inherent in all tragedy: the residual tension, rifts, and contradictions between the order-imposing coherence of the artistic form and the order-resisting chaos of the experience that such art represents.’

24. Eur. Her 1331-39. Euripides’ version downplays Heracles’ eventual deification and highlights his mortal suffering, friendship with Theseus, and burial concerns (of his children, Megara, and Amphitryon).

25. Cf. Mader (2014b), 127 n.8, for an overview of the scholars who favor these opposing views.

26. Cf. Watson (2003b), 420: ‘But is not Achilles, to whom Chiron in the exemplum proffers similar advice, an unsettling, even inappropriate, choice of paradigm, given that his fate was not susceptible of improvement, doomed as he was to die young in war, a prospect which Horace's companions very possibly face in the immediate future?’

27. Lowrie (1992), 429.

28. Lushkov (2016) finds additional moments in which Hercules’ situation evokes Oedipus and Theban history more generally.

29. Seo (2013) explores this tendency in characterization in Roman poetry.

30. See Littlewood in this volume on such tension.

31. Gunderson (2015), 139.

32. See Tarrant (1976), ad loc.

33. See Gunderson in this volume for more on this passage.

34. See Beare (1951), 259-66, and Manuwald (2011), 69-73, on the stage curtain.

35. ‘Similar fates’ are mirrored in comparatively similar language. Her verbal connections with Thyestes’ speech (ductor ducum, mille, rates) trump his point of view, and reveal her rhetorical and poetic prowess. Note how paria fata also recalls Aen. 10.740f. where the dying Orodes tells Mezentius to expect reciprocal vengeance.

36. Cf. the title of Shelton's (1983) piece in Seneca Tragicus.

37. In addition to the words underlined, fata may recall the paria fata of 1008, and sequamur may suggest the sequatur of 1003. Sequor is a very important verb for the Agamemnon (40, 81, 144, 391, 453, 742, 747).

38. Tarrant (2012), ad loc. He also comments on the doubling of words in the speech as a whole and the fact that the collocation iam iam appears more frequently in Book 12, and notes, ‘a feeling of urgency is apt as the time to the conclusion grows shorter’.

39. Tarrant (2012), ad 680, notes that the figura etymologica of furere…furorem is unique in Latin and his commentary points out additional features of this speech which might make it stand out in Seneca's mind (e.g. the Stoic character of Virg. Aen. 12.677 and its echo of Cleanthes—which Seneca translates in Ep. 107.11).

40. Schmidt (2014), 535, takes Clytemnestra's morere as an implied stage direction: ‘Cassandra…receives the deadly stroke from the queen, and dying, utters the last half line of the drama.’ If so, the intertextual connection also foreshadows Cassandra's immediate (and staged) death, and makes Cassandra Turnus-like in concluding this work with her death.

41. In the Aeneid, Aeneas is said to have been impelled to sail to Italy by Cassandra who is identified by her divinatory ‘fury’ (Italiam petiit…Cassandrae impulsus furiis , Virg. Aen. 10.67f.).

42. See Shelton (1983), 159f.: ‘The play therefore does not leave us with the impression that justice has been served, but rather that the desire to redress a wrong produces a chain of sordid violence. In addition, the inclusion in the play of so many examples of revenge and expiation serves to emphasize the point that vengeance produces instability and disorder, although the avenger ironically believes that he is restoring stability.’

43. Putnam (1995), 272.

44. Eliot (1950), 78f.

45. Ahl (2000), 166-69, comments on Seneca's wordplay with the name Medea: ‘Jason's final exclamation that Medea's actions and escape are proof that “there exist no gods”…is also a denial of the deity in Medea…’ (167). See Fowler (2000), 276 n.147: ‘Emphasis on the literally last word is rare in stage drama, but there is a famous example in tragedy designed for reading, Sen. Med. 1027.’ One should also note how Medea's final words (ego inter auras aliti curru uehar, ’I shall be transported through the breezes in my winged chariot’, 1025) shows the efficacy of her earlier entreaty to the Sun (da, da per auras curribus patriis uehi ,, ‘Grant, grant that I be transported though the breezes in my father's chariot’, 32) and adds to the ring-composition of the first and last word of this play.

46. For more on Severus, cf. Hollis (2007), 340-67. Cf. Trinacty (2009) for more on the intertextual relationship between Seneca the Younger and the Elder.

47. Note, for example, how often the verb fugere (‘flee’) occurs in Fuscus’ diuisio (‘subject headings’): non esse honestum fugere etiamsi tutum esset; deinde: aeque periculosum esse fugere et pugna; renouissime: periculosius esse fugere: pugnantibus hostes timendos, fugientibus et hostes et suos (‘That it is dishonourable to flee even if it is safe to do so; next: that it is equally dangerous to flee as it is to fight.; lastly: that it is more dangerous to flee: for those who are fighting it is the enemy who is to be feared, but for those who are fleeing it is both the enemy their own people’, Sen. Suas. 2.11). This fight-or-flight question can be seen throughout Seneca's Medea as well, with Medea, characteristically, finding a way to do both (non ibo in hostes?, ‘shall I not go against my enemies?, 27; sic fugere soleo, ‘thus am I accustomed to flee’, 1022).

48. In some sense, this is her last day of human life, as she has removed (extraham, Sen. Med. 1013) any remnants of her humanity just before her final apotheosis. I owe this observation to one of the anonymous readers of Ramus.

49. Note that Medea says of herself that she is using the time that was allowed her (tempore accepto utimur, 1017).

50. Seneca points to the development of Medea's character as an avenger as the language of 399 hearkens back to her earlier self-affirmation: ‘now end your inactive delays: the house that was won by crime must be left by crime’ (rumpe iam segnes moras: / quae scelere parta est, scelere liquenda est domus, Sen. Med. 54f.).

51. Note also the possible intertext with Euripides’ Medea, as the chorus contemplates the various evils that have befallen Jason τῆιδ’ἐν ἡμέραι (Eur. Med. 1231), cf. Eur. Med. 1248 as well.

52. See Fairweather (1981), 312f., for occasional comments on the Elder's admiration for Virgil and the influence of declamation on contemporary poetry. Rolland (1906), 65, details the additional moments of this phrase in Seneca's tragedies. Cf. McGill (2012), 167-77, for further discussion of this passage.

53. See Hollis (2007), 330f., for a discussion of his life and poetry.

54. Cf. Ag. 211: non sola Danais Hector et bello mora (‘not Hector, sole delayer of Danaans and the war’); Phoen. 458: proinde bellum tollite aut belli moram (‘so discard either war or the delay of war’. The phrase also occurs in Ovid (Met. 8.21), Lucan, and several prose authors. Cf. Trinacty (2009), 273f., and Berti (2007), 279-82.

55. None of the commentaries I have consulted mentions this as a parallel for this passage.

56. See Boyle (1994), ad 124, for more on delay in Troades, and Fitch (2004), 64, who elucidates, ‘The obstacle (mora) to her death which she will “spew out” is her spiritus; cf. Pho. 44 spiritum inimicum exspue, Thy. 245 spiritum inimicum exspuat. In the context the phrase perhaps also suggests her spewing-out of laments about the belatedness of her death…’

57. No other play features the repeated anaphora of ite, ite, although it does appear once at HF 1131-32, and Med. 845.

58. Ovid's tale of Procne, Philomela and Tereus wraps up fifteen lines after Philomela throws the head of Itys at Tereus (Met. 6.659), and Shakespeare's Titus Andronicus only survives four lines after revealing the Chiron-and-Demetrius pie.

59. Tarrant (1985), ad loc.

60. For uindex applied to the Furies, cf. Cic. N.D. 3.46: Furiae deae sunt, speculatrices credo et uindices facinorum (‘the Furies are goddesses, and as I believe watchers foro and avengers of criminal acts’); Stat. Theb. 1.80: tu…debita uindex [i.e. Tisiphone] huc ades (‘you…my owed avenger [Tisiphone], be present to us here’. Dodson-Robinson (2012), 60, points out how Atreus’ crime can be seen, in part, as ‘determined by the infecting contagio of Tantalus’ ghost…it is Tantalus, the primal transgressor, who has defined Atreus’ identity, and Thyestes rightly recognizes a family resemblance.’ In addition, Schiesaro (2003), 104, examines how Atreus manifests the power of the Fury: ‘the furor that inspired [Atreus’] actions thus far is now also explicitly presented as a viable source of rational understanding’.

61. See lines 9, 15, 74, 82, 86, 87, 92, 151, 246, 279. The discussions Atreus and Thyestes have near the play's conclusion about the nature of scelus (1051-53, 1095-98, 1103f.) recall the Fury's earlier claims about scelus.

62. See Schiesaro (2003), 152: ‘Thyestes and the chorus display a firm belief (at least, a firm discursive belief) in their [the gods’] existence and importance, while Atreus effectively disposes of them, and replaces them with himself.’

63. Note Thyestes’ earlier cry: ‘if I, the father, desire to bury my sons and give them over to the funeral fire, I must be cremated’ (si natos pater / humare et igni tradere extremo uolo, / ego sum cremandus, Sen. Thy. 1090-92). In addition, the ghost of Tantalus at the play's opening hopes to prevent the crime (arcebo scelus, 95) and does not want to persecute his grandchildren (89f.).

64. See Berti (2007), 322-25, and Contr. 1.1.21 (Thyesteo more). Van Mal-Maeder (2004), 191-94, points out the influence of Seneca's Thyestes in pseudo-Quintilian's major declamations.

65. Gunderson (2003), passim.

66. Goldberg (1996), 277.

67. Goldberg (1996), 283 (my emphasis).

68. Hammond (2009), 112-20, focuses on the way that linguistic boundaries are troubled in the play.

69. Eliot (1950), 54.

70. Cf. Ep. 80.7: ‘this mime show of human life, which assigns parts for us to play badly’ (hic humanae uitae mimus, qui nobis partes quas male agamus adsignat) and the discussion of those who repeatedly change ‘masks’ at Ep. 120.22. Cf. Ker (2009), 115-19, and Edwards (2007), 144-60.

Related content

Powered by UNSILO


  • Christopher V. Trinacty (a1)


Full text views

Total number of HTML views: 0
Total number of PDF views: 0 *
Loading metrics...

Abstract views

Total abstract views: 0 *
Loading metrics...

* Views captured on Cambridge Core between <date>. This data will be updated every 24 hours.

Usage data cannot currently be displayed.