Skip to main content

Ovid's Canace: Dramatic Irony in Heroides 11*

  • Gareth Williams (a1)

Heroides 11 has long enjoyed a favourable reputation among critics, largely because Ovid appears to show a tactful restraint in his description of Canace's last moments and to refrain, for once in the Heroides, from descending into what Jacobson terms ‘nauseating mawkishness’. Despite appearances, however, Ovid's wit is not entirely extinguished in this poem, for a devastating irony accompanies the certainty of Canace's imminent death. My objective is to demonstrate the nature of this irony by adopting a methodological approach which owes much to Kennedy's analysis of Heroides 1 in the light of the later books of the Odyssey. Kennedy's argument – that without knowing it Penelope is about to give her letter to its intended addressee – is based on two premises which are postulated by the epistolary mode of the poem. The first is that we are to imagine Ovid's heroines writing at a specific moment within a dramatic context; the second is that they have a specific motive for writing at that moment. In Kennedy's hands, this approach assumes the privileged position of the reader of Heroides 1 who, through access to the Odyssey, is alive to the ironies which Ovid's Penelope cannot realize. I propose to establish a similarly privileged position for the reader of Heroides 11, a position from which Canace's death can be seen to be both ill-timed and unnecessary. The key to identifying the ironic circumstances of Canace's death lies in reconstructing the background to the Canace and Macareus myth and the possible precedents which Ovid drew on in his treatment of the story. The situation is more complex than in the case of Heroides 1, however, since the literary sources familiar to Ovid and his readers have, in this instance, largely been lost to us and can only be reconstructed from fragmentary evidence.

Hide All

1 Jacobson H., Ovid's Heroides (Princeton, 1974), p. 159.

2 Kennedy D. F., ‘The Epistolary Mode and the First of Ovid's Heroides’, CQ 34 (1984), 413–22.

3 Jacobson, op. tit., pp. 160–1 and nn. 7–9 conveniently summarizes scholarly speculation on Ovid's possible sources (Hellenistic as well as Euripidean).

4 Frr. 14–41 N2. On the fragments and their possible order in the play see Jä kel S., ‘The Aiolos of Euripides’, GB 8 (1979), 101–18.

5 First published by Turner E. G., Rea J., Koenen L. and Pomer J. M. Fernández, edd., The Oxyrhynchus Papyri: Part XXVII (London, 1962), pp. 70–3 (henceforth Turner, op. cit.). See Lloyd-Jones H., Gnomon 35 (1963), 443–4 and Austin C., Nova Fragmenta Euripidea in Papyris Reperta (Berlin, 1968), pp. 88–9 for, respectively, a review and update of Turner's text.

6 See RE 10.1853 s.v. Kanake 1 with Hopkinson N., ed., Callimachus: Hymn to Demeter (Cambridge, 1984), p. 31 and n. 1.

7 Aeolus, ruler of the winds, is the son of Hippotes (cf. Homer, Od. 10.2), while the other Aeolus is the son of Hellen (cf. Hesiod, Eoiai fr. 9 Merkelbach–West); for more references see RE 1.1036–7 and 1039. There is nothing in the hypothesis or extant fragments to suggest that Euripides made a dramatic or poetic point of the conflation.

8 For this dating and brief discussion see Turner, op. cit., p. 71.

9 For basic discussion of the different kinds of tragic hypothesis which survive see Zuntz G., The Political Plays of Euripides (Manchester, 1955), pp. 125–52, where three main categories are distinguished: the hypotheses of Aristophanes of Byzantium (but cf. Brown A. L., ‘The Dramatic Hypotheses Attributed to Aristophanes of Byzantium’. CQ 37 [1987], 427–31 on the possible misattribution of hypotheses assumed to be by Aristophanes), those of often pedantic and verbose Byzantine grammarians, and those which Zuntz terms ‘Tales of Euripides’, hypotheses which give a bare account of a play ‘for the use of readers interested in mythology rather than in poetry’ (op. cit., p. 135). Pap. Oxy. 2457 belongs to the latter category; on Dicaearchus as the author of this and the other Euripidean hypotheses on the same papyrus roll see Haslam M. W., ‘The Authenticity of Euripides, Phoenissae 1–2 and Sophocles, Electro 1’, GRBS 16 (1975), 150–6; but cf. Rusten J., ‘Dicaearchus and the Tales from Euripides’, GRBS 23 (1982), 357–67, who argues against attribution to Dicaearchus. I am grateful to an anonymous referee for drawing my attention to these items of bibliography.

10 I discount the Rhesus because of its disputed authorship, on which see Knox B. M. W., The Cambridge History of Classical Literature, Vol. 1. Greek Literature, edd. Easterling P. E. and Knox B. M. W. (Cambridge, 1985), pp. 342–3.

11 Coles R. A., A New Oxyrhynchus Papyrus: The Hypothesis of Euripides' Alexandros, BICS suppl. 32 (1974).

12 Op. cit., p. 32.

13 Op. cit., pp. 66–70.

I read ἧκʼ (‘loosed’, ‘let fly with’) with Sommerstein in preference to ἦγʼ (Dover/Borthwick, ‘began’) or the ἦσʼ (‘sang’) of the MSS. The scholiast comments γγραπται δρμα Εὐριπδῃ οὕτω λεγμενον, Αἴολος· ν ᾧ παργαγε τν παῖδα Αἰλου Μακαρα ϕθεροντα Κανχην τν δελϕν (Koster I 31 p. 241). Aristophanes parodied Euripides' play in his Aeolosicon (PCG III. 2, pp. 33–41). In describing the violent and lustful effects of drink, Athenaeus (444c–d) cites Antiphanes' parody of the Aeolus: according to Antiphanes, Macareus took advantage of his sister (ἔτυχεν ὧν βολετο, fr. 18.6 Kock) in a drunken act of violence.

15 Ars Rhetorica 9.11 (ii. p. 345 Usener–Radermacher): ν τῷ Aἰλῳ Mακαρες στιν μιλσας τῇ δελϕῇ κα λανθνων, κα συμβουλεων τῷ πατρ τς δελϕς τοῖς δελϕοῖς συνοικσαι, ἵνα τ οἰκεῖον διοικσηται.

16 For the vase painting, dated to c. 415 b.c., see Trendall A. D. and Webster T. B. L., Illustrations of Greek Drama (London, 1971), p. 74 with plate III. 3.4.

17 Cf. Jacobson, op. cit., p. 160 n. 3: ‘It is very difficult to draw conclusions about the literary tradition from the iconographic one’. See also Pickard-Cambridge A. W., ‘South Italian Vase Paintings and Attic Drama’, CQ 43 (1949), 57, who disputes Webster's T. B. L. claim (CQ 42 [1948], 15ff.) that South Italian vases of the mid fourth century are evidence of theatrical arrangements in fifth-century Attic drama. His first objection might extend to the Lucanian vase painting depicting Canace: ‘A number of scenes on the vases are not scenes presented in the plays at all, but are scenes suggested to the painter by descriptions in messengers' speeches, or, quite possibly, by the story dramatized in the play, but not by the Athenian poet's particular treatment of the story.’

18 Moralia 312c–d = Parallel. Min. 28a: Aἴολος, τν κατ Tυρρηναν βασιλες, ἔσχεν ξ Ἀμϕιθας θυγατρας ἓξ κα ἴσους ἄρρενας. Mακαρεὺς δ νεώτατος ἔρωτι ἔϕθειρε μαν, δ παιδον κησεν· τεκοσα δ κα ξϕονς πεμϕθντος ὑπ το πατρς, ἄνομον κρνασα αντν διεχρσατο· μοως δ κα Mακαρες· ὡς Σώστρατος ν δευτρᾳ Tυρρηνικν. Cf. Hyginus, Fabulae 238.3, where Aeolus is listed under the heading ‘Qui filias suas occiderunt’, though we are to imagine that Aeolus had Canace kill herself; Hyginus reports Canace's suicide at 243.6, Macareus' at 242.2.

19 Florilegium 4.20.72 Wachsmuth = F. Jacoby, FGrHist 23 fr. 3: Σωστρτου ν δευτρῳ Tυραννικν· Aἴολος τν κατ Tυρρηναν βασιλεὺς τπων ἔσχεν ξ Ἀμϕιθας θυγατρας ἓξ κα τν αὺτν ριθμν ρρνων παδων, ὧν πρεσβτατος Mακαρεὺς Kανκης τς δελϕς ρασθες βισατο τν προειρημνην. Aἴολος δ περ τοτων μαθὼν τῇ θυγατρ ζϕος ἒπεμψεν· δ ὡς νμον δεξαμνη τν σδηρον αὑτν νεῖλε· Mακαρεὺς δ τν γεννσαντα προεξιλεωσμενος ἔδραμεν εἰς τν θλαμον· εὑρὼν δ τν γαπωμνην αἱμορραγοσαν τῷ αὐτῷ ξϕει τν βον περιγραψε.

20 Palmer, op. cit., p. 381.

21 Op. cit., p. 381.

22 There is no need to infer from lines 101–8 that Canace has knowledge of the marriage arrangement which Macareus makes with Aeolus. So Jacobson, op. cit., p. 161: ‘The marriage torches need reflect no more than Canace's frustrated desire for marriage, especially if we consider how frequently in the Heroides the presence of (wedding) torches is imagined (e.g. 6.45–6). Further, the wish expressed as an imperative “nubite felices Parca meliore sorores” may have reference to a distant and unseen future, a simple wish that her sisters fare better in “marriage” than she has.’

23 Diggle J., ed., Euripides: Phaethon (Cambridge, 1970), pp. 182–4, 186 and 200.

24 Cf. Jacobson, op. cit., p. 110: ‘Ovid incorporates the quite different Medeas of Apollonius and Euripides into a character of his own creation.’

25 See Jacobson, op. cit., pp. 142–6 for a demonstration and discussion of the different sources, Euripidean and Sophoclean, detectable in Heroides 4.

26 The inference is supported if we accept the commonly held view that the author of the Tyrrhenica is the grammarian Sostratus of Nysa; for the identification see RE 3A 1200–1 s.v. Sostratos 7. Sostratus of Nysa was one of the city's ξιλογοι…γραμματικο (Strabo 14.1.48); his father was Menecrates, a pupil of the great Aristarchus; his family seems to have been related by marriage to Posidonius (see Rawson E., Intellectual Life in the Late Roman Republic [Oxford, 1985], p. 68). Given this educational background, it is not absurd to suggest that Sostratus would have been familiar with the plot of one of Euripides' more notorious tragedies.

27 The Nature of Greek Myths (Harmondsworth, 1974), p. 95. Cf. Vernant J.-P., Myth and Society in Ancient Greece (Brighton, 1980), p. 206, stressing the changeability of a myth from one version to another while it retains ‘the balance of the general system’.

28 Clouds 1371–2, Frogs 850 and (more explicitly) 1081.

29 Laws 8.838c, where all three are examples of sexual depravity.

30 Tr. 2.383–4, where the juxtaposition with Hippolytus suggests that Ovid has Euripides in mind; but the fact that Sophocles wrote a Phaedra (frr. 677–93 Radt) may equally suggest a general sphere of reference in the couplet rather than specific reference to Euripides.

31 For dramatic representation at Rome see Lucilius' epigram (Anth. Gr. 11.254), where Canace's role is danced, and Suetonius, Nero 21.3; on artistic representation see Hopkinson, op. cit., p. 31 and n. 5 for references and further bibliography.

32 ‘Aspere et amare dictum: multa enim mala graviora videntur si ante oculos nostros eveniant, quam si audiantur’ (ii, p. 437 Thilo).

33 Cf. Jacobson, op. cit., p. 165 n. 22, who notes the Virgilian echo but offers a more limited interpretation of it: ‘Turnus, about to kill Pallas, wishes the latter's father were present to witness the murder; but in Ovid the father is also the killer.’

34 Jacobson, op. cit., p. 164 n. 21 cites the anonymous De Ulixis Erroribus (332.3 Mυθολγοι Westermann), but he acknowledges that Aeolus' character as defined there (δεινν τινα κα κακτεχνον ἄνδρα) is ‘built around the etymology of the name Aeolus, not around his role as the ruler of the winds’: Aἴολον τοὔνομα ὥσπερ δ κα τν τρπον. For this etymology see n. 37 below.

35 For the violence of the winds as depicted in the conventional νμων στσις, see the extensive examples cited by Nisbet–Hubbard on Horace, C. 1.3.13, Bömer on Ovid, M. 1.58ff., Tarrant on Seneca, Agamemnon 476 and Morford M. P. O., The Poet Lucan: Studies in Rhetorical Epic (Oxford, 1967), pp. 40–2 on 5.597–620.

36 See Otto , Sprichwörter, p. 364 s.v. ‘ventus’ 1, adding Propertius 2.12.5–8 and Ovid, Am. 2.9b.33–4. The shifting winds also proverbially scatter the forlorn lover's empty words; see Otto, op. cit., pp. 17–18 s.v. ‘amare’ 4 and p. 364 s.v. ‘ventus’ 2. Ovid applies the analogy of the fickle winds to faithless friends in the exile poetry (e.g. P. 4.3.33), and ‘ventosus’ regularly indicates human fickleness (see OLD s.v. 5). The comparison of man to the wind goes back at least as far as Eupolis, to whom Photius attributes the phrase ἄνεμος κα λεθρος ἄνθρωπος (Eupolis fr. 406 Kassel–Austin, PCG V p. 519), which Photius explains as meaning that man resembles the wind as being ϕερμενον νμον δκην κα λώμενον κα ββαιον.

37 See also 7.51, 18.185–6, and cf. 21.75–6.

38 The adjective αἰλος defies simple explanation and was variously interpreted by the ancient lexicographers. For the various interpretations and synonyms adduced by these lexicographers (e.g. ποικλος / ταχς / πολτροπος) see TGL s.v. I.i. 1015–18 with Chantraine P., Dictionnaire étymologique de la langue grecque. Histoire des mots I (Paris, 1968), p. 37 s.v. The name Aἲολος possibly reflects the rapid, shifting movement of the winds under Aeolus' control; otherwise, or in addition, it may indicate the resourceful versatility which Tsetzes (on Lycophron, Alex. 4; ii. 10 Scheer) associates with the ruler of the winds. The least that may be inferred is that the etymological associations of αἰλος suggest a mercurial character who is neither uniform in temperament nor predictably inflexible in character.

39 At Od. 10.72–5 Aeolus' hostility towards Odysseus marks a sudden and abrupt change from the hospitality with which he first greeted him (cf. 14ff.).

40 So Fränkel H., Ovid: A Poet Between Two Worlds (Berkeley and Los Angeles, 1945), p. 191 n. 9, Jacobson, op. cit., p. 159 and Verducci , Ovid's Toyshop of the Heart: Epistulae Heroidum (Princeton, 1985), p. 234: ‘She [Canace] is at best a passive accomplice in the action, at worst an uncomprehending victim of it.’

* I owe a debt of thanks to Dr R. Hunter, Dr J. Henderson, Dr N. Hopkinson, Professor M. D. Reeve and Dr B. Harries for help at various stages in the preparation of this paper. All quotations follow the text of A. Palmer, P. Ovidi Nasonis Heroides (Oxford, 1898).

Recommend this journal

Email your librarian or administrator to recommend adding this journal to your organisation's collection.

The Classical Quarterly
  • ISSN: 0009-8388
  • EISSN: 1471-6844
  • URL: /core/journals/classical-quarterly
Please enter your name
Please enter a valid email address
Who would you like to send this to? *


Full text views

Total number of HTML views: 0
Total number of PDF views: 8 *
Loading metrics...

Abstract views

Total abstract views: 110 *
Loading metrics...

* Views captured on Cambridge Core between September 2016 - 19th January 2018. This data will be updated every 24 hours.