2. Palmer (n.1 above), 351; cf. id., P. Ovidii Nasonis Heroides XIV (London 1874), xiv-xvGoogle Scholar: ‘[The Hermione] treats of an uninteresting subject in an uninteresting manner.’ See also Jacobson, H., Ovid’s Heroides (Princeton 1974), 55Google Scholar: ‘This poem…is not very successful.’

3. Knox, P.E. (ed.), Ovid, Heroides: Select Epistles (Cambridge 1995), 8-11Google Scholar, esp. 8, for the ‘objective’ case, carefully worded, for disputing Ovidian authorship: ‘…If the number of questionable words, phrases, or lines exceeds the tolerance level [but where to set that level?], doubts may legitimately harden to conviction.’

4. durat Heinsius, Bentley; errat (read with most MSS by Palmer [n.1 above]; cf. 355 ad loc), ‘makes fate undetermined’ (Reeve, M.D., CR 24 [1974], 62Google Scholar, reviewing Dörrie, H., ed., P. Ovidii Nasonis Epistulae Heroidum [Berlin and New York 1971]Google Scholar).

5. aequora Palmer (n. 1 above), but see Reeve, M.D., ‘Notes on Ovid’s Heroides’, CQ 29 (1973), 327.Google Scholar

6. On the awkward elision in the first foot of line 71, the hiatus and the fourth-foot spondee see Palmer (n.1 above), 356 ad loc., with Courtney, E., ‘Ovidian and non-Ovidian Heroides’, BICS 12 (1965), 65Google Scholar, and Sabot, A.-F., ‘Les Héroides d’Ovide: préciosité, rhétorique et poésie’, ANRW II 31.4 (Berlin and New York 1981), 2556fGoogle Scholar.; Knox (n.3 above), 9, is further troubled by the ‘impossibly awkward’ repetition of Taenaris (72) in line 73.

7. I.e. ‘Hermione and Hippodamia were not matres, ‘matrons’, when carried off, but puellae’ (Palmer [n.1 above], 355 ad loc).

8. Knox (n.3 above), 10.

9. So e.g. If., an introductory couplet printed in some early editions but (pace Kirfel, E.-A., Untersuchungen zur Briefform der Heroides Ovids [Noctes Romanae 11, Bern and Stuttgart 1969], 65-67Google Scholar) rightly excised by most modern editors, Palmer (n.1 above) among them (but his numeration of the poem begins at line 3); after 19, two lines given in some late MSS (see Housman, A.E., ‘Ovid’s Heroides’, CR 11 [1897], 200Google Scholar = Diggle, J. and Goodyear, F.R.D. [edd.], The Classical Papers of A.E. Housman [Cambridge 1972], 388f.Google Scholar, with Sicherl, M., ‘Vermeintliche Versinterpolationen in Ovids Heroides’, Hermes 91 [1963], 192fGoogle Scholar. and 196-201); 59 obiecit, unparallelled without an accusative ‘of the thing which is the matter of reproach’ (Palmer [n.1 above), 354 ad loc; cf. Housman, 204 = Diggle and Goodyear, 395); 75-81 (Palmer [n.1 above], 356f.). See also Knox (n.3 above), 11, on lexical anomalies in lines 9, 24 and 115; the fact that Hermione’s letter is not among those which Ovid claims as his work at Am. 2.18.19-34 is significant for Knox (n.3 above, 5-8 and 1 If.), but cf. in response Casali’s, S. review (CJ 92 [1996-97], 305Google Scholar: ‘The list found in Am. 2.18 has no importance at all in coming to a decision about the authenticity of the Heroides.’).

10. Cf. the stimulating approach briefly explored by Mack, S., Ovid (New Haven and London 1988), 4Google Scholar: ‘…Ovid sees, as Euripides saw before him, that Hermione is what she is largely because she had Helen for a mother. It would be bad enough to have the most beautiful woman of all time for one’s mother (mother-daughter relations are difficult, at best), but when that mother not only owns the face that launched a thousand ships but also abandoned the daughter when she did so, psychological problems are obviously in store.’

11. See esp. Rahn, H., ‘Ovids elegische Epistel’, A&A 7 (1958), 105-20Google Scholar, and now P.A. Rosenmeyer, ‘Ovid’s Heroides and Tristia: Voices from Exile’ (29-56 above) with Williams, G.D., The Curse of Exile: A Study of Ovid’s Ibis (PCPS Suppl. Vol. 19: Cambridge 1996), 30f.Google Scholar

12. So Rosenmeyer (n. 11 above), 29.

13. See Jacobson (n.2 above), 371f.

14. Jacobson’s term (n.2 above, 372), used in connection with Hermione’s physical powerless-ness and (hence) her expressive weeping in lines 59-63.

15. See on the erotic elegiac influence Nagle, B.R., The Poetics of Exile: Program and Polemic in the Tristia and Epistulae ex Ponto of Ovid (Coll. Latomus 170: Brussels 1980), 42-70Google Scholar, and cf. Spoth, F., Ovids Heroides als Elegien (Zetemata 89: Munich 1992), 126Google Scholar and n.1, on inclusam (4) as implicated in the exclusus amator motif (the puella inclusa balances the equation…).

16. Ransmayr’s, C.The Last World (London 1990Google Scholar) = Die heme Welt (Nordlingen 1988Google Scholar) sets an example; for a more pedestrian approach see Williams (n.1 1 above), 87-89.

17. See further for the motif of mental vision and travel Nagle (n.1 5 above), 93 n. 44.

18. Harris, M., The Loss that is Forever: The Lifelong Impact of the Early Death of a Mother or Father (Harmondsworth 1995), 96fGoogle ScholarPubMed. Bereavement may cause less pain than abandonment; cf. Edelman, H., Motherless Daughters: The Legacy of Loss (Reading MA 1994), 82fGoogle Scholar.: ‘The mother who abandons her daughter leaves a pile of questions behind: Who was she? Who is she? Where is she? Why did she leave? Her presence created the daughter’s life, but her absence defines it. Like the child whose mother dies, the abandoned daughter lives with a loss, but she also struggles with the knowledge that her mother is alive yet inaccessible and out of touch. Death has a finality to it that abandonment does not…A daughter whose mother chose to leave her or was incapable of mothering may feel like a member of the emotional underclass…As a result, she often develops a sense of degradation and unworthiness even more profound than that of the daughter whose mother has died’ (my emphasis).

19. Sabot, A.-F., Ovide poète de I’amour dans ses oeuvres dejeunesse (Paris 1976), 344Google Scholar, surely underestimates Hermione’s intense feeling of unilateral loss by portraying Helen and her as equally pained by their separation: ‘Au début de chaque distique (91ff), les negations expriment la frustration dont Hermione et sa mere furent victimes.’

20. She thus constructs a personal identity; cf. Edelman (n.18 above), 84: ‘…The [abandoned] daughter grows up in a motherless limbo, left to piece together a feminine identity based on scraps of memory, idealised images, and whatever nuggets of information she can uncover from family members and friends.’

21. Knox (n.3 above), 10.

22. Jacobson (n.2 above), 54 n. 18.

23. Jacobson (n.2 above), 53.

24. Jacobson (n.2 above), 53.

25. For Hermione as beautiful like Helen cf. Od. 4.13f. with Eustath. Comm. in Od. 4.9 (= Weigel i.141), Prop. 1.4.6; and see Nardi, C., ‘Ermione o l’epifania del femminino: Omero e Giovanni Crisostomo’, Prometheus 19 (1993), 255fGoogle Scholar. Hermione perhaps also regrets the fact that in her years of absence Helen never helped to dress her daughter’s hair; cf. for this maternal role Her. 21.87-90.

26. Cf. OLD s.v. paro 6; but for parata praeda also as an idiom for ‘ready prey’ cf. Curt. 4.14.11, coniuges…et liberi…, parata hostibus praeda (‘wives…and children…, ready prey for the enemy’), and 9.10.27, both cited with other examples of the participial adj. at TLL X.1 426.66-79. On the theme of ‘the ineluctability of one’s inherited nature’ in the Heroides see Jacobson (n.2 above), 374 and n.9.

27. si medios numeres (‘if you count those in between’) Palmer (n.1 above) after Nodellius, Heinsius, Bentley; my preferred reading is defended below (cf. also n.32).

28. sed tu quid faceres? induit ilia pater (‘But what were you to do? Your father put them upon you’) Palmer (n. J above); for Housman’s conjecture as printed here see op. cit. (n.9 above), 204 = Diggle and Goodyear (n.9 above), 394 with CR 13 (1899), 178Google ScholarPubMed (a review of Palmer [n.1 above]) = Diggle and Goodyear (n.9 above), 480, and then cf. Palmer (n.1 above), lvi n.1.

29. Trans. Lattimore, R., The Iliad of Homer (Chicago and London 1951Google Scholar); Jacobson (n.2 above), 47 n.9, detects the Homeric echo.

30. On the general motif of ‘worthy ancestors’ in the Heroides see Jacobson (n.2 above), 399, with Sabot (n.19 above), 233; cf. also the contest of descent from Jupiter played out by Ajax and Ulysses at Met. 13.21-28 and 140-45, and see Bettini, M., ‘Heroides 8,47 e i gradi di parentela in Ovidion’, RFIC 118 (1990), 425.Google Scholar

31. Palmer (n.1 above), 353. Although Palmer evidently reads tu quoque per proauum (‘you too, through your grandfather…’) in line 47 (see 353 ad loc), the text printed is tu quoque habes proauum (‘you too have a grandfather in Pelops…’; but cf. the corrigenda sheet appended to p. lix), a reading to which he objects (353f.) because (i) quoque is apparently solecistic (‘Pelops was not Pyrrhus’ proauus’) and (ii) ‘it would be a very poor boast for Orestes that the perjured Pelops and the impious Tantalus were ancestors of his’; also, (iii) ‘Tantalus could not be proauus to Orestes if Pelops was’. In answer to (i), however, Hermione may want to imply through quoque that Orestes shares with Neoptolemus such disreputable ancestors as Pelops and Tantalus; for precisely because of (ii), i.e. the embarrassment of having such ancestors, she suggests that Neoptolemus can make the same dubious boast as Orestes, and she thus contrives a stalemate of sorts between them; (iii) is irrelevant (so Housman [n.28 above], 174 = Diggle and Goodyear [n.9 above], 473, also answering Palmer’s other two points but on grounds different from those given here). Perhaps quoque is to be defended as the preferred reading after all; but for a different approach to the couplet see Bettini (n.30 above), 418-22.

32. The emendation medios, printed in the text of Palmer’s edition (n. 1 above), is surely wrong. If it means the four ancestors who come between Jupiter and Orestes, then Hermione is drawing attention to a respect in which Orestes is markedly inferior to Neoptolemus (between Jupiter and whom there are only three ancestors) when she should be urging parity.

33. Palmer (n.1 above), 354.

34. Pace Kirfel (n.9 above), 65-67.

35. Hermione’s idiosyncratic approach to matters of genealogy is perhaps reflected in her unsual choice of patronymic; Achillides, not found before Ovid, is parallelled only at Ibis 301, where the reference is in fact to Pyrrhus I, the king of Epirus who claimed descent from Achilles (see Williams [n.11 above], 94).

36. After line 19 some editors suspect a lacuna; cf. n.9 above.

37. Hermione’s idealising tendency can again (cf. pp.116f. above) perhaps be illuminated by modern observation. Cf. e.g. Dally, A., Inventing Motherhood: The Consequences of an Ideal (New York 1982), 94Google Scholar (summarising Rycroft, C., Imagination and Reality [London 1968], 29-41Google Scholar [‘On idealisation, illusion and catastrophic disillusion’]): ‘The child loves the mother, invests much emotion in her, and needs her love. He also needs her to provide the stimuli enabling him to develop his love. If she fails to satisfy him or fails to provide the right stimuli, his love for her begins to wane. This is intolerable for him for without this love life is meaningless. So he withdraws his love and feeling away from her as a real person and directs it inwards, towards the image that he has of her in his mind. He splits this image into good and bad and directs his attention and his love only onto what is good, which he elaborates in his imagination. What is bad he ignores, denies or projects elsewhere. Thus he is in love with the ‘idealised” image of his mother in his own mind rather than with his mother as a real person.’

38. Palmer (n.1 above), 353 ad loc.

39. So Helen counters Paris’ profoundly ironic rhetorical question tot prius abductis ecqua est repetita per arma? (‘Of so many who have been abducted before, has any ever been reclaimed by arms?’, Her. 16.343). Hermione, however, can answer Paris more specifically (Her. 8.19) and expects to be extending the list herself (sic quoque eram repetenda, 25 ‘I too could have been reclaimed in that military way’).

40. See on this terminology Palmer (n.1 above), 351 (on 7 and 8), and Jacobson (n.2 above), 56 n.22.

41. The version of Hermione’s double betrothal as given in lines 31-34 suggests that Ovid was at least partly indebted to Sophocles’ Hermione, perhaps with Pacuvius’ Hermiona as an intermediary (cf. fr. 167 R² [Orestes to Neoptolemus] prius data est quam tibi dari dicta, aut quam reditum est Pergamo, ‘she was given as a wife before she was said to be given to you, or before the return from Troy’); see Stevens, P.T. (ed.), Euripides: Andromache (Oxford 1971), 4Google Scholar, with Jacobson (n.2 above), 44f., and D’Anna, I. (ed.), M. Pacuvii Fragmenta (Rome 1967), 97-102Google Scholar; for the Sophoclean plot see Radt, S. (ed.), TrGF IV (Göttingen 1977), 192f.Google Scholar

42. See Sabot (n.6 above), 2606f., for Hermione’s legal terminology as an important factor in rendering her letter ‘un bon plaidoyer’ and ‘une merveille de logique’.

43. For further examples of Helen’s legal terminology see Kenney, E.J., ed., Ovid: Heroides XV1-XXI (Cambridge 1996), 134Google Scholar on 107, 140 on 194 and 141 on 206. Such terminology is by no means limited to these two epistles within the collection, but it is interesting that both mother and daughter set out what is in part a legal case for themselves.

44. So Virg. Aen. 3.327-29; cf. Eur. And. 24-31, 1243-46 (where Andromache is handed over to Helenus only after Neoptolemus’ death at Orestes’ hands). As such, the comparison can be seen as helping to locate Hermione within a wider context of collective female victimisation (cf. femineae…manus, 6, Tantalides matres, 66) from which she draws much of her self-confidence.

45. On the reading in line 131 see Garvie, A.F. (ed.), Aeschylus: Choephoroi (Oxford 1986), 77fGoogle Scholar. ad loc. (the version of lines 130f. given below is Garvie’s).

46. Cf. Soph. El. 110-17, Eur. El. 332-38. Interestingly, Ovid passed over the obvious idea of including in this collection a letter from Electra to Orestes in favour of one using the less familiar Hermione. Perhaps Electra’s various tragic personalities were too well known for Ovid to want to add to or rival them.

47. For frater so used cf. Met. 13.31 (Ajax on Achilles); soror, cf. Her. 14.123 (Hypermestra to Lynceus);frater patruelis is more usual (cf. the examples given at TLL VI. 1 1255.3-17).

48. Cf. Isbell, H., Ovid: Heroides (Harmondsworth 1990), 73fGoogle Scholar. on line 35 cum tibi nubebam, nulli mea taeda nocebat, ‘when I was wedded to you, my marriage brought harm to no one’: although ‘it seems reasonable to understand the term ‘married” as ‘betrothed’”, Hermione ‘has every incentive to argue for the strictest possible understanding of a marriage contract’ with Orestes. And on 41f.: ‘Again, Hermione insists that a marriage contract negotiated and entered into by her father and presumably consummated by her and Pyrrhus can be abrogated simply by the fact that her desires have not been met…She does not hesitate to see Pyrrhus and Paris as fulfilling similar roles. She does, however, close her eyes to significant facts which make such a comparison dubious at best.’ From a Roman perspective Hermione’s legal interpretations are open to grave doubt. Her grandfather’s seniority (31) gives him no legal precedence over his son’s unique privileges as father. It seems that Hermione could have contracted a legal marriage to Orestes in her father’s absence at war, but only if his assent could be assumed (see Treggiari, S., Roman Marriage [Oxford 1991], 170-76Google Scholar). As he was at this time contracting her to Pyrrhus, such an assumption would be difficult to sustain.

49. See Jacobson (n.2 above), 47, and cf. ibid. 379 on 83-86 (‘Hermione’s letter virtually asks to be seen in the light of Briseis’.’).

50. Jacobson (n.2 above), 47f.

51. Jacobson (n.2 above), 48 n. 10.

52. I.e. Alexander = ‘Aλἑξανδρoς, ‘warding off men’ (see Kenney [n.43 above], 123 ad loc).

53. For rumpor used of bursting with passion cf. Her. 16.22, cited with other examples at OLD s.v. 2b; tumescunt, OLD s.v. 2a (tumeo is more common; OLD s.v. 3a). For ignis used of literal fever cf. OLD s.v. 8a (including Met. 7.555), of hot passion s.v. 9; for uror of fever, cf. OLD s.v. 5a, of passion s.v. 6a; for the ill effects of suppressed (inclusis) passion cf. Tr. 5.1.63, strangulat in-clusus dolor atque exaestuat intus, ‘suppressed grief chokes me and seethes within’ (with rumpor, Hermione can no more contain her feelings than can the poet in exile; cf. Williams [n. 11 above], 32). We can see from Her. 8.57f that inclusam (4) already expresses Hermione’s psychological isolation as well as her physical detention by Neoptolemus.

54. Given Ovid’s extensive and obvious familiarity with Euripidean tragedy, most evident in the Metamorphoses and Tr. 2.381-408, it may well be supposed (but it is not essential for my argument) that he knew the Orestes. Cf. Jacobson (n.2 above), 50: ‘…we must allow for the possibility of Euripidean influence [i.e. on Ovid’s characterisation of Orestes in Heroides 8], an Orestes free from supernatural persecution but plagued by his own conscience and reluctant to take bold measures,’ later asking (n.15), ‘Is Ovid’s an extension of the Euripidean characterisation [sc. in the Orestes]? For one thing, this would explain why Hermione’s emphatic praise of Tyndareus (3If.) might be expected to intensify her pressure on Orestes (cf. Eur. Or. 459-69).

55. For the fragment so placed in the reconstructed play see conveniently Warmington, E.H. (ed.), Remains of Old Latin II (Cambridge MA and London 1932), 227Google Scholar, with Argenio, R., ‘Crise, Duloreste, Ermione, Iliona e Medo di Pacuvio’, RSC 6 (1958), 311 f.Google Scholar

56. Murray’s, G. text (OCT III²: Oxford 1913Google Scholar), with coordination supplied by Rieske’s <τ’= in 35. Line 34 as given here is defended by Willink, C.W., Euripides: Orestes (Oxford 1986), 86Google Scholarad loc, but J. Diggle (OCT 1994) obelises 34f. ().

57. Jacobson (n.2 above), 50 n. 14.

58. Here no doubt including the genital membrum; for the plural with this specific sexual sense see J.N. Adams, The Latin Sexual Vocabulary (London 1982), 46. Similarly, her reference to pollutas…manus (114) suggests revulsion at the sexual act (see OLD s.v. polluo 4 with Adams ibid. 223).

59. Cf. Ibis 303f. with Williams (n.11 above), 94, Hyg. Fab. 123.2.

60. Cf. with emphasis on the polluted hand(s) e.g. Aesch. Eum. 41f., 102, Eur. Or. 517, 831 -33, ; (‘what affliction…is greater in the world than taking the blood-guilt of matricide on one’s hand?’). But cf. Jacobson (n.2 above), 396, on the general motif in the Heroides of ‘groping, grasping, touching while asleep’.

61. Jacobson (n.2 above), 56. Line 115 is of course ambiguous, although we are ‘obviously’ meant to read the sense-break at the end of the fourth foot. The potential ambiguity, however, only compounds Hermione’s errorem uocis.

62. Jacobson (n.2 above), 48.

63. Jacobson (n.2 above), 49.

64. Cf. Mack (n. 10 above), 3: ‘Even though there is no ‘Phaethon complex” named after Ovid’s teenager…we are all familiar with the phenomenon of sons trying to prove their worth (and their identity) by competing with their fathers;’ cf. Glenn, EM., The Metamorphoses: Ovid’s Roman Games (Lanham MD and London 1986), 17.Google Scholar

65. This psychological emphasis on the search for true paternity singles out one feature of the plot of Euripides’ Phaethon to the neglect of others (see Diggle, J. [ed.], Euripides: Phaethon [Cambridge 1970], 181-83Google Scholar, with Galinsky, G.K., Ovid’s Metamorphoses: An Introduction to the Basic Aspects [Berkeley and Los Angeles 1975], 49-51Google Scholar). Other sources for Ovid are unknown, and without fresh evidence need not be supposed ever to have existed.

66. Cf. Rosenmeyer (n.1l above), 31: ‘In each letter he [sc. Ovid] has a fresh opportunity to reinvent himself…Ovid’s stance as an unhappy exile…fluctuates from letter to letter…, as his relationship with his hosts changes over time.’