This article has had an unusually long incubation period, and I would like to thank many people who have helped it along the way. Thanks to Mary Bachvarova, who first suggested that I develop the topic, and provided many helpful suggestions. Thanks also to Ed Harris, for encouragement and conversation. Audiences at the Ohio Classical Conference, the University of Wisconsin-Madison, and the American School for Classical Studies in Athens asked many helpful questions and helped sharpen my thinking. Anne Feltovich, Sara Lima, Christina Kolb, Matt Sears, Amanda Flaata, Jamie Donati and Jake Butera were particularly supportive. Daniel Levine made a number of helpful comments and saved me from an egregious error. Helen Morales read the manuscript at a late stage, improved the argument in numerous places, and provided much needed encouragement in bringing the piece to its final form. Finally, I should note that this piece began over twenty years ago in Jack Winkler’s last seminar; most of what is useful here is due to his incisive thinking, unflagging patience, and inspiration. Any errors that remain are mine alone.

1. Text of Aesop is taken from Perry, B.E., Aesopica (Urbana 1952), 105f.Google Scholar

2. In Longus’ Daphnis and Chloe, both hero and heroine of the novel discover sexual experience as a form of education (see esp. 3.17–19). This aspect of the novel has been brilliantly discussed by Winkler, J.J., The Constraints of Desire: The Anthropology of Sex and Gender in Ancient Greece (New York 1990), 101–26Google Scholar. Here, however, the promise of education is used rather callously by the man in order to enjoy sex with the young woman who remains fundamentally ignorant. I was reminded of this connection by Helen Morales.

3. Selden, D., ‘The Genre of Genre’, in Tatum, J. (ed), The Search for the Ancient Novel (Baltimore 1994), 39–64Google Scholar, discusses syllepsis as the defining trope of the Greek romances.

4. Again, this is strikingly different from Daphnis and Chloe, where the experience of sex is described as bloody and painful for the woman (3.19); see esp. Winkler (n.2 above), 121, 124–26.

5. At times the word parthenos appears to refer to a social status, i.e., young and unmarried, rather than a state of experience. Sissa is correct, however, in arguing that being a parthenos was a ‘sexual, not a sociological state’: Sissa, G., Greek Virginity, tr. Goldhammer, A. (Cambridge MA 1990), 86Google Scholar. See further discussion below.

6. See LSJ. The verb is used in just this sense at Herodotus 4.168, where the author is describing the bizarre marriage customs of the Libyans.

7. Foucault, M., The History of Sexuality, Vol. 3: The Care of the Self, tr. Hurley, R. (New York 1986), 228–32Google Scholar. Konstan, Both D., Sexual Symmetry: Love in the Ancient Novel and Related Genres (Princeton 1994Google Scholar), and Goldhill, S., Foucault’s Virginity: Ancient Erotic Fiction and the History of Sexuality (Cambridge 1995CrossRefGoogle Scholar), offer important modifications to Foucault’s formulation, which I will discuss at some length. Morales, H., Vision and Narrative in Achilles Tatius’ Leucippe and Clitophon (Cambridge 2004), 212–20Google Scholar, presents a brilliant analysis of Leucippe’s paradoxical status as ‘a virgin/whore, with one aspect overlaying the other’ (quotation at 218).

8. Foucault (n.7 above), 228.

9. For the sake of clarity, when 1 am referring to Romance as a literary genre as opposed to the designation of an erotic sentiment between two lovers, I will capitalise the word. Though the word ‘Romance’ has become problematic in ancient novel studies, I find the formulation of Doody, M.A., The True Story of the Novel (New Brunswick 1996), 15Google Scholar, to be essentially correct: ‘Romance and the Novel are one. The separation between them is part of a problem, not part of a solution.’

10. J J. Winkler, ‘The Invention of Romance’, in Tatum (n.3 above), 23–38, at 33.

11. Winkler (n. 10 above), 37.

12. B. Egger, ‘Women and Marriage in the Greek Novels: The Boundaries of Romance’, in Tatum (n.3 above), 260–80, provides a clear discussion of the difficulty inherent in trying to read the novels as reflecting social reality. See esp. 268–71 on contemporary marriage laws. This is also treated well by Haynes, K., Fashioning the Feminine in the Greek Novel (New York 2003), 23–30Google Scholar.

13. This observation is supported by the observations of Morales (n.7 above) regarding Achilles Tatius’ use of various animals as analogues for his novel’s heroine; see esp.193–96,199.

14. We do not see virginity as a common theme in Greek love poetry until the Hellenistic period. The Palatine Anthology contains a number of poems in which the speaker urges his love-object to surrender her virginity, or in which he contemplates taking it from her: cf. AP 5.79, 5.85, 5.294.19–24. Some epitaphs have the speaker mention the loss of virginity: AP 7.164, 7.183. The daughters of Lycambes speak out against Archilochus’ slanders against their virginity: AP 7351.1–4, 7352. Similarly, Philomela writes to Procne about Tereus’ theft of her virginity and her voice, AP 9.452.1–5. Not all of these examples speak of virginity in specifically physical terms, though some do: at AP 7.164.3f., for example, the speaker says that her husband ‘first untied the untouched knot of my virginity’.

15. Page, D., Sappho and Alcaeus (London 1955), 122 with n.2.Google Scholar

16. Sissa (n.5 above), 73–86. See also Konstan (n.7 above), 156 n.38; Loraux, N., The Children of Athena: Athenian Ideas about Citizenship and the Division Between the Sexes, tr. Levine, C. (Princeton 1993), 224–27Google Scholar,224 n.183; Viitaniemi, L., ‘Parthenia: Remarks on Virginity and its Meanings in the Religious Context of Ancient Greece’, in Lovén, L.L. and Strömberg, A. (eds.), Aspects of Women in Antiquity (Jonsered 1998), 44–57Google Scholar, at 49–51.

17. See discussion in Sissa (n.5 above), lOOf. Several other examples are discussed at 78–80.

18. Sissa (n.5 above), 82f.

19. Sissa (n.5 above), 105–23; contra see Hanson, A.E., ‘The Medical Writers’ Woman’, in Halperin, D.M., Winkler, JJ. and Zeitlin, F.I. (eds.), Before Sexuality: The Construction of Erotic Experience in the Ancient Greek World (Princeton 1990), 309–38Google Scholar, at 324–30. Goldhill (n.7 above), 116, and 172 n.9, points out that some Christian texts speak of a physical integrity that may refer to the hymen. In some of the discussion that follows I will suggest that virginity is less a physical state than a personal one, that is, a state of innocence/ignorance.

20. Sissa (n.5 above), 355; Goldhill (n.7 above), 38. The relevant passage is Soranus 1.17.

21. Quoted and translated in Morales (n.7 above), 213. Morales’s discussion of Achilles Tatius’ novel as a fleshing out in narrative of these sorts of legal quandaries is instructive: see 213f.

22. Viitaniemi (n. 16 above) provides a brief but interesting discussion.

23. Sissa (n.5 above), 88.

24. Pregnancy is often the only sure sign of lost virginity, as G. Sissa, ‘Maidenhood Without Maidenhead: The Female Body in Ancient Greece’, in Halperin, Winkler and Zeitlin (n.19 above), 339–64, at 348, notes: ‘By swelling the belly of its mother or simply by appearing unexpectedly, the virgin’s child becomes the unique sign of a truth that is otherwise impossible to grasp.’ See, e.g., Pausanias 4.9.6–8.

25. Sissa (n.5 above), 116f.: ‘The difference between the Greeks and us therefore lies not in the contrast between a social status and a physical status…but in the value of a body as opposed to that of a sign.’

26. Doody (n.9 above), 72.

27. Hägg, T., The Novel in Antiquity (Berkeley 1983), 41fGoogle Scholar. For a thorough discussion of the problems that arise in trying to determine the genre of these long prose works, see Selden (n.3 above).

28. Konstan (n.7 above), 55. See 48–55 for the fullest expression of this argument.

29. Morales (n.7 above), 83f., points out that Thersander’s critical stance here is one of a realist, and, as such, a rejection of the novel’s standard tropes. In this way, his desire to do violence to Leucippe is paralleled by his violence to the text itself: ‘Thersander’s reading of the world is just dangerous, striking at the very heart of the genre which contains him…. The realist is a bully: to read with expectations of realism is to do violence to the text.’

30. See Goldhill (n.7 above), 118: ‘From the first deferral of sex…Achilles has both preserved Leucippe’s virginity and repeatedly played with the idea of losing such physical integrity.’ See Konstan (n.7 above), 53 n.53, for a brief discussion of this passage. He argues that what is paramount here is ‘not so much the purity of her body as her pride in her free status…’ but 1 have difficulty seeing this here.

31. Haynes (n.12 above), 58, 171 n.38. See also Morales (n.7 above), 200f., 224. She points out (224) that Melite, in contrast to Leucippe, ‘…demonstrates a formidable command of language throughout the narrative’. Melite is, of course, also sexually experienced.

32. Morales (n.7 above), 216.

33. Winkler (n.10 above), 28.

34. Carson, A., Eros the Bittersweet (Princeton 1986), 81Google Scholar: ‘…the intention to consummate desire puts the lovers at odds with the novelist, whose novel will end unless he can subvert them.’ See also Anderson, M., ‘The Sophrosyne of Persinna and the Romantic Strategy of Heliodorus’ Aethiopica’, CP 92 (1997), 303–22Google Scholar, at 311: ‘By regulating the progress of love, chastity helps sustain the attention of the reader as the narrative progresses slowly towards its inevitable goal.’ At the same time, as Whitmarsh points out, a contradictory desire pushes both characters and readers to the eventual satisfactory ending: ‘This desire of the characters for sexual gratification mimics the reader’s desire for narrative conclusion.’ See Whitmarsh, T., ‘Desire and the End of the Greek Novel’, in Nilsson, I. (ed.), Plotting with Eros: Essays on the Poetics of Love and the Erotics of Reading (Copenhagen 2009), 135–52Google Scholar, at 142.

35. See Morgan’s discussion of the narrowly-averted sacrifice of Chariclea in Heliodorus Aethiopica 10.7: ‘At one level the reader, for whom the hero and heroine are repositories of a value-system to which the novel as a whole has led him to subscribe, will be deeply dismayed by this serious and subtle threat to everything that is good…. And yet at the same time, the reader, as a reader, is enjoying an exciting story and will welcome the prospect of further plot-complication, with its promise of thrills, suspense, and pleasure’: Morgan, J.R., ‘Reader and Audiences in the Aethiopica of Heliodorus’, in Hofmann, H. (ed.), Groningen Colloquia on the Novel Vol. 4 (Groningen 1991), 85–103Google Scholar, at 94. So Lateiner, D., ‘Abduction Marriage in Heliodorus’ Aethiopica’, GRBS 38 (1997), 409–39Google Scholar, at 422 (speaking of Heliodorus): ’Here chastity means much more than the preservation of virginity: it is essential to the excitements of their travels, contact with pirates, incarceration, disguises, near-death, etc.’

36. Goldhill (n.7 above), 121 (emphasis original).

37. Again, Morales (n.7 above), 193–96, is particularly sharp on the way that the figure of the phoenix at 3.25 figures as an analogue for Leucippe, submitting to a physical scrutiny that cannot be applied to the heroine.

38. Sissa(n.24 above),343.

39. Sissa (n.24 above), 347, 360. In this regard it is significant that Callirhoe in Chariton’s novel becomes a woman (gune instead of parthenos) not when she has sex, but when she gives birth; cf. Ach. Tat 3.83 and Elsom, H., ‘Callirhoe: Displaying the Phallic Woman’, in Richlin, A. (ed.), Pornography and Representation in Ancient Greece and Rome (New York 1992), 212–31Google Scholar, at 223.

40. See for example Hägg (n.27 above), 81–108. Stephens and Bowie provide good summaries of previous scholarship on the historical readers of the novels. Both argue, in different ways, against the notions of the novels as ‘popular’ literature, and both argue against a primary female readership. S. Stephens, ‘Who Read the Ancient Novels?’, in Tatum (n.3 above), 405–418; E. Bowie, ‘The Readership of Greek Novels in the Ancient World,’ in Tatum (n.3 above), 435–59. Haynes (n.12 above), 9–17 also summarises the arguments against female readers.

41. For an acute reading of the rise of domestic fiction, see Armstrong, N.A., Desire and Domestic Fiction: A Political History of the Novel (New York 1987Google Scholar). H. Montague, ‘From Interlude in Arcady to Daphnis and Chloe: Two Thousand Years of Erotic Fantasy’, in latum (n.3 above), 391–401, at 392f., lists a number of generic similarities between the ancient novels and modern Harlequins, but also admits that ‘no argument about the ancient texts can be grounded in a discussion of the practices of specific readers’ (392). Haynes (n.12 above), 2–17, 75–77, provides a useful overview of the problem.

42. Haynes (n.12 above), 60f., 78–80; Balot, R., ‘Foucault, Chariton, and the Masculine Self’, Helios 25 (1998), 139–62Google Scholar, at 155.

43. Doody (n.9 above), 18–26 (quotation from 24).

44. See Haynes (n.12 above), 75–77; Stephens (n.40 above), 405f.

45. Stephens (n.40 above), 410f.,415f.

46. Stephens (n.40 above),407; Bowie (n.40 above),454 n.12.

47. Bowie (n.40 above), 438,453.

48. See especially Radway, J., Reading the Romance: Women, Patriarchy and Popular Literature (Chapel Hill 1984Google Scholar).

49. Rabine, L., ‘Romance in the Age of Electronics: Harlequin Enterprises’, in Warhol, R. and Price Herndl, D. (eds.), Feminisms: An Anthology of Literary Theory and Criticism (New Brunswick NJ 1991), 878–93CrossRefGoogle Scholar, at 888–90.

50. Morales (n.7 above), 95. Whitmarsh, T., ‘Reading for Pleasure: Narrative, Irony, and Eroticism in Achilles Tatius’, in Panayotakis, S., Zimmerman, M. and Keulen, W. (eds.), The Ancient Novel and Beyond (Leiden 2003), 191–206CrossRefGoogle Scholar, similarly provides a narratological reading of Clitophon and Kleinias as different types of readers of the novel form.

51. Winkler (n.2 above), 127.

52. Morgan (n.35 above) is particularly helpful in showing Heliodorus’ various techniques for directing the external reader through the reactions of the internal narratees. As Morgan shows, Heliodorus does not rely on simple identification, or indeed on only one strategy. At times our reactions are allied with those of the internal listeners, and at other times we a meant to react in a manner opposite (Morgan’s ‘response by antithesis’, 102).

53. Winkler (n.7 above), 111.

54. Winkler (n.7 above), 111.

55. See especially Goldhill (n.7 above), 14: ‘How stained, how dirtied, is the reader by an inability to read innocence innocently?’

56. Bowie (n.40 above), 436, points to several passages that suggest a masculine readership, including the ‘long debate in Achilles Tatius 2.36–38 on the respective delights for a male of homosexual and heterosexual activity….’ (emphasis original).

57. Leucippe does defend her virginity to her mother at Ach. Tat. 2.25.

58. This discussion does not preclude the fact that the novels may strike modern female readers (and perhaps struck the ancient female readers as well) differently than they do men. As a vast body of feminist criticism points out, female readers are culturally situated differently than men, and sometimes identify differently with the characters and expect different resolutions than men do. This, however, takes us far from my present inquiry. Montague (n.41 above) sees much that is admirable in Chariclea, and argues for the recuperation of some of the ancient novels’ heroines.

59. Konstan (n.7 above), 85, points out that the primary couple do not have premarital sex in any of the extant novels. Morales (n.7 above), 206, notes, ‘Leucippe herself is only too willing to sleep with Clitophon for the first half of the narrative. Her subsequent refusal to do so is not due to a sincere change of heart or a sudden discovery of moral inhibitions, but because an external force, Artemis, who visits her in a dream, commands it (4.13–5).’

60. See especially Goldhill (n.7 above), 118; Morales (n.7 above), 156, 182.

61. Goldhill (n.7 above), 85f.

62. Morales (n.7 above), 152f.

63. Carson (n.34 above) remains the best commentary on the role of the reader’s desire for virginity and for the novelistic plot; see esp. 78–81. Morales (n.7 above), 182–99, brilliantly explores different modes that the novel uses to manipulate the reader’s desire for an investigation of Leucippe’s integrity.

64. Goldhill (n.7 above), 94–98.

65. Foucault (n.7 above), 231. See the brief but acute comments on this passage and Foucault’s reading of it in Whitmarsh, T. and Morales, H. (tr. and intr.), Achilles Tatius: Leucippe and Clitophon (Oxford 2001), xxixfGoogle Scholar.

66. Konstan (n.7 above), 53.

67. Similarly, Morales (n.7 above), 206f., discusses Melite’s trickery in passing a test of fidelity later in the novel (Ach. Tat. 8.14) : ‘The reader is at no stage invited to disapprove of the moral standards here displayed, an omission which encourages us instead to join the conspiracy (a conspiracy which excludes Leucippe) and appreciate its humour.’

68. Goldhill (n.7 above), 98. Goldhill provides here a brilliant analysis of the way that the verb φιλoσoφέω (‘to do philosophy’) has slipped in meaning over the course of the past three books.

69. Sissa (n.24 above), 343.

70. This story is told as well in Longus’ Daphnis and Chloe at 234; subsequently, hero and heroine enact the story in dance, with Daphnis playing the part of Pan and Chloe that of Syrinx.

71. So also Morales (n.7 above), 216–18, on Leucippe as that pervasive male fantasy, the ‘virgin/whore’.

72. Pace Segal, C., ‘The Trials at the End of Achilles Tatius’ Clitophon and Leucippe: Doublets and Complementaries’, SIFC 2 (1984), 83–91Google Scholar, at 87: ‘Melite’s victory…also serves to suggest a kind of purity for the hero…that is parallel to the actual virginity of the heroine.’

73. For full discussion of the structural similarities, see Segal (n.72 above).

74. For the competition between Artemis and Aphrodite, see Cueva, E., ‘The Analogue of the Hero of Heliodorus’ Aethiopica’, SyllClass 9 (1998), 103–13Google Scholar. Cueva argues that Clitophon’s model in the novel is Hippolytos, whose story he manages to re-shape into successful form.

75. Winkler (n.10 above) remains for me the most persuasive reading of Heliodorus, showing the full range of Heliodorus’ skill and playfulness as a narrator who both parodies and presents the Romantic plot. Dowden, K., ‘Heliodoros: Serious Intentions’, CQ 46 (1996), 267–85CrossRefGoogle Scholar, has called for a more serious reading of Heliodorus, one infused with ‘authoritative statement and firm content’ (267).

76. Foucault (n.7 above), 230.

77. See especially Foucault, M., The History of Sexuality, Vol. 2: The Use of Pleasure, tr. Hurley, R. (New York 1985), 63–77Google Scholar.

78. See especially Dowden (n.75 above); also Anderson, G., Eros Sophistes: Ancient Novelists at Play (Chico CA 1982), 33Google Scholar: ‘He is certainly committed to the heroine’s chastity, which in this case is positively scintillating.’

79. Lateiner (n.35 above), 429, seems to think that the gridiron will kill those who are not virgins who stand on it. But at 10.8 the risk seems only to involve burned feet and perhaps injury.

80. This suggestion of a ruse, made in this offhand manner, should give us pause. Mary Bachvarova points out in correspondence that Chariclea may be protected from the gridiron by her pantarbē stone, one of the signs given to her by her mother when she was abandoned. Although the stone apparently did save Chariclea from actual fire in Book 8, its magic powers are not brought up in this passage. Still, are we meant to remember it here? Is she really a virgin, or is it just that she carries a magical jewel? As Anderson (n.34 above), 313, points out, the stone is constructed by the text of the ribbon that accompanies it as parallel to Chariclea’s virginity. Both are ‘precious treasure[s]’, and both seem to be working together to protect Chariclea from burning. Perhaps we should think of the pantarbē stone as the physical embodiment of Persinna’s advice, which Chariclea has sometimes unknowingly followed.

81. Carson (n.34 above), 85.

82. Anderson (n.34 above), 318–21, explores the implications of the story of Andromeda.

83. See the useful discussion of Anderson (n.34 above), 310–22.

84. Anderson (n.34 above), 313.

85. See Lateiner (n.35 above), 430–37, for a helpful discussion of Chariclea’s ambiguous status. Lateiner sees Chariclea as a savvy manipulator of her virginity and its significance; see esp. 433.

86. See Patterson, C., ‘Those Athenian Bastards’, ClAnt 9 (1990), 40–73Google ScholarPubMed, for a full discussion.

87. Cf. LSJ s.v. vóθoς, II.I. and II.2.

88. We should recall Leucippe here, who was unashamed to speak directly to her father once her virginity was clearly established (Ach. Tat. 8.15). Cf. Segal (n.72 above), 89.

89. Haynes (n.12 above), 118, reads this passage as an example of Persinna’s inability to understand Chariclea’s emotional state. Chariclea’s status, however, is far from clear, and I prefer to see this scene as a deliberate rift in the text, in which Persinna and Chariclea will come to understand each other, but in a way that is not on display to the external reader.

90. See Anderson (n.34 above), 316–18, for a discussion of Persinna’s exemplary tact when it comes to discussing sexual matters.

91. Sissa(n.24 above),347.

92. Winkler (n.7 above), 95.

93. Exactly when in the ceremony the anakaluptēria took place is a matter of some debate. For discussion, see Sissa (n.5 above), 93–99; Oakley, J. and Sinos, R., The Wedding in Ancient Athens (Madison WI 1993),25fGoogle Scholar.

94. Morgan (n.35 above), 91f., points out that in this scene ’|t|he narrative is continually interrupted to inform us of the crowd’s reaction to what is happening’. Morgan argues that this technique functions to ally our response to that of the crowd. While I agree, I also want to emphasise that our ignorance is also forced to parallel that of the crowd; neither they nor we see inside the tent.