1. Smith, B. H., ‘Afterthoughts on Narrative III: Narrative Versions, Narrative Theories’, Critical Inquiry 7.1 (1980), 233Google Scholar.

2. Prince, G., Narratology: The Form and Functioning of Narrative (Berlin, New York, Amsterdam 1982), 160CrossRefGoogle Scholar, explains why an unsolicited narrative must first awaken desire in terms of the dynamics of suspense, and of a contract betwen narrator and narratee, which sometimes is emphasised as in The Arabian Nights. Scholes, R., Fabulation and Metafiction (Urbana 1979), 26Google Scholar: ‘The archetype of all fiction is the sexual act.’ On 27 he notes that both are also reciprocal.

3. Otis, Brooks, Ovid as an Epic Poet (2nd ed., Cambridge 1970), 78fGoogle Scholar., reviews briefly Ovid’s variations on ‘the motif of the determined virgin courted by the passionate god (Daphne and Apollo)’, which is introduced for the first time in the first book. Davis, Gregson, The Death ofProcris: ‘Amor’ and the Hunt in Ovid’s Metamorphoses (Rome 1983)Google Scholar, uses ‘Apollo and Daphne’ as the paradigm for a set of tales which overlaps somewhat with the group I am considering in the first part of this paper, but his contribution is a semiotic interpretation of the interconnected motifs of love and hunting, in terms of the concept of hunting as a ‘narrative code’. Curran, Leo, ‘Rape and Rape Victims in the Metamorphoses’, Arethusa 11 (1978), 213–37Google Scholar, refers to the ‘programmatic Daphne’ (231) in considering Ovid’s peculiarly anti-sexual nymphs (230f.); he also considers ‘the recurrent motif of flight and pursuit, of chase and the attempt to escape’, discussing as ‘two salient instances’ (223) Daphne and Arethusa (233–35). By using the word ‘seduction’ rather than ‘rape’ throughout, I do not intend to euphemise or trivialise, but rather to explore the narrative semantics of this motif in a way more like Davis’s. While there can (and, according to Prince and Scholes, must) be narrative seduction, narrative rape seems to me to be an impossibility. Ahl, Frederick, Metaformations: Soundplay and Wordplay in Ovid and Other Classical Poets (Ithaca 1985), 137Google Scholar, has recently given a Varronian reading of the phrase about Apollo’s first love: ‘There is also, probably, a hint of the metaMORPHOsis to come in the first three words, viewed differently: Primus aMOR PHOeW.’

4. Ahl (n. 3 above), 136, remarks that ‘… Phoebus’ words to Daphne throughout the race are replete with — from his point of view unintentionally self-revealing — doubles entendres.’ Gross, Nicholas P., in the first part of ‘Rhetorical Wit and Amatory Persuasion in Ovid’, CJ 74 (1979), 305–08Google Scholar, points out the ridiculousness of Apollo’s rhetoric and the incongruity between it and his situation.

5. This third case occurs within Orpheus’ cycle. Earlier in that sequence of songs, the motif of Cupid’s arrows is negated, when that god is said to have denied that his tela (10.311) caused Myrrha’s incestuous passion.

6. The fact that ‘Pan and Syrinx’ is a doublet of ‘Apollo and Daphne’ has been recognised since Frankel, Hermann, Ovid: A Poet Between two Worlds (Berkeley and Los Angeles 1945), 85 and 216 n. 43Google Scholar. For a consideration of another such instance of similar tales told by Ovid in propria persona and by a character (in this case, Orpheus), see Nagle, Betty Rose, ‘Byblis and Myrrha: Two Incest Narratives in the Metamorphoses’, CJ 78 (1983), 301–15Google Scholar. Another such pair consists of Ovid’s narrative of the battle of Perseus against Phineus and his followers in Book 5, and Nestor’s account of the battle of the Lapiths and Centaurs in Book 11. The most closely juxtaposed pair is Ovid’s account of Apollo’s tragic loss of Cyparissus, at the end of the catalogue of trees which assemble to listen to Orpheus, and Orpheus’ own account of Apollo’s tragic loss of Hyacinthus; Cyparissus’ story ends at 10.142; Hyacinthus’ begins at 10.162.

7. On the pastoral landscape as the setting for rape and other acts of violence, see Parry, Hugh, ‘Ovid’s Metamorphoses: Violence in a Pastoral Landscapes’, TAPA 95 (1964), 268–82Google Scholar. A clear early instance of the erotic connotations of pugnare and vincere appears in Jupiter’s rape of Callisto (2.436–8): ilia quidem pugnat: sed quern superare puellajquisve Iovem poterat? superum petit aethera victor/Iuppiter (‘Indeed that one fights: but whom could a girl overcome, or who could overcome Jupiter? Jupiter the conqueror heads for heaven on high’). Salmacis, Tereus and Myrrha’s nurse all use vincere in their exclamations of erotic triumph (4.356; 6.513; 10.443) at various stages, but all before the actual consummation: Salmacis, when Hermaphroditus, after rejecting her overtures, dives into her pool; Tereus, when he is sailing back to Thrace with Philomena on board; and the nurse in reporting her successful arrangement of an assignation with Cinyras. On the parodic aspects of that third passage (gaude, vicimus), see Clarke, W. M., ‘Myrrha’s Nurse: The Marathon Runner in Ovid?’, CP 68 (1973), 55fGoogle Scholar. The Salmacis episode also contains an instance of pugnare: when she has Hermaphroditus in her clutches, she taunts him (4.370f.): ‘pugnes licet, inprobe,’ dixit/non tamen effugies …’(‘ “You may struggle, scoundrel,” she said, “but you will not get away …” ’).

8. Ahl (n. 3 above), 126, comments on the dammis capreisque fiigacibus (‘swift hinds and roes’, 1.442), which had been the prey of Phoebus before his conquest of the Python, noting a Varronian play on cibus in fugaCIBUS, for which he cites two passages in which the play is explicit (6.480 on facibus and 15.476 on fallacibus).

9. During this rest from hunting, so long as Venus can hold Adonis’ narrative (and erotic) interest, he actually is safe, that is, he is deterred from hunting dangerous prey not by the warning itself but by the duration of its narrative. This is a variation on the life-or-death significance of prolonging narrative in The Arabian Nights: there Scheherezade herself will stay alive so long as she keeps her audience’s interest; in the case of Venus and Adonis, the audience stays alive only for so long as the narrator keeps talking.

10. See n. 7 above.

11. Anderson, W. S. (ed.), Ovid’s Metamorphoses: Books 6–10 (Norman 1972), 522Google Scholarad 10.560–63, comments: ‘The introduction to the story teases the audience, for Ovid leaves names out until 565 and talks vaguely of Atalanta as aliquam …’

12. At 3.225 Ovid breaks off his 19-line 33-dog catalogue with the phrase quosque referre mora est (‘and those whom it would take too long to name’). In a rather different fashion, Jupiter spares the divine council, and himself, the details of the depravity he found when he went to earth in disguise to investigate the rumour about Lycaon: longa mora est, quantum noxae sit ubique repertumj’ enumerare: minor juit ipsa infamia vero (‘It would take too long to count up the amount of evil found on all sides; the awful report itself was less than the truth’, 1.214f.). The word mora does not appear in Venus’ self-interruption, except in the phrase immediately preceding, in which Venus claims she impeded Atalanta oneris pariter gravitate moraque (‘as much by the weight of her burden as by the loss of time’, 10.678).

13. For the concept of ‘duration’ as it applies to narrative, see Gennette, Gérard, Narrative Discourse: An Essay in Method tr. Jane E. Lewin (Ithaca 1980), 86–112Google Scholar. For a brief account of Gennette’s three categories of ‘order, duration and frequency’, see Chatman, Seymour, Story and Discourse: Narrative Structure in Fiction and Film (Ithaca 1978), 63–79Google Scholar.

14. A third category is the cautionary tale. Gieseking, Kurt, Die Rahmenerzählungen in Ovids Metamorphosen (Diss. Tubingen 1965)Google Scholar, uses a different scheme to classify the framed episodes in the poem into three groups, viz. aition-episodes, exemplum-episodes, and simple reports; he outlines this scheme (30–33), develops it (33–58) and summarises his conclusions (59–66). The aitiion-episodes are solicited, and supply a causa; by using the Latin word causa, Ovid himself alludes to the Greek term aition. The exemplum-episodes originate with the speaker’s desire to prove or illustrate a point (this category can include the cautionary tale). The simple reports provide additional information. The difference between Gieseking’s categories and mine is minor in comparison with our disagreement on a larger point; for him a framed tale is signalled by a sudden interruption, regardless of whether the poet or a character is the narrator (11); my object of study is the embedded tale, for which the essential criterion is a change in narrator.

15. Leach, Eleanor Winsor, ‘Ekphrasis and the Theme of Artistic Failure in Ovid’s Metamorphoses’, Ramus 3 (1974), 128–30CrossRefGoogle Scholar, reads the Theocritean burlesque as the product and the reflection of the Cyclops’ emotion, and as the failure of art to sublimate, civilise or tame violent emotion; while she notes that the song is reported by Galatea (128), this fact does not enter into her interpretation of it. At 141 n. 46 she concurs with the suggestion by Dorrie, Heinrich, Die schöne Galatea (Munich 1968), 54–57Google Scholar, that Ovid invented the character of Acis. This idea appears also in Dorrie, , ‘Der verliebte Kyklop’, AU 12 (1969), 81Google Scholar; Dorrie enticingly raises and then sidesteps the issue of the influence of both the narrator and the audience on the report of the song. He observes that it is important that it is told to ‘der mannerfeindlichen Scylla’, that it is likewise important whether Galatea has told everything — although he maintains that one may not ask that question (83) — and simply concludes with a generalisation that this is a stylisation by Ovid of what used to be called ‘girl talk’, i.e. the stereotypical way women talk to one another about men (83f.).

16. One should note that the phrase si fugeres omnes (‘if you fled everyone’, 860) associates Galatea in the Ovidian reader’s mind with the type of the determined virgin. Parry (n. 7 above), 271 (his section on ‘Flight and Pursuit’), comments: ‘Polyphemus … finds the nymph’s most culpable failing to be her fleetness of foot, which exceeds even that of a stag pursued by baying hounds (13.805f.).’ Indeed, it is fugacior aura (‘swifter than the wind’, 807); ftigacior is the last of those incessant comparatives with which the serenade begins. The shift from complimentary — candidior (789) — to critical — saevior (798) — also hinges on the notion of flight — et, si non fugias, riguo formosior horto (‘and more beautiful than a garden bright with sinuous rills — if only you would not flee’, 797). Polyphemus’ gradual realisation of the impossibility of an amor mutuus between himself and Galatea, precluded of course by the existence of Acis, is an important part of Dörrie’s interpretation (n. 15 above, 1969). The reader plays an intriguing role in his discussion of this episode; since Galatea remains silent, the reader must, Dörrie asserts, take her role in reacting to the Cyclops’ favourable self-representation (89). Later he claims that what happens to the reader in reading this narrative is closely associated to the characterisation of the Cyclops (95). Also, Ovid is playing with the expectations of a ‘surprise-ending’ which a reader of the Theocritean versions would have (94f.).

17. To the phrase with which Galatea expresses her actions as an audience, auribus hausi (lit., ‘I drank in with my ears’, 13.787), cf. multa auribus hausi (‘I drank in many things with my ears’, 14.309), with reference to his (narrative) experiences during the year on Circe’s island as part of Ulysses’ crew.

18. See Littlefield, David J., ‘Pomona and Vertumnus: A Fruition of History in Ovid’s Metamorphoses‘, Arion 4 (1965), 465–73Google Scholar; Fantazzi, Carl, ‘The Revindication of Roman Myth in the Pomona-Vertumnus Tale’ in Barba, N.et al. (edd.),, Ovidianum (Bucharest 1976), 283–93Google Scholar. Davis (n. 3 above), 67, calls it ‘the final love story’ for which the others he has discussed ‘function as foil’.

19. In his discussion of Pomona’s ‘transcendence of the norm’ of the ‘nymph-huntress who remains impervious to amor’ (66f.), Davis (n. 3 above), 70, notes that her ‘defensive retreat to the garden appears in a sympathetic light when we learn who her “suitors” are: the rough and erotically truculent numina of the countryside … In fine, though the Latin nymph “rejects many Csuitors” according to the standard formula, she does so out of self-protection rather than inner frigidity of character.’

20. Others include: the Lycian farmers (6.317–81), told by e quibus unus (‘one of them [sc. those impressed by the story of Niobe]’, 317; also called nescio quis, ‘someone or other’, 382); Marsyas’ defeat and punishment by Apollo (6.383–400), told by a ‘second’ (alter, 383) anonymous Lycian; Aesacus’ attempted rape of Hesperia (11.751–95), recounted by aliquis senior (‘some old man’, 749); and the foundation of Croton (15.12–57), related by e senioribus unus/… veteris non inscius aevi (‘one of the old men, knowledgeable about ancient times’, 10f.).

21. See Davis (n. 3 above), 63–66, on ‘Deviation from the norm: the case of Salmacis’. This is all the more interesting, given the fact that the anti-sexual nymph is Ovid’s own conception, ‘a narrative type which Ovid develops with some care’, as Davis puts it (63). For an earlier discussion of the significance of this ‘seemingly eccentric’ treatment of nymphs by Ovid, see Curran (n. 3 above), 230f.

22. This term I take from Otis (n. 3 above), who refers to ‘the motif of the determined virgin courted by the passionate god’ (78). Curran (n. 3 above)), 230, uses the phrase ‘totally dedicated virgins’ in his discussion of Ovid’s peculiarly anti-sexual nymphs. Curran’s use of the word ‘dedicated’ is perhaps more accurate, since this choice of sexual orientation is presented by Ovid as literal dedication to and imitation — even to the point of physical resemblance — of the virgin Diana.

23. Davis (n. 3 above), 63ff., discusses Salmacis as a deviant from the norm of the huntress-nymph; he does not pick out her running for special comment, simply including it under the more general rubric of ‘athletics’, in noting ‘her disinterestedness in, and unsuitability for … the related pursuits of hunting and athletics’ (64). Ahl, however (n.3 above), 145, makes the intriguing comment that ‘running is natural activity for the daughters, such as Io or Daphne, of running streams’. This remark is prompted by the verb fugiebat in 1.597 of Io; thus Ahl is not so precise as I am in distinguishing between currere and fugere. Perhaps, however, if his interpretation is valid, Salmacis does not ‘run’ because as a pool rather than a stream she is not ‘running’ water. The spring Arethusa has become running water; she refers to her efforts at escape both as fleeing (Jugio, 5.601; ut fitgere, 605, in a simile) and as running (currebam, 606; currere sustinui, 609; tolerare diu cursus, 610; cucurri, 613). Curran (n. 3 above), 234, makes much of the way Ceres phrases the question which leads to this narrative quae tibi causa jugae (573). Parry (n. 7 above), 270, considers the vocabulary of flight (and pursuit) only briefly, observing that ‘threats to virginity … almost always involve a literal chase, with frequent recourse to the verbs petere and fugere, and other related words of hostility and violence. Occasionally the interchangeability of hunting and sexual terminology is underlined by the poet.’ Given Parry’s focus on the hunt (literal and sexual) and on violence, it is understandable that he not remark on forms of currere and cursus.

24. This technique of ‘surrogation’, as I call it, is responsible for what scholars have labelled Ovid’s ‘Iliad’ and Aeneid’. For a sophisticated consideration of both these, see Ellsworth, James D., ‘Ovid’s Iliad (Metamorphoses 12.1–13.622)’, Prudentia 12 (1980), 23–29Google Scholar, and ‘Ovid’s “Aeneid” Reconsidered (Met. 13.623–14.608)’, Vergilius 32 (1986), 27–32Google Scholar; the latter is structured around the three love-triangles (Galatea/Acis/Polyphemus; Scylla/Glaucus/Circe; Picus/Canens/Circe) which comprise this section of the Metamorphoses. An earlier discussion of Ovid’s ‘Aeneid’ appears in the section so titled of Galinsky, G. Karl, Ovid’s Metamorphoses: An Introduction to the Basic Aspects (Berkeley and Los Angeles 1975), 217–51Google Scholar. In ‘Ovid’s Iliad…’, Ellsworth (24 and n. 5) remarks on ‘the frequent appearance of the phrase “Ovid’s Aeneid” ’ in Ovidian criticism’, and cites two examples in addition to Galinsky; he begins ‘Ovid’s “Aeneid” …’ (1 and n. 4) with a brief survey of recent opinions, those of Otis, Segal and Galinsky. Perhaps the first occurence of a phrase of this type in the current revival of interest in Ovid is buried in a footnote; Frahkel (n. 6 above) uses the phrase matter-of-factly and without quotation marks in a note (222 n. 81) to his discussion of the Lapiths and Centaurs: ‘While starting out on his own miniature Iliad, Ovid takes his cue from the First Book of the original Iliad (262ff.), where Nestor tells of the Lapiths and their fight with the Centaurs.’ Speaking with reference specifically to Ovid's use of Macareus as a narrator, Due, Otto Steen, Changing Forms: Studies in the Metamorphoses of Ovid (Copenhagen 1974), 84Google Scholar, gives a good account of ‘surrogation’: ‘It should be noted that the function of the Achaemenides episode in Ovid is to introduce Macareus as a narrator, i.e. to escape from Vergilian subject-matter into a Homeric one. In these four last books of the Metamorphoses both Homer and Vergil are used as frames, and to contemporary readers, and at least to some modern classical scholars, the very dexterity of the poet in filling these frames with anything but what they were originally designed to contain would be appreciated as an important element of the experience.’ Referring to the same episode, Dörrie (n. 15 above, 1969), 82f., observes that in Book 13 Ovid begins the wanderings of Odysseus, but that he avoids direct competition with Homer by introducing characters as eye-witnesses, as for instance Achaemenides at the beginning of Book 14; except for the notion that Ovid shunned competition with Homer (or Virgil), I concur.

25. Ahl (n. 3 above), 55f., uses 14.312 to illustrate the first principle of Varronian etymological wordplay, namely that the basic unit for purpose of play is not the word but the syllable, remarking that Macareus ‘comments aptly cum duce … meo Circe dum solA MORatur’.

26. Bömer, Franz, P. Ovidius Naso Metamorphosen (Heidelberg 1976), ii.219ad 4.766f.Google Scholar, explains cultusque genusque locorum represents ‘das Neue … das der Fremde von dem Einheimischen (oder umgekehrt …) wissen will’, but he does not produce any parallel instances, either from the Metamorphoses or from other works, of the stranger interrogating the native. On the same page ad 4.765f., he cites the prototype of the tales of Odysseus, and the parallel of Dido and Aeneas, for the fact that ‘nach dem Mahle kommen Erzählungen und Berichte zu ihrem Recht’. Gieseking (n. 14 above), also discusses banquets (and conversations generally) as the frame for tales in the Metamorphoses (36–40), and banquets as frames in Ovid’s predecessors Homer, Apollonius and Callimachus (67–73). The Ovidian versions of the heroes Perseus, Theseus, Ulysses and Aeneas are peculiarly reticent about themselves, especially in comparison with Homer’s Odysseus and Virgil’s Aeneas. For the latter two, Ovid substitutes other fictional narrators’ accounts of the adventures or events contemporary with them, in that technique I call surrogation.

27. For this idea, and the insight about the phrase non falsa pericula, I am indebted to Professor Sara Mack, both for the MS. of her work in progress on Ovid for the Ttale University Press ‘Hermes’ series which she generously showed me, and for conversations we have had about various aspects of the Metamorphoses.