¹ For a stimulating reconsideration of the masculinity of epic, best understood as a creative tension between the stereotypical definitions of epic thematics provided by Roman poets in programmatic passages, on the one hand, and their actual poetic practice, on the other, which typically honours those definitions in the breach, see Hinds, S. J., ‘Essential epic: genre and gender from Macer to Statius’, in M. Depew and D. Obbink, Matrices of Genre: Authors, Canons and Society (2000), 221–44Google Scholar.

² Hardie, P. R., Virgil, Aeneid Book IX (1994)Google Scholar, ad loc: ‘V.'s readers will take sinite arma uiris in the further sense of a command to leave the world of martial epic.’

³ Debrohun, J. B., ‘Redressing elegy's puella: Propertius IV and the rhetoric of fashion’, JRS 84 (1994), 41–63, at 48Google Scholar.

⁴ Fowler, D. P., ‘God the father (himself) in Vergil’, PVS 22 (1996), 35–52, at 41Google Scholar; reprinted in D. P. Fowler, Roman Constructions: Readings in Postmodern Latin (2000), 218–34, at 225.

⁵ Hardie, P. R., The Epic Successors of Virgil: A Study in the Dynamics of a Tradition (1993), 102Google Scholar.

⁶ Fowler, op. cit. (n. 4), 41/224. Cf. Laird, A. J., Powers of Expression, Expressions of Power: Speech Presentation and Latin Literature (1999), 192–205Google Scholar.

⁷ cf. A. Schiesaro, ‘Rhetoric, politics and didaxis in Lucretius’, forthcoming. Schiesaro finds in the divergent models of instruction visible in the DRN a reflection of developments in contemporary educational practice, as the ‘Catonian’ paradigm of absolutely authoritative communication from father to son, which corresponds closely to Epicurus' unilateral transmission of patria praecepta (3.9–10) to the infinitely receptive Lucretius, began to encounter competition from a less starkly prescriptive paedagogical model where the teaching might be assessed on its own merits by a pupil possessed of a certain degree of autonomy, the latter being a better account of the relationship between Lucretius and his pupils, Memmius and the reader. Epic in general, needless to say, clung to the older paradigm.

⁸ Keith, A., ‘Versions of epic masculinity in Ovid's Metamorphoses’, in Hardie, P. R., Barchiesi, A. and Hinds, S. J. (eds), Ovidian Transformations: Essays on Ovid's Metamorphoses and its Reception, PCPhS Supp. 23 (1999), 214–39, at 227–8Google Scholar.

⁹ Keith, op. cit. (n. 8), 228; Horsfall, N., ‘Epic and burlesque in Ovid, Met. VIII.260 ff.,’ CJ 74 (1979), 319–32, at 330Google Scholar.

¹⁰ Adams, J. N., The Latin Sexual Vocabulary (1990), 14–22Google Scholar (telum), 57 (opus), and 69–70 (uirilis, uirilia).

¹¹ Adams, op. cit. (n. 10), 93 n. 3.

¹² Kennedy, D. F., The Arts of Love: Five Studies in the Discourse of Roman Love Elegy (1993), 59–60Google Scholar.

¹³ Bettini, M., ‘A proposito dei versi sotadei, greci e romani: con alcuni capitoli di “analisi metrica lineare”’, MD 9 (1982), 59–105, at 85–6Google Scholar. In such sexual double entendre, as well as in its persistent parody of epic, the Metamorphoses owes a lot to the sotadean tradition which is Bettini's subject. Compare Demetrius, Eloc. 189 on the sotadean line itself: μεταμεμορϕωμένῳ ἔοικεν ὁ στίχος, ὥσπερ ὁι μυθευόμενοι ἐξ ἀρρένων μεταβάλλειν εἰς θηλείας, ‘the line is as if it has changed its shape, like those characters in stories who change from male to female’.

¹⁴ Demetrius identifies it as an example of σεμναὶ χάριτες καὶ κεγάλαι, ‘charm of a dignified and impressive kind’. Valerius Probus criticizes Virgil for ‘barely conveying’ (‘exigue secutus sit’) ‘the flower’ of Homer's whole simile, than which ‘no greater nor more comprehensive praise of beauty could be expressed’ (‘nulla maior cumulatiorque pulcritudinis laus dici potuerit’).

¹⁵ We should be clear, however, that whilst Ovid may construct the Aeneid as the exemplar of epic norms, it is often the case that Ovid works with tensions inherent in the epic mastertext. Here in Aeneid 1 the Diana simile is already generically hazardous. Philip Hardie per litteras suspects the influence of Cornelius Gallus on this first glimpse of a Diana-like beauty: cf. P. R. Hardie, Ovid's Poetics of Illusion (2002), 44. See Hinds, op. cit. (n. 1), 223–4 for Roman poets' increasing tendency ‘to appeal to unmixed, essentialized and unchanging conceptions of the genre in their poetological policy statements’ the more they ‘mix, blur and hybridize generic categories in their poetic practice’.

¹⁶ cf. Feeney, D. C., The Gods in Epic: Poets and Critics of the Classical Tradition (1991), 72–3Google Scholar on the dangers for epic of too insistent a concentration on ‘the physicality of the gods’.

¹⁷ Arist., Poet. 1448a26–7, I448b34–5 (where the content of epic, τὰ σπουδαῖα, stands in contrast to τὸ τῆς κωμῳδίας σχῆμα), 1449b9–10.

¹⁸ On this quality of ingenium, consistently marked as a characteristic of Ovid by the ancients, see Kenney, E. J. in Kenney, E. J. and Clausen, W. (eds), The Cambridge History of Classical Literature, Vol. 2: Latin Literature (1982), 440CrossRefGoogle Scholar: ‘a quality which continually pervades and informs the Metamorphoses: what the Romans called ingenium and the English Augustans wit.’ Cf. Sen., Contr. 2.2.9, 12; Quint., Inst. 10.1.89, 98.

¹⁹ I am assuming that what survives in our texts of the Naturales Quaestiones is not the full quotation as intended by Seneca but the opening and closing clauses of a longer quotation, from between which an usque ad (vel sim.) has fallen out in the course of the tradition. This would remove the clumsiness of the unexplained juxtaposition of 285 and 290 in the MSS (the second of which is not even a complete hexameter), and allow some grounds for Seneca's assertion that Ovid ‘tantae confusionis imaginem cepit’ in the quotation. It is otherwise a striking coincidence that the two lines in the MSS bracket so clearly defined a section of Ovid's account. The text has apparently suffered other corruption: Seneca's text has labant at 290, Ovid's latent.

²⁰ The quotation comes from his Preface to Jacob Tonson's collection of translations of the Heroides, Ovid's Epistles (1680): Hooker, E. N. and Swedenberg, H. T., The Works of Dryden, Vol. 1: Poems 1649–1680 (1961), 112Google Scholar. The failure to do what is fitting is an implication also of Seneca's term ineptiae, which are literally instances of behaviour betraying no sense of what is appropriate. Beck, M., ‘Inepta loci (Sen. Contr. 1, 2, 22): ein Ovidianum?’, Hermes 129 (2001), 95–105Google Scholar, has argued for a new interpretation and reading of the supposed quotation of Ovid at Sen., Contr. 1.2.22, where (according to the received text) Scaurus greets a rhetorical solecism by Murredius with ‘Ouidianum illud: “inepta loci”’. The expression inepta loci is only found at Priap. 3.8, in a text which is unlikely to be by Ovid. Beck's emendation, ‘Ovidianum illud inepti loci’, which puts the whole expression in Scaurus' mouth, and means ‘That Ovid-like remark of yours is rhetorically inappropriate (belongs to an inappropriate rhetorical locus)’ (104), is persuasive. It could be added that it is not only the obscenity of Murredius’ remark that marks it as Ovidian (104 n. 45) but also the very fact that it is not the appropriate thing to say, something for which Ovid was notorious. Cf. Seneca the Elder's comments at Contr. 2.2.9 on Ovid's tendency as a declaimer to ‘run through the topoi in no fixed order’ (‘sine certo ordine per locos discurrebat’) and at 2.2.12 to the effect that Ovid (again when declaiming) found all argumentatio tiresome (‘molesta illi erat omnis argumentatio’), argumentatio denoting a structured exposition of the case. Ordered structure, clearly, was always anathema to Ovid, but as we shall investigate later, disorder in the Metamorphoses is always accompanied by its opposing principle. Cf. Solodow, J. B., The World of Ovid's Metamorphoses (1988), 34Google Scholar: The poem invites us to look for structures within it and makes a number of proposals, and then it systematically defeats them all … It clearly strives for order, and in many different ways, but it never consistently achieves it: the poem might claim as a motto its own phrase, discors concordia (1.433, ‘an inharmonious harmony’). Instead it conveys a sense of dis-order, of orderings undone.

²¹ This last passage is considered in greater depth in Section VI below.

²² Jugemens des savans sur les principaux ouvrages des auteurs, revûs, corrigez, et augmentez par Mr. de la Monnoye (1722), Vol. 4, 136–7Google Scholar. Cf. D. Hopkins, ‘Dryden and Ovid's “Wit out of Season”’, in C. Martindale (ed.), Ovid Renewed: Ovidian Influences on Literature and Art from the Middle Ages to the Twentieth Century (1988), 167–90, at 170.

²³ Kinsley, J., The Poems of John Dryden (1958), IV. 1450Google Scholar: ‘The Manners of the Poets were not unlike. Both of them were well-bred, well-natured, amorous and libertine, at least in their writings, it may be also in their lives.’

²⁴ Kinsley, op. cit. (n. 23), IV. 1460: ‘I prefer, in our countryman, far above all his other stories, the noble poem of Palamon and Arcite, which is of the epic kind, and perhaps not much inferior to the Ilias, or the Aeneis.’

²⁵ Kinsley, op. cit. (n. 23), IV. 1451; cf. Hopkins, op. cit. (n. 22), 169.

²⁶ Kinsley, op. cit. (n. 23), IV. 1452.

²⁷ Elliott, A. G., ‘Ovid and the critics: Seneca, Quintilian and “seriousness”’, Helios 12 (1985), 9–20, at 10–11Google Scholar.

²⁸ cf. M. Coffey and R. Mayer, Seneca, Phaedra (1990), 13–14.

²⁹ Hopkins, op. cit. (n. 22), 170. In his version of the flood, for example, from Examen Poeticum (1693), Dryden out-Ovids Ovid. Seneca would have been extremely disappointed by Dryden's embellishments of 285–90, for example, where the superbly ironic pun on ‘household’ at Dryden 397 and the oxymoronic ‘watery wall’ threatening real masonry at 399 find no precedent in Ovid, but are in instinctive sympathy with the spirit of the Latin: ‘sapped by floods,/ Their houses fell upon their household gods,/ The solid piles, too strongly built to fall,/ High o'er their heads behold a watery wall.’

³⁰ D. E. Hill, Ovid, Metamorphoses I–IV (1985), ad loc: ‘Ovid's deliberately crude and explicit comparison … pokes fun both at divine and human pretensions.’

³¹ Hopkins, op. cit. (n. 22), 170 and n. 12. For example Addison polishes away Ovid's non-ending to Met. 2 with a bizarre climax which nevertheless fails to provide the climax the reader is looking for: on the Ovidian passage see D. P. Fowler, op. cit. (n. 4), 96–7/ 259; for Addison's embellishment, A. C. Guthkelch The Miscellaneous Works of Joseph Addison, Vol. 1: Poems and Plays (1914), 98.

³² Guthkelch, op. cit. (n. 31), 138.

³³ N. Vance, ‘Ovid and the nineteenth century’, in Martindale, op. cit. (n. 22), 215–31, at 215.

³⁴ On such movement along the ‘vertical axis’ in the Aeneid see, suggestively, P. R. Hardie, Virgil's Aeneid: Cosmos and Imperium (1986), 267–85.

³⁵ Nicoll, W. S. M., ‘Cupid, Apollo and Daphne (Ovid, Metamorphoses 1.452ff.)’, CQ 30 (1980), 174–82CrossRefGoogle Scholar.

³⁶ Apollo was the archetypal ephebe, in fact, according to Isidore (Orig. 11.2.10): ‘ephebi, id est a Phoebo dicti …’

³⁷ Saeuitia is an epic characteristic too, of course, most memorably embodied by the saeua Iuno of the Aeneid. The Cupid of the Amores initiates an elegiac programme of epic dimensions.

³⁸ Hardie, op. cit. (n. 2), ad 654–5. Ovid's arquitenens is a wonderfully resonant epithet, originating in the Saturnians of Naevius' Belium Punicum (frg. 30 Morel): ‘dein pollens sagittis inclutus arquitenens/ sanctus Ioue prognatus Pythius Apollo.’ Naevius' terms continue to find echoes as Ovid's episode unfolds.

³⁹ Hardie, op. cit. (n. 2), 199.

⁴⁰ On this issue see Holzberg, N., ‘Ter quinque uolumina as carmen perpetuum: the division into books in Ovid's Metamorphoses’, MD 40 (1998), 77–98Google Scholar.

⁴¹ R. Brown, ‘The Palace of the Sun in Ovid's Metamorphoses’, in M. Whitby, P. R. Hardie and M. Whitby (eds), Homo Viator: Classical Essays for John Bramble (1987), 211–20. On the parallels between Books 1 and 2 see especially 215–17.

⁴² For parallels see A. Sharrock, Seduction and Repetition in Ovid's Ars Amatoria II (1994), 115–17.

⁴³ cf. A. Zissos and I. Gildenhard, ‘Problems of time in Metamorphoses 2’, in Hardie, Barchiesi and Hinds, op. cit. (n. 8), 31–47, at 35 n. 17 for further metapoetic terminology in Phaethon's chariot ride.

⁴⁴ cf. E. J. Kenney, ‘The style of the Metamorphoses’, in J. W. Binns (ed.), Ovid (1973), 116–53, at 117: poem. ‘Smoothness and speed are likewise the salient characteristics of Ovid's hexameter. Critics who merely miss in Ovid the weight, sonority and expressiveness of Virgil are failing to recognize the great difference, not only between the two poets, but between their two undertakings’ (my italics). A referee suggests that pondus carries asexual innuendo as well.45 M. S. Santirocco, Unity and Design in Horace's Odes (1986), 27–30, with further bibliography on the poem.

⁴⁶ Santirocco, op. cit. (n. 45), 29; cf. Elder, J. P., ‘Horace, C., 1, 3’, AJPh 73 (1952), 140–58, at 156Google Scholar: ‘Heroism is nobility, but is also folly, but a folly well worth the ultimate suffering.’

⁴⁷ Sharrock, op. cit. (n. 42), 114.

⁴⁸ Sharrock, op. cit. (n. 42), 123–4.

⁴⁹ Sharrock, op. cit. (n. 42), 87–195: Horace's allusions to the myth are discussed at 112–26.

⁵⁰ A. S. Hollis, Ovid, Metamorphoses Book VIII (1970) has an excellent note at 224, ‘caelique cupidine tactus’, positing ‘a proverb for extravagant ambition’ underlying Ovid's expression, Hor., Carm. 1.3.38, Virg., Geo. 4.325, and Rhianus fr. 1.15 Powell.

⁵¹ Ars Am. 2.77–8: ‘hos aliquis, tremula dum captat harundine pisces,/ uidit et inceptum dextra reliquit opus’; Met. 8.217–20: ‘hos aliquis, tremula dum captat harundine pisces,/ aut pastor baculo stiuaque innixus arator/ uidit et obstipuit, quique aethera carpere possent,/ credidit esse deos’. The two versions, which the identical opening line insists that we compare, are a brilliantly subtle exercise in metrical differentiation. The first beautifully exploits the tendency of the elegiac couplet to expend energy in the hexameter and relax in the pentameter (Am. 1.1.17–18, 27): in this case the fisherman acts energetically in the hexameter and is paralysed with astonishment in the pentameter. The second passage (as Sharrock, op. cit. (n. 42), 180–1 suggests) exploits, by contrast, the expansive possibilities of continuous hexameters.

⁵² Sharrock, op. cit. (n. 42), 142.

⁵³ Sharrock, op. cit. (n. 42), 177.

⁵⁴ cf. M. W. Edwards, Homer, Poet of the Iliad (1987), 109: ‘Fundamentally, a simile … is a technique of expansion’, and as such, like periphrasis, a favourite mode of epic.

⁵⁵ Hollis, op. cit. (n. 50), ad loc.

⁵⁶ cf. Zanker, G., Realism in Alexandrian Poetry: a Literature and its Audience (1987), 214Google Scholar on Callimachus' Hecale, ‘realistic in a manner that we have come to regard as typical of Callimachus and of the Alexandrian movement as a whole’.

⁵⁷ See Oliensis, E., ‘Return to sender: the rhetoric of nomina in Ovid's Tristia’, Ramus 26 (1997), 172–93, at 182–3 and 192CrossRefGoogle Scholar n. 18 for Ovid's continued metapoetical deployment of the figure of Icarus in his exile poetry.

⁵⁸ On the issue of speech in the Aeneid as a means of assertion of authority, see Laird, op. cit. (n. 6), e.g. on Ascanius (192, ‘As a child, generally in the presence of his father, he is normally seen and not heard’), on the Aeneid in general (196, ‘Hierarchy provides the principal explanation for the frequent occurrence of single, unanswered speeches’), and on epic's kinship with public discourse (204, ‘epic is a discourse of power in some ways comparable to a state decree or military command’).

⁵⁹ Fowler, op. cit. (n. 4), 42/226.

⁶⁰ cf. Wheeler, S. M., Narrative Dynamics in Ovid's Metamorphoses, Classica Monacensia 20 (2000), 69Google Scholar ‘the increasing loss of parental control over the destiny of their children’ in the myth sequence of the second half of Met. 1, which ‘complements the repeated dissolution of cosmic order at the beginning of the Metamorphoses’.

⁶¹ Fuhrer, T., ‘Der Götterhymnus als Prahlrede — zum Spiel mit einer literarischen Form in Ovids Metamorphosen’, Hermes 127 (1999), 356–67, at 359Google Scholar.

⁶² We may note in passing the discreet influence which C/cupid continues to exert on events. Cf. p. 89 and n. 96 below.

⁶³ Hardie, op. cit. (n. 34), 78.

⁶⁴ The noun is common in the seventeenth century, cf. (from 1693) Narcissus Luttrell, A Brief Historicall Relation of State Affairs, 1678–1714 (1857), III.2: ‘On Sunday night last 3 hectors came out of a tavern in Holborn, with their swords drawn, and started to break windows.’

⁶⁵ Hardie, op. cit. (n. 5), 102–3.

⁶⁶ Morgan, Ll., ‘Quantum sat erit: epic, acne and the fourth Eclogue’, LCM 17 (1992), 76–9Google Scholar.

⁶⁷ Nisetich, F., The Poems of Callimachus (2001), 234Google Scholar.

⁶⁸ Hopkinson, N., A Hellenistic Anthology (1988)Google Scholar, ad loc.

⁶⁹ Feeney, op. cit. (n. 16), 78.

⁷⁰ Feeney, op. cit. (n. 16), 81.

⁷¹ Nelis, D., Vergil's Aeneid and the Argonautica of Apollonius Rhodius (2001), 94Google Scholar.

⁷² cf. AP 12.75, 76, 77, and 78 for wings, bow and arrows as the distinguishing features of Eros. Thus 76 (Meleager), ‘If Love had neither bow, nor wings, nor quiver,/ nor barbed arrows of desire dipped in fire,/ never, I swear it by the winged one himself, could you tell/ from their appearance which was Zoilos and which Love’. In these poems, by Asclepiades and possibly Posidippus as well as Meleager, Eros can be confused with the prepubescent object of desire, and Strata's collection of pederastic epigrams, the Μοῦσα Παιδική, suggests another application for imagery of the child, with its own set of power relations, specious and real.

⁷³ cf. Hinds, S. J., The Metamorphosis of Persephone: Ovid and the Self-Conscious Muse (1987), 167 n. 45Google Scholar on ‘Ovid's “restoration” to Cupid of the traditional bow and arrow with which he had operated in the Apollonian version, but which Virgil had suppressed’.

⁷⁴ Zanker, op. cit. (n. 56), 207.

⁷⁵ Feeney, op. cit. (n. 16), 89. Cf. Hunter, R. L., Apollonius of Rhodes, Argonautica Book III (1989)Google Scholar, at 135 on the cosmic quality of the ball which Cypris is offering to Eros.

⁷⁶ Hinds, op. cit. (n. 73), 133–4; cf. Heinze, R., Ovids elegische Erzählung (1919), 7Google Scholar and n. 2.

⁷⁷ Further evidence of Greek sources for childish metapoetical imagery is to be found in the related Latin and Greek ‘game’ terminology to denote poetic activity in the ‘lesser’ genres. Lusus and cognates are a regular presence in such contexts: see Catull. 50.2, Virg., Ecl. 6.1, Geo. 4.565, Ov., Fast. 2.6 (‘cum lusit numeris prima iuuenta suis’), Plin., Ep. 9.25.1 (‘lusus et ineptias nostras’). There is a ready parallel in Greek (and Latin) works entitled Paegnia, Erotopaegnia or Technopaegnia: see RE s.v. παίγνιον (citing, for example, Aelian, Hist. An. 15.19, Θεόκριτος ὁ τῶν νομευτικῶν παιγνίων συνθέτης; E. Courtney, The Fragmentary Latin Poets (1993), 119. For the interest in children evinced in both poetry and art of the Hellenistic period see T. B. L. Webster, Hellenistic Poetry and Art (1964), ch. 8. In both forms it clearly represents a departure from conventional objects of artistic representation.

⁷⁸ Casali, S., ‘Ovidio e la preconoscenza delta critica: qualche generalizzazione a partire da Heroides 14’, Philologus 142 (1998), 93–113, at 96CrossRefGoogle Scholar: ‘Ovidio, come tutti i poeti, sa quello che il lettore-critico andrà a cercare nella sua opera. Quello che distingue Ovidio è che lui, come non tutti i poeti, esibisce questa preconoscenza.’

⁷⁹ Casali, S., ‘Apollo, Ovid and the foreknowledge of criticism (Ars 2.493–512)’, CJ 93 (1997–1998), 19–27, at 25Google Scholar.

⁸⁰ Solodow, op. cit. (n. 20), 27.

⁸¹ E. J. Kenney in Kenney and Clausen, op. cit. (n. 18), 432.

⁸² Farrell, J., ‘The Ovidian corpus: poetic body and poetic text’, in Hardie, Barchiesi and Hinds, op. cit. (n. 8), 127–41Google Scholar. See especially 130–1 on literary applications of the term corpus.

⁸³ Barchiesi, A., Il poeta e il principe. Ovidio e il discorso Augusteo (1994), 248Google Scholar. There is an English version of the relevant section of the book at ‘Endgames: Ovid's Metamorphoses 15 and Fasti 6’, in Roberts, D. H., Dunn, F. M. and Fowler, D. P. (eds), Classical Closure: Reading the End in Greek and Latin Literature (1997), 181–208Google Scholar, with the ‘unclear boundaries’ at 181–3.

⁸⁴ Feeney, D. C., ‘Mea Tempora; patterning of time in the Metamorphoses’, in Hardie, Barchiesi and Hinds, op. cit. (n. 8), 13–30, at 18Google Scholar.

⁸⁵ Fowler, D. P., ‘First thoughts on closure: problems and prospects’, MD 22 (1989), 75–122 at 81–2Google Scholar, and Roman Constructions: Readings in Postmodern Latin (2000), 239–83, at 245, citing B. H. Smith, Poetic Closure: A Study of How Poems End (1968), 172–82.

⁸⁶ Holzberg, N., Ovid. Dichter und Werk (1997), 136Google Scholar.

⁸⁷ Farrell, op. cit. (n. 82), 131.

⁸⁸ ‘Technisch erfolgt der Übergang zu Pelops “gewaltsam”’, in Bömer, F., P. Ovidius Naso, Metamorphosen: Kommentar, Buck VI–VII (1976), ad 6.401–11Google Scholar, citing Kraus, RE 18.1.1942.43.

⁸⁹ Solodow's transitio per absentem: op. cit. (n. 20), 43–4.

⁹⁰ Ivory, with its misleading similarity to flesh (it was of course an ivory statue which deceived Pygmalion, Met. 10.243–97), might itself be considered an intrinsically dubious substance. It was the material of the gates of deception, after all (Hom., Od. 19.564–5; Virg., Aen. 6.895–6), and Homer's play on words, οἳ μέν κ΄ ἔλθωσι διὰ πριστοῦ ἐλέϕαντος, οἵ ῥ΄ ἐλεϕαίρονται, ‘those dreams that pass through sawn ivory deceive’, will have been influential.

⁹¹ Thomas, R. F., Virgil, Georgics (1988), ad loc.Google Scholar

⁹² Solodow, op. cit. (n. 20), 46. Cf. with this reading of Ovid's Pelops Steiner, D., Images in Mind: Statues in Archaic and Classical Greek Literature and Thought (2001), 282–6Google Scholar on the self-reflexive potential of Pindar's account of Pelops in Olympian 1.

⁹³ cf. Feeney, op. cit. (n. 16), 206.

⁹⁴ Connors, C., ‘Seeing cypresses in Virgil’, CJ 88 (1992–1993), 1–17. at 4–12Google Scholar.

⁹⁵ For a discussion of how Gérôme's representations of Pygmalion reflect on his own artistic enterprise, ‘an obsessive and narcissistic concern with the nature and power of his own art’, see Hardie, op. cit. (n. 15), 206–26.

⁹⁶ For other intimations of Cupid's agency see his apparent defeat by Medea's sense of ‘Right, Duty and Shame’ at Met. 7.72–3, the heedlessness which causes the Venus and Adonis episode at 10.525–6, his exculpation from responsibility for the incestuous impulse of Myrrha, 10.311–12, less than convincing after his involvement in Byblis' passion for her twin brother (9.543–4), and the confirmation at 10.26–9 that the extension of Cupid's power to the Underworld envisaged in Book 5 has been achieved. Ovid's influence in this respect shows in Statius' Achilleid, a text much indebted to the Metamorphoses, which I interpret as a fragment ‘by design’, a poem which selfconsciously fails to attain to epic status and gives out, not coincidentally, just as its protagonist, ‘nee adhuc maturus Achilles’ (1.440), threatens to reach adulthood: when the text fails at 2.167 Achilles is leaving Scyros on the way to Troy, and has just completed an account of his childhood. On the Ovidianism of the Achilleid see G. Rosati, ‘Momenti e forme della fortuna antica di Ovidio: l'Achilleide di Stazio’, in M. Picone and B. Zimmermann (eds), Ovidius Redivivus (1994), 43–62; S. J. Hinds, Allusion and Intertext: Dynamics of Appropriation in Roman Poetry (1998), 135–44; and Hinds, op. cit. (n. 1), 236–44, esp. 242–4, for a clever reading of the ‘shipboard storytelling’ between Achilles and Ulysses as an adumbration of ‘two ways other than Statius' of beginning an Achilles epic’.

⁹⁷ K. Galinsky, ‘Ovid's Metamorphoses and Augustan cultural thematics’, in Hardie, Barchiesi and Hinds, op. cit. (n. 8), 103–11, at III.

⁹⁸ Boyle, A. J., The Imperial Muse: Ramus Essays on Roman Literature of the Empire, Vol. 1 (1988), 1Google Scholar.

⁹⁹ Martindale, op. cit. (n. 22), 2

¹⁰⁰ See Cunningham, M. P., ‘Ovid's poetics’, CJ 53 (1958), 253–9Google Scholar, who parallels Controv. 2.2.12 with Am. 3.1.10, where the attractiveness of Elegy's blemish makes a virtue of the imbalance of the elegiac couplet. Elegy, with this inherent flaw, always remained Ovid's dominant aesthetic.

¹⁰¹ cf. Tissol, G., The Face of Nature: Wit, Narrative and Cosmic Origins in Ovid's Metamorphoses (1997), 11–88Google Scholar, on the ‘transgressive language’ of the poem, and 14–15 on the ‘semantic short-circuit’ of the self-cancelling paradox, a trope very prominent in the work, and which is a key, as Tissol argues, to Ovid's quite profound disruption of the reader's cultural expectations.

¹⁰² Nicoll, op. cit. (n. 35), 181, citing Prop. 4.6.35–6.

¹⁰³ Fowler, op. cit. (n. 4), 46–7/230–1.

¹⁰⁴ Reproduction at T. Clark, Art and Propaganda in the Twentieth Century: the Political Image in the Age of Mass Culture (1997), 6.

¹⁰⁵ Fowler, op. cit. (n. 4), 41/225.

¹⁰⁶ Fowler, op. cit. (n. 4), 49/233–4.

¹⁰⁷ Elliott, op. cit. (n. 27), 10–11.