Versions of this paper have been delivered at the Courtauld Institute and at the American Philological Association's Annual Conference in New Orleans in 1992, as well as at the Cambridge Philological Society. I am grateful to all those present for their comments on these occasions, especially to John Henderson and Ian DuQuesnay for various discussions of Petronius, to John Bodel for sending me an unpublished paper, and to Catherine Connors, Simon Goldhill and David Konstan for their careful readings of an early draft. My deepest thanks are due to Richard Hunter's excellent advice, Michael Reeve's very incisive observations and to Froma Zeitlin, who commented on my reading in a missive which was insightful, critical, helpful and lengthy far beyond the call of academic duty.

1. See further the editors' introduction to Elsner, J. and Masters, J. (eds.), Reflections of Nero: culture, history and representation (London, 1994, forthcoming)Google Scholar.

2. On Petronius and the Satyrica generally, see Sullivan, J. P., The Satyricon of Petronius (London, 1968)Google Scholar, the articles collected in Arion 5 no. 3 (1966); Slater, N., Reading Petronius (Baltimore and London, 1990)Google Scholar. On the Satyrica as a satiric text, I have found most useful: J. P. Sullivan, ‘Satire and realism in Petronius’, in idem (ed.), Critical essays on Roman literature: satire (London, 1963) 73–92; Zeitlin, F. I., ‘Petronius as paradox: anarchy and artistic integrity’, TAPA 102 (1971) 631–84Google Scholar; Slater (1990). On the Satyrica specifically as a parody of the ancient novel, see Heinze, R., ‘Petron und der Greichische Roman’ Hermes 34 (1899) 494–519Google Scholar; Courtney, E., ‘Parody and literary allusion in the Menippean satire’, Philologus 106 (1962) 86–100, esp. 92-100CrossRefGoogle Scholar. On the uses and dangers of satire in the writing of ancient history, see Braund, S. H. (ed.), Satire and society in ancient Rome (Exeter, 1989) Introduction (esp. 1–3)Google Scholar.

3. On the art gallery episode of the Satyrica as a parody of themes in the romances, see Walsh, P. G., The Roman novel, (Cambridge, 1970) 93–7Google Scholar; Slater, N., ‘“Against interpretation”: Petronius and art criticism’, Ramus 16 (1987) 165-76, 169Google Scholar.

4. This description has sufficiently convinced art historians that many regard the gallery as having existed; see especially Lehmann-Hartleben, K., ‘The Imagines of the Elder Philostratus’, Art Bulletin 23 (1941) 16–44CrossRefGoogle Scholar, with the discussion of Bryson, N., ‘Philostratus and the imaginary museum’, in Goldhill, S. and Osborne, R. (eds.), Art and text in ancient Greek culture, (Cambridge, 1994), forthcoming)Google Scholar.

5. On the device of the introductory painting, see especially von Fleschenberg, O. Schissel, ‘Die Technik des Bildeinsatzes’, Philologus 72 (1913) 83–114Google Scholar; Bartsch, S., Decoding the ancient novel (Princeton, 1989) 40–79Google Scholar. On Longus, see Hunter, R., A study of Daphnis & Chloe (Cambridge, 1983) 38–51 on the prologueCrossRefGoogle Scholar; MacQueen, B. D., Myth, rhetoric and fiction: a reading of Longus's Daphnis and Chloe (Lincoln, Nebraska, and London, 1990) 15–30 on the prologueGoogle Scholar; Zeitlin, F. I., ‘The poetics of Eros: nature, art and imitation in Longus' Daphnis and Chloe’, in Halperin, D. M., Winkler, J. J. and Zeitlin, F. I. (eds.), Before sexuality: the construction of erotic experience in the ancient Greek world (Princeton, 1990) 417–64Google Scholar.

6. My thanks to J. Henderson for his interest in these awful puns.

7. My thanks to M. Reeve for this suggestion. Delz's alternative, monoglenon, would lose this joke while preserving the Apelles/appellant pun (see Delz, J., ‘Petron Satyrica ed. K. Muller’, Gnomon 34 (1962) 676-84, 678)Google Scholar.

8. As suggested by N. Hopkinson in discussion at the Philological Society.

9. Subsequently suggested by J. Henderson.

10. ergo amor etiam deos tangit: this is as crass a truism as Encolpius ever expresses, but also a parody of Vergil's et mentem mortalia tangunt (Aeneid 1.462) also from an ecphrastic passage, which will be evoked again later in Eumolpus' poem, the Troiae halosis, on which see below. On the many relations of the gallery episode to Vergil, see Zeitlin, F. I., ‘Romanus Petronius’, Latomus 30 (1971) 56–82Google Scholar.

11. Not quite the story given by the Homeric Hymn to Aphrodite 202–17. On the theme of Ganymede, with a number of parallel ‘transumptions’ of homoeroticism as art and art as homoeroticism in humanist writing and painting, see Barkan, L., Transuming passion: Ganymede and the erotics of humanism (Stanford, 1991) 48–74Google Scholar.

12. For an earlier reductio ad absurdum of naturalism, compare Trimalchio's comments at 52.1:

Myself, I have a great passion for silver. I own about a hundred four-gallon cups engraved with Cassandra killing her sons [Trimalchio has of course confused Cassandra with Medea], and the boys lying there dead – but you would think they were alive!

13. On Encolpius' reactions to the paintings, see esp. Slater (n. 2) 220–30 and idem (n. 3) 168–71.

14. Zeitlin (n. 5) 432.

15. Daphnis and Chloe proem 1–2, with the discussions of Zeitlin, (n. 5) 430–6 on ‘art and eros’ and Hunter (n. 5) 38–51; Achilles Tatius 1.1–2; 3.6–7; 5.3.

16. However, on the theme of homosexuality in the novels, see Effe, B., ‘Der Greichische Liebesroman und die Homoerotik’, Philologus 131 (1987) 95–108CrossRefGoogle Scholar.

17. Cf. Heinze (n. 2).

18. See e.g. Clarke, J. R., ‘The decor of the House of Jupiter and Ganymede at Ostia Antica: private residence turned gay hotel?’, in Gazda, E. K. (ed.), Roman art in the private sphere (Ann Arbor, 1991) 89–104Google Scholar.

19. On the Tabula, see Praechter, C., Cebetis Tabula quanam aetate conscripta esse videatur (Marburg, 1885)Google Scholar; Joly, R., Le Tableau de Cébès et la philosophie religieuse (Brussels, 1968)Google Scholar; and (for an annotated translation) Fitzberald, J. T. and White, L. M., The Tabula of Cebes (Chico, Calif., 1983)Google Scholar.

20. For a discussion of this issue see Elsner, J., Art and the Roman viewer: the transformation of art from the pagan world to Christianity (Cambridge, 1994, forthcoming)Google Scholar ch. 1 ‘Viewing and “the Real”: the Imagines of Philostratus and the Tabula of Cebes’.

21. On phantasia as the criterion for truth in Stoic thought, see Diogenes Laertius, Lives of eminent philosophers 7.49 on the Stoic Zeno, with Rist, J. M., Stoic philosophy (Cambridge, 1969) 133–55Google Scholar, Annas, J., ‘Truth and knowledge’, in Schofield, M., Burnyeat, M. and Barnes, J. (eds.), Doubt and dogmatism (Oxford, 1980) 84–104Google Scholar; C. C. W. Taylor, ‘All perceptions are true’, in Schofield, Burnyeat and Barnes, op. cit. 105–24; Watson, G., Phantasia in classical thought (Galway, 1988) 38–58Google Scholar. On earlier Greek views of phantasia, see Watson (1988) 1–37; Silverman, A., ‘Plato on Phantasia’, Classical Antiquity 10 (1991) 123–47CrossRefGoogle Scholar; Schofield, M., ‘Aristotle on the imagination’, in Lloyd, G. E. R. and Owen, G. E. L., Aristotle on mind and the senses (Cambridge, 1978) 99–140Google Scholar.

22. Cicero, Lucullus 77, with Ioppolo, A. M., ‘Presentation and assent: a physical and cognitive problem in early Stoicism’, CQ n.s. 40 (1990) 433–49 (esp. 433–41)CrossRefGoogle Scholar. See also Long, A. A., Hellenistic philosophy (London, 1974) 121–31Google Scholar; Imbert, C., ‘Stoic logic and Alexandrian poetics; in Schofield, Burnyeat and Barnes (n. 21) 183–216Google Scholar; Pollitt, J. J., The ancient view of Greek art (New Haven and London, 1974) 52–5, 61–3 and 293–7Google Scholar.

23. On ‘the transformation of phantasia’, see esp. Watson (n. 21) 59–95.

24. Longinus, De sublimitate 15.1:

Another thing which is very productive of grandeur, magnificence and urgency, my young friend, is visualisation (phantasia). I use this word for what some people call image production. The term phantasia is used generally for any thing which in any way suggests a thought productive of speech; but the word has also come into fashion for the situation in which enthusiasm and emotion make the speaker see what he is saying and bring it visually before his audience.

Quintilian, Inst. or. 6.2.29.

There are certain experiences which the Greeks call phantasiai and the Romans visiones, whereby things absent are presented to our imagination with such extreme vividness that they seem actually to be before our very eyes.

Quintilian, Inst. or. 10.7.15:

Those vivid conceptions of which I spoke and which, as I remarked, are called phantasiai, … must be kept clearly before our eyes and admitted into our hearts: for it is feeling and force of imagination that makes us eloquent.

Not only do these passages emphasise the creative importance of phantasia in inspiring the orator's eloquence, but that of Longinus also hints at a shift in meaning of the word in precisely the century when Petronius too was writing. The implication is that once phantasia moved from being a precise term in Stoic epistemology, it gradually became a concept for creative imagination which could be generally applied to a large number of creative activities.

25. Cicero, Orator 2.9:

Surely the great sculptor [Phidias], while making the image of Jupiter or Minerva, did not look at any person he was using as a model, but in his own mind there dwelt a surpassing vision of beauty; at this he gazed and all intent on this he guided his artist's hand to produce the likeness of the god. Accordingly … there is some thing perfect and surpassing in the case of sculpture and painting – an intellectual ideal by reference to which the artist represents those objects which do not themselves appear to the eye …

Cf. the Elder Seneca, Contr. 8.2 and 10.5.8; Quintilian, Inst. or. 12.10.9; Chrysostom, Dio, The Olympic discourse (Oratio 12) 70–1Google Scholar; Philostratus, Vit. Apoll. 6.13; Plotinus, Enneads 5.8.1. On the theme of phantasia and the artist, see Pollitt (n. 22) 52–5 and 203–5.

26. Here I am in conflict with the generally excellent account of Slater (n. 2), who tentatively suggests that Petroius may have held the Stoic doctrine of phantasia, or at least used it to attack simple views of mimesis (228–30). While I agree that mimesis is being ‘systematically satirized’, I suggest that Stoic doctrines are no less immune from satiric undermining. (In general, Petronius is as happy to take up a Platonic position for satiric purposes, such as attacking the Stoics, as he is to adopt any other position – including ones which attack Platonic views.) On the Stoicism present in Petronius' rhetoric, see Barnes, E. J., ‘Petronius, Philo and Stoic rhetoric’, Latomus 32 (1973) 787–98Google Scholar.

27. On Eumolpus, see particularly Walsh, P. G., ‘Eumolpus, the Halosis Troiae and the De Bello Civili’, Classical Philology 63 (1968) 208–12CrossRefGoogle Scholar; Zeitlin (n. 10); Beck, R., ‘Eumolpus Poeta, Eumolpus Fabulator: a study of characterization in the Satyricon’, Phoenix 33 (1979) 239–53CrossRefGoogle Scholar. Brief accounts of his characterisation include Sullivan, 1968 (n. 2) 230–1; Cameron, A. M., ‘Myth and meaning in Petronius’, Latomus 29 (1970) 397–425, esp. 415Google Scholar; Slater (n. 2) 91–5.

28. On the Eleusinian Eumolpus, see Richardson, N. J., The Homeric Hymn to Demeter (Oxford, 1974) 197–8Google Scholar.

29. On this context for Eumolpus, see the brief accounts of Schissel von Fleschenberg (n. 5) 103–5 and Courtney (n. 2) 97.

30. See Fitzgerald and White (n. 19) 20–7 for a summary of the eclectic philosophising presented there.

31. Indeed, Eumolpus will later try to seduce Giton from Encolpius, just as he seduced the Pergamene boy (92–100).

32. E.g. Slater (n. 2) 93.

33. Contrast for instance the Elder Philostratus, whose self-accreditations include four years of study with the painter and writer Aristodemus of Caria (Imagines 1 proem 3), and the Younger Philostratus, who explicitly writes in the tradition of his uncle (Imagines proem 1–2).

34. Cf. Zeitlin (n. 10) 61; ‘The Milesian tale of the Pergamene boy … provides a relevant contrast to Encolpius in that it takes up the story of a successful and clever seduction of a young boy.’

35. Cf. Slater (n. 2) 95: ‘The message thus far is clear: Eumolpus is just as much of an intellectual fraud as Encolpius.’

36. For a discussion of ‘Realism in Petronius’, see Jones, F. M. in Hofmann, H. (ed.), Groningen Colloquia on the novel 4 (Groningen, 1991) 105–20 (esp. 110–12Google Scholar on ‘visualism’, 112–14 on decorum and its inversions, 118–19 on the disruptions of verisimilitude). For the problems of realism and symbolism in the Satyrica, in connection with a riveting account of the paintings of Trimalchio's life (29.3–6), see now Bodel, J., ‘Trimalchio's Underworld’ in Tatum, J. (ed.), The search for the ancient novel (Baltimore, 1993), forthcoming)Google Scholar. For Trimalchio's dog, see Slater (n. 3) 167, 169–70.

37. On Zeuxis and Parrhasius, see Bryson, N., Looking at the overlooked (London, 1990) 30–2Google Scholar; Bann, S., The true vine: visual representation and Western tradition (Cambridge, 1989) ch. 1Google Scholar.

38. See Fitzgerald and White (n. 19) 7–8.

39. On ‘this comically inaccurate survey of the philosophical and artistic geniuses of the past’, see Walsh (n. 3) 96–7.

40. See Walsh (n. 3) 96 and Slater (n. 3) 171.

41. On the Troiae halosis and its introduction in Eumolpus' discourse, see Sullivan, 1968 (n. 2) 165–89, Zeitlin (n. 10) 58–67 and Slater (n. 2) 95–101 and 186–90.

42. Cf. Slater (n. 3) 172: ‘The Troiae Halosis is entirely unsatisfactory as an interpretation of a painting. It elucidates neither the visual nor the historical dimensions of the painting. The painting is merely an excuse …’

43. See e.g. Imagines 1.4 (4) – ‘Let us catch the blood, my boy …’; 1.6 (5) – ‘Let not yonder hare escape us …’; 1.28 (1) – ‘Do not rush past us, ye hunters …’ For some reflections on this ecphrastic strategy – particularly its involvement of the viewer in a voyeuristic fantasy of engagement – see Elsner (n. 20) ch. 1.

44. Cf. Jones (n. 36) 118–19, esp. 119: ‘It is hard to believe there is a centre beneath the masks, and beyond this level, in the narrative itself there is no accessible reality, only layers of imitation, and only relative criteria to judge between different genres.’

45. As Slater points out: (n. 3) 167, (n. 2) 216.

46. This strategy of reading an emotional interpretation into art is specifically what a sophist like the Elder Philostratus aims to teach, see Elsner (n. 20) ch. 1.

47. If one sees illusionism and deception as central themes of the Satyrica, then Eumolpus as priest, poet and paradigm of Petronian deception becomes the embodiment of the strains and contradictions of the Satyrica as a whole. For another angle on the deceptions of Eumolpus – as the victim of false deaths – see C. Connors, ‘Famous last words: authorship and death in the Satyricon and Neronian Rome’, in Elsner and Masters (n. 1).