Thesis, Chapter 3: World of fictions

CHAPTER 3: WORLDS OF FICTIONS Introduction In this chapter I look at the world in which the third book of the Aeneid takes place, and argue that the ancient εediterranean of Aeneas’ wanderings is a specifically literary world created by Vergil’s poetic predecessorsέ In entering it, Aeneas seems paradoxically to enter the story world of other narrators, as if they somehow existed, above and beyond the texts that create them, in some world of fictions. I begin with a (necessarily compressed) overview of the book’s monstrous episodes and then show how the Aeneid moves away from romantic narratives of monsters and heroes. The second half of the chapter is a detailed analysis of the Achaemenides episode – one of the poem’s most fascinating experiments in fictionέ Monstrosity I ended the previous chapter by interpreting the wooden horse as a figure for the antitotality of Vergil’s style of fictionέ We might see this aspect of the horse as part of its monstrosity. There are clear hints in Book 2 at its monstral features and its emblematic multiplicity, resisting incorporation within a single organic whole, represents a monstrous distortion of (and challenge to) Aristotelian standards of narrative unity and order, which were drawn from zoological concepts of organic unity.1 Book 3 continues 1 On the monstrosity of the horse, note 2.245 monstrum infelix; it is of huge, uncanny size (Priam calls it immanis at 150). Just as the monsters of myth were typically multiform confusions of human and animal (Celaeno for instance), so the horse is figured as a perverse, boundary-breaking conflation of inanimate machine made from dead wood and creature endowed with a life-giving womb (uterumque armato milite complent, 20; scandit fatalis machina muros / feta armis, 237-8; arduus armatos mediis in moenibus astans / fundit equus, 328-9). See Clausen (1987) 33-4 on the doubleness of the horseέ τn the horse’s doubled ambiguity as both stratagem and divine portent (cf. innuptae donum exitiale Mineruae, 31) see Austin (1964) ad 2.15; Harries (1989) 139-40 who argues that it is ambiguously situated between the ‘rationalizing’ tradition (just a stratagem) and the ‘religious’ (Frangoulidis (1λλ2) 2ι-9 says much the same). On metaphors of organic unity, and monstrous narration as a challenge to it, see the discussion of Citroni (2009) 19-21 of the hybrid beast pictured at the start of the Ars Poetica; also Arist. Poet. 1458a20 102 the thematization of monstrosity as not just the depiction of contradictory doubleness, but a narrative mode that is itself multiform and contradictory. Stories of monsterslaying typically (and especially in tragedy) fail to define and preserve the opposing terms of their narrative, such that the hero often symbolically merges with, or even perpetuates, the monstrosity he is supposed to eradicate; Aeneas is no exception.2 As well as the confusion as to whether Vergil’s monsters are absurd, meaningless excrescences or signs of divine, as yet unrevealed wisdom, they also confuse the Aeneid’s fantastical and historical elements, such that we can read in them garbled, faintly transmitted signals of the Roman futureέ So the discovery of Polydorus’ body during Aeneas’ foundation of a new colony ironically but indistinctly foreshadows the discovery of a severed head during the laying of the Capitol’s foundations (δivy 1έηηέ16; Serv. 8.345); similarly Horsfall notes that the language of 3.58-9 (delectos populi ad proceres primumque parentem / monstra deum refero, et quae sit sententia posco) figures Aeneas as a ‘Roman magistrate who refers a portent to the Senate for discussion’, at that point in the narrative where Rome’s eventual foundation is at its most distant and uncertain.3 The monstrous episodes of the Aeneid, and Achaemenides too, represent fairly clear-cut examples of narrative incorporation: these are fragments from other narratives that are being framed for the first time within the Aeneas tradition, and the novelty of this is hinted at in the noua proelia with the Harpies (240) and the noua forma of Achaemenides (591).4 Moreover, harking back to arguments made in the first chapter, it is a new reframing but of material that belongs markedly to an archaic and out-of-time and Pfeiffer (1968) 68-9. See Schwindt (2009) for an interesting discussion of thaumata as paradoxes that resist thematic unity. 2 On tragedy see Segal (1986) 22, Dodds (1973). Putnam (1995) 52 has a convincing interpretation of the Polydorus episode in which Aeneas becomes ‘the wanderer desirous of steadying knowledge acquired even through violence, the sacrificer who pollutes’ν see also Thomas (1λκκ)έ In the fight against the Harpies the remarkable word choice at 3.24, obscenas pelagi ferro foedare uolucris, equates the Trojans with the similarly foul Harpies (contactuque omnia foedant / immundo, 227-8; et uestigia foeda reliquunt, 244): rather than cleansing, they are polluting the polluters, furthering the monstrosity rather than overcoming it; see Labate (2009) 138, and cf. perhaps Aeneas’ belated wish that the Trojans had attacked the Wooden Horse: si fata deum, si mens non laeua fuisset / impulerat ferro Argolicas foedare latebras (2.54-6). 3 Horsfall (1989) 16-17. Regarding other ironic foreshadowings, Labate (2009) 137 notes the anachronism of Aeneas’ behaving like a real-world, historical Roman general in the fight against the mythical Harpies; the description of the gathering Cyclopes as a concilium horrendum (3.679) is both an ironic reflection of Homer’s uncivilized giants who know no such a thing as a concilium (a0gorai\ boulhfo/roi, Od. 9.112) and of the Roman polity that is at this point only a garbled, evanescent possibility. 4 τn Vergil’s additions to the tradition of Aeneas’ wanderings see δloyd (1ληι), (1ληιb)έ 103 world. Aeneas’ encounter with Achaemenides and Polyphemus is, as Jolivet has pointed out, the point of closest proximity between the narrative of the Aeneid and the world of the Odyssey. The Vergilian Cyclopeia is, he says, en somme le moment où le héros côtoie au plus près le monde fabuleux d’Ulysse, marchant dans les traces mêmes du Grec quelques semaines seulement après son aventure chez Polyphème.5 The philosopher and poet Xenophanes of Colophon, writing in the late sixth century BCE, had urged singers at symposia to songs with fantastical themes: ou1 ti ma/xav die/pein Tith/nwn ou0de Giga/ntwn ou0de/ <te> Kentau/rwn, pla/smata tw=n prote/rwn, h2 sta/siav sfedana/v, toi=s 0 ou0de\n xrhsto\n e1nesti. DK fr. B1.21-2 = Athenaeus 11, 462c Titans, Giants, and Centaurs are the ‘fictions of men of old’έ Indeed, Vergil emphasizes that the fictional world in which Aeneas moves in Book 3 is not only historically primitive, but also literarily archaic. In each of the monstrous episodes Aeneas is too late to act as the archetypal monster-slayer or heroic avenger. Euripides has already shown Hecuba’s vengeance upon Polymestorν when Aeneas reaches the Strophades Apollonius’ Phineus has already been relieved by the Boreads and left the island, and instead haunts the later representation of Achaemenides (3.590-4)ν and the latter’s prior escape from Polyphemus, alongside Odysseus, offers Aeneas no real opportunity for heroic deeds.6 Not only does all this reinforce the secondary nature of the Aeneid, coming at the end of an already long literary tradition, it also makes clear that Aeneas must find an opening in a new, refreshed story-world away from ‘the fictions of men of old’, and join a different paradigm of heroesέ So not only are narratives of monsters multiform, ambiguous, and antitotalizing, but their incorporated presence within the Aeneid makes Vergil’s poem similarly multiplicitous, containing a self-enclosed world that it cannot fully integrate within a single totalizing structure. Below I will argue that a peculiarity of Book 3 is its vision of a fictional world that is inter-fictional, comprising a narrative that seems to 5 Jolivet (2005) 48. τn Vergil’s Strophades as a direct continuation from where Apollonius had left the story, see δabate (2009) 135-6; see below for my argument on the Achaemenides as an Odyssey continuation. 6 104 enter into the created world of other literary works. The literary world that Book 3 enters into, and the mode of fiction that the Aeneid comes to reject (I argue in the next section), seems closely related to the Cyclic poems. Aeneas has just finished relating to Dido the fall of Troy in a compressed retelling of the Cyclic Parva Ilias and Iliu Persis. Even before his arrival in Carthage she was familiar with an Aeneas that we know from the Aethiopis, as depicted in the friezes in the temple of Juno: after an Iliadic panel showing the death of Hector, the gaze of the viewers shifts to a new frieze depicting Aeneas amidst the exoticism and romance of the Aethiopis (Penthesilea and Memnon).7 In Book 3, much more so than in Book 2, Aeneas goes all out to indulge his host and satisfy her expectations of him by presenting himself in romantic mode. Indeed, in a meta-literary nod Aeneas continues his tale by following the Cyclic progression, from Aethiopis in Book 1 to Ilias Parva and Iliu Persis in Book 2, with the implication that Book 3 will take us into the story-world of the Nostoi and the Odyssey. We noted in the previous chapter how Aeneas-narrator does not necessarily give an account of the ‘real’ Aeneas, but rather of an aestheticized ‘Aeneas’ who belongs to the worlds of art and story; following on from this, we might say the abrupt change in narrative mode and tenor between Books 2 and 3 marks not just a change in the sort of experiences that the real Aeneas had, but also a new shaping by Aeneas-narrator of his literary self as he reconstitutes himself as a hero of romance – albeit, as we will see, a dysfunctional one. Aeneas’ wanderings, while a departure from his home country, form a nostos because of their resemblance to the wanderings of Odysseus and other Greek heroes, and because the Trojans are returning to the original home of their ancestor Dardanus.8 7 For the presence of Cyclic influence in the Aeneid, and on Book 2 as an Iliu Persis, see Kopff (1981) 921, 928-31; on each scene of the friezes coordinating with a Cyclic or Iliadic scene see Stanley (1965). 8 For the resemblances to the nostoi of Greek heroes other than Odysseus see Tracey (1968); cf. Nelis (2001) 23. Heinze’s conjecture, (1λλ3) κί, that the account of Polydorus’ body goes back to a hero unbeatable in close combat, who could only be killed by a long-range volley of spears, certainly accords with the Cyclic penchant for heroes of fantastical, sometimes magical strengthέ τn the ‘monsters, miracles, metamorphoses’ typical to the Cyclic poems see Griffin (1λιι)μ his characterization, 4θ, of the conspiracy in the Cypria to drown Palamedes (recounted by Sinon at Aen. 2έκ2ffέ) as a ‘most unHomeric’ episode – a ‘treacherous murder of an ally for selfish reasons’ – suggests obvious similarities to Polymestor’s murder of Polydorusέ It is possible that alongside the Euripidean model for the episode, Vergil is also trying to evoke the lurid world of Cyclic romance. 105 New monstrous fictions There are various possible thematizations of ‘anomaly’ in Book 3μ the bodily anomaly of the monsters’ anatomyν their indeterminate status as either aberrations or divine messengers; the unepic language used to describe them; their not belonging to the traditional Aeneas legend; and their literary-historical inappropriateness within the Aeneid itselfέ Gordon Williams revives the old theory that Book 3 is literally ‘out of place’ within the Aeneid as we have it, claiming that it was originally written in the third person and intended as the poem’s first book – the beginning of the chronological story and the narrative.9 A useful philosophical thematization of ‘out-of-placeness’ is found in what the Neo-Platonists called atopia. As well as signifying something disgusting or inappropriate (such as the Harpies’ excrement), atopon for the Neo-Platonists also signifies that which seems absurd because it is allegorically expressed and therefore requires special interpretation: Il apparaît donc que, sans qu’il soit formulé de façon théorique, l’exégèse de Porphyre obéit en fait à un principle selon lequel l’absurdité du sens littéral est l’indice de l’intention allégorique de poète et trouve dans l’interprétation allégorique son explicationέ10 The absurd, or even that which totally escapes our comprehension, such as a singularly odd oracle, are in fact forms of higher wisdom awaiting the correct interpretation.11 So, Anchises possesses sufficient powers of understanding and interpretation to realise that the flames that engulf Ascanius’ head (2έθ80-91) are not the random excrescence of an irrational world, but the higher language of divine command. The ambiguity between monstrum and monster which we see in Book 3 rests upon this narrow liminality between the two meanings of atopon: romance fiction is not sophisticated enough to ask how, through a sort of redemptive exegesis, atopon (monstrosity) could be interpreted as atopon (concealed divine wisdom). Rather it consistently presents a confused and unresolved mixture of the two. Monstrosity is not either divine message or fictional absurdity: it is the state of the two existing simultaneously. Indeed, at no point in Book 3 is the existence of its various monsters rationalized: no god or seer has decreed that Polydorus’ body should remain the way it is (but contrast the Sibyl’s prophecy of the 9 Williams (1983) 262ff. Pépin (1λθη) 2η2έ For a discussion of the earlier Plutarch’s use of the word see Hardie (1λλ2b) 4ιηί-1. 11 Pépin (1965) 254. 10 106 death and afterlife of Palinurus, 6.373-81); no genealogy is given for the Harpies, nor is their connection to Jupiter mentioned, which in Apollonius’ account had saved them from the pursuing Boreads; nor, finally, does Vergil follow Homer by making his Polyphemus the son of Neptune. There is less emphasis on a causal chain which, if followed to its source, could rationalize or even excuse the monsters as oddities within a world that is still, nevertheless, divinely governed. They present an irresolvable contradiction. There are other ways, I think, in which Aeneas’ narration of the wanderings is ‘out of place’έ It is not simply that the fictional world he depicts contains anomalies and resists interpretation, it is also presented as the wrong fictional world, that is to say, a fictional world belonging to earlier literary texts which the Aeneid stumbles in to. As argued above, the ancient Mediterranean traversed by Aeneas is a belated world, still populated by the creations of earlier poets, which the passage of time has made anomalous. For Labate, Book 3 is a series of near misses with a world that is alien to the ‘historical’ aims of the Aeneid: In the cases of Scylla, Polyphemus, and Circe, Aeneas repeatedly verges on coming into contact with extraordinary realities: the narrative diaphragm separating Aeneas from the various monstrous adversaries of Ulysses is thin, but it is sufficient to safeguard his quality as a hero of ‘history’έ This diaphragm becomes much more substantial in the case of adversaries and exploits like those on which the fame of Hercules was based.12 This is, of course, an intertextual illusion: fictional worlds are unique imaginative creations, which do not have to share in the reality of any other fictional world. If I come up with a story about Victorian London I can populate it as I like, and there is no danger that Sherlock Holmes will intrude upon it, as if he had some sort of real existence ‘there’ in all fictional worlds of Victorian London. I discuss this further in my treatment of Achaemenidesέ Vergil’s choosing to make his fictional world of the wanderings a literary-fictional world, that is, a fictional world that belongs to the description of that world in a specific literary work, comes down, I argue, to a comment Labate (2009) 128. Compare Jolivet (2005) 48, who also resorts to metaphor to describe Aeneid 3’s tangential relationship with another literary-fictional world imagined as spatially present within the Aeneidμ ‘en somme le moment où le héros côtoie au plus près le monde fabuleux d’Ulysse, marchant dans les traces memes du Grec quelques semaines seulement après son aventure chez Polyphèmeέ’ 12 107 he wishes to make on the fittingness of that world within his own vision of the world of the Aeneid. The poor fit can also be explained within the fiction of the inset narrator, as though it were another blunder made by Aeneas as narrator, and another example of the inconcinnities between his local narrative and the global frame: Aeneas the character doesn’t know the overall plan that Vergil has for his epic, so there is always the risk that he might lead the poem astray by telling the wrong sort of storyέ Furthermore, Aeneas’ self-reinvention might clash with the overarching frame of the poem, but it is perfectly fitting with his situation and motivation as a character to tell a story to Dido of a heroic Aeneas facing marvels and monsters in a world just like that of the Cyclic poems. I argue in the ecphrasis chapter that Vergil creates a clever illusion of Aeneas stepping from the friezes and into the midst of the Carthaginian court, like a fantasy prince come true; we have already seen how Aeneas’ narration suggests a continuation of Cyclic themes and an indulging of Dido’s desire to hear tales of legendέ Part of Aeneas’ developing awareness of his alternative existence as a literary character, whose destiny he will ultimately be unable to control, is his tendency to miscast himself into the wrong story and to be reborn into an inappropriate story world. There is a suggestion of this in Andromache’s misapprehension when she sees him in Buthrotumμ sollemnis cum forte dapes et tristia dona ante urbem in luco falsi Simoentis ad undam libabat cineri Andromache manisque uocabat Hectoreum ad tumulum, uiridi quem caespite inanem et geminas, causam lacrimis, sacrauerat aras. ut me conspexit uenientem et Troia circum arma amens uidit, magnis exterrita monstris deriguit uisu in medio, calor ossa reliquit, labitur, et longo uix tandem tempore fatur: ‘uerane te facies, uerus mihi nuntius adfers, nate dea? uiuisne? aut, si lux alma recessit, Hector ubi estς’ dixit, 3.301-12 Not only is Aeneas mistaken for a visitor from the underworld – itself very much a world generated by and made up of fictions, as I argue in the next chapter – but there this is a sense too of his emergence into another fictional world, that is, into the substitute Troy of Andromache and Helenusέ As well as the ‘false Simois’ he sees later the other features of their fictional Troy: 108 procedo et paruam Troiam simulataque magnis Pergama et arentem Xanthi cognomine riuum agnosco, Scaeaeque amplector limina portae 349-51 For Hexter, the parua Troia of Buthrotum is an allusion to the Ilias Parva of the Cycle, a literary exemplar that Vergil here anathematizesμ ‘whatever type of Homeric imitation Vergil is engaged in, he indicates, it will be far different from that of a cyclic poet, the archetypal continuator and epigoneέ’13 Aeneas steps into, or seems to step into, a fictional world that has already been explored by poetry and already exists as a literary reality; once again it is also the psychagogic fictional world shaped by an addictive female imagination, and one wonders if Dido sees any of herself in Aeneas’ portrayal of Andromache.14 One wonders too if the use of inanis for empty fictions is a thematic link here: in Book 1 Aeneas steps from the pictura inanis into Carthage and Dido’s dreams made true, and in Buthrotum he seems to materialize from the tumulus inanis, the complement to Andromache’s dreams of Troy rebornέ In an intriguing meta-literary play on the Omphale topos, Vergil imagines his hero trapped and travestied within the make-believe world that stories supposedly create in female minds. Andromache’s obsession with the dead Troy has led a number of critics to interpret Buthrotum as a parallel to the Underworld in Book 6.15 Going further, we 13 Hexter (1999) 76-7. On the Augustan rejection of the Cyclic poems as overdone and trite see Hor. AP 136ff. (ut scriptor cyclicus...) and Severyns (1928) 155-9 on Aristonicus, the scholiast teaching in Rome around the time of Vergil and critic of the banality of cyclical style. 14 Particularly her creepy fascination with Ascanius (3.339-43, 488-91). Both women are portrayed as having an emotional reliance on surrogates for lost realities – Andromache and the mei Astyanactis imago, Dido and her paruulus Aeneas (4.328-9) – and the supposed feminine susceptibility to fictions is also represented by Dido’s being utterly hoodwinked by the Ascanius doppelgangerέ τn this belief generally, fairly widely attested in antiquity, see Plato Gorgias 527a; Cic. ND 2.5; Plut. Quomodo 16f; Apul. Met. 4.27; Seneca De Ben. 1.4; Quintilian, 1.8.19. Cf. Brisson (2004) 17, Buxton (1994) 161. On Andromache’s nostalgia as an emotional pact with the past bringing sterility see Bettini (1997) 16. For Andromache’s Troy as a series of empty simulacra no longer with any essential unity see σelson (1λθ1) 344. 15 One immediate reason to make this association is that the Odyssey’s journey to the land of the dead (nekyia) takes place in Odysseus’ account of his wanderings to the Phaeaciansέ It is argued that the necrotic aspects of Buthrotum correspond with the nekyia because an intertextual rule of analogy implies it. Lloyd (1957b) 137-8 and 149-50, who provides a very useful account of Vergil’s likely sources for the wanderings and his adaptations of them, sees Helenus’ prophecy as a prologue to the grand revelations of the catabasis. Hershkowitz (1991) reads Book 3 as a miniaturized intratextual allegory of the poem as a whole, with Buthrotum and the catabasis corresponding with one another in the suggested schema. The similar approach of Bright (1981) 40-ι sees Book 3 as a condensed intertext of τdysseus’ wanderings in Od. 9-12. He argues that the structures of Od. 9-12 and Aen. 3 are as inverted images of each other; according to this rule of congruence Vergil uses Buthrotum as a substitute for the nekyia, rather than 109 might see the whole book as a parallel to the catabasis. The basic plots of both consist of journeys into the unknown, conducted by fate in Book 3 (dare fatis uela 3.9) and by the Sibyl in Book 6 (doceas iter et sacra ostia pandas 6.109), while in both Aeneas follows the literary example of τdysseus’ wanderings in Od. 9-12. Both journeys see Aeneas reunited with figures from his own past (Anchises’ amicus Anius, and Helenus and Andromache in Book 3; Palinurus, Dido, Deiphobus and others in Book 6). More generally both place the hero among characters, places, and topoi of the Homeric poems. Perhaps the stand-out link between the two books, however, is that they are the Aeneid’s most sustained instances of non-realistic, marvellous fiction: they are both episodes containing fictional worlds in which things happen that do not – cannot? – happen elsewhere in the poem. Both mark the episodic parameters of their fictional worlds, I argue, by likening them to the alternative realities of dreams. The catabasis is figured as a dream about both the future and the past, while the wanderings can be seen as a nightmare vision of the archaic past that also, in Helenus’ prophecy, contains glimpses of the future.16 The monstrous episodes are strange and nightmarish, and the phrasing of lines 4-5 (diuersa exsilia et desertas quaerere terras / auguriis diuum) anticipates closely Dido’s nightmare in Book 4μ agit ipse furentem in somnis ferus Aeneas, semperque relinqui sola sibi, semper longam incomitata uidetur ire uiam et Tyrios deserta quaerere terra. 4.465-8 Just as Book 2 opens with an unwitting ironic allusion to the ‘wound’ of love that Aeneas inflicts on the listening queen, so the next half of his narration unknowingly looks ahead to the nightmare obsession that Aeneas’ story is sowing in Didoέ The dream association is accented further by setting the Polydorus episode in the first landfall: he also begins Euripides’ Hecuba, announcing in the prologue that he is appearing to his mother in frightening dreams (lines 31-4; at 54 Hecuba is described by him as fa/ntasma deimai/nous 0 e0mo/n). We should be wary of reading too much into the verbal reminiscence of deserta quaerere terra in Dido’s nightmare, but it is at least a noteworthy coincidence that Book 3 begins with these words (albeit in a different simply removing the nekyia from the wanderings altogether and transplanting it to Book 6. Cf. also Quint (1993) 56-8. 16 A further distinction between the two is that, as discussed above, Book 3 often presents garbled intimations of the Roman future, in contrast to the much clearer vision provided by Anchises in Book 6. 110 grammatical arrangement) and, almost immediately after, a figure taken from a nightmare in a tragedy. All the more noteworthy, of course, because in a remarkable simile Dido’s nightmare is compared to scenes one might see on the tragic stageμ Eumenidum ueluti demens uidet agmina Pentheus et solem geminum et duplices se ostendere Thebas, aut Agamemnonius scaenis agitatus Orestes, armatam facibus matrem et serpentibus atris cum fugit ultricesque sedent in limine Dirae. 4.469-73 Does this simile also shed some light on the opening to Book 3, as an intratextual correspondence, or is any correspondence more likely a philological fiction? Looking back to the discussion of ‘runaway’ similes in the first chapter, we might see this as an instance of deviant exemplification: although the stage versions of Pentheus or Orestes are cited not for their truth value (i.e. they are not asserted to exist) but for their abstract exemplary value, because they do exist in our real world as literary constructs, and because we cannot override the fact of this existence and read them as only exemplary, the simile introduces a world that is alien to the fiction of the Aeneid and creates a clash of ontological orders. Why does the narrator cite Pentheus and Orestes as artificial characters in the real world’s theatre, when at all other times his gaze is fixed on the mythical world within which they move as real figures? After all, his protagonist probably met τrestes’ father at some point.17 The question that could be asked is whether Vergil associates the disorder of dreams and nightmares – and specifically their confusion of falsehoods for realities – with those instances in which his fiction lapses into ontological disorder and treats what is normally false within its world as really existing within its world.18 This is, it could be argued, a meta-literary trope for nightmarish misapprehension, involving the disorienting intrusion of extra-fictional, literary reality on to the fictional world. This would certainly concord with the exaggerated meta-literary features of the third book: its posited fictional world is consistently invaded by the fictions of other Compare my analysis of Aeneas’ hypothetical Greek narrator of his story (above pp.66ff.): that was another instance of a hypothetical situation being too tendentiously close to a specific reality to avoid an ironic suggestion of a referential, as well as a hypothetical function. 18 False, that is, in the way that it is false within Aeneid-world that Achates is a literary construct (his comrades presumably think he is a real person). Aeneas, as I have argued, does seem to know, or at least suspect that he has two simultaneous existences, in one of which he is a literary construct. 17 111 literary textsέ Aeneas’ visit to a simulated parua Troia, or Achaemenides’ ‘re-reading’ his captain’s footsteps (relegens, 3.690 – discussed below) break the fourth wall almost as much as the tragic scaenae in Book 4. More broadly, certain constituents of its fictional world, the monsters, are also characterized by their unframable, irresolvable contradictions. This is not only in their confusion of the divine and the absurd, but also in their vague and misleading adumbration of a historical, Roman world in which, paradoxically, they cannot exist. Book 3 presents the nightmare of an embryonic fiction whose terms (hero vs. monster e.g.) keep lapsing into impossible, monstrous contradictions, and whose make-believe world is porous. The passing of the monstrous world is not achieved by its defeat; the Trojans fend off but do not overcome the Harpies in any conventional heroic way, and come nowhere near daring a confrontation with Polyphemus comparable to τdysseus’έ By the time the Trojans reach Italy, the Odyssean world seems far removed: Jolivet notes how the Cyclopean aspect of the description of εezentius ‘établit un lien entre deux univers différents, de temporalités différentes’έ19 The link bridging the two worlds, Jolivet notes, is a figurative one, and this is, I think, important. In putting the adunaton of the table-eating into Celaeno’s mouth, another addition to the tradition, Vergil affirms the sense of monstrosity posing a challenge of paradoxical impossibility which the narrative must resolve. So it is telling for the Aeneid’s subsequent treatment of monstrosity that the dire-sounding prophecy (uos dira fames nostraeque iniuria caedis / ambesas subigat malis absumere mensas, 3.256-7) is defused by the realization at 7.107ff. that it represents more a troublesome and riddling statement (due to the ambiguity of the referent of mensa) than a troublesome thing (eating a table would be awful).20 Monstrosity in the Aeneid is largely rejected by the shift to a mode of fiction that no longer believes in monsters, and instead presents them as figurative entities or qualifies them as merely intensional constructs. To deal with monstrosity is not to deal with a paradoxical thing that exists, in a fictional or any world, but to deal with a paradox that only exists as a function of poetic representation.21 The dissolving of the literal 19 Jolivet (2005) 47. Compare the rationalizations of Palaephatus above, pp.29-30. 21 As well as Jolivet, cited above, Hardie especially has established Vergil’s use of gigantomachy as a metaphorical figure for describing the enemies of Rome. See, for instance, Hardie (1986) 118-19 on the Turnus’ flame-wreathed helmet at 7.785f. and its reminiscence of Enceladus’ Etnan fire-breathing. On the monsters that gives names to the ships in the race in Book 5 (Chimera, Scylla, and Centaur, 5.114-23) – again turning monsters into mere linguistic signs – see Hardie (1λκι)έ σote also the ship called ‘Centaur’ 20 112 monsters of the wanderings into figurative abstractions is seen most decisively in the vestibule to Hades in Book 6: the monsters that Aeneas finds there are revealed to be no more than formae, shapes or likenesses rather than real things. As I argue in the next chapter, they seem to represent figments, rather than anything that could exist outside of poetic description. Similarly the creations of Circe (7.21-4) are described but not encountered – except figuratively, in the form of humans bestialized by their passions (especially Turnus) or wearing animal-skins (in the catalogue of Italians) – while the multi-form monsters of Antony’s armies (omnigenumque deum monstra, 8.698) no more than figurative monsters pictured on the artifice of Vulcan’s shieldέ22 The story of Hercules and Cacus is the most sustained depiction of a monster outside the third book, and it contains the same theme of heroism imitating the monstrosity that it vanquishes.23 It is in some ways, however, a much clearer story than anything encountered in Book 3. Most notably, the story Vergil chooses is one featuring names that can be etymologized as emblematic – ‘good man’ Evander and ‘bad man’ Cacus – and these immediately remove any claims to realism: the reader immediately infers that the names taken together refer to characters whose significance is here limited intensionally to their function in the exemplary story that follows, rather than extending to more open-ended historical figures.24 The clearer markers of the episode as a framed piece of story-telling, possibly allegorical, has led some critics to see in the episode an allegory for Augustus’ civilizing missionέ25 Whatever the validity of these interpretations, it certainly seems true that the ontological and thematic relationship between fictional and historical worlds is much more readily definable here than in the in the catalogue of the Etruscans (10.195), named after its figurehead (see Harrison (1991) ad loc.): as monstrous and threatening as this centaur is, Centaurus is here merely a verbal representation of an artistic representation of a monster. By this point in the narrative, any sense of the reality of monsters is far removed. 22 Note also, in contrast to the confusion of the Odyssean monsters, the perfect separation of quasidivinity (Augustus) and monstrosity in this ecphrasis; see Quint (1993) 29-31. 23 See Lyne (1987) 27-32, and Hardie (1986) 115-16 who reads the episode as an optimistic portrayal of the boundary-breaking heroμ ‘this inversion may be understood within the wider context of the progression within the Aeneid from helplessness in the face of natural or mythological catastrophe to the ability to withstand and ultimately control such forces.’ For an excellent account of Hercules as the ‘great interstitial, paradoxical figure’ see Feeney (1λκθb)έ 24 On emblematic names see Riffaterre (1990) 33-6. Enrico Montanari, writing in the EV, also notes how the episode is placed in a set-aside, ahistorical worldμ ‘Un problema è dunque stabilire in base a quali meccanismi, nell’operare la più grandiose rievocazione dei primordiale compiuta in età augustea, V. abbia inteso superare la verosimiglianza topografica, trasformando quella che presenta tutti i tratti di una primordiale divinità ‘nazionale’ romana … in un monstro infernale, escluso dal pomerium e dal ‘tempo’ di Romaέ’ τn the ‘other-worldly atmosphere’ of Book κ see Bacon (1λ3λ) λιέ 25 See Drew (1927) 35ff.; Zarker (1972), esp. 35-6 and 45. 113 Book 3 episodes, and that it is much easier for us not to believe in Cacus if we do not want to. The transformation of the Trojan ships into water-nymphs in Book 9 is, in a sense, a double metamorphosis: as well as the actual transformation, the eventual fate of the fleet and the story of its genesis on Ida send us back to the first mention of its raising in Book 3 (classemque sub ipsa / Antandro et Phrygiae molimur montibus Idae, 5-6) and to the Polydorus episode that follows it: there has been a metamorphosis, too, in the way the Aeneid narrates such prodigies. While both episodes may be Vergilian inventions, the passing of the narrative into the harder, more real world of Italy and war conversely provides more definite boundaries for the removed and unreal fictional world in which metamorphoses take placeέ As Gordon Williams puts itμ ‘σot only are we not being asked to believeν we are also being asked not to believeέ’26 Like the Cacus episode, we are given enough signposts by the poet to attribute the event to literary unreality – it is simply a fama with no deed. This shift underlines, I argue, the Aeneid’s wider, subtle emphasis on paradoxical fiction as the uttering of impossible statements and the intimation of logically impossible modes of thought, rather than the description of an envisaged world populated by impossible things. In the new, more sophisticated fiction that Vergil proposes, the anomalies of fiction are portrayed, self-reflexively, as an internal problem with how literary fiction structures the world and then complicates or contradicts the structures it posits. It is all too easy for mythical monsters to excite a sense of wonder at what can be possible of the make-believe world in which the reader is immersed, rather than at what can be possible of fictional utterances manipulated by a skilful poet. The rest of this chapter will look at Achaemenides as a character who wanders between the frameworks of two different narrative worlds, thus showing that no single narrative can achieve a unique and absolute framing of the world but rather exists alongside, and is permeable by, other narratives. Although this chapter has argued for 26 Williams (1983) 28. See my argument in the first chapter (pp.21-2 above) for the importance of this episode in asserting Vergil’s poetics of a fiction that accords to no original reality. On metamorphoses as typical traditional territory for muthos / fabula, see Strabo 1.2.11: e0kei=nou w0keano\n kai\ 3Adhn kai\ 9Hli/ou bo/av kai\ para\ qeai=v kai\ metamorfw/seiv kai\ mege/qh Kuklw/pwn kai\ Laistrugo/nwn kai\ morfh\n Sku/llhv kai\ diasth/mata plou= kai\ a1lla plei/w toiau=ta teratografou=ntov fanerw=v. 114 Vergil’s working towards a more coherent, and less ambiguous, form of fiction, it still entails the imitation and incorporation of what is rejected. Again we gain a sense of the Aeneid’s frame and, in its rejection of narrative modes that are not commensurate to it, its grandeur; but again it can only be disclosed through its dysfunction and by bringing it to near-failureέ We see also that Vergil’s meta-poetic narrative of founding the right sort of epic fiction, and rejecting the wrong sort, raises questions about the firmness and validity of his narrative frame on the world: it suggests that the Aeneid’s structuring of its theme, and of reality, is a work still in progress, and that poem has not come to world a completed and absolute vision but one still in the process of establishing itself.27 In a way this is an interesting twist on the age-old focus on the poem’s incompleteness at the time of Vergil’s deathμ the poet himself includes a self-alluding narrative of his poem’s struggle to reach a finalized form. It also implies that while his theme is an eternal empire, and while his literary work fixes forever the words of a transitory mortal poet, the work itself is not removed from the passage of time: the process over time of composition, which happens largely ‘offstage’ in Homeric epic, is inscribed into the narrative. In a temporal paradox, the text of the Aeneid fixes forever a process of change. Achaemenides The culmination of the abortive venture into the romance mode is Achaemenides’ entrance into the story and the re-running of τdysseus’ encounter with Polyphemus from Book 9 of the Odyssey. The Trojans do not come face to face with any of the Cyclopes, so the episode is a less marked incursion into the world of romance than the battle with the Harpies. On the other hand the meeting with a character who has only just stepped out of the narrative of the Odyssey represents the point of closest contact between the worlds of the Aeneid and the Odyssey. Although my priorities and conclusions are different from his, Hinds’ excellent interpretation of the Achaemenides episode largely structures the approach I take here.28 Like Hinds, I argue that it is a On the metapoetic aspects of Aeneas’ attempts at founding in Book 3 (especially at 3.18, Aeneadasque meo nomen de nomine fingo) see Hexter (1999) 68-9. 28 Hinds (1998) 111-16. Following on from Hinds, Papaioannou (2005) 79-111 also looks at the Achaemenides episode with a heavy emphasis on intertextuality. On intertextuality between story worlds 27 115 virtuoso, and strikingly modern, execution of intertextual craft; I differ in arguing that the figure of Achaemenides thematizes not only the structures of poetic influence that bind Homer, Vergil, and Ovid, but also the overlapping fictional worlds created by each text, which Vergil and Ovid bring into paradoxical collision. This overlap will lead to a further exploration of issues already broached in this thesis, particularly the question of the possibility of identity across fictional worlds, and the applicability of speech acts across intertextual boundaries. A thread running subtly through Aeneas’ narration in Books 2 and 3 is the question of whether he will come face to face with Odysseus, his rival epic hero and narrator. As they are leaving the Strophades the Trojans deliberately avoid going anywhere near τdysseus’ home island of Ithacaμ iam medio apparet fluctu nemorosa Zacynthos Dulichiumque Sameque et Neritos ardua saxis. effugimus scopulos Ithacae, Laertia regna, et terram altricem saeui exsecramur Vlixi. 3.270-3 While this is partly common sense – the run-in with the Harpies was bad, and there is no point risking further danger by meeting with former enemies (counsel confirmed by Helenus’ later advice to avoid εagna Graecia (3έ3λθ-8)) – the bypassing of Ithaca is preceded by a narrative pattern of tantalizing near-encounters or fleeting brushes with Odysseus, starting in Book 2 and capped at the end of Book 3 by the discovery of one of his companionsέ Aeneas’ account of the fall of Troy ensures that τdysseus always appears either offstage or only obliquelyμ depicted in indirect speech in Sinon’s account (and this is Sinon’s fake τdysseus anyway)ν inexplicably glimpsed by Aeneas leaving the Wooden Horse at 2.261 (or is this actually from a story or painting and, as discussed above, not eye-witnessed reality?); present only by synecdoche in the wound he has given Pelias at 436; and, finally, seen guarding looted treasures by Aeneas when reentering Troy to find Creusa, presumably clandestinely and from afar as Aeneas is alone at this point (761-3). At no point is he represented by direct speech; the most that he is depicted as doing in direct narrative is standing statically as a guard, and even this feels like more of the factual summary of a state of affairs than a description of an agent as a standard feature of narrative discourse – and not necessarily a marked ‘special effect’, as I treat it here – see Laird (1993) 152-3. 116 doing something in a scene. The marked avoidance of Odysseus, and specifically the use of Achaemenides as his proxy, is interpreted by Heinze and Lloyd as a form of meta-poetic decorum: to appropriate the Homeric hero as a fully-formed speaking character within the Aeneid would be an affront to the supremacy of Vergil’s predecessor.29 Barchiesi, however, thinks it is the other way roundμ through an ‘anxiety of influence’ Vergil diverts Aeneas away from τdyssean encounters as a means of preserving the autonomy of his own text and preventing it from becoming an epigonal re-run of the Odyssey.30 So is it said by some that an appearance by Odysseus in the Aeneid would constitute a Vergilian intrusion upon territory that properly belongs to Homer; by another, that it would constitute a Homeric intrusion upon Vergil’s new narrative terrain. This raises questions about which text has the authority to incorporate the other; and further questions, as I will show, about precisely which fictional world frames the narrative at this point – Homer’s or Vergil’sς We can already see this ambiguity in what I have written above: in writing about Odysseus as he is narrated in the Aeneid should I talk about Ulysses, the hero as he exists Latinized in his instantiation in Roman poetry, or about Odysseus, the hero who belongs to Homer’s Greek epic and who is only a foreign, fleeting visitor to the Aeneid? It is hard to be sure whether τdysseus is fully incorporated into Vergil’s narrative, and therefore to be known by the naturalized name of Ulysses, or whether he is, to borrow Hinds’ term, a ‘stray’ originating from and belonging to Homeric epic who pops up anomalously and intrusively in the Aeneid. The question surrounding Achaemenides, then, is this: which writer does he belong to, and who of them is doing the encroaching: Vergil, by taking a background character from the Odyssey and rewriting it, or Homer by interpolating a background character onto the world of the Aeneid? As a character is he Homeric, or Vergilian, or both simultaneously; and is the comes infelicis Vlixi (3.613) the same character as the unnamed e9tai=rov, or a new creation of Vergil’s, or again both simultaneously? 29 30 Lloyd (1957b) 397, Heinze (1993) 84. Barchiesi (1986) 88-9. 117 Worlds of myth and worlds of narrative We discussed in the first chapter some of the features of fictional world-making. The difficulty with a character like Aeneas is that although his existence, and the attributes that attach to the name ‘Aeneas’, are intensional – dependent on and altered by the statements made about them – he is still reified in our imaginations. As a result we think literary versions of Aeneas instantiate some external ‘Aeneas’ that is ‘out there’, and no one particular instantiation is the sole cause of his existence. Also, we allow for Aeneas to exist in a generalized myth-world that is not marked as the creation of any one particular textέ So, for instance, Hercules’ presence at Aeneid 10.464-5, lamenting the inevitable death of Pallas, does not signal the introduction of a character from another narrative, a Hercules-as-described by some other author. This is simply the Hercules whom we accept as habitually present in the world of myth; we have already, in reading and knowing of other narratives, granted him a fictional existence, which carries over into further narratives that take place in the same mythical world. We are also still left with an ontological problem: what thing in common, what quiddity, allows us to identify all the versions as that same external, sort-of-reified Aeneas, when it is the case that there is no quiddity, just an endlessly mutating existence in words? One reason for using Aeneas or Sherlock Holmes to think about fictions is that their migration across ontological borders is what makes them interesting. Minor literary characters who are only ever described in one literary work, and who largely only receive a make-believe existence when this work is read, are more simply categorized as artificial constructs. We can be far more definite in saying that Col. εusgrave of the ‘Five τrange Pips’ (or any other randomly picked minor character from the Holmes stories) belongs only to the world we make-believe for that literary text and is tied to the descriptions of him in that text. This makes him a less rich and interesting character than Holmes, but on the other hand it makes all the more interesting the hypothetical question of what would happen if he were perversely treated as a Holmes-type character who could hop between fictional worlds – and what is going on with Achaemenides. The question is approached by Lamarque and Olsen, but left somewhat open-ended: 118 Suppose we have a minor character, say in a whodunit, described merely as a suspicious-looking housemaid with no active causal role in the plot, and a narrative function largely conventional. Could this same character crop up in different whodunits? That might seem odd if the stories are unconnected, yet if a character is constituted by its salient qualities we must accept that consequence. Again we need the notion of a charactertype, particularly when describing a genre of fiction. The same type, we want to say, can appear repeatedly in unconnected narratives such that each narrative presents distinct tokens of that type. Yet how is it possible if characters already belong in the ontological category of types? Perhaps the best response is that whereas character-type is determined solely by clusters of qualities, a character-token is individuated both by charactertype and by a rootedness is some particular narrative. To determine the identity of a token we need to trace the character presentation back to an act of story-telling.31 This highlights the crux of the problem – the original act of story-telling and the extent to which its scope extends to the new narrative – but does not quite fit the interrelation between Odyssey 9 and Aeneid 3 because the original Homeric crewman has so few ‘salient qualities’, and Vergil adds so many, that we might say that they are two different character types (i.e. the original is a nameless, unindividuated background character, whereas Vergil names him and gives him a personal history etc.). Currie, in a chapter focusing on the ‘transworld identity’ of characters, points out that it is difficult to establish identity ‘for characters very underdescribed by the stories in which they occur.’32 What do we recognize when we come across Achaemenides that we have already seen in the Odyssey, in which he is highly underdescribed? We might take the pastoral world of the Eclogues as an example of how, in a different genre, there can be identity across various manifestations of a purely intensional construct. The poems, as the name states, stand alone each separate from the other rather than form a single continuous narrative, but the recurrence of characters and names of characters between the poems implies, or can be made to imply, an unnarrated, or extradiegetic, world in which the shepherds move and interact. The shadowy figure of Phyllis, who recurs throughout the Eclogues, helps illustrate this complex situation: we have too little information about her, too few attributes attach to the name ‘Phyllis’, for her to be identifiable by antonomastic paraphrase (iέeέ some nugget of information about her beginning illa quae would not allow us to marry the 31 32 Lamarque and Olsen (1994) 133. Currie (1990) 138; see also 133-5. 119 information to the aforementioned Phyllis); the act of giving her a name is therefore a condition of her continuity of character throughout the various poems, because her name is one of the few things we can consistently recognize in her. On the other hand, however, because we have so little information to check against the name, we cannot know for certain whether it is the same Phyllis each time or whether a new, different Phyllis is being described. The name might be consistent, but because it might be empty or almost empty of meaning, it really just consistently tells us nothing or very little. We say that it is the same Phyllis, probably because no obvious purpose would be served by the author having us create a different one each time: we have to resort to the presumed intention of Vergil the author, as if each use of the name were a speech act stipulating identity, or the intention for identity, cutting across any quibbles about her limited ontology. So although we can see the ontological problem of identity between Phyllides, a pragmatic solution is to imagine an authorial diktat overruling them. We can also overcome the shallowness of the pastoral world’s inhabitants by positing an imaginary world in response to Vergil’s empty description of Phyllisν within this world the term ‘Phyllis’ extends to a single person, and it is this single ‘Phyllis’ who crops up piecemeal in various poems. To change examples: Meliboeus has a dialogue with his friend, Tityrus, in the first poem; he is mentioned again in the third (with the suggestion that the situation has come to a head offstage, and he has now left herding altogether) and therefore clearly is or was a herding acquaintance of Damoetas and Menalcas, the amoebaean singers in that poem; Menalcas in turn is a lover whom Corydon rejects for Alexis in Ecl. 2 (15-16), and in the seventh poem Corydon (still seemingly in love with Alexis, 55-6) takes part in a signing contest witnessed by Meliboeus, and so on. There is no continuous narrative demonstrating the precise nature of the relations between the shepherds, but the reader can go beyond the narrative to posit a possible world in which further inferences can be made about them. Although I pay more attention to the blocking of inferences, this is surely an instance of Vergil inviting the reader to build up the make-believe world by supplementing it with further inferences. The one Menalcas or the one Meliboeus that links all of these mentions is 120 the one we posit as the extension of Menalcas or Meliboeus, in the fictional world we conjure up within which these terms can have extension.33 There are problems with this, however. All of the speech acts that create and identify Phyllis are made by the same author, whereas we have the impression that Achaemenides is created as an anonymous crewman by a speech act of Homer, and then identified later by a different speech act of Vergil. Similarly, the fictional pastoral world that we make up to house the characters of the Eclogues can be an ad hoc one in response to Vergil’s descriptions – we could, if we wanted, make it the same fictional world as that which housed Theocritus’ shepherds, but we do not have toέ Achaemenides, however, seems to come fresh from a specific event described in another narrative: he cannot belong only to the world we create for the Aeneid, as he is implied to have been present in the world of the Odyssey tooέ If we say he is the ‘same’ character across both manifestations of him, we have to say that he is somehow a hybrid created by the speech acts of two authors, belonging to two simultaneously valid fictional worlds. Because of the intertextual element, possibly three worlds: what if Achaemenides walks not from the make-believe world of the Odyssey, but steps out from the pages of the literary-textual Odyssey as a fictional character, strange and unreal? So the question is – what world are we in as readers when we read the Achaemenides episode? Of course, perhaps the insubstantiality of names in the pastoral world is not meant to be resolved: perhaps Phyllis is obviously artificial and refers not to any character existing in any world, made-up or otherwise, but simply marks the pastoral genre in which Vergil is writing. Perhaps too the paradox of Achaemenides is not an ontological problem to be solved but, as Hinds puts it, a ‘cipher’ whose inevitable artificiality (we cannot make him real in any one fictional world) forces us to see him as a statement about the artificial literary world.34 But while we might expect this from a Payne (2ίίι) argues for Theocritus’ bucolic poems as the first example in literature of a literary ‘fictional world’ν see espέ 1ηffέ While I do not dispute his thesis, he does seem to underplay the importance of myth as a story world within and from which transductions can take place. See also above pp.40ff. 34 Doležel in his theory of narrative uses the term transduction to refer to the continuation of one narrative into another, new narrative which typically tendentiously rewrites the previous one. It is, in other words, an artificial programmatic statement. On the post-modern re-writing and ‘correcting’ of texts, see Doležel (1998) 206; the tendentiousness of intertextual remakes of predecessors is a consistent theme throughout Hinds (1998), but see esp. 1ίίffέ, and 113 on Achaemenides as a ‘cipher’έ 33 121 ‘lower’ genre like pastoral, which openly makes throw-away truth-claims for the sake of self-signalling, it is perhaps more difficult in epic, which generally narrates people and events already supposed to exist in the world of myth, and which tends decorously to avoid topicalizing its own truth claims. If Achaemenides is different from Phyllis in the Eclogues in that he is a minor character crossing an intertextual, rather than intratextual, boundary, both characters are different again from Odysseus, who is no minor literary character but a major figure of myth. Odysseus appears obliquely in Book 2 but he does so without triggering any sort of intertext that we know of: this is because as a hero of the ancient world he moves, in one instantiation, within the world of myth and in another within the more specific world of the narrative created by the Odyssey. Myth, though it is produced by a collection of individual narratives, forms an overarching story world of its own which texts can instantiate without entering into a literary relationship. Furthermore there are multiple Odysseuses and so it is not necessarily the case that any depiction of Odysseus would railroad the narrative of the Odyssey straight through the Aeneid: as a mythical figure he enjoys the privilege of a reified existence that transcends and is not dependent on any single portrayal of him, such that Homer does not have an exclusive claim on him.35 However, were Vergil to have Aeneas meet Odysseus in a scene that closely imitates Homer’s Cyclopeia (as Achaemenides’ account does) then this would be τdysseus as instantiated by Homer’s text, and we would be dealing with a collision of literary narrativesέ Vergil’s shying away from τdysseus is perhaps because of a fear that any meeting between the Ithacan and his own epic hero, in an epic poem (and an epic poem that borrows heavily from Homer), would inevitably suggest the two meeting as literary as well as mythical peers: there is simply too much metapoetic pressure from other parts of the Aeneid for readers to interpret the two as crossing each other only in the myth world, and to suppress any considerations of the Odysseus of the literary Homeric world. The fictional world would fail to drown out or keep from the frame the real literary world, in which it is true of Homer’s character τdysseus that he too went to 35 There are plenty of non-Homeric Odysseuses and Ulysses that Vergil could pick from, such the Odysseus who is a co-founder of Rome with Aeneas, as told by Hellanicus and Lycophron, for example (FGH 4 F 84 = DH 1.72.2; Lycoph. 1243-5). 122 the island of the Cyclopes. Also, any hypothetical meeting would have most naturally fitted into Book 3, for example, where it could not have been depicted without inevitably thematizing the literary similarity between the two heroes’ wanderings, thus turning them into ciphers. In other words: there is little doubt, given the peripheral presence of Odysseus in Books 2 and 3 and then the Achaemenides episode, that Vergil probably wanted to allude to his poem’s relationship with the Odyssey, but also that he would surely have wanted to do so without embroiling the protagonists of both in an artificial and trivializing literary rendezvous.36 The bind he is in, which makes equally unfavourable both options of a meet-up in the narrative world and in the mythical world, leads to Achaemenides as a compromise. The skirting around Odysseus, however, still adumbrates his presence and gives a foretaste of the fictional game at the end of Book 3, in which it is proposed that fictional constructs from one narrative can be present within another. Achaemenides and simultaneous narration Achaemenides is unique in the Aeneid in his being a character marking an intertext between narratives who is, as far as we know, also non-traditional and non-mythical: therefore unlike Aeneas or Odysseus (or Sherlock Holmes) he exists only as an inference, or the result of an inference, from another story. He is not an instantiation of a figure who exists in the mythical or historical or real worlds, but totally artificial, the illusory extension of Homeric postulates into the Aeneid’s fictional, and I will argue paradoxical, world.37 One might object that as any retelling of the Cyclopeia story would have Odysseus accompanied by his crewmen, Achaemenides cannot with any certainty be said to be exclusively the continuation of a Homeric e9tai=rov: the archetypal myth of Odysseus and Polyphemus would always include the potential for an Achaemenides figure. Vergil gives a number of signs, however, that he is picking up directly where Homer left offμ Achaemenides’ introduction of himself as comes infelicis Compare δabate (2ίίλ) 12ι, who sees Vergil’s avoidance strategies as a means of ensuring his Aeneas remains a credible historical figure, distanced from the mythical world of the Odyssey. See also Nelis (2001) 61. On Vergil avoiding Homer in the Underworld, and choosing Musaeus instead see Deremetz (2005) 113. 37 Cf. Ramminger (1991) 69, who searches long and hard for an intertextual exemplar other than the unnamed crewmanμ ‘Achaemenides’ [outward] appearance is … another example of Vergil’s use of traditional literary images, without reference to any one text in particularέ’ 36 123 Vlixi (613) at the start of the Aeneid’s Cyclopeia is a segue from the e9tai/roi (9.566) which had been the closing word of the earlier Homeric version.38 Vergil then follows Homer by repeating comes infelicis Vlixi at 691 as the closural note to his own rendering of the episode. That Achaemenides is a not a generalized Odyssean crewman but a Homeric one is also signalled by his emphatic claim to have witnessed scenes directly lifted from Odyssey 9: uidi egomet duo de numero cum corpora nostro prensa manu magna medio resupinus in antro frangeret ad saxum, sanieque aspersa natarent limina; uidi atro cum membra fluentia tabo manderet et tepidi tremerent sub dentibus artus. 3.623-7 The proximity of this scene to Polyphemus’ dashing out the brains of his victims in the Odyssey is important: it establishes that he was present as the events of the Odyssey, and the Odyssey specifically, were taking place.39 This, as the emphatic egomet and anaphora of uidi underline, is an eye-witness, inside-track account of the story that was told in another text. There are problems here. The crewman in the Aeneid has a name, a family background (we learn the name of his father and that he was poor, 614-15), and a description of his appearance is given (590-4). These are all invented by Vergil: there is no crewman described thus in the Odyssey, so can this character be the same as one of the unnamed e9tai/roi, and consequently can his claim to have been present still be a plausible fiction? One could quickly answer ‘yes’μ Vergil is pretending that his character, which he has invented, is the same as an anonymous member of a collective group of characters in the Odyssey, and he is simply making further inferences about how one of the original crewmen could possibly be realized. No, Homer’s collective crewmen cannot be extended to any real world referents, because they do not exist and 38 Though note that Od. 9.566 refers, formulaically, to comrades who have been lost to death. For Vergil to identify his comes with one of the e9tai/roi at 566 he must either postulate that Odysseus mistakenly believes Achaemenides to be dead (hence his abandonment) or, in an intriguing twist on the infernal aspects of Book 3, that Achaemenides is dead and what we encounter is his ghost, moving between fictional worlds. See the next chapter, where I discuss death and fictive transformation. 39 Od. 9.287-90: w$j e0fa&mhn, o( de/ m' ou)de\n a)mei/beto nhle/i qumw?|, / a)ll' o# g' a)nai5caj e9ta&roij e0pi\ xei=raj i1alle, / su_n de\ du&w ma&ryaj w#j te sku&lakaj poti\ gai/h| / ko&pt': e0k d' e0gke/faloj xama&dij r(e/e, deu~e de\ gai=an. 124 nor does Homer give any hints as to how they might possibly correlate to real people, but Vergil can rewire the semantics of e9tai/roi by hypothesizing a possible but unreal world in which it could extend to people with names and backgrounds and so on.40 However, is this further imagined world Homer’s augmented (and thus exclusive from the Aeneid’s world), or a new Vergilian world that replaces it entirely (and thus exclusive from the Odyssey’s world)? The illusion is of an impossible continuity in which a possible world of the Odyssey extends into the world created by the Aeneid – of Achaemenides achieving existence through Homer but then passing over into a narrative told by Vergil. Again, I should stress that Achaemenides’ being a minor character is crucial hereέ An epic poem can instantiate characters from traditional myth without ‘letting in’ characters from other narrators’ worlds because belief in Helenus’ role in Aeneas’ journey from Troy to Italy, say, had already been established by a series of earlier narratives: a meeting with Helenus (somewhere – other sources placed it in Dodona not Buthrotum) does not entail a clash of simultaneous narrators as the meeting was no longer thought of as an episode tendentiously narrated by a rival narrator: it is an established, general truth of myth that Vergil can draw on. Believing that Achaemenides is the same as one of τdysseus’ crewmen, however, entails that a character who exists only by means of one narrator’s, Homer’s, speech act – that is, who has no external existence like Helenus – can continue to exist in a narrative which is supposedly constituted only of the speech acts of another narrator, Vergil. Two narrative worlds come together in the same text and we have the illusion of two simultaneous narrative worlds, which is completely paradoxical. So to whom does Achaemenides belong? The notion of continuity between two narratives is impossible, because what happens and who exists in a narrative is a result of description in that narrative: narrating a story is an exclusive business, and no two narrators can share the same narrative world (though they can both participate in the See Doležel (1λλκb) ικι-8 for the argument that the creation of possible worlds is a means of getting round the non-referentiality of fictional utterances. See also the section on psychagogic worlds in the introductory chapter. 40 125 same story world, such as myth).41 Moreover we know the continuity is impossible because Homer’s ‘companion’ and Vergil’s Achaemenides are only descriptive constructs relating to no existing person common to the two descriptions: to change the description of a person who only exists as a description, as Vergil does with τdysseus’ unnamed companion, is not to give a new instantiation of the person but to create him all over again: it is a new character, rather than a sequel to the old one. Unlike Aeneas or σapoleon, there is no superordinate ‘Achaemenides’ of which fictional descriptions constitute an instance: the archetype is created and recreated each time. As Hinds notes, Vergil seems to be aware of the pragmatic impossibility of his narrative game when he describes his creation as a noua forma (591): Virgil’s ignoti nova forma viri … is simultaneously an arch denial of Achaemenides’ status as a derivative character, and a literally true designation of a particular Odyssean crewman who never achieved named individuality back in Homer’s epicέ42 Achaemenides, going by my interpretation, is ‘new’ because, really, the expansion of his character entails replacing entirely the old dead-end reference to a e9tai=rov with suppositions and inferences that go above and beyond what could be inferred from the Odyssey.43 But, according to the illusion, he is simultaneously a derivative character, albeit one overhauled in the Aeneid, because his existence in its narrative world derives from an original Homeric postulateέ Similarly, Achaemenides’ eye-witness account gives us more than enough information to confirm that he is a recurrence of the ‘same’ Homeric crewman from one narrative to the other, but to accept this as plausible we have to accept also that a character can be the simultaneous result of two narrators him describing him into existence at once, and that two narrative worlds subsist parallel in the one narrative. Alternatively, if we say, not unreasonably, that this is impossible and that Achaemenides must be a new descriptive construct generated solely by the Aeneid, which completely displaces the original, then the fiction of his presence during the Cyclopeia becomes flawed and implausible. By having him cross the borders, or the shorelines, between narratives, Vergil makes him an impossible paradox. On the impossibility of simultaneous narratives see Hutcheon (1996) 216-1ιν Doležel (1λλκ) 1θ-18. σote too how we might see this as an experimental development of Vergil’s Alexandrian habit of giving simultaneous, even conflicting variants of the same myth. 42 Hinds (1998) 114. 43 Though Homer did not invite these inferences – Vergil is being a rebellious reader and pursuing a blocked inference. 41 126 There are, as E. L. Harrison nicely points out, subtle hints that a character who was present at the confrontation with Polyphemus cannot possibly explain his continued presence on the same island in a new narrative.44 Achaemenides’ account of the episode is quite faithful to the Odyssean original and so after the description of the braining, the drunken vomiting, and the blinding we would expect a re-run of the story of how τdysseus contrived his crew’s escape by clinging on to the bellies of the Cyclops’ sheep. He breaks off his account abruptly, however, at 639: sed fugite, o miseri, fugite atque ab litore funem / rumpite. As Harrison argues, Vergil here seems to be acknowledging that he is heading for a crux, from which he must turn back: in the Odyssey all of the surviving crewmen were safely extracted from the cave once Polyphemus had been blinded, and no mention is made of one either not making it out or getting lost en route back to the ships. Achaemenides halts, seemingly arbitrarily, because he realizes he is about to spill to the Trojans the secret of his impossibility as a characterμ if he is still on the island then he cannot have been one of τdysseus’ crewmen. It is a neat paradox: his ability to recount the brush with Polyphemus so closely entails that he must have been there with the hero when he happened; but his being one of the hero’s companions should also entail that he escaped and left the island with him, so his ability to recount the story to the Trojans at all when they visit Sicily must also entail that he was not one of τdysseus’ companionsέ45 Of course, we could counter here that as Aeneas’ Cyclopeia takes place in a possible world extended from the Odyssey, which Vergil is free to develop as he wants, he could plausibly fill one of the gaps in Homer’s narrative with a new story of Achaemenides becoming detached from the main group of his companions, thus reconciling the two texts: say at Od. 9.461-7, where the account of the journey back to the ships after exiting the cave is given in a relatively undetailed summary, which could be read inferentially as containing the elided story of Achaemenides, the crewman who went missing. That he does not do this is another indication of the limitations on how far he is willing to take the narrative game he is playing: creating a possible world which contains an alternative, almost contradictory version of the Odyssey would be too indecorously 44 45 Harrison (1986). It is interesting that this is also the paradoxical condition of recounting a catabasis; see below p.p.142- 3. 127 playful. And perhaps too, looking at it another way, Vergil does not want to pretend that the two narratives segue smoothly: perhaps he enjoys the paradox of it all. Other aspects of the episode imply a direct temporal continuity between the events of the Odyssey and the Aeneid, what Horsfall calls a ‘filling in’ of the story between the two.46 It is again, however, paradoxical, involving a clearly impossible mixing of ontological orders (the ‘reality’ of the internal world of the Aeneid and the ‘falsity’ of a character from an external work of fiction)έ For Hinds, the ‘mixing’ of worlds is between the mythological and the literary: For poets who handle mythological themes, occasions for negotiation between the time-frames of the narrated world and the time-frames of their own poetic traditions will tend to arise again and again.47 So, he says, regarding the spruced-up Achaemenides who appears in Ovid Met. 14, the changes wrought on the Vergilian crewman record not only the passage of time since he was rescued by Aeneas, but also the time between the writing of the Aeneid and of the Metamorphoses.48 As well as the metapoetic effects Hinds detects in Ovid and in Vergil, which I follow up below, we also see a narratological effect in which an underlying progression of story-world time is threaded through the narratives of both the Aeneid and Odyssey. The first glimpse of Achaemenides illustrates this: cum subito e siluis macie confecta suprema ignoti noua forma uiri miserandaque cultu procedit supplexque manus ad litora tendit. Aen. 3.590-2 As we find out later, this is not the way he looks constantly, nor is it an index of his attributes as a character; rather the dirt and untidiness is the result of a storyline that began in the Odyssey and has continued through into the Aeneid: he was left on the island and regressed to primitivism over the next three months as he lived off the land and tried to avoid the Cyclopes (645ff.). The portrayal of Polyphemus eyeless in the aftermath of the attack on him (655-65) is another instance of the Vergilian ‘sequel’ 46 Horsfall (2006) 408. Hinds (1998) 115. 48 Hinds (1998) 113. 47 128 participating in a shared time continuum.49 The word that follows the above passage is respicimus; its stark position at the start of the line in a one-word sentence makes one wonder if Vergil is highlighting the fact that the origins of Achaemenides’ plight lie in a previous narrative: upon seeing the state of him, the Trojans immediately thumb through their Odysseys to see where this man came from and how he ended up in the state he is in. That Achaemenides is a literary character who has broken free from his narrative bounds (thanks to Vergil’s rogue inferential work) and is now anomalously and intrusively continuing his storyline in the ‘real’ world of the Aeneid is signalled again by the extraordinary metapoetic word relegens: talia monstrabat relegens errata retrorsus litora Achaemenides, comes infelicis Vlixi. 3.690-1 His memories consist not in the experiences and senses of the real world, but in the narrative of the literary text which begot him and then interpolated him into this new narrativeμ as he looks back at his strange trajectory he ‘rereads’ his textualized past to the Trojans.50 The suddenness of his appearance (subito, 590) is also perhaps a sign of an unexpected development in the narrative: the fact that a man appears from some woods is not a particularly dramatic or arresting happening, nor the sort of sudden change in action marked by Vergil’s use of subito in comparable contexts.51 The sudden break marked by subito is perhaps not a development in direct narrative action, but the meta-narrative break of his trespass from the Odyssey into the precincts of Vergil’s 49 It also raises interesting questions about the archetypal form in which mythical figures are stored in the collective consciousness, especially if they feature in a myth in which an irreversible change befalls them. The mental image most people would have of Polyphemus would be a giant with one eye (I will make my readers modern but only to avoid the vexing question of whether the Homeric Polyphemus was monocular or not – see Mondi (1983) 31-6 on the folkloric origins of this)έ But shouldn’t it really be a giant with no eyes at all? It is not, because the conception of mythical (perhaps all) figures is often timeless: the stories in which they participate do not have the causal power to affect the snapshot general idea by which the mythical figure is tagged in our cultural memory. Perhaps it is trivially true that all memory involves a synecdochic selection of what seems to be most appropriately representative of the figure, rather than a true record of what ultimately became of them (otherwise our mental image for all historical figures would be a corpse). Nevertheless, it seems fair to say that representations of Polyphemus point to the generalized ancient idea of him as being statively one-eyed or subject to a process of eye-losingμ Vergil’s statively eyeless Polyphemus is another example of his showing us the lesser-displayed flipside of the story. See Clare (1998); on causal changes to heroes remaining after death see below, pp.177, 184. 50 On this see Hinds (1998) 114 n.24. 51 Cf. e.g. the drastic and ominous nightfall during the tempest: eripiunt subito nubes caelumque diemque / Teucrorum ex oculis (1.88-9); the menacing sound of the approaching enemy as Troy falls: subito cum creber ad auris / uisus adesse pedum sonitus (2.731-2). 129 epic.52 The jerky abruptness, the arbitrary suddenness of his apparition and (as noted above) of his account itself are both markers of the incongruousness and artificiality of his role in the Aeneid. Also in question is the geographical continuity between Homer’s unnamed island of the Cyclopes and Vergil’s Sicilyμ is this the same crewman and are the Trojans even on the same island?53 One interpretation could even go as far as positing that Achaemenides comes from a Cyclopeia set entirely in an alternative possible world, in which the Ithacans went to Sicily, not Homer’s island, and in which one crew member was lost: if the Trojans went to the right island they would find the Polyphemus of Homer’s possible world but no missing crewmen, as no one went missing in his version.54 One thing I would like to note is that setting the encounter on a shoreline constitutes a fitting spatial representation of the paradoxical meeting of narrative worlds: the boundary line between land and water (though we should remember the meeting takes place with both parties on land) is also the boundary line between the enclosed, contained world of the Aeneid’s narrative and the external, uncontained world of ‘other’ storiesέ55 In both Odyssey 9 and Aeneid 3 journeys by water are opportunities to narrate strange encounters with a monstrous ‘other’, though the relationship is often 52 Note Khan (1998) 235-8 and the interesting observation that the use of subito here associates Achaemenides with the sudden occurrence of prodigies (as, e.g., the plague in Crete, 3.137; the apparition of Anchises, 5.723; the (false) omen of the attacking eagle, 12.249). Note also, below, the portentous aspects of the ecphrases (pp.206-7). subito also identifies Achaemenides as an irruptive, adventitious fictive episode that the poem must incorporate. For an equally obvious, almost self-mockingly transparent introduction of a new topic, cf. the segue into the Iolaus story at Ov. Met. 9.394-7: compescuit omnem / res noua tristitiam: both episodes are only sudden or new artificially, as new topics in the narrative. 53 It seems that Theocέ θέ2 is the first location of Polyphemus in Sicily, with Homer’s Cyclops on an unnamed island. Jolivet puts forward a convincing argument for Vergil’s use of scholiastic and mythographic sources to ‘correct’ Homer by specifying Sicily – Jolivet (2005) 44-7. As Jolivet notes, however, some critics – Aristarchus and later Apollodorus – still located the Odyssean wanderings in an entirely make-believe geographyέ εondi (1λκ3), who assesses Polyphemus’ origins in folklore, speculates a mythical, and not folkloric, origin for Homer’s idyllic, golden-age island, in which it was given to the Cyclopes by Zeus as a reward for their help in defeating the Titans. So it is likely that despite Vergil’s choice of Sicily Polyphemus’ island still had some connotation as a marvellous, fictional nonplace. Note too that journeys to non-places were not limited to archaic myth: the early 3rd-century BCE Euhemerus had told of travelling to a fictional archipelago where he saw an inscription telling the true history of the gods. 54 Or, conversely, Achaemenides might prompt us to ask what is often the most challenging question of any fictionμ if Vergil’s crewman were coincidentally exactly the same as Homer’s and had all the same attributes, would they still be the same? If not, then we have to find criteria of identity other than the sharing of attributes (e.g. the signposted or intimated intention for identity – but isn’t this unsatisfyingly arbitrary? And so on). See Lamarque and Olsen (1994) 50-2, and especially Lewis (1978) 41. 55 Note how the notion of narrative trespass, the Trojans blundering into the narrative world of the Odyssey, is signaled metaphorically with their heedless journeying into the land of Cyclopes: ignarique uiae Cyclopum adlabimur oris (3.569). 130 the other way round, with monstrous otherness found pocketed in remote islands and the sea a place of refuge and safety. Passage through water is frequently used in the Aeneid, and elsewhere, to figure transition into some sort of fictional otherness.56 Metapoetic rewriting What, then, is some of the wider significance of all this? If Achaemenides is to the Odyssey as Tom Stoppard’s Rozencrantz and Guildenstern are to Hamlet, then we need to ask what is glimpsed through this new perspective that could not be seen before – what, to use Hinds’ term, is tendentious about it, and how is Achaemenides given new thematic significanceς Doležel notes that postmodern rewrites of narratives tend to inscribe new values on the older text, and there is plenty of evidence of that in this instance.57 Putnam notes how the shift from saeui Vlixi at 273 to infelicis Vlixi at 613 and 691 marks the creation of a new, more humane Roman epic hero.58 Also of interest is Hinds’ notion of a series of intertextual redemptions of Achaemenidesμ Vergil takes an unnamed Homeric background character, a cardboard cut-out who has no attributes or human depth, and makes him into a living character; in the process however, Hinds argues, he really makes him a literary cipher for the intertextual relationship between himself and Homerέ τvid’s description of Achaemenides as iam suus in his own ‘remake’ of the Aeneid (Met. 14.166) is seen as especially loaded: Equally suggestive in the Ovidian passage is iam suus, ‘now his own man’έ… Does τvid thus imply that Achaemenides was a mere cipher in Vergil, and has become a fully achieved character only in the Metamorphoses?59 56 An excellent, and apposite, cinematic example of a river encounter as a brush between narrative worlds can also be found in, all of places, the spoof comedy Hot Shots! Part Deux. A character played by Charlie Sheen is on a river patrol boat in south-east Asia: the scene evokes the general fictional world of Hollywood’s Vietnam, and parodically incorporates elements from the film Apocalypse Now. Sheen stands to inspect an identical boat approaching his, and as it passes sees that it contains his father Martin Sheen in his guise as Willard, his character from Apocalypse Now, repeating the scene from that film in which he sailed in a patrol boat on a south-east Asian river. The switch from a general story world within which the film takes place (Hollywood’s Vietnam) to a specific narrative that intrudes into the film itself, and seems to be taking place at the same time, is exquisiteέ Whatever one’s opinion of the film, Donnelly (1994) 254-η is right to praise the ‘theoretical approach to cinema’ it embodiesέ 57 Doležel (1λλκ) 2ίθ. 58 Putnam (1995) 64; Hexter (1999) 68 refers to Vergil’s ‘agon’ with his predecessor, imitating closely a Homeric episode so as to ‘correct’ itέ 59 Hinds (1998) 113. 131 According to my interpretation of the episode, there is perhaps more we can say about iam suusέ First, however, it is worth seeing what else we can do with Hinds’ idea of a series of tendentious, ‘corrective’ intertextsέ Vergil is a highly tendentious, resistant reader of Homer in rewriting the anonymous background character, as the original is a deliberately inference-blocked figure; that is, the description of him is deliberately sparse and invites no further inference because he plays no relevant part in furthering the themes of the narrative. For Vergil then to do the opposite, and explore the possible extensions of the crewman, defiantly negates Homer’s attempts to guide his reader’s attention to what he thinks is relevant and worthy of further meditation in his narrative. Recall, however, the scholiasts’ tendency to read further logical inferences into passages of Homer, in such a way that they ended up commenting on their own hypothesized version of the text (see above pp.34ff.)ν Vergil’s exploration of a possible Homeric crewman employs a similar method, and we might see this as an example of Vergil correcting Homer via critically-derived methods. Vergil acts like a scholiast by reading against the grain and bringing attention to the flatness of Homer’s portrayal of τdysseus’ companions, which leaves untold and obscures the humanity of some of the characters inhabiting his fictional world.60 The technicolour virtuosity of the description of Achaemenides as he appears to the reader for the first time, with its close attention to detail, is thus a reproof of Homer’s curtailment of him as a characterέ While in the Odyssey, the protagonist, definitely a ‘somebody’ in the world of myth as his rash boast to Polyphemus claims (9.502-η), becomes a ‘nobody’ in a game of deceit (ou]tiv, 3667), Vergil asserts a new, inverted emphasis on the lost and hidden aspects of narratives by constructing a fictive game in which a Homeric ‘nobody’, the nameless crew member, becomes a named and delineated ‘somebody’ in the Aeneid. While, as stated above, for decorum’s sake he refrains from redrafting too drastically the original story, in this instance Vergil subtly imprints this change on Homer’s original accountμ upon re-reading the Odyssey we cannot but think to ourselves that any unnamed, non-descript ‘companion’ might well be Vergil’s Achaemenides lurking in the shadowsέ Just as the episode is a Homeric intrusion upon the Aeneid, so too in turn the Achaemenides episode intrudes upon our subsequent re-reading of the Odyssey. 60 For a modern, and markedly adverse, critical view of Homer as an author unalive to the nuances of the implicit see Auerbach (2003) 6-7: ‘never is there a form left fragmentary or half-illuminated, never a lacuna, never a gap, never a glimpse of unplumbed depthέ’ 132 If we follow Hinds in seeing the episode as full of metaphors signalling the intertextual relation between the two poets, we might see further critique of Homer’s fictional technique in the story that Vergil tells: as Achaemenides is abandoned by τdysseus and then rescued by Aeneas, so Vergil ‘rescues’ him from the fictional oblivion in which Homer had left him stranded. As we have discovered, however, he has delivered him into a paradoxical, impossible existence in which he straddles two narrative worlds, belonging to neither. To continue the metaphor, then, in rescuing Achaemenides, Aeneas has also exposed him to the further dangers awaiting their departure from Sicily: Scylla and Charybdis and the storm that opens the poem. Likewise, Vergil has rescued him from a truthless, permanently embryonic existence in Homer, but delivered him into a narrative from which he disappears almost as soon as he has entered it; and, further, into a nebulous fictive world to which he belongs as little more than a fictional conjuring trickέ τvid’s retelling of the Aeneas legend in Metamorphoses 14 picks up on this recursive pattern: as Vergil followed his predecessor by correcting the gaps left in his account, so Ovid follows Vergil by correcting the gaps in the latter’s corrected versionέ61 His first improvement is to assure his reader that Achaemenides did not, as Vergil had made out, simply disappear into the fictional ether: talia quaerenti iam non hirsutus amictu, iam suus et spinis concerto tegmine nullis fatur Achaemenides. Met. 14.165-7 τvid repeats Vergil’s trick of weaving his Achaemenides into a temporal continuum that stretches across the separate narratives, the iam pointing back to the timeframe of the Aeneid’s narrative worldέ62 As already stated, the correction that Hinds sees implied in iam suus, ‘now his own man’, is to make Achaemenides a full character after Vergil’s treatment of him as no more than a literary cipher. But it also raises again the question of who, which narrative, Achaemenides belongs to: Ovid implies that he has resolved the problem of his bifurcated existence between two worlds and two authors by making him whole and self-contained, iam suus, in the Metamorphoses: he now inhabits fully σote how τvid’s Achaemenides includes the detail of Polyphemus’ barrage against τdysseus’ ship (14.181-2)μ something Vergil’s version chose to omit because of the questions surrounding Achaemenides’ whereabouts during the departureέ 62 Hinds (1998) 113. 61 133 the narrative world of the Aeneid, and the role of interstitial straddler is passed on to his ‘double’ εacareusέ63 Finally, the theme of the bifurcation and subsequent unity of the self is important to wider questions we raised in the previous chapter about fictional portrayal. I return to the notion that Achaemenides (and Phyllis from the Eclogues) are not discrete, self-contained entities but constructs that are recreated anew each time they are described: the fact that any narrative or artistic representation of a person (or thing) entails, to varying degrees, the transformation of that person into a new descriptive or representative construct has already occurred to Aeneas as a bewildering corollary of his narrating his own storyέ He was forced to realize that his narrated self is not the ‘same’ as the narrating self, but a discontinuous bifurcated or dissociated version: his eventual destiny is to cease existing as the single, flesh-and-bone Aeneas and become refracted into the Aeneases of thousands of different manifestations. Bearing this in mind, I wonder if this is what underlies Achaemenides’ curious statement that his captain overcomes the Cyclops by not ‘forgetting himself’μ haud impune quidem, nec talia passus Vlixes oblitusue sui est Ithacus discrimine tanto. 3.628-9 What exactly is the ‘self’ that he remembers and that he lives up toς If Odysseus sometimes could potentially not act like Odysseus, sometimes not conform to the set of attributes that attach to his name, then there is, it is implied, some manifestation of him that we are unaware of, some other mode in which he exists, which belongs in an undisclosed possible world. There is Odysseus, and there is also Odysseus-in-the-story, and here he must remember to live up to the latterέ δike Tennyson’s Ulysses, he has ‘become a name’μ but like Achaemenides, or Rosencrantz and Guildenstern, behind this name, this legendary existence, is an implied offstage existence, and he too ducks and weaves in and out of stories.64 τvid seems to pick up on this sense of ‘forgetting’ when his Achaemenides recalls Polyphemus’ attack on the Greek shipsμ I leave unanswered the question of whether τvid’s Achaemenides belongs solely to a limited narrative world (as Vergil’s did) or whether his appearance in the Aeneid, due its achievement of almost immediate canonicity, conferred upon him a place in the myth world of Aeneas’ wanderingsέ 64 For a further metapoetic forgetting cf. 3.616-18: hic me … / immemores socii uasto Cyclopis in antro, / deseruere. Not only is he a forgotten background character (and also therefore unknown and unknowable, 63 134 et ne deprimeret fluctusue lapisue carinam pertimui, iam me non esse oblitus in illa. 14.185-6 He too hints at confusion of his multiple self-manifestations, forgetting that he is no longer in the Odyssey, in whose fictional world he would be on the boat, but in the fictional world of the Aeneid / Metamorphoses, in which he was left stranded. Conclusion I consider now some of the ramifications of this paradoxical simultaneous narration for Books 2 and 3, and its relevance to the narrative framework of the Aeneid as a whole. Some of the issues considered can be clarified through a comparison with the sleeping Red King in δewis Carroll’s Through the Looking Glass (that thesaurus of language paradoxes): ‘He’s dreaming now,’ said Tweedledeeμ ‘and what do you think he’s dreaming aboutς’ Alice said ‘σobody can guess thatέ’ ‘Why about YτU!’ Tweedledee exclaimed, clapping his hands triumphantlyέ ‘And if he left off dreaming about you, where do you suppose you’d beς’ Where I am now, of course,’ said Aliceέ ‘σot you!’ Tweedledee retorted contemptuouslyέ ‘You’d be nowhereέ Why, you’re only a sort of thing in his dream!’ ‘If that there King was to wake,’ added Tweedledum, ‘you’d go out – bang! – just like a candle!’ ‘I shouldn’t!’ Alice exclaimed indignantlyέ ‘Besides, if I’ε only a sort of thing in his dream, what are YτU, I should like to knowς’ ‘Ditto’ said Tweedledumέ ‘Ditto, ditto’ cried Tweedledee.65 The novel ends with Alice returning to the question of whether she was part of the Red King’s dream, or vice versa, or both: ‘σow, Kitty, let’s consider who it was that dreamed it allέ… You see, Kitty, it MUST have been either me or the Red King. He was part of my dream, of course – but then I was part of his dream too! WAS it the Red King, Kittyς’66 ignoti 591), Vergil’s retelling can also be figured as the recovery of a forgotten variant in which a crewman is left behind. 65 Carroll (1960) 165. 66 Carroll (1960) 236-7. 135 Tweedledum and Tweedledee confound Alice by presenting her with a paradox of ‘inter-subjectivism’έ67 Dreaming is a state of mind exclusive to the dreamer, or subject, such that everything contained in the dream is put there by the dreamer with no other dreamer able to access it: there can be no convergence between the dreams of different dreamers. The twins cannot be telling the truth, because Alice can only be in the Red King’s dream as a construct of his dreamμ she cannot be in the dream and be the same, real Alice at the same time. Nor can the events of the narrative have been the single dream of both Alice and the Red King, as she speculates: they could have both had a dream that was exactly the same in content, but they would still have been separate mental events and neither could have accessed the dream of the other. Though dreams have important associations with fictions, I am citing Alice’s aporia only as a demonstration of the same paradox found in the Achaemenides episode, and I hope it is fairly clear by now how the analogy works. Just as everything that occurs in a dream is dependent upon the dreamer for its existence, so everything in a fictional narrative is present in that narrative because, absolutely exclusively, the narrator has put it there by describing it. And just as there can be no such thing as an intermediate, shared dream world, in which dreams mingle and interact, so there can be no shared narrative world, in which separate narratives meet one another or narrate simultaneously. The Odyssean crewman Achaemenides is in the Aeneid because Vergil describes him as being so: a truly Homeric character can no more walk into the Aeneid than Alice can walk into the dream of the Red King: it is really a Homeric character under Vergilian description. Similarly, as Alice will disappear ‘bang! – just like a candle!’ if the Red King stops dreaming her, so unreified minor characters like Achaemenides disappear when they are no longer narrated as, in a normal, compliant reading of the text they belong to, they have no extension into any other world which sustains their existence.68 To pretend that this is not the case, and that, like Tweedledum and Tweedledee’s paradoxical dreams, narratives can overlap each other has consequences: it entails that Vergil is also pretending that the narrative world he creates is not the exclusive domain of him as narrator, and that his narrative is not under his control. Also brought into 67 For inter-subjectivism in Carroll see Agassi (1975) 6. On the notion of Alice being trapped in a world constituted by language and language games, and for a thorough-going philosophical interpretation of Through the Looking Glass, see Sacksteder (1967). 68 For an ancient paradox of dreams continuous with reality see Martial 6.53: Hermocrates the doctor is so dangerous that seeing him in a dream can kill in reality. 136 question is the literary compendiousness of the Aeneid, its status as an allcomprehending summation of poetic influences, the literary component to its putative totalization of history and the cosmosμ rather than ‘incorporating’ this intertext and subordinating it to the narrator’s design, it creates the impression of two narrators equally participating in constructing a narrative worldέ Doležel notes the potential for a collapse of the narrative’s defining terms and structuresμ If beings from different ontological orders co-occur, then the ontological status of all inhabitants of the fictional world is problematized. Who is the author and who is the authored, who is the fiction maker and who is a fictional being?69 The totalizing power of the narrative is also undermined by the illusion of a narrative world that can be decoupled from the narrative speech acts that create it. While the Aeneid is a compendious incorporator of other utterances, epic decorum entails a lack of means of governing and framing them coherently. Homer’s crewman is non-reified and exists only by virtue of a speech act making him exist, a speech act tied to the locutionary context of the Odyssey. 70 That same speech act could only be incorporated in the Aeneid by saying that the locutionary context of the Odyssey is also the locutionary context of the Aeneid – that is, implying an impossible merger of the two poems and the unframable contradiction of being in two contexts at once. Now the contradiction could be avoided by Vergil’s stating that he is only quoting, or commenting about, Homer’s original creation of the crewman and not perpetuating it on the same grounds (which is largely his purpose anyway), but to do so would be to admit he is just a poet making comments about the speech acts of other poets that can be found in libraries in the real world – it would admit as false the illusion of his presence within the fictional world he narrates, and so epic decorum forbids it. He is therefore forced into the madness of a narrative that is in two places at once.71 Doležel (1λλκ) 22ίέ Similarly characters invented, say, for a hypothetical story are never instantiated or reified – they exist for an ad hoc, provisional purpose and are only valid regarding that purpose. I would therefore be surprised if the same ‘John Smith’ that I made up to illustrate an argument about politics, or whatever, turned up soon afterwards in a short story. The same applies to minor characters. 71 Let us use one of Austin’s examples of a speech act as an illustrationμ naming a boat (Austin (1975) 56). At a launch that I witnessed, the queen named a boat when she said ‘I name this ship Brenda’ – the act of saying it caused the naming to happen. When, asked later that day what I had seen and heard, I repeat ‘I name this ship Brenda’, I do not repeat the queen’s act and name the ship all over againέ εy speech act is one of repetition, and what I repeat is a framed speech act which lost its illocutionary force with the loss of its context of utterance. I cannot be the Queen at the same time as imitating her speech act, or repeat her action as I say the words: that would entail the madness of being, simultaneously, me repeating 69 70 137 τne thing I have not touched on so far is the Vergilian Cyclopeia’s occurrence within Aeneas’ inset narrativeέ As he is an unreliable narrator, possibly given to embellishing his story to Dido with scraps he has picked up from literary and artistic sources, we might see the episode’s literary nature, both in terms of verbal style and ontologically, as an addition he, and not Vergil, has made in order to impress on Dido how he moves within the world of heroism and romance.72 More fitting with my earlier argument of ‘narratorial lapse’, however, is to propose Achaemenides’ entrance as another collapse of narrative integrity in Aeneas’ speechμ that is, just as the poem’s structural framework fails to keep Aeneas and Sinon in their subordinate position as inset speakers, and instead lets them at times intrude as master narrators, so too at the end of Book 3 the structural framework gives way again and ‘lets in’ a character from a rival narrative world, rather than incorporating and dominating the intertextέ Aeneas’ inability in Book 2 to control Sinon, resulting in the Greek speaking at times with the authority of the omniscient narrator, is followed by a similar loss of control regarding Achaemenides in Book 3. In fact we can review the often noted correspondence between the two characters within the terms of the approach taken in this section.73 Both Greek strangers exploit the notion of multifarious possible truths and possible worlds. Both make us question who is narrating (Vergil / Aeneas / Sinon? Homer / Vergil / bothς) and subvert in different ways the ‘story within a story’ deviceμ Sinon by trespassing the borders that separate the concentric narrations, Achaemenides by making the device ‘two stories simultaneously’ν and in doing so both raise serious questions about the validity of the poem’s narrative frameέ Finally, both of these subversive elements deal with the underside of fictional narratives: Sinon is drawn to the hidden ‘truths’ and secrets that can be revealed (or created) by telling a narrative, while Achaemenides is himself born from a hidden, unnarrated recess of Homer’s narrative that Vergil brings to light. now and the Queen naming then. I would in be two realities at once, so in no reality at all. Vergil is forced into the crazy aporia of pretending to do the Homeric speech act that he imitates because he cannot frame the original act of creation within his secondary act of repetition. 72 Contrast Gordon Williams’ opinion that the highly poeticized description of Achaemenides, and the fellow-feeling implied by infelicis both argue for the voice of Vergil, not Aeneas. Radically, he attributes this to a change in plan, not fully executed, in which the third book had originally been written in thirdperson narrative. See Williams (1983) 262-4. 73 See Ramminger (1991) 53-9 for an overview. 138 The encounter with Achaemenides also brings a fitting conclusion to Vergil’s experiment with and rejection of the romance mode. It is the point at which the Aeneid and the world of romance heroism come into closest proximity, but this proximity is made possible only by a narrative device that is itself impossible and paradoxical: this underlines that there is no way in which the old myths of heroes and monsters can be at home in Vergil’s epicν they are always in some way ironized or problematizedέ Similarly the episode shows a move towards a novel form of epic thaumastia: the description of the huge Cyclopes amassing as the Trojans leave is a powerful and hyperbolical depiction of an anomaly, but equally important in surprising and confounding the reader are the artificial paradoxes engineered by Vergil’s fictionsέ 139