1 2 3 4 Richard Hunter 5 6 7 The Garland of Hippolytus 8 Abstract: This article discusses a set of remarkable scholia on the dedicatory ad- 9 dress and prayer which Hippolytus offers to Artemis as he places a garland at her 10 statue (Euripides, Hippolytus 73–87); the scholia consider a variety of allegorical 11 interpretations for the garland and for Hippolytus’ moral elitism. The article sets 12 these scholia within the context of the poetic interpretation of later criticism 13 and traces their roots in the language of classical poetry itself. The affiliations of Hippolytus’ language and why it attracted the notice of the scholiasts is 14 also explored, as is the way in which this scholiastic interest points us also to 15 a very important strand of the play’s meaning. 16 17 Keywords: allegorical interpretation, Euripides, Pindar, Orphics, scholia 18 19 One of the most celebrated Euripidean passages is the dedicatory address 20 and prayer which Hippolytus offers to Artemis as he places a garland at 21 her statue, immediately after the hymn which he and his fellow-hunts- 22 men have sung to her as they enter: 23 24 so· t|mde pkejt¹m st]vamom 1n !jgq\tou _ § 25 keil mor, d]spoima, josl^sar v]qy, 5mh’ oute poilµm !nio? v]qbeim bot± 75 26 out’ Gkh] py s_dgqor, !kk’ !j^qatom 27 l]kissa keil_m’ Aqimµ di]qwetai, 28 aQd½rd³ potal_aisi jgpe}ei dq|soir, 29 fsoir didajt¹m lgd³m !kk’ 1mt/i v}sei ¹ ? U 1 ± ! 30 t syvqome mekgwem rt p\mt’ e_,80 to}toir dq]peshai, to?r jajo?si d’ oq h]lir. 31 !kk’, § v_kg d]spoima, wqus]ar j|lgr 32 !m\dgla d]nai weiq¹reqseboOr %po. 33 l|myi c\q 1sti toOt’ 1lo· c]qar bqot_m· 34 so· ja· n}meili ja· k|coir !le_bolai,85 ³ q / e q b _ ¹ 35 jk}ym l ma d r, lla d’ o w q mt s|m. t]kor d³ j\lxail’ ¦speq Aqn\lgm b_ou. 36 37 38 I am grateful to Hans Bernsdorff and audiences in Cambridge, Frankfurt, Ioannina, 39 Leiden, Sydney, Thessaloniki and Washington D.C. for helpful discussion of earlier 40 versions.

Trends in Classics, vol. 1, pp. 18 – 35 DOI 10.1515/tcs.2009.003 © Walter de Gruyter 2009 The Garland of Hippolytus 19

1 Mistress, I bring you this woven garland which I have fashioned from an 2 unravaged meadow, where no herdsman chooses to graze his animals nor 3 has iron ever passed there, but in the springtime the bee traverses the un- ravaged meadow and Aids nurtures it with river waters; those who have 4 no share in the taught, but in their natures sphrosynÞ has its place in all 5 things for all time – these may pluck [from the meadow], but for the 6 wicked it is not permitted. Mistress of mine, receive from a pious hand a 7 wreath to bind your golden hair. Alone of men do I enjoy this privilege, 8 for I keep company with you and converse with you, hearing your voice, though I do not see your face. May I end my life as I have begun it. 9 Euripides, Hippolytus 73 –87 10 11 The extant scholia on these famous verses offer a compilation of detailed 12 and rather remarkable readings, extracts from which deserve to be quot- 1 13 ed at length: 14 15 Scholia on v. 73: ‘This is a notorious problem (zÞtÞma). Some suppose that 16 Hippolytus garlands Artemis with a garland of flowers, but others suppose that Hippolytus is saying this about himself, namely “Goddess, I dedicate 17 myself as a garland to you”, that is as the most blooming ornament (kosmos), 18 for it is an ornament to the virgin to pass time with the most sphrn of the 19 young men. Others say that the poet is not riddling (aQm_tteshai) or allego- 20 rising at all, but using words in their straightforward sense (juq_yr k]ceim) 21 and Hippolytus is in fact carrying a garland which he derived from a mead- ow in which it is not holy (fsiom) for us to pluck flowers. ‘Iron has never 22 entered it’ (v. 76) indicates that the meadow has never been cropped or 23 worked by anyone. Others say that Euripides metaphorically (tqopi- 24 j~teqom) calls the hymn to Artemis a garland, for it would be remarkably 25 strange to imagine that there was a flowery meadow where flowers were 26 picked and it was of such a kind that those who entered were examined 27 as to whether their sphrosunÞ was taught or naturally acquired and the meadow was irrigated by aids. Like a philosopher he says that he is bring- 28 ing a woven garland to the statue, a hymn to the god. ‘From an unravaged 29 [meadow]’ means ‘from my mind (di\moia) which lacks deceit and corrup- 30 tion’. 31 Alternatively: Poets quite reasonably liken their own natures to meadows or 32 rivers or bees, and their poetry to garlands: the flowers indicate the variety 33 and beauty of poetry, the rivers its mass and the impetus (bql^) to creation, 34 the bees its sweetness, and the garlands the honour (kosmos) of the subjects 35 of song. The poet has combined all of these things and thus made the na- ture of his allegory more brilliant (1va_dqume). ‘From an unravaged mead- 36 ow’ indicates that someone who is to practise mousikÞ must have a soul 37 which is pure and unravaged, unstained by any evil, and most of all partakes 38 39 1 I generally follow Schwartz’s text, though more work clearly needs to be done 40 on the text of the scholia. 20 Richard Hunter

1 of aids. It is because of the importance of aids that they represent the 2 Muses, who are most fertile (comil~tatai), as virgins. 3 Alternatively: … He calls the hymn a woven garland because they compose 4 hymns by putting together words as in weaving. The unravaged meadow 5 from where the flowers are woven into the garland and where not even a shepherd thinks it proper to graze his animals is an allegory for a virginal 6 and undeceitful intention (5mmoia). The flowers of this meadow are the re- 7 sults of wisdom and virtue. No iron has come to cut this meadow and crop 8 its flowers; by ‘iron’ he means either evil meddlesomeness (vikopqaclom_a) 9 and wrong-doing or the corruption of shameful pleasures, and in this way 10 he makes clear Hippolytus’ virginal and guileless character. The bee, how- 11 ever, is an allegory of the soul itself, for the bee is the purest of creatures (whence poets call priestesses ‘bees’). He calls it2 ‘of the springtime’ either 12 because bees rejoice in the spring because of the flowers or because pure 13 souls are always blooming, and spring is when flowers are produced. 14 Scholia on v. 78: This cannot be understood if one wants to understand it 15 literally (juq_yr) as being about gardens. Therefore there is an allegory 16 here. Poets reasonably liken their own natures to bees and rivers and mead- 17 ows, and poetry itself to garlands; the flowers indicate the variety and beau- 18 ty of poetry, the rivers its mass and the impetus to creation, the bees the ¹1 3 19 labour (t pilek]r) and concentrated effort involved, as well as the sweet- ness of the poems, and the garlands indicate that those who are praised win 20 glory through them. Euripides has combined all of these things and thus 21 made more brilliant the allegory through which he wished to describe 22 his hymn to Artemis; other poets use these devices (tq|poi) in a scattered 23 fashion. Plucking from unravaged meadows indicates that a poetic soul 24 must be pure and unravaged, and unstained by any evil. Those who are 25 going to practise poiÞtikÞ must most of all partake of aids. For this reason some call the Muses too virgins. 26 27 Scholia on v. 79: A quality which does not derive from nature, but is ach- 28 ieved by constant practice (lek]tg), is ‘learned’ (didajt|m). Philosophers call bad things ‘learned’ and good things ‘natural’ … 29 30 Although the whole of Hippolytus’ speech here eventually comes under 31 the scholiastic microscope, the ‘notorious problem’ is introduced as that 32 of the garland: is it a real or an allegorical garland? The scholars whose 33 work lies behind the scholia presumably knew, and may even have been 34 prompted to their interpretations by, the epithet Stevam_ar or 35 Stevamgv|qor which was attached to this Hippolytus by at least the 36 37 2 The reference is either to the bee or the meadow, depending on which reading 38 is adopted. 39 3 Reading t¹ sumtetal]mom for the transmitted t¹ sumtetacl]mom, cf. LSJ s.v. 40 sumte_my I2. The Garland of Hippolytus 21

1 time of Aristophanes of Byzantium.4 Our scholia on Euripides go back 2 ultimately to the work of Hellenistic scholars in Alexandria, most nota- 3 bly Aristophanes and Aristarchus,5 though we may find it hard to imag- 4 ine either of these figures behind the metaphorical readings of the scho- 5 lia. The very stark interpretative choice that the scholia offer between 6 ‘allegorical’ or ‘riddling’ readings on the one hand and ‘literal’ 7 (juq_yr) readings on the other is, of course, very familiar in ancient 8 criticism, and these scholia are an excellent illustration of one turn of 9 the scholiastic mind: interpretation begins from the question ‘What 10 do the verses say?’, and if the answer is ‘something which cannot be 11 meant literally’ (after all, aids is not ‘literally’ a gardener), then one 12 must seek other explanations in ‘troped’ language and ‘allegory’. This 13 latter term covers a very wide range of phenomena,6 and in this instance 14 we are dealing with a set of interpretations which largely appeal, not – as 15 do many ancient ‘allegorical’ readings – to a scheme of the order of the 16 cosmos, as for example does Porphyry in his famous discussion of Hom- 17 er’s ‘Cave of the Nymphs’,7 but rather more simply to a metaphorical 18 system which ancient readers tended to think of as inherent in the art 19 of poetry itself. Before turning to the question of how, if at all, these 20 scholia can help us to understand the Hippolytos,8 we should investigate 21 the intellectual affiliations of the scholia in rather greater detail. 22 The principal individual elements of the ‘troped’, poetological in- 23 terpretations (the poet as bee, the ‘garland’ of song, the meadow of 24 the Muses etc.) are very familiar and familiar from poetry well before 25 9 26 Euripides. Behind these scholia lies a very long tradition of high poetic 27 metaphors for song; Simonides is reported to have called Hesiod a gar- 28 dener and Homer a garland-weaver because the former ‘planted the 29 mythologies of gods and heroes’ and the latter ‘wove from them the gar- 30 31 32 4 Cf. Barrett 1964, 10 n.1. 33 5 Cf., e.g., Pfeiffer 1968, 222–4, Dickey 2007, 32. 34 6 Struck 2004, Pontani 2005, 26–40 and the contributions to Boys-Stones 2003 35 offer an excellent introduction to this subject. 36 7 Nauck 1886, 56–81; translation and discussion in Lamberton 1983. 37 8 Commentators on the play have (perhaps unsurprisingly) paid these scholia scant attention; unless I am mistaken, Barrett’s only reference to them (n. on 38 76–7) is to label ‘absurd’ the ‘allegorisation’ of the bee as really referring to 39 the soul. 40 9 Cf., e.g., Steiner 1986, 35–9, Nünlist 1998, 60–3, 206–23. 22 Richard Hunter

1 land of the Iliad and the Odyssey’ (Simonides T 47k Campbell).10 Un- 2 surprisingly, it is Pindar, whose victorious patrons receive both songs 3 and (literal) garlands, who supplies our richest source of such figures 4 (and this itself is a fact of some significance for the Hippolytus).11 5 When Pindar asks the eponymous nymph of Akragas to ‘receive this 6 garland from Pytho’ (Pythian 12.5), it is hard not to recall Hippolytus’ 7 prayer to Artemis. In Nemean 7 the song is a highly wrought and pre- 8 cious crown,12 9 10 eUqeim stev\mour 1kavq|m, !mab\keo· Lo?s\ toi 11 jokk÷i wqus¹m 5m te keuj¹m 1k]vamh’ "l÷ · % r ? 1 12 ja ke_qiom mhelom pomt_ar veko s’ ]qsar. 13 It is not difficult to weave garlands – strike up the prelude! The Muse binds 14 together gold and white ivory together with the lily flower she has re- 15 moved from the sea’s dew Pindar, Nemean 7.77 –9 16 17 and at Nemean 8.15 the song is a ‘Lydian headband embroidered with 18 resounding music’,13 where the scholia note that the poet is speaking ‘al- 19 legorically’. Nemean 3 offers a particularly elaborate ‘cocktail of song’: 20 21 1c½ t|de toi _ 22 p]lpy lelicl]mom l]ki keuj i s»m c\kajti jiqmal]ma d 5eqs !lv]pei 23 , ’ ’ , p|l’ !o_dilom AQok_ssim 1m pmoa?sim aqk_m jtk. 24 25 I send you this honey mingled with white milk, attended by the foam which has been stirred, a drink of song among the Aeolian breaths of 26 pipes … 27 Pindar, Nemean 3.76 –9 28 29 Here the scholia connect milk with the natural talent, the phusis, needed 30 for poetry and the honey with the p|mor of bees, and this is precisely the 31 realm of ideas in which the Euripidean scholia also move. 32 Poets freely used such images for their own work, but these were 33 also the very stuff of how poetry was explained, and this intimate link 34 with the imagery of poetry itself is fundamental for understanding the 35 36 10 The story obviously implies the chronological priority of Hesiod, but to what 37 extent it provides firm evidence for Simonides’ view of the matter may be de- bated. 38 11 Cf. below p. 28–9. 39 12 Cf. further below p. 29. 40 13 To the commentators add Kurke 1991, 190–1, Ford 2002, 117–18. The Garland of Hippolytus 23

1 language of ancient poetic criticism; it is telling that two of Quintilian’s 2 three Latin examples of ‘allegory through metaphor (allegoria continuatis 3 tralationibus)’ are poetological images from Lucretius and Virgil (Quintil- 4 ian 8.6.45). In the present case, however, what stands out is the Platonic 5 background of the Euripidean scholia. Like the scholia, many modern 6 critics have stressed the analogy between the ‘unravaged meadow’ and 7 Hippolytus’ virginal soul, but crucial here is a famous passage of Plato’s 8 Phaedrus which was very important for later, particularly of course neo- 9 Platonic, discussions of poetry:14 10 11 tq_tg d³!p¹ Lous_m jatojyw^ te ja· lam_a, kaboOsa "pakµmja·%batom 12 xuw^m, 1ce_qousa ja·1jbajwe}ousa jat\ te ¡id±rja· jat± tµm %kkgm 13 po_gsim, luq_a t_m pakai_m 5qca josloOsa to»r 1picicmol]mour d # % _ 1 · ± ! 14 paide}ei· rd’ m meu lam_ar Lous m p poigtij r h}qar v_jgtai, peishe·r ¢r %qa 1j t]wmgr Rjam¹r poigtµr 1s|lemor, !tekµraqt|r te ja·B 15 po_gsir rp¹ t/rt_m laimol]mym B toO syvqomoOmtor Avam_shg. 16 17 There is a third sort of possession and madness which comes from the Muses. It takes hold of a tender and untrodden soul, and by rousing it 18 and inducing a state of Bacchic possession in song and other forms of po- 19 etry, it educates future generations by celebrating the countless deeds of 20 men of old. But whoever comes to the doors of poetry without madness 21 from the Muses, in the belief that craft (technÞ) will make him a good 15 22 poet, both he and his poetry, the poetry of a sane man, will be incomplete 23 and eclipsed by the poetry of the mad. Plato, Phaedrus 245a 24 25 In his commentary on the Phaedrus, Proclus explains that the soul which 26 is to receive the divine inspiration of the Muses must be clear of all other 27 distracting influences and ideas, including (we may assume) over-subtle 28 intellectual calculations;16 we are here not far from the scholiastic ex- 29 planation that the rejected ‘iron’ of Hippolytus’ speech stands for ‘evil 30 meddlesomeness (vikopqaclom_a)’.17 Be that as it may, the soul 31 which, in Proclus’ words, is !pahµrja·%dejtor ja·!lic^r to every- 32 thing except the ‘breath of the divine’ (1.181.16–17 Kroll) is at least 33 34 14 It is intriguing to find part at least of this passage cited already in Satyrus’ Life of 35 Euripides (F 6 fr. 16 col. I Schorn), perhaps in a contrast between Euripides and 36 truly ‘inspired’ poetry (? Aeschylus), cf. Schorn 2004, 193–4. !tek^r 37 15 Commentators rightly note that both means ‘uncompleted’ and also suggests ‘uninitiated’. 38 16 Commentary on Plato’s Republic 1.181.2 –17 Kroll. 39 17 For scholarly and intellectual ‘meddlesomeness’ cf. Hunter forthcoming (b), 40 Struck 2004, 72. 24 Richard Hunter

1 how Hippolytus sees himself, even if, of course, his Artemis is much 2 more associated with sphrosynÞ than with mania; the language of poetic 3 inspiration and the language of mystical religious devotion are here, as 4 so often, very close. 5 Plato’s %bator ‘untrodden’ for a young man’s soul is a word, like 6 Hippolytus’ !j^qator, which can have sacral resonance – it is used 7 for a holy place (such as a meadow) which may not be entered except 8 under special circumstances; whereas, however, the sexual resonance of 9 !j^qator and related words is very well attested, %bator in the sense 10 ‘(sexually) unmounted’ is only found in a humorous context in Lucian 11 (Lexiphanes 19). Nevertheless, it is easy enough to see how any reader 12 would feel this resonance in the Platonic passage, particularly when 13 %bator is put together with "pak|r (and particularly in the context of 14 the 1qytij¹r k|cor of the Phaedrus as a whole), and here perhaps lies 15 part of the origin of the scholiastic stress upon the purity of soul needed 16 by those who wish to practise mousikÞ or poiÞtikÞ. So too, although Hip- 17 polytus uses josl^sar (v. 74) in the sense ‘arranging, putting together 18 [i.e. the garland]’, it is clear that the scholia felt that the word contrib- 19 uted importantly to the ‘metaphorical’ sense of the passage, and the ex- 20 planation that poets compare their poems to garlands to indicate ‘the 21 honour (kosmos) of the subjects of song’ (p. 13. 23 Schwartz) picks up 22 Plato’s claim that possession from the Muses ‘celebrates (josloOsa) 23 the countless deeds of the ancients’ (Phaedrus 245a4). 24 In choosing %bator Plato was also, as often, imitating in language 25 the subject of his discourse. ‘Untrodden’ to describe a soul is, to put 26 it simply, the kind of ‘metaphor’ which one might expect to find in po- 27 etry;18 in describing the possession which comes from the Muses, Soc- 28 rates speaks like one possessed in just that way. The most significant 29 analogy here, as for the passage of the Phaedrus itself, is Socrates’ famous 30 account in the Ion of poetic inspiration and of why poetry is precisely 31 the result of ecstatic inspiration rather than technÞ (533c8 –535a2). 32 Here too poets are like bees and the language imitates their alleged 33 34 ‘flights’: 35 Poets tell us that, like bees (l]kittai), they are bringing us songs (l]kg) 36 which they have gathered (dqep|lemoi) from springs flowing with honey 37 (lekiqq}tym) in gardens and groves of the Muses, and they do this in flight. 38 39 18 The discussion of Plato’s style at Dion. Hal. Dem. 5–7 is obviously relevant 40 here. The Garland of Hippolytus 25

1 They speak the truth: for a poet is a light and winged and holy thing, and 2 unable to compose before the god is inside him and he becomes out of his 3 senses and his mind no longer resides in him. Plato, Ion 534a7 –6 4 5 It is precisely poetic imagery and metaphor, of a kind very close to Hip- 6 polytus’ imagery, which ‘proves’ the irrational nature of poetic compo- 7 sition.19 Aristotle more than once stressed that ‘metaphor’ was the most 8 important aspect of poetic language and that making metaphors was a 9 natural gift: 10 11 It is important to use each of the elements I have mentioned appropriately, 12 including double nouns and glosses, but by far the most important aspect of diction is the metaphorical. This is the only aspect which cannot be ac- 13 quired from another and it is a sign of natural gifts (eqvu_a), for to make 14 good metaphors is to observe similarity. 15 Aristotle, Poetics 1459a4 –820 16 Although this is not the same point as Plato’s insistence that poetry is the 17 result of inspiration, not technÞ, they could clearly be seen to stand in the 18 19 same tradition, particularly as it is metaphorical language which Plato 20 uses in the Ion to illustrate the irrational nature of poetic composition. 21 From the perspective of this later tradition, Hippolytus’ highly meta- 22 phorical address to Artemis would illustrate the very lesson he teaches, 23 namely the primacy of phusis over ‘taught qualities’, for only someone eqvu_a 24 with a very special , who does not in any sense rely on what 25 he has ‘learned’, could ‘make metaphor’ like this. We will see that Euri- 26 pides certainly had other reasons as well for making Hippolytus speak 21 27 like this, but it is perhaps not utterly idle to wonder whether the tra- 28 dition of reflection upon the nature of poetic metaphor which we have 29 found in Plato and Aristotle had roots already in fifth-century discussion 30 of poetry and is reflected in Hippolytus’ opening speech. 31 Before proceeding, it may be as well to cast a quick glance at the po- 32 etological ideas themselves which the scholia display. Many are, as we 33 have noted, very familiar and not to be traced to any particular intellectual 34 tradition, but we may suspect that much can again be traced back to Pla- 35 to’s Ion. The notion that poets and poetry are likened to rivers because of 36 37 19 We might compare the comparison of poetic inspiration to the workings of a magnet (Ion 533d-e) which some have seen as (pointedly) an adoption of the 38 mode of epic simile. 39 20 Cf. Rhetoric 3.1405a8 –10. 40 21 Cf. below p. 33. 26 Richard Hunter

1 ‘mass (pkgh}r) and the impetus (bql^) to creation’ might seem, on one 2 hand, to pick up Socrates’ claim that poets and rhapsodes only perform 3 in that one ‘genre’ ‘towards which the Muse impels (¦qlgsem)them’ 4 (Ion 534c1–2).22 On the other hand, however, the reference to ‘mass’, 5 suggestive of epic grandeur or the raging and swollen mountain torrent 6 which is Horace’s vision of Pindar (Odes 4.2),23 might seem untrue to 7 the apparent exclusivity of the pot\liai dq|soi which water Hippolytus’ 8 garden of Aids, a source perhaps more Callimachean than epic.24 The 9 scholia are, of course, nothing if not eclectic. If rivers denote the rushing 10 power of poetry, the bee indicates, as it does for Horace in the same 11 poem, the labor plurimus involved in making operosa … carmina;25 in the 12 scholiast’s t¹1pilek³rja· t¹ sumtetal]mom we are not far from ‘Callima- 13 chean’ ideals, and Aratus’ s}mtomor !cqupm_g (Callimachus, Epigram 27.4 14 Pfeiffer), if that is the right reading, may particularly come to mind. 15 As compilations, the scholia are of course less concerned with a con- 16 sistent poetic ‘program’ than with the very overload of poetological im- 17 agery which they find in the Euripidean verses. For us that imagery 18 looks both forward and back. Callimachus’ famous image for his 19 poem at the end of the Hymn to Apollo shares more than one element 20 with Hippolytus’ remarkable prayer: 21 22 Dgo? d’ oqj !p¹ pamt¹r vdyq voq]ousi l]kissai, ! F ·! ! 23 kk’ tir jahaq^ te ja wq\amtor m]qpei p_dajor 1n Req/r ak_cg kib±r %jqom %ytom. 24 25 To Deo the bees do not carry water from every source, but only from that which rises up pure and untainted, a tiny trickle from a holy spring, the 26 height of perfection. 27 Callimachus, Hymn to Apollo 110–12 28 29 30 31 22 Murray 1996, 119 notes that Plato’s expression here picks up the Homeric bqlghe·rheoO 32 (Odyssey 8. 499 of Demodocus, where, though many modern ed- itors take a different view, the scholia note the bql^ from the god); pace Mur- 33 ray, however, it is far from clear that Proclus, Commentary on Plato’s Republic 34 1.184.27 –8 Kroll, who notes Homeric influence on Plato here, is actually 35 thinking of this passage of the Odyssey. 36 23 On this imagery cf., e. g., Hunter 2003, 220–3. Somewhere behind the scho- pkgh»mdû oqj #m 1c½ luh^solai jtk 37 liast’s language is probably Iliad 2.488 . 24 Cf. below. 38 25 The etymological play in the scholia on l]kissa and 1pilek]r is not, to my 39 knowledge, found elsewhere, though it might be thought that the Horatian 40 passage implies it. The Garland of Hippolytus 27

1 The bee image (the Euripidean scholia note the usage of ‘bee’ as ‘priest- 2 ess’ which must be part of Callimachus’ image)26 and the stress on a sa- 3 cral purity and exclusivity strongly recall Hippolytus’ attitudes, the met- 4 aphorical language in which he expresses them, and the explanations of 5 the scholia.27 Modern criticism has tended to write ‘religion’ out of Cal- 6 limachus’ poetry, with the result that his sacral language is seen as ‘purely 7 literary’, but Hippolytus’ prayer should make us pause. If the scholia 8 offer as one interpretation that Hippolytus’ garland is in fact the song 9 in the goddess’ honour, the Callimachean Hymn to Apollo is indeed an 10 offering to the god, and one which we know that he accepts.28 Callima- 11 chus draws the sacral boundaries in much the same terms as does Hip- 12 polytus: 13 14 ¢p|kkym oq pamt· vae_metai, !kk’ ftir 1shk|r· 15 fr lim Udgi, l]car oxtor, droqj Ude, kit¹r 1je?mor. 16 ax|leh’, §:j\eqce, ja·1ss|leh’ oupote kito_. 17 Apollo does not appear to everyone, but to he who is good; he who sees 18 him, this man is great, he who does not see him, that man is of no value; 19 we shall see [you], Far-Worker, and we shall never be of no value. 20 Callimachus, Hymn to Apollo 9–11 21 If Hippolytus knows that he will never actually ‘see’ his goddess (cf. 22 vv. 85–6, 1391–6), it is nevertheless the jajo_ – those whom Callima- 23 chus would call the !kitqo_ (v. 2), the oqj 1shko_ (cf. v.9), and the kito_ 24 (vv. 10–11) – who may not enter the meadow.29 If viewed through a 25 Callimachean lens, the ‘metaphorical’ interpretation of Hippolytus’ 26 speech which we find in the scholia becomes, if not necessarily easier 27 to accept, at least firmly contextualised. As we have seen, there are im- 28 portant differences between the various elements of the pattern. Where- 29 as Hippolytus, like Pindar before him (cf., e.g., Olympian 9.100–4), re- 30 jects ‘the taught’ in favour of natural gifts,30 the scholia seem to ac- 31 32 26 For discussion cf. Williams’ note on v. 110; of particular importance is Supple- 33 mentum Hellenisticum 990.2. 34 27 jahaq|r is a particularly good example of the seepage between sacral and critical 35 language, cf., e.g., ‘Longinus’, On the Sublime 33.2. 36 28 For these ideas in Hellenistic and Roman poetry cf. Hunter 2006, 14–15. 37 29 On a second century AD inscription from Attica members of a club are to be tested to see eU1sti "cm¹rja· eqsebµrja·!cah|r (Sokolowski 1969, no. 53, line 38 33). There is a helpful discussion of the mystical aspect of Hippolytus’ language 39 in Asper 1997, 51–3. 40 30 Barrett calls the idea ‘a commonplace of old aristocratic thought’ (1964, 173). 28 Richard Hunter

1 knowledge both as important poetic ideas; if Callimachus does not ex- 2 plicitly (but cf. vv. 42–6) stress technÞ in the Hymn to Apollo and the 3 image of the pure spring would seem to foreground the gifts of divine 4 nature, nevertheless, his emphasis on this elsewhere is well known, and 5 it can be argued that the ‘Reply to the Telchines’ precisely lays claim to 6 both technÞ and the divine inspiration of the Ion.31 7 Hippolytus’ mode of speech looks back also. As we have seen, some 8 of the closest parallels are to be found in Pindar, and it is in Pindar too 9 where the sharpest lines are drawn between those who can and cannot 10 understand. A famous passage of Pindar’s Second Olympian asserts the 11 special nature of what Pindar has to say: 12 13 pokk\ loi rp’ 14 !cj_mor ¡j]a b]kg 5 1 · 15 mdom mt vaq]tqar vym\emta sumeto?sim· 1rd³ t¹ p±m 2qlam]ym 16 wat_fei. sov¹r b pokk± eQd½rvu÷i· 17 lah|mter d³ k\bqoi 18 pacckyss_ô j|qajer ¤r %jqamta caqu]tym 19 Di¹rpq¹r eqmiwa he?om· 20 I have under my arm many swift arrows inside their quiver which speak to 21 those who understand; in general, however, they require interpreters. Wise 22 is the man who naturally knows many things. Those who have learned are 23 unruly and their words spill out; they are like a pair of crows who caw in 24 vain against the divine bird of Zeus. 25 Pindar, Olympian 2.82 –9 26 Eustathius took this passage as programmatic of Pindar’s poetry as a 27 whole, and to ancient scholars (at least from Aristarchus on),32 confronted 28 – as in Euripides’ Hippolytus – with a passage where a ‘non-allegorical’ 29 reading was simply not possible (Pindar does not ‘literally’ have arrows 30 and a quiver, any more than Aids is a market-gardener), where there 31 is an explicit contrast between ‘the wise man who knows much by na- 32 ture’ and ‘those who have learned’, and which followed directly on a pas- 33 sage of apparently mystical eschatology, it was clear that Pindar was assert- 34 ing that his difficult poems required ‘interpreters’ (i.e. commentators) for 35 ordinary people (‘common folk’, ‘the non-specialists’, ‘the many’). It was 36 then entirely ‘natural’ to see the crows and the eagle as ‘riddling’ referen- 37 38 31 Cf. Hunter 1989. 39 32 Aristarchus is cited by the scholia on v. 85 (Drachmann p. 98). For Eustathius 40 cf. Drachmann III.287.1–8. The Garland of Hippolytus 29

1 ces to (respectively) Simonides and Bacchylides and Pindar himself. The 2 text itself seemed to direct the scholars to read ‘riddlingly’. Such dichot- 3 omies in the potential audience either originally arose in or were con- 4 firmed by sacral or mystical contexts; Hippolytus’ exclusivity suggests 5 this, and indeed any claim to purity implies a group of the ‘impure’, as 6 we can see, for example, on the gold leaves of the Underworld.33 7 These dichotomies soon found their way, however, into the exegesis 8 of texts, particularly, though not exclusively, what we might call ‘allego- 9 rical’ exegesis, for such interpretation inevitably constructs a dual reader- 10 ship – the ‘few’, the ‘wise’, the ‘initiated’ on the one hand, and ‘the 11 many’, ‘the vulgar’, the ‘uninitiated’ on the other.34 The process is now 12 most familiar from the Derveni papyrus, where the commentator seems 13 to distinguish (the text is unfortunately broken) between ‘the many’ 14 and ‘those pure (?) of hearing’ (col. VII.10–11); given the nature of 15 that text, the process of transition from ‘religious’ to ‘literary’ exegesis 16 is here starkly exposed. ‘Metaphorical’ and ‘riddling’ language creates 17 boundaries and displays them openly. On the elaborate ‘crown of song’ 18 at Nemean 7.77–9, Andrew Ford comments that this image ‘is a form 19 of kenning … but it is also a form of knowing, a mode of addressing 20 the sophoi’;35 just the same could be said of Hippolytus’ images. 21 Another, though closely related, distinction drawn within archaic 22 poetry is also relevant here. The language of the !cah|r and the 23 jaj|r, and indeed of syvqos}mg, is of course most familiar from the 24 socio-political world of sympotic elegy. Theognis describes a world 25 turned upside down: 26 27 mOmd³ t± t_m !cah_m jaj± c_metai 1shk± jajo?sim 28 !mdq_m· Bc]omtai d’ 1jtqap]koisi m|loir· 29 aQd½rl³mc±q ekykem, !maide_g d³ ja·vbqir 30 mij^sasa d_jgm c/m jat± p÷sam 5wei. 31 But now good men’s evils have become virtues for the base; they rejoice in 32 customs turned upside down. Aids has perished, and shamelessness and 33 outrage have defeated justice and hold sway over the whole land. 34 Theognis 289–92 35 36 33 Cf., e.g., texts 5–7 and 9 in Graf-Johnston 2007. 37 34 Through Philodemus we can see traces of these dichotomies in Hellenistic lit- erary criticism, cf. Fantuzzi-Hunter 2004, 452. [Plutarch], De Homero 92 also 38 explicitly refers to the two classes of Homeric audience, the vikolahoOmter 39 and the !lahe?r, cf. Pontani 2005, 32–3. 40 35 Ford 2002, 123. 30 Richard Hunter

1 aQd~r is as much a catch-word for the self-appointed !caho_ in Theog- 2 nis’ world of aristocratic power and values, as it is in Hippolytus’ dom- 3 inating sense of self; elsewhere the same point is made explicitly: 4 5 !mdq\si to?s’ !caho?s’ 6petai cm~lg te ja· aQd~r· T O 1 ? ! a 6 o m m m pokko r tqej]yr k_coi. 7 Judgement and aids attend the good; now they are really few among 8 many. 9 Theognis 635–6 10 In another well known passage which concludes one of the fullest early 11 examples of the ‘ship of state’ allegory, the now familiar language of the 12 !cah|r (or the 1shk|r) and the jaj|r is combined with an appeal to the 13 ‘decoding’ of poetic imagery:36 14 15 voqtgco· d’ %qwousi, jajo· d’ !cah_m jah}peqhem. 16 deila_my, l^ pyr maOm jat± jOla p_gi. 17 taOt\ loi Aim_why jejqull]ma to?s’ !caho?sim· cim~sjoi d %m tir ja· jaj|r #m sov¹r Gi 18 ’ , . 19 The cargo-carriers are in charge, and the base are above the good; I am 20 afraid that a wave will swallow up the ship. Let these be my veiled riddles for the good; even a base man, if he is wise, would know the meaning. 21 Theognis 679–82 22 23 If we then ask about the resonances of Hippolytus’ extraordinary im- 24 agery for an Athenian audience in the late fifth century, there will of 25 course be more than one answer, but prominent among them will be 26 not just the sacral, but also the world of the aristocratic, perhaps now 27 ‘old-fashioned’, symposium and the poetry which accompanied it; 28 this is one of the important truths to which the neglected scholia di- 29 rect us. When at the start of the Homeric Problems ‘Heraclitus’ illus- 30 trates what allÞgoriaACHTUNGRE is, the three examples he chooses are now famous 31 instances of archaic poetry – Archilochus (fr. 105 W) and Alcaeus 32 (frr.208,6V)describingstorms,whichare‘infact’warandinternal 33 strife, and Anacreon (PMG 417) addressing a Thracian filly, who is 34 really a lovely girl. The setting for all such poems was very probably 35 a male gathering such as the symposium, i.e. a closed ‘reception con- 36 text’, a gathering of ‘those who know’, and one in which, as we have 37 seen, both ‘coded’ modes of speech, such as the riddle and the eikn, 38 and (at least during the later fifth century) what we now call ‘poetic 39 40 36 Cf. further Ford 2002, 75–6, Hunter forthcoming (a). The Garland of Hippolytus 31

1 criticism’ flourished. The gradual disappearance of this style of fig- 2 ured speech is a major issue of literary history; how archaic this 3 style was already felt to be as Hippolytus spoke is a question to 4 which the scholia direct us. 5 Hippolytus’ prayer takes us, of course, in other (related) interpre- 6 tative directions as well. We may wish (rightly) to set this speech with- 7 in an epistemological pattern whereby the three central characters of 8 the tragedy are each characterised by a different form of knowledge 9 which orders (but eventually undermines) their world: Phaedra, par- 10 ticularlyofcourseinhergreatspeechtothechorusat373ff.,by 11 moral reflectiveness leading to clear ethical principles, Theseus by a 12 straightforward reliance upon perception and inherited values, and 13 Hippolytus by a ‘revealed’ truth and certain sense of self; the very 14 way he speaks shuts out ‘the many’. So too, critics have long discussed 15 the battle for control of language in this play, for example for control 16 of the meaning of syvqos}mg or Phaedra’s struggle with the semantic 17 range of aQd~r;37 Hippolytus’ speech, with its claims to the control of 18 metaphor and by its juxtaposition to his exchange with the servant, 19 who insists upon a kind of aqho]peia while also revealing the traps lan- 20 guage sets for us (the ambiguity of selm|r etc.), introduces this theme 21 to powerful effect. 22 The division of the world into ‘those who understand’ and ‘those 23 who do not’ which is implied in Hippolytus’ prayer and which, as we 24 25 have seen, is a prominent feature not just of forms of religious worship 26 but also of the world of archaic poetry and its exegesis, resurfaces, as do 27 Hippolytus’ claims to syvqos}mg (vv. 995, 1007, 1013, 1034–5) and Q 28 a d~r (v. 998), in the speech of self-defence which he makes to his fa- 29 ther. He begins with a very striking proemium: 30 1c½ d’ %jolxor eQr ewkom doOmai k|com, 31 1r Fkijar d³ j¡k_cour sov~teqor· 32 5wei d³ lo?qam ja· t|d’· oR c±q 1m sovo?r 33 vaOkoi paq’ ewkyi lousij~teqoi k]ceim. 34 flyr d’ !m\cjg, nulvoq÷r !vicl]mgr, 990 _ ! ? 35 ck ss\m l’ ve mai. 36 I am not clever at speaking to the rabble, but more skilled before my equals 37 and a small audience. This is only reasonable. Those who fail before the 38 39 40 37 Cf., e.g., Goldhill 1986, 132–7, Gill 1990. 32 Richard Hunter

1 wise have more success with speaking in front of the rabble. Nevertheless, 2 in this present misfortune, I must let loose my tongue. 3 Euripides, Hippolytus 986–91 4 This ‘tactless … contempt for his audience’ (Barrett) might seem a truly 5 remarkable form of the ‘unaccustomed as I am’ topos, but much is at 6 stake here. Barrett notes that Hippolytus’ reference to the ochlos is ‘espe- 7 cially tactless since although there is of course a crowd gathered round 8 … it is only to Theseus that his arguments are addressed’, but we may 9 wonder if this is not one of those places in tragedy where the audience 10 may well feel itself involved, if not specifically addressed; ochlos is (un- 11 surprisingly) one of the terms for the audience used in the famous ac- 12 count of Athenian theatrical history offered by Plato, yet another élitist 13 (Laws 3.700a-1b). It is the fact of public ‘performance’, as well as the 14 ignorance of the broad audience, which Hippolytus rejects. Words mat- 15 ter to him, ‘letting loose the tongue’ (991) is not a mode which he fa- 16 vours; as we know from the violence of his reaction to the Nurse’s at- 17 tempt to win him over (653–5), even hearing words of a morally cor- 18 rupt kind threatens to make him jaj|r and stains his purity ("cme}eim)so 19 that he will need to wash out his ears with ‘water from running streams’. 20 Here it is now very hard not to remember (again) the distinction in the 21 Derveni commentary between ‘the many’ and ‘those pure (?) of hear- 22 ing’ (col. VII.10–11), in a chapter precisely about the exegesis of an ‘al- 23 legorical’ text; Tsantsanaglou’s tµ]m !joµm ["cme}o]mtar is there almost 24 universally accepted. 25 As the rejected crows of Pindar’s Second Olympian are indiscriminate 26 in their choice of language (k\bqoi pacckyss_ai), so the exercise of 27 linguistic choice is the activity of the sophos. The Plutarchan treatise 28 ‘On the Education of Children’ quotes Hippolytus 986–9 in support 29 of the need to expose children only to the right kind of education, 30 and the distinctions which Plutarch draws make the passage worth quot- 31 ing at length: 32 33 I say again that parents must cling to the uncorrupted and healthy education 34 and must take their sons as far away as possible from the rubbish of public 35 speeches (t_m pamgcuqij_m k^qym), for to give pleasure to the many (oR R 36 pokko_) is to displease the wise (o sovo_). Euripides supports this … Hippo- 37 lytus 986–9 … I see that those whose practice it is to speak in a manner which pleases and wins favour with the vulgar rabble (to?r suqvet~desim 38 ewkoir) turn out generally to be dissolute in their lifestyle and fond of pleas- 39 ure. This is just what we would expect. If as they provide pleasure for oth- 40 ers they neglect what is honourable (toO jakoO), they would be slow in- The Garland of Hippolytus 33

1 deed to place what is morally correct and healthy (t¹aqh¹mja·rci]r) above ¹ _ 2 the pursuit of their own luxurious pleasures or what is modest (t s vqom) toO teqpmoO 3 above the delightful ( ). Plutarch, On the education of children 6a–c 4 5 The context here is quite different from that of the Hippolytus, but Plu- 6 tarch too is the spokesman for a self-appointed élite, the sovo_, whose 7 authority depends upon a shared body of knowledge (paideia) which ex- 8 cludes the ‘uninitiated’; like Hippolytus, Plutarch equates verbal excess 9 and facility with a morally impure life and an absence of sphrosunÞ. 10 From the outside such claims, whether those of a Hippolytus or of a 11 Plutarch, are always open to charges of hypocrisy (as, for example, Lu- 12 cian knew only too well). Thus Theseus famously throws in Hippolytus’ 13 face the charge of hypocritical allegiance to ‘Orphic’ behaviour: 14 15 Edg mum auwei ja· di’ !x}wou boq÷r 16 s_toir jap^keu’ iqv]a t’ %majt’ 5wym _ _ 17 b\jweue pokk m cqall\tym til m japmo}r· 1pe_ c’ 1k^vhgr. 18 19 Now hold your high opinions and with your lifeless food make a show of 20 your diet; with Orpheus as your leader revel on and honour writings, in- substantial as smoke. You have been found out! 21 Euripides, Hippolytus 952–5 22 23 What is important here is not whether or not Hippolytus was really an 24 Orphic, but rather the familiar and much commented upon phenomen- 25 on of the association of ‘Orphics’ with ‘books’ (cf. Plato, Rep. 2.364e); 26 here, if anywhere, were Greeks with ‘sacred books’ to be honoured (v. 38 27 954) and, as the Derveni papyrus has shown us, interpreted. Such 28 books offered a kind of knowledge not (to be) widely available and 29 one which both seemed to invite, and may perhaps have exploited, al- 30 lÞgoria.ACHTUNGRE Texts which are intended for and/or taken up as privileged by 31 particular groups are always fertile ground for ‘metaphorical’ or ‘allego- 32 rical’ reading, for this is precisely one of the ways in which the special- 33 ness of the text is preserved. In principle, of course, this may also apply 34 to oral ‘texts’, as we see not just in pre- or partially literate societies, but 35 in, say, the ‘secret knowledge’ of closed societies (fraternities, Masons 36 etc.) in highly literate contexts. Committing knowledge to writing 37 risks its promulgation among the ‘profane’, and if this must be done, 38 the knowledge must therefore be ‘encoded’ in such a way that it is of 39 40 38 Cf. Henrichs 2003 for a discussion of the general phenomenon. 34 Richard Hunter

1 no use if it falls into the wrong hands; metaphor and ‘allegory’ are forms 2 of literary code. In antiquity the idea of religious ‘mysteries’ is never far 3 away in this context: the Platonic Socrates seems to link ‘allegory’ with 4 eschatological rites (Phaedo 69c), the critic Demetrius tells us that ‘the 5 mysteries are conducted through allegory to increase their power to in- 6 stil amazement and terror … for allegorical language is like darkness and 7 night’ (On Style 101), and the Hippocratic ‘Law’ concludes by noting 8 that the holy facts of medicine are to be revealed only to those who 9 have been initiated through knowledge (CMG I.i.8).39 10 The metaphorical and mystical mode of Hippolytus’ opening speech 11 may thus be seen to prepare us for his terrible fate. Theseus’ angry words 12 – ironically placed in the mouth of a ‘tyrant’ – reflect the ‘democratic’ 13 suspicion that those who hide things in books have something (perhaps 14 risible) to hide; metaphorical language may, by its very nature, seem an- 15 tithetical to the proclaimed transparency of democratic principles.40 16 Hippolytus, who of course surrounded himself with %qistoi v_koi (v. 17 1018), is thus damned both ways: on the one hand, his language and 18 behaviour suggest the closed circle of the aristocratic symposium, pre- 19 dominantly of course an oral culture, and on the other he can be assimi- 20 lated to suspect sects who claimed to find revealed truth in writings. 21 Both frames testify to the very singularity of this character and the strug- 22 gle to find the appropriate categories for him. That singularity was strik- 23 ingly signalled by his opening dedicatory address to Artemis which in- 24 vites ‘interpretation’, whether we call this ‘allegorical’ or prefer (with 25 Barrett) to speak of ‘transparent symbolism’.41 It is this invitation to in- 26 terpretation upon which the ancient scholiasts focused, and so should 27 we. 28 The scholiasts sought to understand the intellectual structure which 29 lay behind Hippolytus’ words; we may not wish to follow the path they 30 trod, but their curiosity is something we should ponder hard before 31 going our own way. If there is a continuing tradition of criticism 32 from antiquity to the present, then it is one of debate and struggle, 33 and the authors of many of our scholia knew that the texts of the past 34 mattered and were worth struggling over. Old trends are often the best. 35 36 37 38 39 On allegory and the Mysteries cf., e.g., Pontani 2005, 34–6. 39 40 Cf. the remarks of Ford 2002, 87. 40 41 Barrett 1964, 172. The Garland of Hippolytus 35

1 Bibliography 2 3 Asper, M. (1997), Onomata allotria. Zur Genese, Struktur und Funktion poetolo- 4 gischer Metaphern bei Kallimachos, Stuttgart. Barrett, W.S. (1964), Euripides, Hippolytos, Oxford. 5 Boys-Stones, G.R. (ed.) (2003), Metaphor, Allegory and the Classical Tradition, 6 Oxford. 7 Dickey, E. (2007), Ancient Greek Scholarship, Oxford. 8 Fantuzzi, M. – Hunter, R. (2004), Tradition and Innovation in Hellenistic Poetry, 9 Cambridge. Ford, A. (2002), The Origins of Criticism, Princeton. 10 Gill, C. (1990), ‘The articulation of the self in Euripides’ Hippolytus’ in: A. Po- 11 well (ed.), Euripides, Women, and Sexuality (London/New York) 76 –107. 12 Goldhill, S. (1986), Reading Greek Tragedy, Cambridge. 13 Graf, F. – Johnston, S.I. (2007), Ritual Texts for the Afterlife, London. 14 Henrichs, A. (2003), ‘Hieroi logoi and hierai bibloi: the (un)written margins of the sacred in ancient Greece’, in: Harvard Studies in Classical Philology 101, 15 207–66. 16 Hunter, R. (1989), ‘Winged Callimachus’, in: Zeitschrift fr Papyrologie und Ep- 17 igraphik 76, 1–2. [= Hunter 2008, 1.86 –8] 18 – (2003), ‘Reflecting on writing and culture. Theocritus and the style of cul- 19 tural change’, in: H. Yunis (ed.), Written Texts and the Rise of Literate Cul- ture in Ancient Greece, Cambridge, 213–34. [=Hunter 2008, 1.434–56] 20 – (2006), The Shadow of Callimachus, Cambridge. 21 – (2008), On Coming After. Studies in Post-Classical Greek Literature and its Re- 22 ception, Berlin/New York. 23 – forthcoming (a), ‘Language and Interpretation in Greek Epigram’, in: M. 24 Baumbach, A. Petrovic and I. Petrovic (eds.), Archaic and Classical Greek Epigram. Contextualisation and Literarisation, Cambridge. 25 – forthcoming (b), ‘The curious incident …: polypragmosyne and the ancient 26 novel’, in: S.J. Harrison et al. (eds.), Readers and Writers in the Ancient Novel, 27 Groningen. [= Hunter 2008, 2.884–96] 28 Kurke, L. (1991), The Traffic in Praise, Ithaca/London. 29 Lamberton, R. (1983), Porphyry. On the Cave of the Nymphs, Barrytown, NY. Murray, P. (1996), Plato on Poetry, Cambridge. 30 Nauck, A. (1886), Porphyrii Philosophi Platonici Opuscula Selecta, Leipzig. 31 Nünlist, R. (1998), Poetologische Bildersprache in der frhgriechischen Dichtung, 32 Stuttgart/Leipzig. 33 Pfeiffer, R. (1968), History of Classical Scholarship, Oxford. 34 Pontani, F. (2005), Eraclito. Questioni omeriche sulle allegorie di Omero in merito agli di, Pisa. 35 Schorn, S. (2004), Satyrus aus Kallatis, Basel. 36 Sokolowski, F. (1969), Lois sacres des cits grecques, Paris. 37 Steiner, D. (1986), The Crown of Song, London. 38 Struck, P.T. (2004), Birth of the Symbol, Princeton. 39 40