Desktop versionMobile version

Roman grec et poésie

 | 
Michel Briand
, 
Michèle Biraud

2. Références, lectures, réécritures

Poetic elements in the Greek novelists’ prose

Ewen Bowie

Abstract

This paper explores the extent of Longus’ poetic vocabulary, discussing classical, hellenistic and imperial poetic uses of words claimed by Valley as poetic in Longus. It concludes that only rarely does he aim to give general poetic colour or evoke a particular genre: most cases involve intertextuality with specific poets or specific passages in their works. Three tables documenting words claimed as poetic in Daphnis and Chloe are followed by two analysing for comparison small samples from Achilles Tatius and Heliodorus, two tabulating how far the novels share the vocabulary of two epigraphically preserved poems, and one doing likewise for a tiny sample of pseudo-Oppian. It concludes tentatively that in the 2nd century the vocabularies of prose and poetry continued to be distinct.

Full text

  • 1 See especially Harrison, Paschalis and Frangoulidis (éd.) 2005.

1The focus of this paper is strictly the vocabulary of the Greek novels – not the syntax, nor indeed the choice of metaphors which may be demonstrably poetic, a subject that has been given attention elsewhere and is discussed in this volume1. Occasionally, for particular reasons, I have registered that a metaphorical use of a word in the novels seems to be poetic, but the great majority of the information distributed through my tables simply concerns vocabulary.

  • 2 Conca, De Carli and Zanetto 1963‑1997.
  • 3 Valley 1926.
  • 4 Feuillâtre 1966.
  • 5 Paulsen 1992.
  • 6 Hernandez Lara 1994.

2So far as I have been able to discover, remarkably little analytic work has been done specifically on the vocabulary of the novelists, despite the availability of the Lessico dei romanizeri greci, the fourth and last volume since 19972. The novelists’ vocabulary has not been wholly neglected, but most of the scholars who have given it their attention have done so in order to establish a particular intertextual relation between a part of a novelistic text and an earlier poetic text (or texts), whether the earlier text(s) in question be Homeric epic, Attic tragedy, Hellenistic pastoral or some other poetic genre. Investigations of such intertextuality have been numerous and fruitful throughout the modern history of scholarship on the novel, and I do not attempt to register them here. It may be that I have missed a discussion of poetic usage in the Greek novels that addressed the issue independently of an attempt to establish some specific intertextuality, but so far as I can discover only the lexicon of Longus has been systematically studied, by Gunnar Valley3. No close discussion of the lexicon of Heliodorus can be found, for example, in Feuillâtre4, despite his acknowledging help from Chantraine in his preface, nor in Paulsen5. For Chariton there is nothing on poetic usage comparable to Hernandez Lara6 on Chariton’s supposed Atticism.

  • 7 Valley 1926, p. 56‑58.
  • 8 Valley 1926, p. 58‑59.

3What is needed, indeed, is thorough study of the linguistic texture of each of the novelists, a study that would require years and that would fill at least a substantial volume, or perhaps better an electronic data-base. What I offer here is something much less ambitious and hence much less satisfactory. I present a number of preliminary sondages which I hope may help determine how much poetic vocabulary there is in three of our five complete Greek texts, which may give some indication of the classical and Hellenistic ancestry of that vocabulary, and which will also take account of words’ use by some poets who were near contemporaries of the novelists. These sondages, represented by my eight tables, are not all of the same kind. The first five tabulate words in three novelists, Longus, Achilles Tatius, and Heliodorus. I begin with Longus – my tables 1, 2 and 3 – because for Longus one has the systematic study by Valley as a starting point. Thus my table 1 presents words in Longus listed by Valley7, as poetic in the classical period but as also found in prose of the Hellenistic and imperial periods; my table 2 presents words listed by Valley8, as poetic without qualification (poetische Ausdrücke); my table 3 presents some words in Longus that seem to me to have a claim to be “poetic” but which are not in Valley’s two lists.

4My tables 4 and 5 are far from systematic: they offer simply an indication of the classical, hellenistic and contemporary poetic uses of words that have a prima facie claim to be poetic in the opening chapters of Achilles Tatius and of Heliodorus. I have attempted nothing similar for Chariton and Xenophon.

5All these first five tables take the text of a novel as the main axis of investigation. tables 6, 7, and 8 start from contemporary poetry. Table 6 lists the 18 words with interesting poetic ancestry that are also used by at least one of the five novelists in the lexicon of one of Marcellus’ poems on Regilla from the Via Appia, Rome, the 59 lines on stele A. Table 7 assembles some hopefully diagnostic words in the lexicon of the poet who composed five elegiac epigrams for the tomb-obelisk of Sacerdos of Nicaea, ca. AD 130, AP XV.4‑8, 36 lines in all. Finally table 8 takes a small number of poetic words from the index verborum of Pseudo-Oppian’s Cynegetica and tabulates their use by the novelists, but does not attempt to sketch their earlier poetic appearances.

6As I have said, these sondages are heterogeneous, but this heterogeneity may be a strength as well as a weakness. If different types of data seem to point to the same conclusion, then that conclusion has a stronger claim to credibility. Nevertheless they are only sondages, and systematic study might lead to different results.

  • 9 οἱ ποιηταὶ τὰ πνεύματα ἀήτας καλοῦσι, Plato, Cratylus 410b: cf. Phrynichus, Ecloga 294 on χθιζός a (...)

7One generalisation does, however, seem to be permitted: just as in the archaic and classical period, as is shown by a particular case observed by a speaker in Plato’s Cratylus in the fourth century BC9, so too in this period some words are found only, or predominantly, in poetry, others only, or predominantly, in prose. That seems to follow both from the large proportion of the words in the poetic texts tabulated in tables 6, 7, and 8 that do not appear in the novels and from the low proportion of apparently poetic words in the admittedly small samples of the writing of Achilles Tatius and Heliodorus.

8But these are, indeed, small samples. What emerges concerning the lexicon of Longus from the two lists of words drawn up by Valley and from the further information I present about their use in earlier and in contemporary poetry? I begin with two general points. First, the number of poetic words identified by Valley might be thought surprisingly small for a text that runs to 65 Teubner pages. Secondly, many of these words – and other words not registered by Valley, like ἀκριδοθήρα or ἀκριδοθήκη at I.10.2 – do indeed seem to be used to trigger a reader’s awareness of a particular intertextuality, in the case of ἀκριδοθήρα or ἀκριδοθήκη with Theocritus 1.45‑54. Given how much we have lost of the Greek literature written before AD 200 and still available to readers of the high Roman empire, it is quite possible, perhaps even likely, that some of the words we can simply identify as “poetic” were expected by Longus to guide his readers to a particular earlier text that is unidentifiable by us.

9Bearing this caveat in mind, I now explore the question whether there are nevertheless cases where Longus’ apparently poetic words might be seen as used to evoke not particular texts but genres. The outcome of this exploration will be largely negative – i.e. in only a very few cases may we be at all confident that Longus is aiming to give a general poetic colour or to evoke a particular genre: the vast majority of cases do seem to be best explained as modes of intertextuality with specific poets or even with specific passages in their works.

  • 10 Valley 1926.
  • 11 μακρὰς διαντλοῦσ’ ἐν δόμοις οἰκουρίας.
  • 12 Antonius 10.5, Cicero 41.6, Coriolanus 35.2, Coniugalia praecepta 142d7, Quaestiones graecae 271b5 (...)
  • 13 Heroicus 35.7, 43.6, Apollonius II.40, III.36, Vitae sophistarum I.21, 516.25, 533.

10One further caveat is needed. Many of Valley’s words, both in my table 1 and my table 2, have a very weak claim to be seen as poetic. Thus I exclude ἐνορμίζομαι at II.12.3, which is hardly poetic, pace Valley10: before Dionysius of Halicarnassus I.36, it appears only in Theognidea 1274 (if indeed Theognidea 1274 precedes Dionysius of Halicarnassus) and Pollux I.102 seems to treat it as a prose word. Similarly the single use of εὐλίμενος by Euripides in his Helen 1463 does not suffice to show as poetic a word that Pollux I.100 includes in his list of terms of praise for a harbour. I also exclude οἰκουρία: its claim to be a poetic word rests on Euripides, Hercules Furens 1373, where indeed its conjunction with διαντλῶ (used by Longus at IV.9.1) might point to Longus’ recalling that line of Euripides11. But οἰκουρία is much used by Plutarch, which counts against it being perceived by Longus or his readers as poetic12. Similarly the claim of οὐκ ἀθεεί (II.26.5) to be poetic is fragile – a single appearance in the Odyssey at XVIII.353. Soon after Longus Philostratus uses οὐκ ἀθεεί or μὴ ἀθεεί six times13. I also exclude the verb ὑπερηφανῶ (Longus III.30.5 and IV.19.5) which is found often in late prose, including e.g. Lucian, Nigrinus 31, though never in Aelius Aristides, and in the other novelists at Xenophon I.4.5, V.16.5, II.5.5 and Achilles Tatius V.11.6. Its only poetic use seems to be at Iliad XI.694.

Terms evoking epic?

11Epic colour might be thought especially to be conveyed by ἀγέρωχος, ἀπηνής, ἄσχετος, δρεπάνη, καλαῦροψ, πόρπη, and πυροφόρος in my table 1; by μεσαιπόλιος, πρωθήβης, τάρσος and χθιζός in my table 2; and perhaps by ἐνδινεύοντα in my table 3. But many of these cases seem to justify a different conclusion when subjected to close scrutiny.

  • 14 Eustathius, In Homerum 314.42, citing also Alcman (the adjective appears at both frg. 5, frg. 1 (b (...)
  • 15 For Longus’ techniques of contrasting his world with that of Attic tragedy see Bowie 2007.

12The epithet ἀγέρωχος is well-established as epic by its appearances in Homer and Hesiod, and is certainly not tragic, but we know from Eustathius that it was used by Archilochus, Alcman and Alcaeus14. Perhaps its use by the Mytilenean Alcaeus suggested to Longus that he should use it of his Mytilenean characters (unnamed shepherd-lads at I.28.2; Philetas’ feisty son Tityrus at II.32.1; the self-assertive Lampis at IV.7.1). One might argue that each of these bearers of the term ἀγέρωχος is being presented by Longus as readier to act self-assertively than Daphnis and that this presentation is meant to recall the conflicted world of epic and of Alcaeus from which, as from Attic tragedy, Longus seeks to distance his own work15. One might also argue that in applying it twice negatively and once positively Longus is picking up a contemporary philological debate about its use in early poetry, a debate we know was already under way by the time of Suetonius de blasphemiis. But such arguments might be quite erroneous: if we had the whole poem of Alcaeus from which the single word that constitutes frg. 402 Campbell comes, we might reach a quite different interpretation.

  • 16 Also in some epigrams, as registered in table 1 (from which I exclude as being later AP VII.568, b (...)
  • 17 Winkler 1990, p. 124‑126.

13Ἀπηνής is also clearly an epic word – found not only in Homer, but also in Apollonius, Oppian, pseudo-Oppian and Quintus of Smyrna16. The lexicon of Liddell and Scott (10th edition) notes that ἀπηνής is also frequent in “later prose”; but it is never used by the careful Atticist Aelius Aristides, nor indeed by the novelists Chariton, Xenophon or Achilles Tatius. On the other hand the epithet ἀπηνής is much favoured by Heliodorus, who uses it nine times. Might Heliodorus be influenced by Longus’ decision to use it? The issue is complicated by the problem of interpreting the sentence in which it appears in Daphnis and Chloe – the wedding guests sing their hymenaion σκληρᾷ καὶ ἀπηνεῖ τῇ φωνῇ καθάπερ τριαίναις γῆν ἀναρρηγνύντες οὐχ ὑμέναιον ἄιδοντες (IV.40.2). Winkler17 notoriously argued that this “amazing detail of attendant discord, unexplained roughness in the song”, together with the mention of Lycaenion (IV.40.3), recalls Lycaenion’s “careful description of defloration as trauma” (III.19.2‑3) and invites us to read Chloe’s imminent matrimonial defloration as painful and traumatic. That I doubt; and the matter is further complicated by the fact that the combination of σκληρός and ἀπηνής is found in Ps‑Aristides, Ars rhetorica II.3.1: ἔστι δὲ οὐ μόνον τὸ εἶδος τοῦ ἦθος δοκεῖν παρέχεσθαι ἐν τῷ λόγῳ τῷ κατὰ τὴν ἁπλότητα τῆς διανοίας, ἀλλὰ καὶ κατὰ τὸ ἐναντίον τούτου, οἷον σκληρότερον καὶ ἀπηνῆ περιθεῖναι λόγον, ὅταν τοιοῦτόν τι ὑποτεθῇ. For the moment I believe that Longus is doing something particular by using the word ἀπηνής, but I don’t know what that something is.

  • 18 Its currency in imperial Greek poetry is also established by the two Anthology poems in which it a (...)
  • 19 The date of the Halieutica seems to be AD 177‑180, and Book II may perhaps be tied down to 178, cf (...)

14What about the epithet ἄσχετος, used by Longus of the speed with which the boat of the Methymnan élite tourists is carried away by wind and wave at II.14.2? Again the word is very clearly epic, with Homer and Hesiod reinforced by Apollonius, the two Oppians, Quintus, and even Dionysius Periegetes18. But like ἀγέρωχος, ἄσχετος is also attested as having been used by Alcaeus, in this case in frg. 364, 1 Campbell. This seems to have been in a quite different context, a gnome preserved in Stobaeus linking Πενία with Ἀμαχανία: ἀργάλεον Πενία κακὸν ἄσχετον, ἂ μέγαν | δάμνα λᾶον Ἀμαχανίᾳ σὺν ἀδελφέᾳ. Does Longus expect his readers to know this gnome from Alcaeus and to reflect on the contrast between the wealth of the Methymnan jeunesse dorée and the πενία and ἀμαχανία of Daphnis, a πενία and ἀμαχανία that will be resolved by the purse of 3000 drachmas washed ashore precisely from this pleasure boat (III.27)? Or is the Odyssean paternity of ἄσχετος the key, and are we meant to contrast the miniature seaborne adventures of the Methymnan youths with the serious, large-scale epic voyages and shipwreck of Odysseus? And does the closeness of Longus’ phrase to ἀσχέτῳ ὁρμῇ in Oppian Halieutica I.492 show that Oppian has been reading Longus, or that Longus has been reading Oppian, or neither of these things19?

  • 20 III.7.8, IV.12.1.
  • 21 Gow and Page 1968, lines 615‑620 = Antipater XCVI.
  • 22 Gow and Page 1968, lines 2757‑2764 = Philip XIX.

15Δρεπάνη at II.2.2, however, may be meant to lead us to a single passage, Homer’s Shield of Achilles, where it first appears (at Iliad XVIII.551). Longus may have chosen the poetic form δρεπάνη, rather than the prose δρέπανον (used by Achilles Tatius)20, to take us precisely to that part of the Iliad where, some lines later, Greek poetry presented one of its earliest surviving images of sheep and shepherds (XVIII.587‑9). Or has Longus simply been reading the Garland of Philip and been reminded of the poetic form δρεπάνη by its appearance in AP XI.3721, a poem by Antipater of Thessalonice circulating in the Garland of Philip and depicting the autumn vintage shortly to be followed by winter (a sequence that chimes with Longus’ attention to the succession of the seasons), as well as in a poem by the Garland’s creator, Philip of Thessalonice, AP VI.10422? In all this, however, caution is counselled: Pollux at I.245 and X.128 gives no indication that δρεπάνη has a different colour from δρέπανον.

  • 23 Theocritus IV.49 with Gow’s commentary ad loc. VII.128.
  • 24 Gow and Page 1965, lines 1972‑7 HE = Leonidas IV.
  • 25 Gow and Page 1968, lines 3452‑3457 = Zonas III.

16Perhaps the frequent use of καλαῦροψ has a similar purpose. Καλαῦροψ is used only once in the Iliad, at XXIII.845, but that one use suffices to establish it as the epic vox propria for a stick thrown by herdsmen, partly to head off their cattle. Theocritus equipped his herdsmen not with a καλαῦροψ but with a λαγωβόλον23, a term that Longus never uses. Leonidas also chose the term λαγωβόλον in a dedication to Pan, AP VI.18824, whereas Diodorus Zonas opted for καλαῦροψ, AP VI.10625. Longus may of course choose this term because it and λαγωβόλον are the only voces propriae, and he simply has to opt for one, but in choosing καλαῦροψ he may be taking sides in a dispute about poetic pedigree and reminding us that Homer’s Iliad uses only καλαῦροψ.

  • 26 But it may be relevant too that these accessories have a different function: ὁ δὲ σχιστὸς χιτὼν πε (...)

17One cannot, however, suggest that Longus’ choice of the predominantly poetic term πόρπη for the pin found among Daphnis’ tokens – rather than περόνη, found in prose as well as poetry – has much to do with its Iliadic attestation, since the Iliad has both πόρπη and περόνη26. But that the appearance of πόρπη in the Iliad is again, like δρεπάνη, in the scene where Hephaestus makes Achilles’ shield may be relevant to Longus’ choice. That Shield, after all, was the earliest and most influential extended description of a work of art in Greek literature: it was thus a significant literary ancestor of the painting Longus describes in his proem and that he professed to be the source of his whole narrative.

  • 27 τὸ γὰρ πρωθήβης ποιητικόν, Pollux II.9.
  • 28 The other of the word’s only two appearances in the Odyssey is at I.431 of the nubile, young Euryc (...)
  • 29 A rarity even in poetry from Homer to Nicander, πρωθήβης is even rarer in the imperial period: not (...)

18Of the epic words in my table 2 πρωθήβης is explicitly pronounced by the lexicographer Pollux to be poetic27. We are perhaps expected to notice that Longus’s only use of πρωθήβης, at II.5.3, is in the mouth of Eros addressing Philetas at a time when Philetas himself is already a γέρων, whereas the Iliad’s only use, at VIII.518 (Hector is speaking), couples it in the same hexameter with γέροντας. But equally we might be expected to think of the only place where the Odyssey uses it of young men, the virtuoso Phaeacian dancers on Scherie (VIII.263), whom we are perhaps again meant to recall when Dryas dances his ballet at II.3628. It is in quoting these lines that Pollux’s compatriot Athenaeus offers his only use of πρωθήβης (I.27)29. Whether we adopt either of these explanations, or neither, the epithet πρωθήβης confers on Philetas an epic stature.

  • 30 Vitae sophistarum II.3.568 and II.18.599.

19The same can be claimed of Longus’ only use of μεσαιπόλιος, at IV.13.2. This gives Dionysophanes a general epic patina rather than likening him in any specific way to the Cretan hero Idomeneus, who is termed μεσαιπόλιος in the Iliad’s only use of the word (XIII.361). That μεσαιπόλιος still had a poetic colour in the second century is indicated by its appearance in Mesomedes (6.2), and Tryphiodorus (168): Philostratus’ two uses in his Lives of the sophists may perhaps also be seen as poetic30.

  • 31 For Longus’ systematic and careful reworking of Theocritus see Bowie 2013.

20Ταρσός invites a quite different explanation. The word’s only use in Homeric epic is at Odyssey IX.219, of the cheese-baskets of the Cyclops. That use is picked up by the love-sick Cyclops of Theocritus XI.37, and Longus is surely picking up both, and in using Theocritus XI he evokes a poem he exploits many times (e.g. I.17.3)31. It is perhaps no accident that we first encounter Daphnis’ ταρσοί at III.33.2, when he has just been assured that he will be allowed to marry Chloe, and that their second appearance, at IV.4.4, is when he is making preparations for Dionysophanes’ tour of inspection, a tour that will result first in Daphnis’ recognition by his true parents and then in turn in his marriage to Chloe. The window-allusion to the brutish Cyclops of Odyssey IX, the first closely observed herdsman in Greek poetry, via the love-sick Cyclops of Theocritus XI who is never going to get his Galateia, prompts us to reflect how much more satisfying a species of young herdsfolk Longus has created. But this is inevitably speculative. It remains possible that ταρσός is not a specifically poetic term at all: Pollux thrice includes it in a list of kit related to cheese-making (I.231, VII.173, X.130).

21Χθιζός may take a reader in a similar direction. Its second use, at III.17.2, is in the mouth of Lycaenion, who has just impudently played a Penelope figure in fabricating a false tale about her twenty geese (III.16.1‑2). It is fitting that an inverted Penelope should now use epic language when she has lured Daphnis into the woods with his likewise Homeric καλαῦροψ (III.17.1), but at the same time these epic terms may draw our attention to how un‑Penelope‑like is Lycaenion, how un-Odysseus‑like is Daphnis. If this is how III.17.1‑2 should be taken, then it helps us see the point of the earlier use of χθιζός at III.11.1, when it is used by the narrator to contrast the couple’s sexual catch on the second day of Daphnis’ winter hunting with the small game he got on the first.

  • 32 Note too Herodianus, de orthographia III.2.574 line 23 s.v. πρῳζόν.
  • 33 χθιζινός, Aristophanes Vespae 281, Ranae 981; Aelius Aristides, Oratio XXVIII, Alciphron III.61.
  • 34 Plutarch, Quomodo quis suos in virtute sentiat profectus 75e5, Quaestiones convivales 688b10, 696e (...)

22But that χθιζός is poetic at all may be questioned. For the lexicographer Moeris (χ 6) it is simply Attic – χθές καὶ χθιζόν Ἀττικοί· ἐχθές καὶ ἐχθεσινόν Ἕλληνες – and Pollux I.66 recommends it without any hint that it has poetic colour32. Only Phrynichus dismisses it as poetic – χθιζόν ἀποβλητέον ὅτι ποιητικόν, ἀντὶ δὲ τοῦ χθιζὸν ἐροῦμεν χθεσινόν, Ecloga 294 – and notes its Homeric colour (τὸ δὲ χθιζόν Ὅμηρος, praeparatio sophistica 127.9‑10). But of Phrynichus’ preferred terms χθιζινός and χθεσινός the latter seems to be koine, and only χθιζινός is supported by Aristophanes and used by Atticists such as Aelius Aristides and Alciphron33. Contrary to Phrynichus’ (later) recommendation, χθιζός is used without any hint of poetic colour by Plutarch and Lucian34.

23Whether one believes that Longus’ use of χθιζός involves sophisticated intertextuality with Homer, or is simply his use of a word he takes to be regular in prose, its appearance does not constitute a case of a word being employed to give general or generic poetic colour.

24If most of the suggestions made above are accepted, then the number of cases of epic poetic terms used by Longus simply to give epic colour is very drastically reduced.

Terms evoking early melic poetry

  • 35 εὐμορφωτέρα Μνασιδίκα τὰς ἀπάλας Γυρίννως. The reference to the name Gyrinna by Maximus of Tyre, O (...)

25We have already seen that part of the explanation of ἀγέρωχος and ἄσχετος may lie in their use by Alcaeus. It seems likely that εὔμορφος, used in the comparative εὐμορφότερος of Dorcon at I.18.2 and of Chloe at IV.32.1, in each case focalised through Daphnis, is picking up its comparative use in a line of Sappho, frg. 82(a) Campbell35, quoted by the second-century metrical writer Hephaestion – i.e. a line that was known to at least some second-century readers – rather than its uses at Aeschylus, Agamemnon 416 and 454 or Choephori 490. But the fact that εὔμορφος had already been used several times by Chariton, Xenophon and Achilles Tatius may mean that Longus did not see it as either markedly poetic in general or Sapphic in particular.

  • 36 See Bowie 2013.
  • 37 AP VII.214 = Gow and Page 1968, lines 3724‑3731 = Archias XXII.
  • 38 ἐπτοημένος and ἐπτόημαι Pollux V.123, πτοούμενος I.197.

26Another term that perhaps evokes Sappho is πτοοῦμαι, used at I.22.2 of goats. Its first known uses are in two fragments of the poetry of Lesbos’ great poet Sappho, whose songs (like those of Alcaeus) are exploited by Longus to construct “his” Lesbos36, and there, of course (at frg. 22, 14 and frg. 31, 6), the term refers to the sexual or sexually-related excitement of young women. If Longus indeed expects Sapphic bells to ring – and frg. 31 Campbell was a poem well known in antiquity as it is now – he is perhaps aiming partly at humour (giddy goats, not giddy girls) and partly offering one of a number of suggestions that in the pastoral world humans and animals are closer to each other than urban readers might expect. But the word could also be familiar to him from Callimachus (Hymn 3, 191), from Nicander’s Alexipharmaca 243, or from a poem of Archias in the Garland of Philip37. And once again, its use by Pollux does not hint at any poetic colour at all38.

The impact of epigram

27Some terms found especially in ekphrastic epigrams transmitted by the Palatine anthology suggest that Longus admires the genre’s handling of the locus amoenus, perhaps because, like him, the poets were creating miniatures.

  • 39 Including AP VII.24, Simonides, an epitaph for Anacreon = Gow and Page 1965, lines 3314‑3323 = Pag (...)

28The case of ἀκρεμών (table 1) is striking. Longus uses ἀκρεμών just once, in his ekphrasis of the intertwined myrtle-trees and ivy outside Dryas’ cottage, at III.5.1. The term comes into Greek literature in Euripides, Cyclops 455, used of the olive branch which Odysseus might sharpen to blind the Cyclops. Again, perhaps, a reader of Longus may be asked to contrast his shepherd Daphnis with the mythical shepherd Polyphemus. But ἀκρεμών then reappears in an idyllic scene along with shepherds and cicadas at Theocritus XVI.96, and is used of the foliage nibbled by a goat about to be sacrificed to Apollo Pythius in Theocritus epigram 1.6 Gow. Thereafter ἀκρεμών reappears in a host of ekphrastic epigrams, many of them in the Garland of Philip39. I think it is this plethora of appearances in poetic loci amoeni that moved Longus to choose this term for a branch rather than, for example, the Sapphic term ὄρπαξ (chosen by Theocritus VII.146).

  • 40 Bowie 2007.

29Another term much found in epigrams preserved in the anthology is γοερός, though unsurprisingly it is well-established in Attic tragedy. In this case origin may be less relevant than the special effect at which Longus aims in II.37.3: here Daphnis’ playing of different sorts of tune on Philetas’ syrinx includes one where he created a sound that was γοερὸν ὡς ἐρῶν, expressing the feelings of Pan pursuing Syrinx. As I have argued elsewhere40, the inset tale of Syrinx offers a world of sexual violence endemic in traditional myth and Attic tragedy to which Longus contrasts his own particular fiction and overall his own genre: the use of the tragic term γοερόν points up this contrast, but at the same time Longus’ encounters with the word may have been as much in epigram as in Attic tragedy. But there may be no intertextuality at all: we know from Pollux V.87 that one type of tune played on the aulos was called γοερόν – Pollux offers no hint that the term of musical classification was in any sense poetic.

  • 41 Dio of Prusa, Orationes 2.47, 2.55; 4.22, 4.84, 4.112; 6.15, 32.67, 33.15, 66.25, 70.8; Plutarch, (...)
  • 42 AP V.151, 154, 174, 190, 198 and AP XII.109 are by Meleager; AP V.193 is by Dioscorides.
  • 43 AP VII.222 is by Philodemus; AP IX.232 by Philip; AP V.35 and 66 by Rufinus; AP XII.10 and 208 by (...)
  • 44 AP XII.136: ὄρνιθες ψίθυροι, τί κεκράγατε; μή μ’ ἀνιᾶτε / τὸν τρυφερῇ παιδὸς σαρκὶ χλιαινόμενον, / (...)

30Erotic epigram was very probably main poetic site of Longus’ and his readers’ encounters with τρυφερός in its physical sense of “soft” or “tender”, though it should be kept in mind that τρυφερός is common in second-century AD prose in a moral sense41. Eighteen of the twenty appearances of τρυφερός in the Greek Anthology are in erotic epigrams, many composed by grand masters of the genre: from the hellenistic period Meleager and Dioscorides42, from more recent times Philodemus, Philip, Rufinus and Strato43. Both Meleager and Rufinus use τρυφερός of χρώς, the latter in his splendid parody of the judgement of Paris in which the speaker narrates and describes his assessment of the buttocks of three girls who are presumably hetairai. A single pederastic epigram whose author was unclear to the compiler of the Anthology uses it of σάρξ44. It is tempting to think that Rufinus’ poem, perhaps little more than a century old when Longus wrote, was known to him and might even have suggested some of the themes associated with the judgement of Paris he deploys (Chloe’s κρίσις of Daphnis and Dorcon at I.15.4 – I.17.1; the apple at III.34.1‑2). Even if it was not, the association of τρυφερός with erotic epigram in which an experienced poet applies the epithet to the skin of urban and sometimes professional sexual objects, female and male, adds a frisson to Chloe’s action at I.13.2: in touching herself to see if Daphnis’ skin is τρυφερώτερος than hers, the forever naïve Chloe is presented to the knowing reader as slipping into the sexual responses of sophisticated urban lovers.

Bacchic terms

  • 45 Cf. Pollux IV.117, 118; V.16.

31At least two of the terms in table 1, i.e. νεβρίς and εὐάζω, have a special place in the terminology of Dionysiac cult, and should probably be regarded as having not a poetic but rather a cultic colour that would be well known to a writer who gave so prominent a place to Dionysus and to the civic leader Dionysophanes in his fourth book. But of the two νεβρίς may be regarded simply as the correct and only term for a garment made from a deer’s skin45.

Other technical terms?

  • 46 In poetry ῥινηλατῶ seems only to be found in Sophocles Ichneutae frg. 314.94, one of the places Lo (...)
  • 47 Βληχή is taken by Pollux V.87 to be the correct term for the sound made by sheep.
  • 48 Pollux V.76 seems to offer both as terms acceptable in prose: ὁ δ’ ἄρρην κερωφόρος ἢ κερασφόρος ἢ  (...)
  • 49 Despite its frequent use in the Odyssey συφεός does not appear in any of the later poetic texts in (...)
  • 50 The epigrams in which σκίρτημα is found include two from the Garland of Philip which seems likely (...)
  • 51 Περκάζω is a botanic terminus technicus in Theophrastus and the Geoponica.
  • 52 The two cases in the Anthology are later than Longus: AP VI.167 by Agathias, AP VII.69 by Iulianus
  • 53 Dio of Prusa, Oration 7.3.20, 16; Plutarch, Cimon 18.2; Lucian, Menippus 10.2, Verae Historiae I.3 (...)

32One might, of course, object that the term I have discussed above, ἀκρεμών, belongs in category of words which were virtually technical terms in special areas – terms for hunting, herding and botany: ῥινηλατῶ46, βληχή47, κερασφόρος, κεράστης48, συφεός49, σκίρτημα50, περκάζω51. If these were the only terms he could properly use for the thing or action he wanted to convey, can their use correctly be seen as poetical? ὑλακή constitutes a similar case. Its poetic use is limited – thrice in Apollonius Rhodius, once each in the Halieutica and in Quintus52. Several instances show that it remains the vox propria for the barking of dogs in second-century AD prose53.

  • 54 Including AP V.85, Asclepiades II = Gow and Page 1965, lines 816‑9; AP VI.276, Antipater of Sidon (...)

33The same sort of argument might be offered for his very frequent use of θέλγω and παρθενία. I suspect παρθενία does not merit its place in Valley’s list. In prose writing παρθενία was already well-established, and had appeared in Parthenius, Erotopathemata 26.2 and in previous novelists. Its centrality to the sexual ideology of the novels would make it hard to avoid, which is also one of the reasons for its frequency in erotic epigram54.

34But note that θέλγω is different. As table 1 shows, θέλγω has wide poetic attestation, including very many anthology epigrams, but is not used at all by Chariton, Xenophon and Achilles Tatius, and appears only once in Heliodorus. Longus may be using θέλγω in awareness of its role in the Odyssey and then more prominently in Apollonius of Rhodes and in Hellenistic epigram, but his decision to use it so often must also be a function of the concept’s importance for his narrative of ἔρως.

A poetic residue?

  • 55 Of the five pre-Byzantine epigrammatic uses one might expect Longus especially to know AP IX.64 = (...)
  • 56 Its two uses in epigram are in poems Longus might well have known, AP XVI.200 = Moschus I Gow and (...)
  • 57 Of a river, III.103, or a rainstorm, I.116, or metaphorically, IV.69, 72, V.79, VI.147.
  • 58 ἔνθεος in Pollux I.8, 15, 20, 23, IV.82, III.59, IV.62; ἐνθέως I.16, 22, III.68, IV.52.

35There remain a few words for which no special explanation under my above categories seems appropriate. These include ἔνθεος, found in tragedy and in epigram55; πυροφόρος, found in hexameter poetry from Homer to the Cynegetica56; and ῥόθιος, used by Heliodorus but not by Longus’ predecessors, and well established, both in adjectival and nominal form (ῥόθιον), in several genres of poetry, though it is worth noting that Pollux treats it as a word to be used in prose57. Each of these should perhaps be admitted to be a “poetic” word that might add some “poetic” colour to Longus’ prose, though the frequency with which Pollux uses ἔνθεος (and four times the adverb ἐνθέως) may suggest that by the late second century AD it is simply a prose word58. Στεφανίσκος, however, has a very different profile, found only in Longus among the novelists, prominent in Anacreon and the Anacreontea, but absent from other poetry. It should perhaps be seen not as “poetic” at all. That as a diminutive term it is suitable for Longus’ miniaturising world is arguably more important than its appearance in sympotic contexts in Anacreon and the Anacreontea.

Conclusions concerning Longus

36Where has this circuitous path taken me? As my analysis of several cases has tried to show, there may be many terms in Valley’s lists which are not “simply” poetic but which may rather be chosen to trigger some intertextuality, whether with a particular earlier text or more broadly and loosely with an earlier genre. There are others which have no strong claim to be “poetic”. The remaining “poetic words” that cannot so be explained are not numerous. Longus may be writing a poetic sort of prose in terms of his Theocritean subject, of his rhythmical sentences, and of his preference for coordination over subordination: but the actual language he uses to do this is predominantly the language of prose.

Achilles Tatius

37Very few words in Achilles Tatius’ opening chapter seem to have a claim to be poetic, and not one of these few is used by Homer, Hesiod or Pindar.

  • 59 Anabasis III.5.15. Oeconomicus V.9.3.

38Only θερίζω is used by the tragedians, and thereafter it is used literally both in prose (as by Longus) and in poetry, but metaphorically only in poetry. The particular sense in which it is used by Achilles Tatius, “to summer”, is also found in Xenophon59.

39Like θερίζω, ἐπινήχομαι is found in the Greek Anthology, though only once: but the poem is one by Dioscorides on the disastrous effects on a peasant of the Nile flood, a poem that the apparently Alexandrian Achilles Tatius might well have known, as he could also have known its two uses in Callimachus’ Hymn to Delos and that at Theocritus XXIII.61. That might be enough to make him see ἐπινήχομαι as a poetic word, as too Oppian may have done at Halieutica II.46.

  • 60 I.1.1, 1.4, 1.11, 10.5, 12.2, II.21.5, III.7.3, IV.19.2, VI.7.1, VIII.7.5.

40Prima facie ἠρέμα might seem to have an even stronger claim to be “poetic”, with two appearances in each of Apollonius Rhodius, Theocritus and Nicander, six in the Greek Anthology and two in the Cynegetica. But it is frequent in Plato, a prose writer with whose works Achilles plays in various ways, and even used occasionally by Aristotle. Its frequent use by Achilles himself60, and by Chariton and Heliodorus, together with a couple of appearances in Longus (I.23.2, 25.1), suggest that it was not perceived as a specifically poetic word.

41The same probably holds for κοιλαίνω: found in Achilles Tatius both here and at III.25.4, its few poetic appearances (in Choerilus frg. 10 as well as its use once each by Theocritus, Oppian and Quintus) have to be set against its employment by Herodotus and Thucydides.

42If the four words have some claim to be seen as poetic, therefore, only in the case of ἐπινήχομαι might that claim be pressed.

Heliodorus

43A brief glance at the opening chapter of Heliodorus’ Aethiopica may provisionally point to a similar conclusion. The words I have picked out and included in table 4 as apparently having some claim to be poetic have acquired that status in five cases because of their use in Attic tragedy: but most are little used by the poets which this table shares with tables 1 to 3.

  • 61 XV.37, 40, XVI.387b (a palindrome).

44Thus ἄρδην is much used in tragedy, but in my other diagnostic poets only in the Greek Anthology, and only thrice61. It is used by Xenophon, Plato and Demosthenes in the fourth century BC, and eight times by Aelius Aristides in the second century AD.

  • 62 Phrynichus, praeparatio sophistica 93‑94: ὄρθρος μὲν γάρ ἐστιν ἡ ὥρα τῆς νυκτός, καθ’ ἣν οἱ ἀλεκτρυ (...)

45Διαγελῶ appears in Sophocles (frg. 171 Radt) and Euripides (Bacchae 272, 286, 322) but not in my later poets. It has a respectable pedigree in prose (e.g. Xenophon, Anabasis II.6.26) and we know from that for Phrynichus and his readers, at any rate, it was a normal locution for day-break, precisely Heliodorus’ use62.

  • 63 Partitiones 100, 12, ὠμοβόειον δέρμα, τὸ ἀπὸ νεοσφαγοῦς βοός.

46The compound νεοσφαγής, used by Heliodorus at I.1.2 and VI.12.1, has some marks of a poetic word – five uses in Attic tragedy, two in Nicander: but it seems to be treated by Pseudo-Herodian as a prose term63.

  • 64 [Aeschylus] Prometheus Vinctus 713. In Sophocles frg. 1088 Radt its sense is “ridge”.
  • 65 Gow and Page 1968, lines 2078‑2083 = Diocles I.

47Ῥαχία appears twice in tragedy, and only one of these uses is in the sense “shore” that we find in Heliodorus64. It makes a single appearance in the Greek Anthology65. But it is also much employed in prose – e.g. by Thucydides, Polybius, Strabo and Aelius Aristides, Oration 1 Lenz-Behr (the Panathenaicus) and 25 Keil (the Rhodiacus).

  • 66 I.1.7, 9.8, 13.1, 26.3, II.22.1, 26.3, 33.3, III.10.5, VI.1.2, IX.6.4, 7.2, X.4.6.

48Σαλεύω is certainly used in poetry – five appearances in tragedy. But it is also used by Plato and in later prose, including koine, both in its literal and its metaphorical sense (e.g. each once in Aelius Aristides). Its use in the novels begins with Achilles Tatius IV.14.2, and it becomes a favourite of Heliodorus66.

  • 67 χηρεύω, τὸ στερίσκομαι, partitiones 100, 1.

49Χηρεύω is different: used by Heliodorus at VI.8.4 as well as at I.1.2, it is the only one of the words in this table to appear in Homer – once, at Odyssey IX.124 – and it has three appearances in tragedy as well as two in the Greek Anthology. Again it was used once before Heliodorus by Achilles Tatius (at IV.1.1), and Pseudo-Herodian thinks it unusual enough to need glossing67. Perhaps here indeed there is a poetic word.

  • 68 AP VII.626: contra Gow and Page 1965 on lines 3494‑3501. Ἀκρώρεια also appears as a toponym at The (...)
  • 69 Cf. Herodian, De prosodia catholica III.1 p. 277, 18: ἀκρώρεια ἄκρον ὄρους.

50Only four words in the table have a life in Hellenistic poetry outside epigram. Ἀκρωρεία is used at Callimachus hymn 3.224 and in an epigram ascribed by Cichorius to Antipater68. But ἀκρωρεία is much used in military contexts by Xenophon in his Hellenica and by Polybius. I doubt if a reader of Heliodorus would have perceived it as poetic – Pollux certainly seems not to have done (II.161)69.

  • 70 E.g. Xenophon, Hiero I. 23.1, Cyropaideia I.1.3 and other prose including Callimachus frg407, 48
  • 71 Pollux VI.32, 38, 48, 57, 58, 182, X.87.

51Another of the four words, ἔδεσμα, is used by Callimachus in the same hymn (hymn 3.48) and in a fragment of Nicander (frg80, 2): again the evidence of several uses in prose writing70, augmented by that of Pollux, who uses it seven times to mean “foodstuff”, counts against any poetic colour71.

  • 72 To the 30 instances in the Greek Anthology can be added, e.g., the epigram preserved by Pausanias (...)
  • 73 I.1.4, 22.5, II.1.3, 3.3, V.22.1, VI.7.6, VII.4.3, 18.1.
  • 74 For the second century AD cf. e.g. Pollux VI, 94. IX, 129.

52A word whose poetic attestation starts with Apollonius Rhodius in the third century BC, λείψανον, is indeed much exploited by epigram, almost entirely (and predictably) sepulchral epigram72, and is quite a favourite of Heliodorus73. But it is clear from its use in prose writers that it is in no sense a poetic word74.

  • 75 Gow and Page 1968, lines 583‑590 = Antipater of Thessalonice XCI.
  • 76 I.1.1, II.1.1, V.7.3, 31.2, VI.14.2, VII.7.7, 26.1, IX.22.4.

53There remains καταυγάζω, once in Apollonius (IV.1248), once in epigram (AP IX.58)75. The Homeric uses of the simple verb αὐγάζω (noted by Livrea commenting on Apollonius IV.1248) may strengthen the case for seeing Heliodorus’ recurrent uses of the compound verb as poetic76.

Two epigraphic poets

54The corollary – which I have no space to pursue at length here – is that very few poetic words used by Marcellus, by the Sacerdos poet, or found in my small sample of the Cynegetica, also appear either in Longus or in the other novelists.

  • 77 I shall publish elsewhere the evidence for the poetic pedigree of the other 53 words analysed, evi (...)
  • 78 ἀνηρείψαντο at Iliad XX.324‑325, cited by Achilles Tatius II.36.3; ἴκελος in Ἀρτέμιδι ἰκέλη ἢ χρυσ (...)
  • 79 In poetry it is found at Aristophanes, Aves 454. For elevated prose cf. IGL Syr I.47, column 1, li (...)

55Table 6 sets out some of the poetic pedigree of the 18 terms (out of 71 in all that I have analysed) in the lexicon of the first of Marcellus’ poems on Regilla from the Via Appia, Rome (the 59 lines on stele A) which are also used by at least one of the five novelists77. Of these 18, three are in fact quotations from the Homeric poems78, leaving only around 20 % of Marcellus’ poetic vocabulary as words also found in the novelists, and among these may be a word more common in prose than in poetry79.

  • 80 See Bowie 2014, p. 40‑43.
  • 81 It also appears in AP IX.688, a poem inscribed on a gate of Argos (according to J in the codex Pal (...)
  • 82 E.g. οὐρανομήκης, as well as appearing at Aeschylus, Agamemnon 92, is used by Aristophanes, Nubes (...)

56Table 7 traces the poetic uses of terms that seem to be poetic in five elegiac epigrams for the tomb of Sacerdos of Nicaea, probably composed around AD 130, and preserved in the Greek Anthology, Book XV, poems 4‑880. Two words that seem to be poetic have been excluded since they occur neither in my selected poets nor in the novels: ἀποικεσίη and ἀρχέγονος. Another term, ὀργιοφάντας, has been ruled out of the assessment as being technical, though its poetic use is not limited to the Sacerdos poems81. Of the remaining 14 words with various claims to be seen as poetic82, only 6 are used by the novelists, or only five if one excludes the used of ἀντωπῶ by Heliodorus I.21.3 (whereas AP XV.6.3 has ἀντωπός).

  • 83 Aeschylus, Supplices 1055, Agamemnon 764, frg458 Radt; Sophocles, Trachiniae 144; Euripides, Pho (...)
  • 84 Aeschylus, Eumenides 300; Sophocles, Oedipus Tyrannus 312, 313; Euripides, Andromache 575, Ion 165 (...)
  • 85 XI.256 (Lucillius), 261 (Epigonus of Thessalonice); XVI.383 (late).
  • 86 VI.125 (Mnasalcas), 141(“Anacreon”), 231 (Philip). VII.72 (Menander), 74 (Diodorus), 195 (Meleager (...)
  • 87 VII.607 (Palladas), IX.500 (incertus), XI.166 (incertus), 171 (Lucillius), 294 (Lucian), 384 (Agat (...)

57Five poetic words used by the novelists out of a list of 18 is not a negligeable proportion (28 %), but even of these five, two have only slender claims to be poetic: βεβοημένος with two appearances in the Greek Anthology, ζάκορος with one in Nicander’s Alexipharmaca. Only three are left that have a broadly based poetic pedigree, κλαρόνομος, νεαζόμενος and ῥυσάμενος. Forms of νεάζω / νεαζόμαι are well represented in Attic tragedy83, as are forms of ῥυσάμενος84; νεάζω appears twice in pre-Byzantine Anthology epigrams85, while forms of ῥυσάμενος are frequent in the Greek Anthology86. Finally κλαρόνομος in its Doric form appears in other poetry only in a text that has not otherwise been included in my searches, [Moschus], Epitaphius Bionis 3.96 (in the phrase κλαρονόμος Μοίσας τᾶς Δωρίδος); but there are several cases of κληρονόμος in the Anthology87. Unlike νεαζόμενος and ῥυσάμενος it seems to be absent from archaic and classical poetry.

58The evidence of Marcellus and the Sacerdos poet may point to a slightly greater permeability of the frontier between prose and poetic vocabulary than the evidence from the novelists themselves, or may point to the novelists other than Xenophon (who appears only once in the final column of table 6 and does not appear in the final column of table 7 at all) being ready to use poetic terms. Further close analysis would be needed to establish which explanation should be favoured.

59The small sample of words from Pseudo-Oppian, Cynegetica in table 8 seems to confirm that some poetic words seep into the novelists, with the highest uptake in Heliodorus. Of course the claim of some to be poetic can be questioned: ἀγάλλομαι is much used by poets, but also by Herodotus, Thucydides, Plato and Xenophon; alongside its “poetic” uses ἄγρη/ἄγρα is also found in Herodotus and Plato; αἰθήρ appears in philosophical prose as well as in poetry. Only a careful examination of Heliodorus’ use of each term, comparable to that offered above for Longus, would be able to establish the modes and possible objectives of such apparently “poetic” vocabulary.

Conclusions

60Scrutiny of the “poetic” words listed by Valley for Longus suggested that only a small proportion of these words seemed likely to be employed to give a generally “poetic” colour or even to evoke a particular poetic genre. Many were arguably chosen to trigger intertextuality with particular passages in earlier poetry, and some seemed not to have a good claim to be “poetic”. A brief overview of a small selection of potentially “poetic” words in Achilles Tatius and Heliodorus supported the view that in their writing too only a low proportion of a “poetic” words was to be found, a view that might be thought to be corroborated by the small number of predominantly “poetic” words in Marcellus, the Sacerdos poet and my sample from the Cynegetica that are also found in the novelists. Overall, though much more work remains to be done, my investigation suggests that poets and writers of novelistic prose are still drawing their vocabulary from two different linguistic pools.

Tables

61The abbreviated names of Greek authors and works are as follows:

62A = Aeschylus
Alc = Alcaeus
Alex = Alexipharmaca
Anacreont = Anacreontea
Ant = Antigone
AP = Anthologia Graeca
Ap Rhod = Apollonius Rhodius
Ar = Aristophanes
AT = Achilles Tatius, Leucippe and Cleitophon
Ba = Bacchylides
Ba = Bacchantes
C = Chariton, Chaereas and Callirhoe
Call = Callimachus
Cy = Cyclops
D Per. = Dionysius Periegetes
E = Euripides
Eccl = Ecclesiazusae
El = Electra
Epig = Epigrams
H = Heliodorus, Aethiopica
h = hymn
He = Hecuba

Hec = Hecale
Hel = Helen
Hes = Hesiod
HF = Hercules Furens
Hh = hymni Homerici
H h Ap = Homeric hymn to Apollo
H h Ba = Homeric hymn to Dionysus
H h Her = Homeric hymn to Hermes
Hipp = Hippolytus
Il = Iliad
L = Longus, Daphnis and Chloe
Nic = Nicander
Nu = Nubes
Od = Odyssey
“O” Cyn = Pseudo-Oppian, Cynegetica
Op = Opera et dies
Opp Hal = Oppian, Halieutica
Or = Orestes
OT = Oedipus Tyrannus
P = Pythian
Ph = Phoenissae
Pi = Pindar
PV = Prometheus Vinctus
Qu Sm = Quintus Smyrnaeus, Posthomerica
Rhes = Rhesus
S = Sophocles
Sa = Sappho
Sc = Scutum
Th = Theogony
Theo = Theocritus
Theogn = Theognis
Ther = Theriaca
Vesp = Vespae
X = Xenophon, Ephesiaca

Table 1

63Words in Longus listed by Valley 1926, p. 56‑58, as poetic in the classical period but as also found in prose of the Hellenistic and imperial periods. I exclude instances of usage, e.g. λοιπόν without the article (said by Valley 1926, p. 57 to be poetic until Polybius) or κόλπος/ κόλποι in the sense “lap” (found at I.26.1, 31.1; III.34.3; IV.36.3). None of the words listed appears in Aratus, so I have not included him in this table.

Table 2

64Words listed by Valley 1926, p. 58‑59 as poetic (poetische Ausdrücke) but excluding items of syntax, such as θέλγω + infinitive (II.4) and μέλεσθαι τινί τι (II.27.1), figures of speech (πτερόν =ὄρνις, III.5.2, III.22.1) or phraseology (ἁπαλὸν γελῶ, II.4.4). None of the words listed appears in Aratus, so I have not included him in this table.

Table 3

65Some words in Longus with a claim to be “poetic” but not in Valley’s lists. None of the words listed appears in Hesiod, Aratus, Callimachus, Nicander, the Greek Anthology, Dionysius Periegetes, the two Oppians or Quintus, so I have not included these in this table.

Table 4

66Words with a claim to be “poetic” in Achilles Tatius I, 1 (not e.g. ἐπικάθημαι, 6 times in Achilles Tatius and found in Herodotus, Attic and later prose). None of the words listed appears in Homer, Hesiod, Aratus, or Dionysius Periegetes, so I have not included these in this table.

Table 5

67Words with a claim to be “poetic” in Heliodorus Aethiopica I.1. None of these words appears in Hesiod, Pindar, Aratus, Dionysius Periegetes or Quintus, so I have not given these poets columns in this table.

Table 6

68The lexicon of Marcellus’ poem on Regilla from the Via Appia, Rome: IG xiv, 1389 stele A (= IGUR iii, 1155 A, cf. SEG xxix, 999) c. AD 161. Of 71 words with varying degrees of poetic pedigree only the following 18 are also found in the novelists.

Table 7

69The lexicon of the poet who composed five elegiac epigrams for the tomb-obelisk of Sacerdos of Nicaea, ca. AD 130, AP XV.4‑8 =GVI 1999, some examples.

Table 8

70Ps‑Oppian Cynegetica: some of Ps‑Oppian’s words that do appear in the novelists.

Notes

1 See especially Harrison, Paschalis and Frangoulidis (éd.) 2005.

2 Conca, De Carli and Zanetto 1963‑1997.

3 Valley 1926.

4 Feuillâtre 1966.

5 Paulsen 1992.

6 Hernandez Lara 1994.

7 Valley 1926, p. 56‑58.

8 Valley 1926, p. 58‑59.

9 οἱ ποιηταὶ τὰ πνεύματα ἀήτας καλοῦσι, Plato, Cratylus 410b: cf. Phrynichus, Ecloga 294 on χθιζός and Pollux II.9, πρωθήβης, both cited below in my discussion of these words.

10 Valley 1926.

11 μακρὰς διαντλοῦσ’ ἐν δόμοις οἰκουρίας.

12 Antonius 10.5, Cicero 41.6, Coriolanus 35.2, Coniugalia praecepta 142d7, Quaestiones graecae 271b5 and 271e11, de Iside et Osiride 381e4, An seni respublica gerenda 784a6, 788f10.

13 Heroicus 35.7, 43.6, Apollonius II.40, III.36, Vitae sophistarum I.21, 516.25, 533.

14 Eustathius, In Homerum 314.42, citing also Alcman (the adjective appears at both frg. 5, frg. 1 (b) 4 and 10 (b) 15 Davies), Archilochus frg. 261 West, cf. Etymologicum Gudianum I.10.13 de Stefani, Suetonius, De blasphemiis p. 56 Taillardat, and Alcaeus frg. 402 Campbell.

15 For Longus’ techniques of contrasting his world with that of Attic tragedy see Bowie 2007.

16 Also in some epigrams, as registered in table 1 (from which I exclude as being later AP VII.568, by Agathias).

17 Winkler 1990, p. 124‑126.

18 Its currency in imperial Greek poetry is also established by the two Anthology poems in which it appears – AP XI.285, by Philip and AP XVI.110, the only surviving epigram of Philostratus.

19 The date of the Halieutica seems to be AD 177‑180, and Book II may perhaps be tied down to 178, cf. Fajen 1999, p. viii.

20 III.7.8, IV.12.1.

21 Gow and Page 1968, lines 615‑620 = Antipater XCVI.

22 Gow and Page 1968, lines 2757‑2764 = Philip XIX.

23 Theocritus IV.49 with Gow’s commentary ad loc. VII.128.

24 Gow and Page 1965, lines 1972‑7 HE = Leonidas IV.

25 Gow and Page 1968, lines 3452‑3457 = Zonas III.

26 But it may be relevant too that these accessories have a different function: ὁ δὲ σχιστὸς χιτὼν περόναις κατὰ τοὺς ὤμους διεῖρτο καὶ πόρπῃ κατὰ τὰ στέρνα ἐνῆπτο, Pollux VII.54.

27 τὸ γὰρ πρωθήβης ποιητικόν, Pollux II.9.

28 The other of the word’s only two appearances in the Odyssey is at I.431 of the nubile, young Eurycleia.

29 A rarity even in poetry from Homer to Nicander, πρωθήβης is even rarer in the imperial period: not in poets, nor in Dio, Plutarch, Aristides, Pausanias or Philostratus, and in Lucian only at De dea Syria 35 and Dialogi Mortuorum 15.2.

30 Vitae sophistarum II.3.568 and II.18.599.

31 For Longus’ systematic and careful reworking of Theocritus see Bowie 2013.

32 Note too Herodianus, de orthographia III.2.574 line 23 s.v. πρῳζόν.

33 χθιζινός, Aristophanes Vespae 281, Ranae 981; Aelius Aristides, Oratio XXVIII, Alciphron III.61.

34 Plutarch, Quomodo quis suos in virtute sentiat profectus 75e5, Quaestiones convivales 688b10, 696e6, De sollertia animalium 975c9, Lucian Hermotimus 1; note also [Plutarch], de liberis educandis 13f2.

35 εὐμορφωτέρα Μνασιδίκα τὰς ἀπάλας Γυρίννως. The reference to the name Gyrinna by Maximus of Tyre, Oration 18.9, suggests that this line will have been known to him too.

36 See Bowie 2013.

37 AP VII.214 = Gow and Page 1968, lines 3724‑3731 = Archias XXII.

38 ἐπτοημένος and ἐπτόημαι Pollux V.123, πτοούμενος I.197.

39 Including AP VII.24, Simonides, an epitaph for Anacreon = Gow and Page 1965, lines 3314‑3323 = Page 1981, lines 956‑965; AP VII.385, Philip (on Protesilaus’ tomb) = Gow and Page 1968, lines 2853‑2860; AP IX.3, Antipater (on a nut‑tree) = Gow and Page 1965, lines 669‑674; AP IX.71, Antiphilus (shelter from mid‑day sun) = Gow and Page 1968, lines 985‑990; AP IX.220 Thallus (love under a plane-tree) = Gow and Page 1968, lines 3434‑3439; AP IX.256, Antiphanes (a caterpillar eats an apple) = Gow and Page 1968, lines 741‑746.

40 Bowie 2007.

41 Dio of Prusa, Orationes 2.47, 2.55; 4.22, 4.84, 4.112; 6.15, 32.67, 33.15, 66.25, 70.8; Plutarch, Alexander 22.10, Demosthenes 4.6, Pericles 27.4, Phocion 2.4, Sertorius 13.1, de recta ratione audiendi 41f12, de Pythiis oraculis 397b1, de tranquillitate animi 466b9 (quoting Menander), de vitioso pudore 528e10, Arrian, Dissertationes Epicteti 2.24, 2.28, 3.1, 3.27. Plutarch uses the comparative at de capienda ex inimicis utilitate 89e5, and τρυφερός has a clearly physical sense at Cato Maior 4.5 and de sera numinis vindicta 560c5 (ἐν σαρκὶ τρυφερᾷ).

42 AP V.151, 154, 174, 190, 198 and AP XII.109 are by Meleager; AP V.193 is by Dioscorides.

43 AP VII.222 is by Philodemus; AP IX.232 by Philip; AP V.35 and 66 by Rufinus; AP XII.10 and 208 by Strato.

44 AP XII.136: ὄρνιθες ψίθυροι, τί κεκράγατε; μή μ’ ἀνιᾶτε / τὸν τρυφερῇ παιδὸς σαρκὶ χλιαινόμενον, / ἑζόμεναι πετάλοισιν ἀηδόνες· εἰ δὲ λάληθρον / θῆλυ γένος, δέομαι, μείνατ’ ἐφ’ ἡσυχίης.

45 Cf. Pollux IV.117, 118; V.16.

46 In poetry ῥινηλατῶ seems only to be found in Sophocles Ichneutae frg. 314.94, one of the places Longus may also have encountered κεράστης (frg. 314, 307). Pollux II.74 offers it as the correct term for τὸ τὰς ὀσμὰς ἕλκειν.

47 Βληχή is taken by Pollux V.87 to be the correct term for the sound made by sheep.

48 Pollux V.76 seems to offer both as terms acceptable in prose: ὁ δ’ ἄρρην κερωφόρος ἢ κερασφόρος ἢ κεράστης.

49 Despite its frequent use in the Odyssey συφεός does not appear in any of the later poetic texts in my table, and is used in imperial prose by Dio of Prusa not only (predictably) in his Euboean oration (7.74) but also in 8.25 and 30.33.

50 The epigrams in which σκίρτημα is found include two from the Garland of Philip which seems likely to have been known to Longus, AP VII.217 = Scaevola I Gow and Page 1968, lines 3374‑79 and AP IX.543 by Philip himself, Philip LIV Gow and Page 1968, lines 2995‑3000. But the term is clearly also the vox propria for animal movements in prose of this period, e.g. Plutarch, Artaxerxes 7.5, quaestiones convivales 706e3; Lucian, Alexander 40, Bacchus 5, Asinus 40; Philostratus, Apollonius VIII.7, Vitae sophistarum II.10.587 (Ἑλληνικὸν σκίρτημα).

51 Περκάζω is a botanic terminus technicus in Theophrastus and the Geoponica.

52 The two cases in the Anthology are later than Longus: AP VI.167 by Agathias, AP VII.69 by Iulianus.

53 Dio of Prusa, Oration 7.3.20, 16; Plutarch, Cimon 18.2; Lucian, Menippus 10.2, Verae Historiae I.32, cf. Pollux V.86 φωναὶ ζῴων. κυνῶν μὲν ὑλακή…

54 Including AP V.85, Asclepiades II = Gow and Page 1965, lines 816‑9; AP VI.276, Antipater of Sidon LI = Gow and Page 1965, lines 510‑15; VII.164, Antipater of Sidon, XXI = Gow and Page 1965, lines 302‑311; VII.182, Meleager CXXIII, 123 = Gow and Page 1965, lines 4680‑87; AP VII.183, Parmenion III GP = Gow and Page 1968, lines 2581‑85; AP VII.351, Dioscorides XVIII = Gow and Page 1968, lines 1555‑64; AP VII.352, Meleager CXXXII = Gow and Page 1965, lines 4742‑9; AP VII.352, Mnasalcas X = Gow and Page 1965, lines 2639‑2642; AP VII.491, Pamphilus II = Gow and Page 1965, lines 2843‑6.

55 Of the five pre-Byzantine epigrammatic uses one might expect Longus especially to know AP IX.64 = Asclepiades XLV Gow and Page 1965, lines 1018‑1025, on Hesiod and the Muses; AP XVI.133 = Antipater LXXXVII Gow and Page 1968, lines 557‑566; AP XVI.226 = Alcaeus XX Gow and Page 1965, lines 106‑113, on Pan, Syrinx and nymphs. Note that ἔνθεος is used seven times by Aelius Aristides.

56 Its two uses in epigram are in poems Longus might well have known, AP XVI.200 = Moschus I Gow and Page 1965, lines 2683‑87 and AP VII.176 = Antiphilus XXV, Gow and Page 1968, lines 935‑940.

57 Of a river, III.103, or a rainstorm, I.116, or metaphorically, IV.69, 72, V.79, VI.147.

58 ἔνθεος in Pollux I.8, 15, 20, 23, IV.82, III.59, IV.62; ἐνθέως I.16, 22, III.68, IV.52.

59 Anabasis III.5.15. Oeconomicus V.9.3.

60 I.1.1, 1.4, 1.11, 10.5, 12.2, II.21.5, III.7.3, IV.19.2, VI.7.1, VIII.7.5.

61 XV.37, 40, XVI.387b (a palindrome).

62 Phrynichus, praeparatio sophistica 93‑94: ὄρθρος μὲν γάρ ἐστιν ἡ ὥρα τῆς νυκτός, καθ’ ἣν <οἱ> ἀλεκτρυόνες ᾄδουσιν. ἄρχεται δὲ ἐνάτης ὥρας καὶ τελευτᾷ εἰς διαγελῶσαν ἡμέραν [...] ἕως δὲ τὸ ἀπὸ διαγελώσης ἡμέρας ἄχρις ἡλίου ἐξέχοντος διάστημα.

63 Partitiones 100, 12, ὠμοβόειον δέρμα, τὸ ἀπὸ νεοσφαγοῦς βοός.

64 [Aeschylus] Prometheus Vinctus 713. In Sophocles frg. 1088 Radt its sense is “ridge”.

65 Gow and Page 1968, lines 2078‑2083 = Diocles I.

66 I.1.7, 9.8, 13.1, 26.3, II.22.1, 26.3, 33.3, III.10.5, VI.1.2, IX.6.4, 7.2, X.4.6.

67 χηρεύω, τὸ στερίσκομαι, partitiones 100, 1.

68 AP VII.626: contra Gow and Page 1965 on lines 3494‑3501. Ἀκρώρεια also appears as a toponym at Theocritus 25.31.

69 Cf. Herodian, De prosodia catholica III.1 p. 277, 18: ἀκρώρεια ἄκρον ὄρους.

70 E.g. Xenophon, Hiero I. 23.1, Cyropaideia I.1.3 and other prose including Callimachus frg407, 48.

71 Pollux VI.32, 38, 48, 57, 58, 182, X.87.

72 To the 30 instances in the Greek Anthology can be added, e.g., the epigram preserved by Pausanias V.20.7.

73 I.1.4, 22.5, II.1.3, 3.3, V.22.1, VI.7.6, VII.4.3, 18.1.

74 For the second century AD cf. e.g. Pollux VI, 94. IX, 129.

75 Gow and Page 1968, lines 583‑590 = Antipater of Thessalonice XCI.

76 I.1.1, II.1.1, V.7.3, 31.2, VI.14.2, VII.7.7, 26.1, IX.22.4.

77 I shall publish elsewhere the evidence for the poetic pedigree of the other 53 words analysed, evidence that is not strictly relevant to the argument of this paper.

78 ἀνηρείψαντο at Iliad XX.324‑325, cited by Achilles Tatius II.36.3; ἴκελος in Ἀρτέμιδι ἰκέλη ἢ χρυσείῃ Ἀφροδίτῃ, Odyssey XVII.37 and XIX.54, quoted by Chariton IV.7.5 and in ὄμματα καὶ κεφαλὴν ἴκελος Διὶ τερπικεραύνῳ, Iliad II.478, quoted by Achilles Tatius I.8.7; the epithet καλλίσφυρος quoted by Chariton IV.1.8.

79 In poetry it is found at Aristophanes, Aves 454. For elevated prose cf. IGL Syr I.47, column 1, line 17 (Antiochus of Commagene), ὅ]σ[α] γε [κ]αιρὸς παρεῖδε[ν χρ]όνοις [προτέρ]ο[ι]ς .

80 See Bowie 2014, p. 40‑43.

81 It also appears in AP IX.688, a poem inscribed on a gate of Argos (according to J in the codex Palatinus) by its erector Cleadas sometime in the second or third century AD: τήνδε πύλην λάεσσιν ἐυξέστοις ἀραρυῖαν, / ἀμφότερον κόσμον τε πάτρῃ καὶ θάμβος ὁδίταις, / τεῦξε Κλέης Κλεάδας ἀγανῆς πόσις εὐπατερείης, / Λερναίων ἀδύτων περιώσιος ὀργιοφάντης, / τερπόμενος δώροισιν ἀγασθενέων βασιλήων. Cleadas was also the erector and perhaps poet of IG ii3674 from Attica: Δηοῦς καὶ κούρης θεοΐκελον ἱεροφάντην / κυδαίνων πατέρα στῆσε δόμοις Κλεάδας, / [Κ]εκροπίης σοφὸν ἔρνος Ἐρώτιον· ὧι ῥα καὶ αὐτός / Λερναίων ἀδύτων ἶσον ἔδεκτο γέρας.

82 E.g. οὐρανομήκης, as well as appearing at Aeschylus, Agamemnon 92, is used by Aristophanes, Nubes 357 and 459 (neither Fraenkel on the Agamemnon nor Dover on Clouds comments) and also in prose. Ἀοιδότατος and ἀοιδότερος are entirely poetic – as well as Euripides, Callimachus and the Anthology note IG xii.2.443 (on Αlcibiades, a poet from Byzantium who died in Mytilene) and IK 33.144.

83 Aeschylus, Supplices 1055, Agamemnon 764, frg458 Radt; Sophocles, Trachiniae 144; Euripides, Phoenissae 713.1619. It is also found in Menander and Epicurus, Epistle 3.59.

84 Aeschylus, Eumenides 300; Sophocles, Oedipus Tyrannus 312, 313; Euripides, Andromache 575, Ion 165, Hypsipyle frg60, 28, Helen 925, 1086.

85 XI.256 (Lucillius), 261 (Epigonus of Thessalonice); XVI.383 (late).

86 VI.125 (Mnasalcas), 141(“Anacreon”), 231 (Philip). VII.72 (Menander), 74 (Diodorus), 195 (Meleager), 250 “Simonides” (Page 1981, “Simonides” 12, also quoted by Plutarch, De malignitate Herodoti 39 and Aelius Aristides, Orationes XXVIII.66), 286 (Antipater), IX.40 (Zosimus), 71 (Antiphilus), 178 (Antiphilus), 496 (Athenaeus).

87 VII.607 (Palladas), IX.500 (incertus), XI.166 (incertus), 171 (Lucillius), 294 (Lucian), 384 (Agathias).

List of illustrations

Title Table 1 (1/5)
URL http://books.openedition.org/momeditions/docannexe/image/2276/img-1.jpg
File image/jpeg, 149k
Title Table 1 (2/5)
URL http://books.openedition.org/momeditions/docannexe/image/2276/img-2.jpg
File image/jpeg, 218k
Title Table 1 (3/5)
URL http://books.openedition.org/momeditions/docannexe/image/2276/img-3.jpg
File image/jpeg, 190k
Title Table 1 (4/5)
URL http://books.openedition.org/momeditions/docannexe/image/2276/img-4.jpg
File image/jpeg, 235k
Title Table 1 (5/5)
URL http://books.openedition.org/momeditions/docannexe/image/2276/img-5.jpg
File image/jpeg, 117k
Title Table 2
URL http://books.openedition.org/momeditions/docannexe/image/2276/img-6.jpg
File image/jpeg, 206k
Title Table 3
URL http://books.openedition.org/momeditions/docannexe/image/2276/img-7.jpg
File image/jpeg, 90k
Title Table 4
URL http://books.openedition.org/momeditions/docannexe/image/2276/img-8.jpg
File image/jpeg, 109k
Title Table 5
URL http://books.openedition.org/momeditions/docannexe/image/2276/img-9.jpg
File image/jpeg, 167k
Title Table 6 (1/2)
URL http://books.openedition.org/momeditions/docannexe/image/2276/img-10.jpg
File image/jpeg, 203k
Title Table 6 (2/2)
URL http://books.openedition.org/momeditions/docannexe/image/2276/img-11.jpg
File image/jpeg, 170k
Title Table 7
URL http://books.openedition.org/momeditions/docannexe/image/2276/img-12.jpg
File image/jpeg, 272k
Title Table 8 (1/2)
URL http://books.openedition.org/momeditions/docannexe/image/2276/img-13.jpg
File image/jpeg, 138k
Title Table 8 (2/2)
URL http://books.openedition.org/momeditions/docannexe/image/2276/img-14.jpg
File image/jpeg, 122k

Author

CC-BY-NC-ND-4.0

The text only may be used under licence CC BY-NC-ND 4.0. All other elements (illustrations, imported files) are “All rights reserved”, unless otherwise stated.

Search OpenEdition Search

You will be redirected to OpenEdition Search