Skip to main content Accessibility help
Hostname: page-component-7f7b94f6bd-w6m4b Total loading time: 1.061 Render date: 2022-07-01T01:30:19.541Z Has data issue: true Feature Flags: { "shouldUseShareProductTool": true, "shouldUseHypothesis": true, "isUnsiloEnabled": true, "useRatesEcommerce": false, "useNewApi": true } hasContentIssue true


Published online by Cambridge University Press:  23 April 2021

Filippomaria Pontani*
Università Ca’ Foscari, Venezia
Maria Giovanna Sandri*
Scuola Normale Superiore, Pisa
Rights & Permissions[Opens in a new window]


A treatise on rhetorical tropes is attributed in manuscripts to the first-century grammarian Trypho: this article considers for the first time a fifteenth-century manuscript of this work (Leiden, BPG 74G), which turns out to be the only complete witness of its hitherto unknown original version; this version (very fragmentarily transmitted by a fifth-century papyrus scrap) is also partly found in another fifteenth-century manuscript now kept in Olomouc (M 79). Four interesting poetic fragments are quoted in this newly discovered, fuller version of Ps.-Trypho's De Tropis: some lines from Callimachus’ fifth and fourth Iambi (23–9 and 90–2 respectively: a radically new light is shed by this new witness on the parallel papyrus fragments carrying Callimachus’ text), an epigram dubiously attributed to Simonides (FGE 44 Page, probably to be dated to the Hellenistic period: the text can be now restored to its complete form), and some enigmatic lines of “Hesiod”'s Wedding of Keyx, which the new witness finally makes fully understandable.

Research Article
Creative Commons
Creative Common License - CCCreative Common License - BYCreative Common License - NCCreative Common License - ND
This is an Open Access article, distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives licence (, which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is unaltered and is properly cited. The written permission of Cambridge University Press must be obtained for commercial re-use or in order to create a derivative work.
Copyright © The Author(s), 2021. Published by Cambridge University Press on behalf of The Classical Association


Recentiores, non deteriores. This is the story of a hitherto neglected Greek manuscript of philosophical and rhetorical content, which was copied as late as the High Renaissance, and happens to be our only witness for a fuller version of four poetic fragments, on whose wording and meaning it casts an entirely new light.

M.L. West's edition of the rhetorical treatise On Tropes ascribed in the manuscript tradition to the first-century grammarian TryphoFootnote 1 was based on just eight out of the twenty-one extant medieval codices and on one papyrus (the fifth-century P.Vindob. 29332),Footnote 2 which West, following an insight by Paul Maas,Footnote 3 believed to carry the very same work handed down by the medieval witnesses. A closer analysis of the entire manuscript tradition, carried out by M.G. Sandri for a new edition of the ancient Greek and Byzantine treatises on tropes, now reveals the special importance of MS Leiden, Universiteitsbibliotheek BPG 74 G (here Z), a codex from the collection of the eighteenth-century traveller Antonios Triphilis.Footnote 4 This manuscript is familiar to students of ancient mathematics and philosophy as a witness of Aristotle's Physics (fols. 67–144) and of Nicomachus of Gerasa's Eisagoge (fols. 4–48, together with John Philoponus’ commentary on it, fols. 52–65): its last folia, however, have a different character, and include (Ps.-)Manuel Chrysoloras's On Anomalous Verbs (fols. 150–7)Footnote 5 and (Ps.-)Trypho's On Tropes (fols. 145–9v). The scribes are not identified, but watermarks throughout the codex consistently point to the first decades of the sixteenth century.Footnote 6

The paths of the manuscript transmission of the texts Περὶ Τρόπων are very complicated; hence we shall leave to another occasion a more thorough consideration of the contribution made by this manuscript to our knowledge of the treatise ascribed to Trypho (‘Trypho II’). Here it will suffice to say that the Leidensis preserves the original form of this treatise, while the rest of the manuscript tradition, embracing twenty codices dated between the late thirteenth and the early eighteenth centuries, carries an epitomized (and sometimes adapted) version. The lost archetype of this shorter version (which itself gave rise to two different families) we shall call α.

As the lone witness of the fuller version of the Περὶ Τρόπων, the Leidensis is of paramount importance for the constitution of the text. In this paper, we shall focus exclusively on four out of five non-Homeric literary quotations appearing in Ps.-Trypho's text:Footnote 7 not only do these offer entirely new (and sound) readings for hotly debated poetical fragments of ‘Hesiod’, ‘Simonides’ and Callimachus, but comparison of the manuscript's readings with papyri containing the same lines (whether the papyri transmit the original poetic works or, in one case, the text of Ps.-Trypho himself) guarantees that the new readings cannot derive from conjectural activity. We shall present each of the four relevant passages in the order of Ps.-Trypho's treatise, with an apparatus criticus; we shall then discuss the impact of this new witness on the text of the fragments concerned.


§1. Ἀλληγορία μὲν οὖν ἐστι φράσις ἕτερον μέν τι κυρίως δηλοῦσα, ἑτέρου δὲ ἔννοιαν παριστῶσα. τότε δὲ καταχρῶνται δεόντως τῇ ἀλληγορίᾳ, ὅταν ἢ δι’ εὐλάβειαν ἢ δι’ αἰσχύνην μὴ δύνωνται φανερῶς ἀπαγγεῖλαι τὸ πρᾶγμα, ὃν τρόπον παρὰ Καλλιμάχῳ ἐν Ἰάμβοις αZ

τὸ πῦρ δὲ τὠνέκαυσας, ἄχρις οὐ πολλῇ

πρόσω κεχώρηκεν φλογί,

ἀλλ’ ἀτρεμίζει κἠπὶ τὴν τέφρην οἰχνεῖ,

κοίμησον· ἴσχε δὲ δρόμου

μαργῶντας ἵππους, μηδὲ δευτέραν κάμψῃς

μή τοι περὶ νύσσῃ δίφρον

ἄξωσιν, ἐκ δὲ κύμβαχος κυβιστήσῃς. αZyCaΠ (Callim. Ia. 5.23–9)

ταῦτα μὲν οὖν οὐ κυρίως εἴρηται· οὔτε γὰρ περὶ πυρὸς οὔτε περὶ ἱπποδρομίας ἐστὶν ὁ λόγος, ἀλλ’ ὥσπερ αἰδούμενος ἐκδήλως εἰπεῖν ὅ βούλεται, ἐχρήσατο τῇ ἀλληγορίᾳ. αZ

MSS Tryphonis: Z = Leidensis BPG 74G; α = consensus codicum praeter Z

ll. 5–11 tradunt etiam CaΠ = P.S.I. 1216; y = Par. gr. 2558, fol. 160v (vide ZPE 213 [2020], 23–7); laudat Choer. = Choerobosci de tropis, pp. 244–56 Spengel

1 κυρίως Z (iam post Stroux add. West): om. α

3 μὴ Z: οὐ α    τὸ πρᾶγμα Ζ, non praeb. α

5 δὲ τὠνέκαυσας Pfeiffer: δετόανέκαυσας CaΠ: δὲ τ᾽ ἀνέκαυσας Choer.: δ᾽ ἀ[.]ὼν (δ᾽ ἑκὼν fort. voluit) ἐκκαύσας y: δ᾽ ἔκαυσας Z: ὅπερ (vel ὅτε) ἀνέκαυσας (vel ἐνέκαυσας) α

ἄχρις οὐ Pf.: άχρισου CaΠ (ἄχρις εὗ olim Pf., Terzaghi, ἄχρι σευ Norsa/Vitelli): ἄχρις οὗ Zy et Choer. MSS aliquot: ἕως (vel ἔρος) οὗ Choer. cett. MSS: om. α πολλῇ πρόσω CaΠy: πρόσω πολλῇ Z: πολλὴν (sed πολλὰ Laur. 87.10) πρόσω α

6 κεχώρηκεν CaΠ: κεχώρηκε Zy Choer. (et Laur. 87.10): κέχρηκε vel κέχρηται (cum πολλὴν … φλόγα) α φλογί Zy Choer.: φλόγι CaΠ: φλόγα α

7 om. α Choer.   ἀτρεμίζει CaΠy (ατρεμιζε a.c. CaΠ): ἔτι ἀτρέμοι Z   κηιπιτην CaΠ, corr. Norsa/Vitelli: κῃπὶ (sc. καὶ εἰ ἐπί) τὴν Pf.: καὶ ἐπὶ τὴν Z: κἤπιος y  ο̣ἰ̣[χ]ν̣εῖ CaΠ (suppl. Norsa/Vitelli): οἰ[κ]εῖ (sive οἴ[κ]ει) Parsons: οἴκοι Z: εἰ[ ̣]ὴ (ex οἰ[ ̣]ὴ corr.) y: ὀκνεῖ Most per litteras

8 κοίμησον CaΠZ: κοιμίσων y: om. α Choer.   δρόμου Zy Choer. (et Pal. gr. 360): δρόμ[ CaΠ: δρόμον α

9 μαργῶντας ἵππους CaΠZy Choer. (et Laur. 87.10): μαργοῦντας ἵππους vel μαργοῦντος ἵππου α, qui reliqua om.  μηδὲ CaΠ: μὴ δὲ Zy: μὴ Choer.  δευτέρ[η]ν̣ suppl. edd. in CaΠ

10 τι Choer. (praeter MS Pal. gr. 40 τοι)  περί CaΠ Z: παρά y Choer.

11 ἄξωσιν post corr. (ἄγωσιν vel ἄγρωσιν scripserat) y   κυβιστήσῃς y Choer. et (]τήσῃς) CaΠ: κυβιστηθείς Z

12 μὲν οὖν Ζ: γὰρ α   οὔτε pr. Z (et Olomuc. M 79): οὐ α   περὶ ambo om. α (prius iam addiderat Walz), praebet Z

13 εἰπεῖν ὅ βούλεται, ἐχρήσατο τῇ ἀλληγορίᾳ Z: ἤλεγξε τὴν ὑπερβολὴν τῆς θρασύτητος α: <εἰπεῖν, διὰ τῆς ἀλληγορίας> ἤλεγξε τὴν ὑπερβολὴν τῆς θρασύτητος ci. West

Allegory is an expression that indicates something in its proper sense, but also presents the meaning of something else. Writers use allegory appropriately when either for reasons of prudence or of decency they refrain from reporting the matter openly, as does Callimachus in his Iambi:

The fire you have lit up, before it has

spread forward with big flames,

but keeps quiet and rests among the embers,

extinguish it: hold back from the race

the raging horses, and don't make a second turn

lest they should crash your chariot

around the turning-post, and headfirst you fall.

These things, however, are not said in their proper sense. In fact, the discourse is not about fire or horse-races, but he uses an allegory, as if ashamed to declare openly what he wants to say.

This fragment has already been discussed in a recent essay after discovering that it is attested (in a much fuller form than in the known manuscripts of Ps.-Trypho and Choeroboscus) in the margins of MS Par. gr. 2558 (y).Footnote 8 While we refer to that article for the discussion of the readings offered by the Parisinus, we emphasize here only a few textual aspects.

The contribution made by Z to the text of Ps.-Trypho is clear (lines 1, 3, 12, 13: the manuscript also admirably confirms conjectures by Walz and West). As for Callimachus’ fragment, the improvements on the text of α, mostly in accordance with the second-century papyrus P.S.I. 1216, are manifold and conspicuous (for example the very existence of lines 7 and 9–11 = Callim. 25 and 17–29; the readings in line 8 = Callim. 26). Aside from orthographical issues (line 5 = Callim. 23 ἄχρις οὗ; line 7 = Callim. 25 καὶ ἐπὶ without the crasis—but Ps.-Trypho must have had κἠπί, see y's κἤπιος) and one word-order blunder (line 5 πρόσω πολλῇ), Z has faulty readings in line 5 = Callim. 23 δ᾽ ἔκαυσας, line 7 = Callim. 25 ἔτι ἀτρέμοι, and line 11 = Callim. 29 κυβιστηθείς (the former two metrically untenable). The issue of line 7 = Callim. 25 οἴκοι is more delicate, as the papyrus traces are uncertain, and the commonly accepted οἰχνεῖ has been restored by Norsa and Vitelli from ο̣ἰ̣[ ̣]ν̣εῖ in the papyrus, while Par. gr. 2558 has εἰ[.]ὴ (corrected from οἰ[ ̣]ὴ): Parsons's conjecture οἰκεῖ, though slightly problematic syntactically, should be carefully considered.

More importantly, the existence of Z now confirms that—as surmised in the aforementioned paper—the quotation in MS y does indeed derive from a lost manuscript witness carrying the fuller version of Ps.-Trypho's treatise on tropes.


§5. Ὑπερβατόν ἐστι φράσις ἀνὰ μέσον τι τῶν ἑξῆς ἔχουσα. γίνονται δὲ τὰ ὑπερβατὰ ἐν εἴδεσι δυσίν, ἤτοι ἐν λέξει ἢ ἐν λόγῳ. […] ἔνιοι δὲ καὶ ἐν συλλαβαῖς ὑπερβατὰ πεποιήκασιν, ὡς καὶ Σιμωνίδης ἐν ἐπιγράμμασιν⋅

Ἑρμῆν τόνδ’ ἀνέθη Δημήτριος Ὀρθιάδου κεν

ἐν προθύροις Δήμη στῆθί τε καὶ μάθε τρος,  ([Sim.] FGE 44 Page =  105 S Sider)

ἄλλοσε προσυπερβιβάσας τοῦ τε “ἀνέθηκεν” καὶ τοῦ “Δήμητρος” τὴν τελευταίαν συλλαβήν. τὸ γὰρ ἑξῆς οὕτως ἀποδίδοται⋅ Ἑρμῆν τόνδ’ ἀνέθηκεν Δημήτριος Ὀρθιάδου· ἐν προθύροις Δήμητρος στῆθί τε καὶ μάθε. ΠZ

MSS Tryphonis: Π = P.Vindob. 29332 (lacunosa); Z = Leidensis BPG 74G; α = consensus codicum praeter Z

1 τι τῶν Z et Marc. gr. 512: τῶν (vel τὰ vel τὸ) α

2 ἤτοι ἐν λέξει ἢ ἐν λόγῳ Z: εἴτε ἐν λέξει, εἴτε ἐν λόγῳ (fere idem coniecerat West) Marc. gr. 512: ἢ ἐν λόγῳ, ἢ ἐν λέξει Barocc. 72: εἴτε ἐν λέξει nec plura α

4–5 Ἑρμῆν τόνδ’ … μάθε τρος post Headlam (qui usque ad προθύροις correxerat) scripsimus: Ἑρμῆν τόνδ’ ὃς ἀνέθηκε Δημήτρϊ, ὄρθια δ’ οὐκ ἐν προθύροις⋅ Δήμητερ στῆθι τε καὶ μάθε Z: Ἑρμῆν τόνδ’ ἀνέθηκε Δημήτριος, ὄρθια δ’ οὐκ ἐν προθύροις, ἀντὶ τοῦ οὐκ ὄρθια δέ α: Ἑρμῆν τόνδ’ ἀνέθηκε Σύρος Δημήτριος, οὐκ εὖ⋅ ὄρθια δ’ οὐ Δήμητρ’ ἔπρεπεν ἐν προθύροις West

6–8 ἄλ\λ/οθε [προ]ϲ̣υπερβ̣[ιβάσας τοῦ τε ἀνέθηκεν καὶ τ]ο̣ῦ̣ Δ̣ή̣μη̣τ̣ρος τὴν̣ τ̣ε̣λευταία̣[ν συλλαβήν. τὸ] γ̣ὰ̣ρ ἑξ̣ῆ̣ς οὕ̣τ̣ω̣[ς ἀπ]ο̣δίδοται⋅ [Ἑρμῆν τόνδ]ε ἀνέθη̣κ̣εν Δη[μή]τ̣ρ̣ιος Ὀρθιά[δου ἐν προθ]ύροις Δήμητρος σ[τῆθι τ]ε̣ καὶ μάθε̣ Π; e MS Z lacunas supplevimus (et ἄλλοσε ipsi correximus, τοῦ τε ex τοῦ δὲ corr. L. Ruggeri per litteras): προσυπερβιβάσας τοῦ δὲ ἀνέθηκε καὶ τῆς Δήμητρος τὴν τελευταίαν συλλαβήν. τὸ γὰρ ἑξῆς οὕτως ἀποδίδοται⋅ Ἑρμῆν τόνδ’ ἀνέθηκε Δημήτριος ὄρθια δ’ οὐκ ἐν προθύροις Δήμητρος στῆθι τε καὶ μάθε Z

Transposition is an expression that presents in the midst some parts of what should follow. Transpositions arise in two forms, either in a word or in a phrase. […] Some have made transpositions also in syllables, as Simonides in his epigrams:

Demetrios, son of Orthiades, dedicat–this herm–ed

in the entrance of the temple of Deme–stand still and learn–ter!

transposing the last syllable of anéthēken and Dēmḗtēr elsewhere. In fact, the sequence must be understood this way: ‘Demetrios the son of Orthiades has dedicated [anéthēken] this herm in the entrance of the temple of Demeter [Dḗmētros]: stand still and learn!’.

This section on hyperbaton is of the utmost importance for the tradition of Ps.-Trypho's treatise, since the verbatim overlap with the Leidensis now confirms beyond any reasonable doubt that the fifth-century Vienna papyrus is indeed, as Paul Maas had understood, a witness of the very same treatise handed down in the medieval codices. The final part of the paragraph, carrying the exegesis of the epigram, is preserved only in the papyrus and in the Leidensis, which rules out the possibility that the scribe of the codex (or his model) could have restored it by way of conjecture.

As for the wording of the lines of ‘Simonides’, MS Z now yields the actual quotation from the epigram, not only its paraphrase in ‘regular’ Greek prose (after τὸ γὰρ ἑξῆς οὕτως ἀποδίδοται), which was the only partly readable section in the Vienna papyrus. The reference to the transposition of the τελευταία συλλαβή proves that the hyperbaton here at stake did involve syllables, and that HeadlamFootnote 9 was therefore on the right path in assuming that the key feature of the first line, however normalized in the manuscript transmission, was the splitting of ἀνέθη–κεν; accordingly, we assume, the second transposition must concern the splitting of Δήμη–τρος.

This is at least how Ps.-Trypho appears to present matters. Some may assume (with Page) a deeper corruption,Footnote 10 and others may even believe that these lines were made up ad hoc by some grammarian.Footnote 11 However, we believe that the authenticity of such an unusual wordplay (where tmesis occurs both times at the caesura and projects the last syllable at line-end, with a subtle game of symmetry) should be seriously considered: ‘Simonidean’ virtuoso pieces are attested (see, for example, FGE 684–5 = CEG 430 with the notorious enjambement Ἀριστο- | γείτων), and precise parallels for similar mots fragmentés, though absent from the corpus of extant Greek lyric, can be found in archaic Latin poetry.Footnote 12 While the corpus of ‘Simonidean’ epigrams offers a limited number of votive pieces,Footnote 13 and while the link between Hermes and Demeter (or Demeter's shrine) is unclear (a pun can easily be imagined with the donor's name), one may assume that the wit in the distich could lie precisely in the bold transposition of syllables (a hyperbaton not imposed on the poet by metrical constrictions with the proper names involved). Thus the final exhortation to ‘stay still and learn’ apparently proceeds from a careful study of the conventions of Greek sepulchral epigrams, urging the passer-by to avoid superficiality, and to pause and pay attention to the stylistic peculiarity for which this epigram stands out, if he wants to learn the names of the donor and the goddess.


§17. Ἀστεϊσμὸς δέ ἐστι φράσις διὰ τῶν ἐναντίων τὸ κρεῖττον ἠθικῶς ἐμφαίνουσα, οἷον εἴ τις πλούσιος ὢν πένης εἶναι λέγει, καὶ ὁ τεχνίτης ἄτεχνος, καὶ ὁ ἀγαθὸς φαῦλος, αUZ ὡς παρὰ Καλλιμάχῳ ἐν Ἰάμβοις τὰ ὑπὸ τῆς ἐλαίας λεγόμενα⋅ UZ

ἐγὼ δὲ φαύλη τ’ εἰμί κοὔτ’ ἔμ’ οἱ μάντεις

οὔθ’ οἱ θύται φορεῦσιν, οὐδ’ ἐπὶ φλιῆς

ἕστηκα· μή με κερτομεῖτε τὴν φαύλην. UZCaΠ (Callim. Ia. 4.90–2)

κατασκευάζει γὰρ αὐτὴν ὡς ἔστι τῆς δάφνης βελτίων τῷ ἑαυτὴν ἀστεϊζομένην φαύλην προσαγορεύεσθαι. UZ ἔνιοι δὲ τὸν ἀστεϊσμὸν ὡρίσαντο προσποίησιν τῆς ἀληθείας. αUZ

MSS Tryphonis: Z = Leidensis BPG 74G; U = Olomucensis M 79; α = consensus cett. codicum praeter UZ

ll. 4–6 in Callimachi iambo IV praebet etiam CaΠ = P.Oxy. 1011, fol. IVv

2 καὶ ὁ τεχνίτης ἄτεχνος, καὶ ὁ ἀγαθὸς φαῦλος α: καὶ ὁ ἀγαθὸς φαῦλος, καὶ ὁ τεχνίτης ἄτεχνος UZ (sed articulos omittit Z)   ὡς Z: καὶ U

4 [ἐγὼ δὲ φαύλη τ᾽ εἰμ] in lac. CaΠ   κοὔτ’ ἐμ’ οἱ Z: κοὔτεμοι U   μάντεις UCaΠ: μάντϊς Z

5 [οὔθ᾽ οἱ θύται φορεῦσι] in lac. CaΠ   οὐδ’ UZ: οὔτ’ CaΠ   φλιῆς Ua.c.ZCaΠ: φλοιῆς Up.c.

6 [ἕστηκα μή με κερ] in lac. CaΠ   ἕστηκα Z: ἔσται καὶ U   κερτομεῖτε Z: κερτόμει U φαύλ̣ην (non δάφνην, ut olim legebatur) etiam CaΠ

7 τῆς δάφνης βελτίων U: τῆς (ex τῇ corr.) δάφνης βελτίωνϊ [sic] Z   τῷ ἑαυτὴν Z: αὐτὴν γὰρ U

8 εἶναι post προσποίησιν praeb. α

Mock-modesty [asteismos] is an expression that gracefully shows what is better through the contrary, as when somebody, albeit being rich, claims he is poor, or the expert unexpert, or the valuable mediocre: as in the words of the olive in Callimachus’ Iambi:

I am mediocre, neither the soothsayers

nor the sacrificers carry me, nor do I stand

on the threshold. Don't sneer at me, as being mediocre. [Callim. Ia. 4.90–2]

He presents her as better than the laurel by having her call herself ‘mediocre’ by way of mock-modesty. Some define the asteismos as a dissimulation of reality.

The manuscript tradition is here clear and easy to analyse: the example from Callimachus’ Iambi has simply been omitted in all witnesses except the Leidensis Z and a second manuscript, an Olomucensis that, like the Leidensis, has not yet been collated and that we shall call U, which otherwise largely agrees with the rest of the tradition (α, see above §0).Footnote 14 In the parallel section on asteismos of the other treatise Περὶ Τρόπων ascribed to Trypho (‘Trypho I’, page 206.16 Sp.),Footnote 15 we do find a reference to a Callimachean line (fr. 93b Schneider = Iamb. 4, fr. 194.13 Pf. †ἐγὼ φαύλη πάντων τῶν δένδρων εἰμί†), which despite its evidently corrupt form has been inserted by Pfeiffer (followed by all subsequent editors) in the large lacuna after line 12 of the fourth Iambus.

Callimachus’ fourth Iambus revolves around the controversy between an olive and a laurel.Footnote 16 The first of the three lines quoted in MSS ZU most probably represents a more correct form of the same line as quoted in the Περὶ Τρόπων ascribed to ‘Trypho I’ (†ἐγὼ φαύλη πάντων τῶν δένδρων εἰμί†, which we think must be a paraphrase of the original verse): despite the absence of any reference to ‘all the trees’ in the line of ZU, and despite the seemingly ‘poetic’ use of the positive φαύλη used for the superlative φαυλοτάτη, it is unlikely that the olive should resort to such an asteismos twice in the same iambus, in virtually the same terms;Footnote 17 indeed, one of the witnesses of Trypho I's treatise, the important MS Marc. gr. 512,Footnote 18 carries this quotation as ἐγὼ δὲ φαύλη τέ εἰμι, which is precisely the incipit of the line as it features in Trypho II's fuller version.Footnote 19 The new find thus suggests that what is quoted by Ps.-Trypho (both I and II) as an outstanding example of mock-modesty is not a claim made by the olive in its opening speech, of which so little is extant,Footnote 20 but rather the concluding outburst of the long tirade (lines 46–92) by which the olive rebuts the laurel's arrogant speech (lines 18–43).

This state of affairs is supported by manuscript evidence from across the centuries. Lines 90–2 of fr. 194 appear in current editions in the following form:

] ̣ικουτεκ̣οι μάντεις
]ν οὔτ᾽ ἐπὶ φλιῆς
φ]ημι τὴν δάφνην

But M.G. Sandri's new inspection of P.Oxy. 1011 has shown that these lines actually read:Footnote 21

]μ̣ί κοὔτ’ ἔμ’ οἱ μ[ά]ντεις
]ν̣, οὔτ’ ἐπὶ φλιῆς
]τομεῖτε τ̣ὴ̣ν φαύλ̣ην

This is fully compatible with the lines as quoted by the Leidensis and the Olomucensis, apart from the trivial oscillation between οὔτ᾽ and οὐδ᾽ in line 91. Indeed, as Pfeiffer had seen, these lines represent an echo of the boastful claims of the laurel about her omnipresence in cult and ritual, in lines 24–5 (τίς δ᾽ οἶκος οὗπερ οὐκ ἐγὼ παρὰ φλιῇ; | τίς δ᾽οὔ με μάντις ἢ τίς οὐ θύτης ἕλκει;).Footnote 22 In her long reply, the olive introduces the dialogue between two crows (lines 64–80), who take on the task of comparing the respective merits of both competing trees.

It has been argued by some scholarsFootnote 23 that lines 90–2, which conclude the olive's speech (line 93 begins ὣς εἶπε), are not spoken by the olive but belong to the dialogue between the crows:Footnote 24 this is now disproved by the new evidence, which indicates that the olive's speech ended on a note of ironical self-deprecation or mock-modesty. It is unlikely that the olive resumed her speech immediately after line 80Footnote 25 or (as Fraser argued) after line 84,Footnote 26 since line 87 καλλίνικος ἡλαίη must still be pronounced by the crows. The fragmentary nature of lines 83–9 prevents us from drawing a firm conclusion on this point,Footnote 27 but it is likely that lines 88–9 contained the first part of the reasoning picked up in line 90 ἐγὼ δέ—perhaps an exhortation to the crows to honour or praise other trees (the pear-tree, line 88 τὴν ὄγχνην, or a better tree, line 89 ] ̣τέρην τιν᾽) as opposed to the olive's own modesty.

There are at least two interesting features in the lines thus recovered: first, the verb κερτομέω (nowhere else in Callimachus), which inscribes the complex relationship between the olive and the crows in the frame of a sophisticated literary game.Footnote 28 Second, the repetition of the adjective φαύλη: first (line 90) in a sort of parodic echo of the laurel's proud statements at line-beginning in line 37 ἱρὴ γάρ εἰμι and line 39 ἁγνὴ γάρ εἰμι;Footnote 29 then (line 92) as the concluding word of the entire speech. It may be argued that this anaphora matches that of δάφνη in the laurel's arrogant words at lines 26–7;Footnote 30 but, more importantly, the place of honour thus attributed to the adjective strengthens Lelli's claim that it represents here an allusion to the particular kind of olive known as φαυλία, while also carrying a stylistic and aesthetic overtone, with reference to the Aristotelian terminology (Poet. 1448a with the opposition between the φαῦλοι and the σπουδαῖοι).Footnote 31 If the parallel between this iambus and the conversation between lady Elegy and lady Tragedy in Ov. Am. 3.1 holds true,Footnote 32 then Ovid's incipit in line 41 (sum leuis, et mecum leuis est, mea cura, Cupido), where Elegy speaks with a similar attitude of understatement and apparent self-depreciation (only to rebound later),Footnote 33 might well be reminiscent of ἐγὼ δὲ φαύλη τ᾽ εἰμί, with a characteristic shift from the ‘humble’ to the ‘tenuous’, from the φαῦλον to the λεπτόν.

4. RIDDLE IN ‘HESIOD’’S WEDDING OF KEYX (FR. 266A.8–11 M.–W. = 204A.8–11 MOST)

§23. Αἴνιγμα δέ ἐστι φράσις διάνοιαν ἀποκεκρυμμένην καὶ σημαινόμενον ἀσύνετον πειρωμένη ποιεῖν, ὡς ἔχει παρ’ Ἡσιόδῳ τὰ περὶ τῆς κύλικος λεγόμενα⋅ αZ

αὐτὰρ ἐπεὶ δαιτὸς μὲν ἐίσης ἐξ ἔρον ἕντο,

οἴνου μητέρα μητρὸς ἐπὶ στόμα χερσὶν ἄγοντο αZHesΠ

ἀζαλέην τε καὶ ὀπταλέην σφετέροισι τέκεσσι

τεθνᾶσιν. ΖHesΠ (Hes. fr. 266a.8–11 M.–W. = 204a.8–11 Most)

οἴνου γὰρ μητέρα μητρὸς λέγει τὴν κύλικα, ἥτις ἐστὶ τὸ ἀνέκαθεν ἡ γῆ, ἥτις καὶ τῆς ἀμπέλου μήτηρ προσαγορεύεται. ὁ δὲ φησὶ Z   ‘ἀζαλέην τε καὶ ὀπταλέην’ ἐπεὶ δοκεῖ πρῶτον ξηραίνεσθαι, εἶτα ὀπτᾶσθαι. ‘σφετέροισι τέκεσσι’, τοῖς ἑαυτῆς τέκνοις, λέγει δὲ τοῖς ξύλοις. τὸ δὲ ‘τεθνᾶσι’, καθὸ δοκεῖ ἐκ τῆς ὕλης ἐκκεκόφθαι. αZ

MSS Tryphonis: Z = Leidensis BPG 74G; α = consensus codicum praeter Z

ll. 3–6 frustula in Hes. Ceycis Nuptiis praebet etiam HesΠ = P.Oxy. 2495, fr. 37

1 διάνοιαν ἀποκεκρυμμένην α: ἀποκεκρυμμένη Z   τὸ (sed τὸ delevimus) σημαινόμενον ἀσύνετον (ἀσύν. iam coniecerat Finckh) Z: σύνθετον α

2 ὡς ἔχει παρ’ Ἡσιόδῳ τὰ Z: ὡς τὰ παρ’ Ἡσιόδῳ α   μηδέ ποτ’ οἰνοχόην τιθέμεν κρητῆρος ὕπερθεν [Hes. Op. 744] post λεγόμενα aliqui codices familiae α praebent

3 ]ἐξ ἔρον ἕντο (p.c.) tantum in fine versus praebet HesΠ   ἐίσης Z et Marc. gr. 512, Laur. 87.10: θίσης vel τεθείσης α (praeter Marc. et Laur.)

4 ].ν ἄγοντο tantum (quod παισ]ὶν ἄγοντο expl. Lobel) in fine versus praebet HesΠ οἴνου Z et Marc. gr. 512: οἷον οὐ α   ἐπὶ στόμα χερσὶν Z: om. α

5 ]τέκεσσι tantum in fine versus praebet HesΠ   ἀζαλέην τε καὶ ὀπταλέην (-έα cod., e l. 8 post West correximus) σφετέροισι τέκεσσι Z: idem iam ci. West

6 τεθνᾶσιν Z: τεθνάναι (sive τεθνάμεναι Bergk) e l. 9 (ubi vide app. crit.) West   ]ον τε καὶ ὄμβρον tantum in fine versu praebet HesΠ

7 οἴνου γὰρ μητέρα … ὁ δὲ φησὶ Z: ἐνταῦθα μητέρα μητρὸς λέγει τὴν βάλανον⋅ ἀπὸ ταύτης γὰρ γίνονται αἱ δρύες, ἀπὸ δὲ τῶν δρυῶν μυθικῶς τοὺς ἀνθρώπους λέγουσι γεγενῆσθαι ci. West

8 τε Z: om. α

9 σφετέροισι vel ἐφ’ ἑτέροισι α: σφετέροις δὲ Z   ἑαυτῆς Kloucek: ἑαυτοῦ αZ

10 ξύλοις Z (iam Cramer): ξένοις α   τεθνᾶσι Z: τεθνάναι α (τεθνάμεναι Bergk) ἐκκεκόφθαι α: κεκόφθαι Z

The riddle is an expression that attempts to conceal a given concept and make a certain meaning unintelligible, as is the case with the words in Hesiod about the wine-cup:

after they had driven away the desire of an equal banquet,

they brought to the mouth with their hands the mother of wine's mother,

dried and baked through its own children

that had died. (Hes. fr. 266a.8–11 M.–W. = 204a.8–11 Most)

In fact, he calls ‘mother of wine's mother’ the wine-cup, which is originally the earth, also called ‘mother of the vineyard’. He says ‘dried and baked’ because it is apparently first dried, and then baked, ‘through its own children’, namely the logs; ‘that had died’, because they had been cut off from the forest.

The Wedding of Keyx is one of the ‘minor’ works attached to Hesiod's name, of which just a handful of fragments remain: its very nature—an autonomous poem or a section of the Catalogue of Women?—has been hotly debated, with the former hypothesis being now more widely accepted.Footnote 34 It probably narrated Heracles’ disembarcation from the Argo at Aphetae on the Pagasaean Gulf and then his unexpected participation in the wedding ceremony of Keyx and Aeolus’ daughter Alcyone at Trachis. In particular, fr. 37 of the second-century P.Oxy. 2495 (fr. 266a M.–W.) has been considered by scholars a witness of this work, because it displays a hexameter-end τρίποδάς τε τραπέζας (the last word is written above the line, just above the original mistaken καθέδρας), which seems to match what Athenaeus says (Deipn. 2.49a; see also Poll. Onom. 6.83 = fr. 266b M.–W.) about the use of the word ‘tripod’ for ‘table’ in the Hesiodic Wedding of Keyx.

Now, lines 8–11 of the same papyrus fragment, though badly flaked, yield a narrative segment that has been convincingly supplemented through the quotation from Hesiod attested in Ps.-Trypho's On Tropes. The Leidensis (Z) now comes to heal the remaining gaps in both Ps.-Trypho's known text and the papyrus, particularly with respect to two points: in line 9 (our line 4), where it guarantees οἴνουFootnote 35 (which no scholar had conjectured from the manuscripts’ οἷον οὐ—the papyrus is lacunose here) and ἐπὶ στόμα χερσίν (all scholars had accepted Lobel's παισίν, and supplemented accordingly);Footnote 36 and in the entire line 10 (our line 5), which had already been brilliantly restored by West (and partly by Finckh) on the basis of Ps.-Trypho's subsequent prose paraphrase. West had understood that the corruption in Ps.-Trypho's text depended on a saut du même au même from one ἀζαλέην (that of the text) to the other (that of the subsequent prose explanation). It is possible that this very mistake prompted the insertion of Hes. Op. 744, which is not in the Leidensis and was probably added in the archetype of one of the two branches of the α-group in order to compensate for the deficiency of the poetic quotation.Footnote 37

However, lacking a wider context for Ps.-Trypho's quotation, West elaborated a totally different explanation for the riddle, assuming:

  • – that περὶ τοῦ κύλικος in Ps.-Trypho's introduction is corrupt (or else refers exclusively to Hes. Op. 744): hence the conjectures περὶ τοῦ ἀκύλου (Merkelbach) and παρὰ τοῦ Κήυκος (West himself);

  • – that the ‘mother's mother’ is the acorn (mother of the oak), and that the solution of the riddle is the Pelasgians (the children of the oaks, according to Stat. Theb. 4.275–81 and other sources), who ‘gathered acorns to die dried and roasted by (for, with) her children’;

  • – that this riddle was proposed by Heracles at the wedding banquet, in the frame of a sympotic contest.Footnote 38

The explanation now available in the Leidensis tells however a different (and more simple) story: it makes clear that the ‘mother of wine's mother’ is indeed—in an ingenious pun—the wine-cup made of clay, that is, of cooked earth, earth being the mother of the vineyard (an idea too obvious to require any parallel), and the vineyard being in its turn the mother of wine.Footnote 39 Along this train of thought, the σφέτερα τέκεα are the wood logs (ξύλα) used for lighting up the fire, which are themselves offspring of the earth (possessive σφέτερα refers to the earth, as the dative is an agent or instrument to the adjectives ἀζαλέην καὶ ὀπταλέην),Footnote 40 but have died upon being chopped away from the trees.Footnote 41 In his Table Talks (730E–F) Plutarch tells us that in the Marriage of Keyx (which he regards as interpolated into Hesiod's corpus by some later poet) there is a riddle alluding to the fact that the fire eats ‘the wood from which it was lit, which was its father and mother’ (fr. 267 M.–W.):Footnote 42 it is hard to imagine that this quotation had nothing to do with the lost poetic context of our fragment,Footnote 43 but specifically what remains of these lines presents the wood rather as an offspring of the earth than as a parent of fire.Footnote 44

Ps.-Trypho's interpretation of the Hesiodic ainigma may be wrong, but we believe that it is methodically wiser to start by taking it seriously. We therefore consider it likely that the narrative inaugurated by these lines (αὐτὰρ ἐπεί) did not belong to Heracles’ intervention (a rather complicated insertion of a narrative-within-a-narrative) but rather to (Ps.-)Hesiod's own voice,Footnote 45 and that the banquet here described is indeed that of Keyx's wedding feast, whose participants simply started drinking wine once they had finished eating.Footnote 46 It is true that there are a number of cases in which riddles are used at agōnes during symposia, but there is no evidence in sources that this should happen in our case; quite the contrary, the references in Plutarch, Athenaeus and Ps.-Trypho, taken at face value, support the idea that the δαίς here implied is precisely that of Keyx. Furthermore, the description of a wine-cup by way of a complex periphrasis is perfectly in keeping with the riddles and kenningar known from Hesiod's poems, such as φερέοικος for ‘snail’ (Op. 571) or the famous periphrastic description of the octopus’ wintry habits ὅτ᾽ ἀνόστεος ὃν πόδα τένδει | ἔν τ᾽ ἀπύρῳ οἴκῳ καὶ ἤθεσι λευγαλέοισιν (Op. 524–5):Footnote 47 this has little bearing on the issue of authenticity (indeed, it might be a good example of how some characters of Hesiod's poetry are picked up and developed in later stages of the epic tradition), but might help better frame a certain use of ainigmata in the context of narrative and sapiential poetry.Footnote 48

We leave to other scholars any further speculation on the consequences of this new find on the structure of the Wedding of Keyx, starting from the problematic presence at the end of line 11 (in P.Oxy. 2495) of the words νιφετ]όν τε καὶ ὄμβρον (suppl. West)—they could indeed refer to the people comfortably sitting around a fireplace while the weather outside is wintry.Footnote 49


Our thanks to Claudio De Stefani, Enrico Magnelli, Glenn W. Most and Luca Ruggeri. Both authors contributed equally to this paper and the responsibility of the contents lies with both, but F. Pontani wrote section 2 and 4 and M.G. Sandri wrote section 1 and 3. The edition of D. Sider, Simonides: Epigrams & Elegies (Oxford, 2020) appeared too late for us to be able to incorporate it fully into our discussion in this article.


1 West, M.L., ‘Tryphon De Tropis’, CQ 15 (1965), 230–48CrossRefGoogle Scholar; since Cramer's edition of some excerpts of this text (‘Anecdota Barocciana’, The Philological Museum 2 [1833], 432–4), the treatise was edited by later editors under the name of the twelfth-century grammarian Gregory of Corinth, although this attribution has no basis in extant manuscripts.

2 Mitteilungen aus der Papyrussammlung der österreichischen Nationalbibliothek in Wien, Neue Serie, vol. 3: H. Oellacher (ed.), Griechische literarische Papyri II (Wien, 1939), 59–61.

3 P. Maas apud H. Oellacher (ed.), Griechische literarische Papyri II (Wien, 1939), 59. But see contra, for example, Conley, T., ‘Byzantine teaching on figures and tropes’, Rhetorica 4 (1986), 335–74CrossRefGoogle Scholar, at 341 n. 14.

4 P. Easterling, ‘From Britain to Byzantium’, in R. Cormack and E. Jeffreys (edd.), Through the Looking-Glass: Byzantium through British Eyes (Aldershot, 2000), 107–20.

5 A. Rollo, Gli Erotemata tra Crisolora e Guarino (Messina, 2012), 83.

6 The watermark of the section of interest in this paper is a Main type Briquet 10750 (Provence, 1529). A description of the manuscript and its contents can be found in K. de Meyier, Codices Bibliothecae Publicae Graeci (Leiden, 1965), 145–7, and particularly in P. Moraux (et al.), Aristoteles Graecus, vol. 1 (Berlin, 1976), 392–3 (available also at The proposed identification of the copyist of fols. 48–149v with the sixteenth-century scribe Michael Kontoleon (handwriting known from Par. gr. 1729; see RGK II.383) is very doubtful for the Aristotle and utterly impossible for the Ps.-Trypho section.

7 The fifth is Trag. Adesp. fr. 569 TrGF, quoted in Ps.-Trypho §4: in this fragment, both the Leidensis and MS Barocci 72 (one of the witnesses of Ps.-Trypho's treatises not previously considered by the editors; it is a grammatical miscellany copied in Crete between the late fifteenth and the early sixteenth century: see H.O. Coxe, Catalogi codicum manuscriptorum Bibliothecae Bodleianae pars prima [Oxford, 1853], 117–25 and P. Krafft, Die handschriftliche Überlieferung von Cornutus’ Theologia Graeca [Heidelberg, 1975], 14–19) carry the participle χρώμενος in the first line (already supplemented by West), which also occurs in the same quotation in the treatises by ‘Trypho I’ (195.16–17 Spengel) and Choeroboscus (247.29 Spengel). In Ps.-Trypho quotations from Homer, the main source of examples for rhetorical tropes throughout the treatise, are also generally fuller and more correct than in other extant witnesses.

8 Pontani, F. and Sandri, M.G., ‘A new manuscript witness of CallimachusIambus 5, ll. 23–29’, ZPE 213 (2020), 23–7Google Scholar.

9 Headlam, W.G., ‘Various conjectures. IV’, Journal of Philology 26 (1898), 92110Google Scholar, at 93. See Page, D.L., Further Greek Epigrams (Cambridge, 1981), 265Google Scholar: ‘The truth is that nothing but Headlam's solution is ever going to make sense of the heading “transposition of syllables”.’

10 It is tempting to surmise that ὄρθια δ᾽ οὔ κεν might conceal a reference to the ὀρθιάζειν of the herm's phallus (see Sider [n. *], ad loc.), but the easiest interpretation of the evidence is a genitive of the masculine name Ὀρθιάδης (attested in the famous late fourth-century inscription from Tenos IG XII 5.2 872 = CIG II 2338: J. Game, Actes de vente dans le monde grec [Lyon, 2008], 133 [§21]).

11 This is S. Timpanaro's suggestion for Ennius’ bold tmesis in ‘Per una nuova edizione critica di Ennio’, SIFC 22 (1947), 179–207, at 196–8 (but he later changed his mind: see next n.).

12 Particularly Ennius, frr. 609 Vahlen saxo cere comminuit brum and 610 Massili portabant iuuenes ad litora tanas (= frr. spur. 5–6 Skutsch, already evoked by W.G. Headlam in the 1898 article (see n. 9 above) and then in his ‘Notes on the Greek lyric poets’, CR 14 [1900], 5–14, at 9). These lines were deemed spurious by O. Skutsch (on the basis of S. Timpanaro, who, however, later changed his mind: Contributi di filologia greca e latina [Firenze, 2005], 232–3) and by several later scholars, but see J. Zetzel, ‘Ennian experiments’, AJPh 95 (1974), 137–40, who offered a number of possible Greek antecedents for both tmesis and apocope, although none as blunt as ours (indeed, M. Fruyt, ‘Mots fragmentés chez Ennius’, Glotta 69 [1991], 243–6 argued that in these fragments the tmesis follows an etymological criterion, for cere-brum and Massili-tanas were word-segmentations mirroring the ancients’ own perception of the structure of those words); see also H. Prinzen, Ennius im Urteil der Antike (Stuttgart, 1998), 307–8.

13 See Bravi, L., Gli epigrammi di Simonide e le vie della tradizione (Rome, 2006), 35–6Google Scholar.

14 The MS Olomouc, Vedecká Knihovna, M 79, written by the well-known Spartan scribe Demetrios Trivolis, active in Greece and at Rome, in the second half of the fifteenth century (RGK I.103: identification by E. Gamillscheg apud A. Guida, ‘Nuovi testimoni di Longo e di Achille Tazio’, Prometheus 7 [1981], 1–10), is more grammatical in nature, since it contains (after a first codicological unit preserving some works attributed to Hesiod) the De Passionibus Dictionum attributed to Trypho (fols. 137–138v), the De Encliticis by Johannes Charax (fols. 138v–141), our Ps.-Trypho's De Tropis (fols. 141–143v) followed by Gregory of Corinth's De Dialectis (fols. 144–145v), and John Philoponus’ Collectio Vocum (fols. 157–159v). This codex has been copied using two different antigraphs, one carrying the epitomized version of the α-group and one carrying the versio plenior found in Z, so that for the first six chapters of the Περὶ Τρόπων U carries the epitomized version, while for chapters 7–19 (chapters 20–6 are missing since the manuscript is mutilated at the end) it preserves the versio plenior: this is precisely the reason why only U helps us in the constitutio of this fragment.

15 Tryph. I Trop. 206.12–17 Spengel: Ἀστεϊσμός ἐστι λόγος ἀφ’ ἑαυτοῦ διασυρτικὸς γενόμενος, ὡς εἴ τις πλουτῶν λέγει, ἐγὼ δέ εἰμι πάντων πενέστατος, καὶ ὁ πάντας καταπαλαίων λέγει ὑπὸ πάντων πίπτειν. παρὰ δὲ Καλλιμάχῳ ἀστεϊζομένη ἡ ἐλαία φησίν⋅ ‘ἐγὼ φαύλη πάντων τῶν δένδρων εἰμί’. καλεῖται δὲ τοῦτο καὶ προσποίησις.

16 The fullest discussion can be found in A. Kerkhecker, CallimachusBook of Iambi (Oxford, 1999), 85–115.

17 Repetition, however, does occur in the laurel's speech, e.g. at the opening of lines 18, 28 and 37 ὤφρων ἐλαίη.

18 Fols. 64–6 (but the treatise is preserved only in its final part, 201.12–206.22 Sp.). This codex also contains Trypho II's Περὶ Τρόπων (fols. 53v–58v).

19 The only varia lectio being τέ instead of τ’. The manuscript tradition of Trypho I's treatise has not yet been investigated exhaustively, but on the basis of preliminary collations we believe that MS Marc. gr. 512 might be the only witness of a peculiar branch of the tradition.

20 Hence (despite the arguments brought by Kerkhecker [n. 16], 88–9) what is now line 13—though of course not necessarily the idea it carries—should probably be removed from its place.

21 In line 90 we read surely ]μ̣ικουτεμ̣οι, not ] ̣ικουτεκ̣οι (the second vertical stroke of the first -μ- is clearly visible, and so is the second -μ-, albeit partly lost in a hole; the -α- of μάντεις, on the contrary, cannot be read); in line 92 ]τομεῖτε is very easy to read, whereas scanty traces remain of the first two letters of τήν; as for φαύλην, we can detect beyond doubt a cup-shaped υ and, on the left under its arch, a smallish α; of the λ, only the upper stroke is visible.

22 On these lines, see Kerkhecker (n. 16), 91–2.

23 From Gallavotti to D'Alessio and Kerkhecker ([n. 16], 101–2 and 107–8, who makes the fullest argument and refers to earlier literature); it is taken for granted by B. Acosta-Hughes, Polyeideia (Berkeley – Los Angeles – London, 2002), 203.

24 Οne of the main arguments for not ascribing these lines to the olive was that they included the word δάφνη, which the olive never speaks elsewhere in the iambus (Kerkhecker [n. 16], 107–8). But δάφνη, as we have seen, should not be read in line 92.

25 This seems to be implied by D.L. Clayman, Callimachus’ Iambi (Leiden, 1980), 25–6.

26 True enough, after the olive's aside in lines 81–2, at least lines 83–4 must still be spoken by the crows, for the form of the question faithfully reproduces that of line 79 τεῦ γάρ.

27 Some tentative hints on these lines are provided by Kerkhecker (n. 16), 108, who believes that they should represent a reply, in reverse order, to the arguments laid out by the laurel in her first speech (but according to Kerkhecker, as we have seen, the lines are spoken by the crows).

28 On the interplay of changing and intertwined perspectives in this Iambus, see R. Scodel, ‘Callimachus and fable’, in B. Acosta-Hughes, L. Lehnus and S. Stephens (edd.), Brill's Companion to Callimachus (Leiden and Boston, 2011), 376–9. The verb in the frame of problematic dialogues is current since as early as Od. 8.153, but its meaning in Homer is hotly debated: Clarke, M., ‘Heartcutting talk: Homeric κερτομέω and related words’, CQ 51 (2001), 329–38CrossRefGoogle Scholar.

29 This is the ordering suggested by Maas, while the papyrus (followed by Pfeiffer) has the two similar incipits following one another in lines 39–40: on the philological problem, see Kerkhecker (n. 16), 94–5.

30 On which Acosta-Hughes (n. 23), 200.

31 See E. Lelli, Critica e polemiche letterarie nei Giambi di Callimaco (Alessandria, 2004), 65–6 and more broadly 47–82 for his complex and ingenious metapoetic reading of the entire fourth iambus—an approach largely discarded in recent scholarship.

32 Acosta-Hughes (n. 23), 192 n. 64; B. Acosta-Hughes and S. Stephens, Callimachus in Context (Cambridge, 2012), 259. On the literary meaning of Ov. Am. 3.1, see M. Wyke, The Roman Mistress (Oxford, 2007), 115–54; J.M. Blanco Mayor, Power Play in Latin Love Elegy (Berlin and New York, 2017), 87 and n. 107.

33 The reference to the φλιή in Callimachus might be consciously reversed by Elegy's insistence on the limen (cf. line 50 liminis adstricti sollicitare fidem).

34 The only available reconstruction is that offered by Merkelbach, R. and West, M.L., ‘The wedding of Ceyx’, RhM 108 (1965), 300–17Google Scholar, to which all subsequent scholars are indebted; for an updated overview, see E. Cingano, ‘The Hesiodic corpus’, in F. Montanari and A. Rengakos (edd.), Brill's Companion to Hesiod (Leiden, 2009), 91–130, at 125–6.

35 This correct reading is also shared by the late thirteenth-century MS Marc. gr. 512.

36 West had conjectured for line 9 δὴ τότε μητέρα μητρὸς ἑοῖς σὺν παισὶν ἄγοντο, Merkelbach καὶ τότε μητέρα μητρὸς ἑῆς ἐπὶ παισὶν ἄγοντο.

37 Bona, M. Della, ‘Gare simposiali di enigmi e indovinelli’, QUCC 104 (2013), 169–82Google Scholar, at 179–80 tries to save the reference to this line as an αἴνιγμα κατ᾽ ὅμοιον (sic), but her explanation looks somewhat strained.

38 West, M.L., ‘Hesiodea’, CQ 11 (1961), 130–45CrossRefGoogle Scholar, at 143–4 (subsumed in the argument made by Merkelbach–West [n. 34], 311–12); Zonas of Sardis in Anth. Pal. 9.312 does not support West's view (the oak is mother of the acorn, not vice versa), and ἐκκεκόφθαι would be a bit odd for a fruit like the acorn (much more appropriate for wood logs). West's explanation (which has bearings on the accommodation of fr. 268 M.–W. on the Fatherless ἀπάτωροι) is inherited by all modern scholarship, including G.B. D'Alessio, ‘Ordered from the Catalogue: Pindar, Bacchylides, and Hesiodic genealogical poetry’, in R. Hunter (ed.), The Hesiodic Catalogue of Women. Constructions and Reconstructions (Cambridge, 2005), 217–38, at 233 n. 65 (an essential overview of the possible reminiscences of the Wedding of Keyx in later poetry); S. Beta, Il labirinto della parola (Turin, 2016), 64–6; and I. Konstantakos, ‘Αίνος, αίνιγμα, μυθόγριφος’, Δόσις Ἀμφιλαφής. Τιμητικός Τόμος Κ. Συνοδινού (Ioannina, 2020), 251–89. On riddle contests at banquets, see Beta (this note), 44–62; Della Bona (n. 37); Monda, S., ‘Gli indovinelli letterari antichi come testimonianza di contesti ludici e agonali’, Enthymema 23 (2019), 390400Google Scholar; Potamiti, A., ‘Playing at riddles in Greek’, GRBS 55 (2015), 133–53Google Scholar.

39 ἄμπελος as the mother of οἶνος is recurrent in Greek poetry from Aesch. Pers. 617 to Pind. Nem. 9.51–2 and Eur. Alc. 757. A different riddle on Dionysus’ mother, taking into account Zeus's μηρός and Semele's story, is in Anon. Anth. Pal. 14.31.1.

40 That the dative might go with ὀπταλέην was already envisaged by Merkelbach–West (n. 34), 312, though discarded by them as hardly compatible with their overall interpretation of the lines.

41 τεθνᾶσιν in the Leidensis would be untenable as an indicative, but could make sense as a dative plural to be taken with τέκεσσι, in the sense of τεθνηῶσιν, along the doctrine of e.g. EM 523.53 Gaisford.

42 On the context of Plutarch's quotation, and the reference to Anaximander's testimonium 30 DK, see Braccini, T. and Pellizer, E., Plutarco. Conversazioni a tavola (Naples, 2014), 167 and 278Google Scholar.

43 Indeed, following Th. Bergk (Kleine philologische Schriften, vol. 2 [Halle, 1886], 746, 752), A. Rzach grouped both Ps.-Trypho's quotation and Plutarch's as fr. 168 of his edition, though of course he had no knowledge of the papyrus.

44 Conclusive arguments against Rzach's solution have been brought by West (n. 38), 143; indeed, μητέρα καὶ πατέρ᾽ in Plutarch's text might well be a quotation from a different line. See, however, Merkelbach–West (n. 34), 313 n. 40.

45 This was denied, in view of the ‘Pelasgian’ theory, by West (n. 38), 144–5. Merkelbach–West (n. 34), 307 added the argument that line 4 of P.Oxy. 2495 fr. 37 ends ]σωσα (‘not likely to be anything but ἔσωσα or a compound’), but the first σ is uncertain, and the letters are also perfectly compatible with a feminine present participle from a verb in -άω. That αὐτὰρ ἐπεί should be taken as the formulaic introduction to the symposium, rather than as Heracles’ words, had been already surmised by Della Bona (n. 37), 179.

46 This was denied, on the ground that ‘no one starts eating at the same time as he finishes’ (but it is drinking, not eating, that is at stake here), by West (n. 38), 145.

47 Beta (n. 38), 25–9 with earlier literature; H. Troxler, Sprache und Wortschatz Hesiods (Diss., Zürich, 1964), 21–8. Not the ainos but the ainigma form: P. Cobetto Ghiggia, ‘Αἶνος e αἴνιγμα nella Grecia classica’, in S. Monda (ed.), Ainigma e griphos (Pisa, 2012), 81–97.

48 On this aspect, see Ercolani, A., ‘Fragments of wisdom, wisdom in fragments’, in Tsagalis, C. (ed.), Poetry in Fragments: Studies on the Hesiodic Corpus and its Afterlife (Berlin and New York, 2017), 2946Google Scholar, at 32–3, who however does not discuss the exact meaning of the fragment.

49 In this direction also Merkelbach–West (n. 34), 313.

You have Access Open access
Cited by

Save article to Kindle

To save this article to your Kindle, first ensure is added to your Approved Personal Document E-mail List under your Personal Document Settings on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part of your Kindle email address below. Find out more about saving to your Kindle.

Note you can select to save to either the or variations. ‘’ emails are free but can only be saved to your device when it is connected to wi-fi. ‘’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.

Find out more about the Kindle Personal Document Service.

Available formats

Save article to Dropbox

To save this article to your Dropbox account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you used this feature, you will be asked to authorise Cambridge Core to connect with your Dropbox account. Find out more about saving content to Dropbox.

Available formats

Save article to Google Drive

To save this article to your Google Drive account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you used this feature, you will be asked to authorise Cambridge Core to connect with your Google Drive account. Find out more about saving content to Google Drive.

Available formats

Reply to: Submit a response

Please enter your response.

Your details

Please enter a valid email address.

Conflicting interests

Do you have any conflicting interests? *