P.OXY. 2438 AND THE ORDER OF BOOKS IN ARISTOPHANES BYZANTIUS’ EDITION OF PINDAR

Two well-known ancient witnesses report that Aristophanes of Byzantium was responsible for the arrangement of Pindar's poems into seventeen book-rolls according to lyric genres (dithyrambs, hymns, etc.). These witnesses form fr. 381 in the edition of Aristophanes’ fragments by W.J. Slater (Aristophanis Byzantii fragmenta [Berlin and New York, 1986]): Vit. Pind. P.Oxy. 2438.35–9 (LDAB 3724, TM 62542; late second/early third century a.d.) δ]ιῄρητα̣ι̣ δὲ α̣ὐ̣τ̣[ο]ῦ̣ τ̣[ὰ ποιήματα ὑπ’ Ἀριστοφάν]ους εἰς βιβλία ιζˊ· διθ̣[υ]ρά̣[μ]βων βˊ [προσοδίω]ν̣ βˊ παιάνων αˊ πα[ρ]θεν[εί]ων γ̣ˊ [ἐπινικίω]ν̣ δˊ ἐγκωμίων αˊ ἐν [ᾧ] κα̣ὶ [σκ]όλ̣[ια ±4 ὕμ]ν̣ων αˊ ὑ[π]ορχημάτων αˊ θρ̣[ήνων. | nisi aliter ind., omnia suppl. Lobel | 2 Ἀριστοφάν]ους Lobel e Vit. Pind. Vat. (q.v.) | 3 ἐν [ᾧ] κα̣ὶ [σκ]όλ̣[ια …. de Kreij : ἐν [ᾧ] κ[αὶ σκόλιά τινα vel ἐν [ᾧ] κ[αὶ σκόλιά ἐστι(ν) Gallo (1968, 73–4; 1969, 107) : ἐν [ᾧ] καὶ [παροίνια D'Alessio (2000) | 4 αˊ post θρ̣[ήνων coni. edd., fort. recte; an θρ̣[- βˊ? Vit. Pind. Vat. page 7.14–17 Dr. ὁ δὲ ἐπινίκιος οὗ ἡ ἀρχὴ “Ἄριστον μὲν ὕδωρ” προτέτακται ὑπὸ Ἀριστοφάνους τοῦ συντάξαντος τὰ Πινδαρικὰ διὰ τὸ περιέχειν τοῦ ἀγῶνος ἐγκώμιον καὶ τὰ περὶ τοῦ Πέλοπος, ὃς πρῶτος ἐν ῎Ηλιδι ἠγωνίσατο. A much-debated question is whether the order of the poems attested by the papyrus can be credited to Aristophanes, since an ordering principle is apparently not to be found in the list transmitted by it. Moreover, there are some disagreements with the list of Pindar's works transmitted by the Vita Ambrosiana, which is generally deemed the most authoritative catalogue of the poet's corpus, on account of its well-recognized criterion of classification: the poems are divided between those for gods (from hymnoi to hyporchēmata) and those for men (from enkōmia to epinikia), and both of these categories of songs are arranged from the most general (respectively, the hymns and the enkōmia) to the most specific (the remaining genres). This catalogue has been thought to reproduce the order established by Aristophanes in his edition of Pindar.

A much-debated question is whether the order of the poems attested by the papyrus can be credited to Aristophanes, since an ordering principle is apparently not to be found in the list transmitted by it. Moreover, there are some disagreements with the list of Pindar's works transmitted by the Vita Ambrosiana, which is generally deemed the most authoritative catalogue of the poet's corpus, on account of its well-recognized criterion of classification: the poems are divided between those for gods (from hymnoi 1 On the papyrus and its content, see M. de Kreij, 'Anonymous, Pindar (1132)', in Die Fragmente der Griechischen Historiker Part IV (2017), <http://dx.doi.org/10.1163/1873-5363_jciv_a1132>; this edition, based on a fresh collation, has been reproduced here. As de Kreij (this note) points out, 'after inspection, cleaning, and conservation of the papyrus', he has been able to revise some readings of the editio princeps and 'to better establish the size of the lacunae'. Previous editions are E. Lobel, Oxyrhynchus Papyri 26 (Oxford, 1961) and I. Gallo, Una nuova biografia di Pindaro (P. Oxy. 2438), (Salerno, 1968). In the critical apparatus 'D'Alessio (2000)' refers to G.B. D'Alessio, 'Review of S. Schröder, Geschichte und Theorie der Gattung Paian (Stuttgart andLeipzig, 1999)', in BMCR 2000.01.24;'Gallo (1969)' refers to I. Gallo, 'Gli σκόλια di Pindaro. Nota critica al catalogo pindarico del Papiro di Ossirinco XXVI, 2438', QUCC 8 (1969 The label Vita Vaticana has been introduced by Gallo instead of the traditional label, Vita Thomana, which is misleading, since the biography cannot be credited to Thomas Magister: it was already known to Eustathius, who not only quoted ample extracts from it in his Introduction to Pindar (25-38 Negri) but also ascribed it to ancient commentators (παλαιοί). The ascription to Thomas is based on a brief note by Demetrius Triclinius (E. Abel [ed.], Scholia recentia in Pindari epinicia, 1 [Budapest and Berlin, 1891], 31), who, however, only says that the text of the biography διωρθώθη ('was corrected' or 'was revised'; cf. LBG s.v.) by Thomas, not that it was compiled by him (γένος Πινδάρου διωρθώθη δὲ παρὰ τοῦ σοφωτάτου Μαγίστρου). On the whole issue, see I. Gallo, 'Eustazio commentatore di Pindaro', QUCC 25 (1977), 43-51, at 50; M. Negri, Pindaro ad Alessandria. Le edizioni e gli editori (Brescia, 2004), 16 n. 3, 23 with n. 4.
The Classical Quarterly 70.2 822-826 © The Author(s), 2020. Published by Cambridge University Press on behalf of The Classical Association. This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution, and reproduction in any medium, provided the original work is properly cited.
to hyporchēmata) and those for men (from enkōmia to epinikia), and both of these categories of songs are arranged from the most general (respectively, the hymns and the enkōmia) to the most specific (the remaining genres). 3 This catalogue has been thought to reproduce the order established by Aristophanes in his edition of Pindar. 4 Here is a comparison between the two lists: 5 P.Oxy. 2438 (17 books) The total number of books (17) is the same in both lists, and the poetic genres are the same. Beside ordering, the other main divergence lies in the number of scrolls containing hyporchēmata: two in the Vita Ambrosiana, only one in P.Oxy. 2438. The papyrus list, as it stands, is one book short of the total number. It is impossible to establish with sufficient probability whether the figure αˊafter ὑπορχημάτων (line 3) is mistaken 6 or whether a reference to the seventeenth book of the edition, corresponding to the second volume of hyporchēmata, is lost in the gap following the mention of thrēnoi (θρ ̣ [ήνων α´ ±3 7 | ±2 ]ων †νεικ[]ο ̣λ ̣ε ̣γ ̣ ων † κα̣ι ̣π ̣[ ̣ ]…[). 8 None the less, two considerations favour the first possibility: in the string of letters following the gap, the expected figure αˊis missing; 3 Vit. Pind. Ambr. page 3.6-9 Dr. γέγραφε δὲ βιβλία ἑπτακαίδεκα· ὕμνους, παιᾶνας, διθυράμβων β′, προσοδίων β′, παρθενείων β′· φέρεται δὲ καὶ γ′ ὃ ἐπιγράφεται κεχωρισμέν<α τ>ῶν (Prodi: κεχωρισμέν<ον τ>ῶν Snell) παρθενείων· ὑπορχημάτων β′, ἐγκώμια, θρήνους, ἐπινίκων δ′. On this list, see E.E. Prodi, 'The list of Pindar's works in the Vita Ambrosiana', RhM 161 (2018), 236-7. 4 Since E. Hiller, 'Die antiken Verzeichnisse der pindarischen Dichtungen', Hermes 21 (1886), 357-71; but see already A. Boeckh, Πινδάρου τὰ σωζώμενα. Pindari opera quae supersunt, vol. 2/2 (Leipzig, 1821), 555-7. 5 I do not take into account either the list offered by Eust. Intr. Pind. 342, since it depends upon the Vita Ambrosiana (cf. Negri [n. 2], 216 with n. 1), or the lists provided by the Vita metrica and the Vita Vaticana, for they exhibit only a selection of the poems. As for the Suda's list (π 1617 A.) and its peculiarities, see Gallo (n. 1 [1968]), 42-5. More generally, on the different lists, see Gallo (n. 1 [1968]), 27-45 and Gallo (n. 1 [1969] 'Gallo believes about 6 letters are missing, but these are all wide letters, as is the number, so the column may have ended after the number; at most 3 letters could be added. It is clear that the right margin was not dead straight in any case (see 6 and 9)'. furthermore, it seems difficult to restore in the gap the name of a poetic genre traditionally associated with Pindar (including the five supplementary titles recorded by Suda). 9 A third, hitherto neglected, possibility is that the figure following θρ ̣ [ήνων could have been β´ instead of α´, but this would introduce a further disagreement between the two lists, which seem otherwise pretty similar: though this option cannot be completely discarded, it does not seems to be very appealing.
As for the differences of ordering, Race and Cameron concluded that there was not a fixed editorial arrangement in antiquity, and that the rolls containing Pindar's poems were to be freely rearranged by scholars. 10 On the other hand, D'Alessio has pointed out that 'Hellenistic and later quotations and papyri seem all to reflect the 17 books division present in the vita Ambrosiana and this must be due to the existence of a single authoritative edition. The attribution of what is practically the same list in P. Oxy. 2438 to [Aristophan]es confirms that it was his work in this field which had set the standard followed by subsequent scholars, scribes and book publishers.' 11 The same stance is held by Negri ([n. 2], 213-25), who explains the divergences between the two lists as a consequence of the different natures and aims of the scholarly works in which they are found. While P.Oxy. 2438 was probably part of a collection of biographies, 12 whose compiler was not interested in accurately reproducing the order of books established by Aristophanes of Byzantium, the Vita Ambrosiana was devised to introduce an edition of Pindar, and therefore the exact order of the book was relevant to its compiler. Other scholars, though accepting the existence of a single authoritative edition by Aristophanes, have also suggested that the different orders of books may reflect different arrangements proposed by ancient scholars: Arrighetti, for instance, traced back the papyrus' list to Didymus 13 and that of the Vita Ambrosiana to Aristophanes, while Gallo ([n. 1 (1968)], 35, 41-2) took the opposite stance. 14 It seems difficult to deny the existence of an authoritative edition in seventeen books, namely the edition provided by Aristophanes, with its own arrangement. 15 Which of the transmitted lists reflects such an arrangement is not easy to ascertain, but the only list explicitly associated with the Alexandrian grammarian is the list provided by P.Oxy. 2438. The book order of this list is generally considered to lack a well-recognizable criterion, but is only apparently chaotic: upon closer scrutiny, it betrays a substantial bipartition between songs εἰς θεούς (lines 36-7: dithyrambs, prosodia, paeans, partheneia) and songs εἰς ἀνθρώπους (lines 37-9: epinikia, enkōmia, thrēnoi), 16 with the only exceptions being hymnoi and hyporchēmata (lines 38-9), both of which are misplaced in the second group of songs. The misplacement, possibly owing to an oversight of the scribe, is, none the less, telling, for it involves two terms that begin with the same letter (υ-). Once these two genres are moved to the end of their category-that is, the songs εἰς θεούς-the resulting order turns out to conform to two clearly recognizable criteria: a hierarchical principle based on the recipients of the song (gods/men), and, within both of these groups, an alphabetic ordering extending no further than the first letter, as seems to have been customary in the Hellenistic period (διθύραμβοι, προσόδια, παιᾶνες, παρθένεια, ὕμνοι, ὑπορχήματα / ἐπινίκια, ἐγκώμια [with σκόλια], θρῆνοι). 17 These two criteria match those which were arguably used in Alexandrian editions of at least some melic poets. The hierarchical principle seems to govern Aristophanes' edition of the four books of Pindar's victory songs (viz. the hierarchy of both Panhellenic games and agonistic disciplines: first the chariot race, then the other equestrian competitions, followed by contact sports and foot races), 18 while the alphabetic principle seems to be recognizable in the editions of Bacchylides (within the book of dithyrambs) and Sappho (as a subordinate criterion governing the arrangement of the poems within each thematic section), 19 which are generally ascribed to Aristophanes. 20 Furthermore, the bipartite structure of Pindar's edition (songs εἰς θεούς and εἰς ἀνθρώπους) seems to be confirmed by the selected list of Pindaric poems given by Horace in Carm. 4.2.10-24 (the celebrated ode Pindarum quisquis studet aemulari), 21 where only dithyrambs (10)(11)(12), hymns (13-16), epinikia (17-20) and thrēnoi (21-4) are mentioned, that is, two genres for each group, in both cases in alphabetical order. 22 It is also 'very interesting that P.Oxy. 2438 […] begins with the dithyramboi and ends with the threnoi exactly as Horace does' (Race [n. 5], 410). These agreements, taken together, suggest that the Roman poet may have had Aristophanes' edition in mind when he outlines the poetic work of his model.
Compared with the papyrus list, the list provided by the Vita Ambrosiana exhibits an order which seems to be more refined, as noted above. It is perhaps not irrelevant to acknowledge here the work of Didymus on lyric poetry-as, for example, Gallo ([n. 1 (1968)], 40-1) did-which is generally considered to be the basis for Proclus' classification (Chrest. 32-3 Severyns). 23 If so, the differences between the two lists could be explained as the result of different attempts to arrange Pindar's poems. There is evidence that the discussion on the classification of the odes continued after Aristophanes' edition, and that 'Didymos himself disagree [d] with the classification of Nemean XI among the Nemeans: it should not even have been inserted among the Epinicia. In this he was following Dionysios of Phaselis. He thought that Nemean XI, composed for the investment ceremony of Aristagoras as a πρύτανις in Tenedos, was not a victory ode but a παροίνιον, a "drinking song"' (D'Alessio [n. 11], 54). As well as in the matter of individual poems, Didymus might have proposed some change of Aristophanes' arrangement in the ordering of poetic genres too.  , 1957], 432-3). 22 Cf. Cazzaniga (n. 14), 5-8; Lehnus (n. 14), 81-3 with n. 102. I follow Lehnus in taking Carm. 4.2.13-16 as a reference to the hymns: compare the expression seu deos regesque canit, deorum | sanguinem with Heliod. in schol. Lond. Dion. Thrax 1, GG I/3 451.6 Hilgard ὕμνος ἐστὶ ποίημα περιέχον θεῶν ἐγκώμια καὶ ἡρώων μετ' εὐχαριστίας. For the bipartition between songs for gods and songs for men, see already Pl. Resp. 10.607a (ὕμνους θεοῖς καὶ ἐγκώμια τοῖς ἀγαθοῖς). 23 On the relationship between Didymus and Proclus, see A. Severyns, Recherches sur la Chrestomathie de Proclos, vol. 2 (Liège and Paris, 1938), 114, who considers Proclus' classification a development of Didymus'; also R. Pfeiffer, History of Classical Scholarship. From the Beginnings to the End of the Hellenistic Age (Oxford, 1968), 184; L.E. Rossi, 'I generi letterari e le loro leggi scritte e non scritte nelle letterature classiche', BICS 18 (1971), 69-94, at 74-5 with n. 20.