Dirk Obbink

“The New Sappho” actually comprises a group of papyrus fragments, quotations, and testimonia for Sappho’s poetry dating back more than two millennia. Scholars who were amazed to learn that Sappho had “composed a new poem” when Edgar Lobel published it a half-century ago—she had, after all, been dead for over 2600 years—would have been even more surprised at finding the same fragmentary poem’s missing line-ends, conjecturally and limpingly supplied by scholars since Lobel, now actually supplied by another papyrus manuscript of the same poem, overlapping but preserving the other ends of the lines, that appeared as late as 2004. The result, like so much else in the archaeology of our textual knowledge of the ancient past, has raised more questions than it has answered about the complex interrelations of language, performance, and context of poetry, as well as its transmission and reception. Not all of these are likely or even able to be solved by the simple constitution of texts, reporting of witnesses, and compilation of suggestions. What follows presents the textual basis for the discussions of the papers in this volume. It has the same basic form and function as David Sider’s edition of Simonides’ elegiac fragments 1–22 W² (Sider 2001). It critically collates the evidence of the manuscript witnesses, both for the ancient editions of Sappho’s poems, and for the testimony, ancient and medieval, relating to these two fragments. As an appendix, the text as originally restored and translated by M. L. West is reproduced exempli gratia, and against which other translations, interpretations, and emendations are critically compared in the contributions to this volume.

Oxyrhynchus Papyrus 1787 was recovered from the ancient land-fill dump on the outskirts of the provincial capital city of Oxyrhynchus in central Egypt by B. P. Grenfell and A. S. Hunt during their excavations there in 1898–1907, and is now in the Papyrology Rooms of the Sackler Library, University of Oxford. As published by E. Lobel in The Oxyrhynchus Papyri vol. XXI, it contains numerous fragments of what was once a professionally produced, critical edition of book 4 of Sappho’s poetry in one edition circulating in the Roman period, as can be seen from the colophon (in fr. 44) containing the fragmentary remains of Sappho’s name and the title of the collection (Mele). The book-number itself is not preserved in the papyrus, but is recoverable by ancient references to the book containing poems solely in this meter. The handwriting enables us to date the copying of the manuscript to the later second century AD. Fr. 2 of this papyrus preserves the line-ends of the verses that came to be known as Sappho fr. 58, together with the line-beginnings of the meager fr. 59.

Cologne Papyrus inv. 21351 was acquired from an antiquities dealer in 2004 by the University of Cologne, and was published by R. W. Daniel and M. Gronewald in Zeitschrift für Papyrologie und Epigraphik in 2004; subsequently, in the same year, an additional fragment (inv. 21376) was identified and placed in the ensemble of fragments, the remains of a papyrus roll containing poems by Sappho and at least one other poet, copied by two different hands in the early part (first quarter or a little later) of the third century BC. The fragmentary papyrus manuscript had been dismantled from painted mummy-cartonnage, viz. covered with a thin layer of gesso plaster and painted, of which the fragments still bear traces. A site in the Fayum, the rich agricultural oasis southwest of Cairo, and the only area in Egypt known certainly to evidence the practice of using recycled papyrus for decorative funerary art, may be assigned as the provenance.

Sappho’s authorship of verses in the Cologne fragments is secured by their overlap with the Oxford manuscript of Sappho in fr. 58: Cologne supplies the earlier portion of the lines (not preserved in the other papyrus), Oxford supplies the ends of the lines, while the two manuscripts overlap for a thin strip of several centimeters in the middle. A new set of verses, referred to below as the “New Fragment,” however, appears before the verses about Tithonus in the Cologne manuscript, entirely different from those in the Oxford one (termed below “Success Poem”) and that stand here in modern editions of Sappho. In both manuscripts there follow the verses held in common by the two manuscripts, called below (in the absence of ancient testimony for a title) the “Tithonus Poem.” After these, in Cologne a series of verses follow that cannot be by any Lesbian/Aeolic poet (for they are in a meter that admits a series of more than two successive short syllables); in Oxford there follow verses which are certainly of Sapphic pedigree, since Athenaeus quotes them (in a defective form partially garbled in our medieval manuscripts) and attributes them to the poetess. In the text below, the former is called “Continuation 1” (its priority reflecting the fact that it is witnessed by the earlier manuscript), and the latter “Continuation 2.”

The terms “New Fragment,” “Success Poem,” and the Continuations 1–2 will become better understood at a later stage of this book, as they are more fully explained by the contributions below. For convenience of reference here, I have numbered each of these sections independently; it should be noted, of course, that the verses of the “Tithonus Poem” continue on after the “New Fragment” and the “Success Poem,” in Cologne and Oxford respectively: i.e. the first verse of the “Tithonus Poem” is actually line 9 of the Cologne papyrus, but is line 11 of the Oxyrhynchus papyrus (and similarly with Continuations 1 and 2).

In both the Cologne and Oxford papyrus manuscripts, the distichs (Asclepiadean verses grouped into pairs) are separated by a short horizontal line (paragraphos) penned in between two verse-beginnings, extending slightly into the blank intercolumnium before and slightly into the column of writing. Here and there, both manuscripts also bear final graphics or coronides, which are known to have marked the ends of poems in ancient manuscripts of Sappho. Where the left margin of a column of writing survives to such an extent as to preserve these, they provide an invaluable control of the point at which ancient editors and readers thought one of these poems ended and a new one began (as at the end of the “Tithonus Poem” in the Cologne manuscript). But where the left margin does not survive to determine the placement of a coronis (as at the end of the “Tithonus Poem” in the Oxford manuscript), we cannot be certain that in this case the ancient reader was warned of a division between poems here, especially when (as in the case of the Oxford and Cologne manuscripts at precisely this point) different sets of verses follow.

Much debate and discussion focuses on considerations of division, whether at the beginning and ending of poems, or on possibilities for the grouping of verses transmitted by both manuscripts into poems, and, of course, the “vexed problem of the ending” (as one of the anonymous press-readers for this volume put it, mildly). It therefore seemed useful, in the critical edition offered below, to represent graphically these lectional signs as they occur in one or both manuscripts at the left edge of the columns of writing. The sign ⊗, by contrast, placed in the right margin at line-end, denotes (as conventionally in modern editions since Voigt) that, in the judgment of the modern editor, a division between poems occurred (or should be posited) at this point, regardless of the corresponding presence, or preservation, of the final graphic (coronis).

The economy of the evidence suggests that, as in West’s reconstruction, the verses transmitted in common by both manuscripts (i.e. the “Tithonus Poem”) ought to constitute a complete poetic unit, in and of itself. And this seems to be borne out for the unusually abrupt conclusion of the “Tithonus Poem” by the presence of a coronis in the Cologne manuscript at this point. However, in the Oxford papyrus witness, the left margin at this point disconcertingly does not survive to confirm whether the verses that follow there (i.e. those quoted by Athenaeus) were signaled in this later manuscript as belonging to a separate poem, or to a continuation of the “Tithonus Poem” entirely unknown, as far as we can tell, to the editors and writers of the Cologne manuscript.

Finally, I have marked the point at which, in the Cologne manuscript, the handwriting changes. This change to a deceptively similar style of script is a subtle and easily overlooked one, but a change of which we can be certain. It is all the more notable, in that it accompanies the transition to a poem that neither Sappho nor any Lesbian poet from antiquity could have written. About this poem much remains to be said, and new readings continue to emerge from scrutiny of this part of the papyrus, which has received little attention thus far by comparison. It will suffice here merely to note that the change of handwriting, together with the change in authorship, marks out the Cologne manuscript as a book-roll produced more informally, at its relatively early date, than the Oxyrhynchus copy four centuries later, and places it in the context of an early anthology of poems constructed for private use or of a copy-book for practice, perhaps in a school-setting.

Sigla: Π^¹ = P.Köln inv. 21351 + 21376 (3rd c. BC papyrus roll); Π^² = P.Oxy. 1787 fr. 1
(2nd c. AD papyrus roll)

Source: Π^¹ (verses lacking in Π^², which has the beginning of fr. 58 in this place) 2 ε̣ὔχ̣ο̣μ̣[ Di Benedetto: ] ̣υχ̣ ̣ ̣[ Gronewald-Daniel 3 γε̣[ Di Benedetto: π̣α̣[ Gronewald-Daniel, West 4 γε̣[νοίμα]ν̣ Gronewald-Daniel, Di Benedetto:
πε ̣[ ̣ ̣ ̣] West 5 κῆ μοιϲοπόλων ἔϲ]λ̣ο̣ν̣ Di Benedetto 6 ψῦχαι (or ϲκίαι)
κέ με θαυμά]ζ̣οιεν Di Benedetto comparing Horace Odes 2.13.25–30:
πάνται δέ με θαυμά]ζ̣οιεν already suggested by West 7 φαίην δὸϲ ἀοίδαν] Di Benedetto (ἀοίδαν already suggested by Gronewald-Daniel)
8 ἔμαιϲι φίλαιϲι(ν) Di Benedetto comparing Sappho fr. 160

Source: Π^² (verses lacking in Π^¹, which has the “New Fragment” in this place)

Sources: Π^¹ Π^² (Π^¹ preserves earlier portions of lines; Π^² preserves ends of lines) 1 ὔμμεϲ πεδὰ West: ὔμμεϲ τάδε Janko: γεραίρετε Di Benedetto: φέρω τάδε Gronewald-Daniel Μοίϲαν ἰοκόλ]πων so already Stiebitz on Π^² 2 ϲπουδάϲδετε West: ϲπουδάζετε would have been expected in Π^¹, cf. 7: χορεύϲατε Di Benedetto: λάβοιϲα πάλιν vel ἔλοιϲα πάλιν Gronewald-Daniel καὶ West: κὰτ F. Ferrari τὰ]ν West: τὰ]μ̣̣ Π^², Gronewald-Daniel φιλάοιδον recognized by Maas in Π^² 3 ἔμοι δ’ ἄπαλον πρίν] West ἔμοι Snell: κέκαρφ’ Gronewald-Daniel δ’ Di Benedetto: μὲν Snell ἄπαλον Gronewald-Daniel πρίν Di Benedetto: μοι Gronewald-Daniel π̣οτ̣’ [ἔ]ο̣ντα Gronewald-Daniel 4 ἐπέλλαβε, (perhaps too short) or κατέϲκεθε, λεῦκαι δ’ West: διώλεϲε Di Benedetto: ὄγμοιϲ(ιν) or ὄγμοι δ’ ἔνι Gronewald-Daniel λεῦκαι Hunt δ’ Lobel: τ’ Hunt ἐγ]ένοντο Hunt ἐκ Π^²: ἐγ Π^¹ 5 θ]ῦμο̣ϲ̣ Gronewald-Daniel 6 νεβρίοιϲι West: -ϲιν Π^¹ Π^² 7 τὰ 〈μὲν〉 West:〈ταῦ〉τα vel 〈ὂν δὲ〉 (i.e. ἀνὰ δὲ) Gronewald-Daniel: τὰ 〈νῦν〉 Janko ϲτεναχίζω Π^¹: ϲτεναχίϲδω West: not preserved in Π^² κεν Π^²: κεμ Π^¹
9 π̣[ο]τ̣α̣ Gronewald-Daniel 10 δ̣ ̣[ ̣]α̣ ̣ειϲαμ Π^¹ (Π^² not preserved here): φ̣ ̣ ̣α̣θ̣ειϲαν West: δ̣έ̣π̣α̣ϲ̣ εἰϲάμ- Gronewald-Daniel: λ̣α̣[λ]ά̣γ̣ειϲαμ Janko βάμεν’ articulation West: εἰϲαμβάμεν’ Gronewald-Daniel, although εἰϲομβάμεν’ would be required in the Aeolic dialect φέροιϲα[ν Stiebitz on Π^² 11 ἔοντα̣ [κ]ά̣λ̣ο̣ν̣ or ἔοντ’ ἄ̣[π]αλον Gronewald-Daniel 12 π̣ό̣λ̣ι̣ο̣ν̣ or π̣ό̣λ̣ι̣ο̣γ̣ Gronewald-Daniel ἔχ̣[ο]ν̣τ̣’ Gronewald-Daniel

“(several words missing) the violet-rich Muses’ fine gifts, children, (several words missing) the clear-voiced song-loving lyre: (several words missing) skin once was soft is withered now, (several words missing) hair has turned white which once was black, my heart has been weighed down, my knees, which once were swift to dance like young fawns, fail me. How often I lament these things. But what can you do? No being that is human can escape old age. For people used to think that Dawn with rosy arms (several words uncertain) Tithonus fine and young to the edges of the earth; yet still grey old age in time did seize him, though he has a deathless wife.”

Source: Π^¹ (verses lacking in Π^², which has “Continuation 2” in this place)
5 ὀμνύω τὸ] Puglia (τὸ] already Gronewald-Daniel) 7–8 ψόγουϲ (e.g.) ἀναίτι|ο]ϲ̣ Puglia 8 θρ̣ή̣[νοιϲα μιμοῦμαι Puglia (μιμουμένα suggested by Gronewald-Daniel) 9–10 τὸν | ἑρ]πετὰ πάντα κ[ηλοῦντα or κ[ηλήϲαντα (ἀοιδᾶι) Puglia 13 πά̣ν̣[των πόνων ἐμῶν Ferrari and Puglia: πᾶ̣ν̣ [ or πά̣ν̣[τα Gronewald-Daniel

Sources: 1–4 Π^² (verses lacking in Π^¹, which has “Continuation 1” in this place) 3–4 line-beginnings restored from the quotation of these verses by Athenaeus 687b (Clearchus fr. 41 Wehrli) 

Source: Π^² (verses lacking in Π^¹, Athenaeus)

The Text and Translation of Sappho frr. 58–59 as restored and translated by M. L. West