I am grateful to Gregory Sifakis, Michalis Tiverios, and Stavros Tsitsiridis for their assistance. Special thanks to Jürgen Hammerstaedt for his fruitful advice and good-natured endurance.
The Book Roll. Former Attempts
Apart from the general interpretations of the painting,3 to which I shall shortly add my own, already since the end of the 19th century there have been attempts to read the text that could be seen in the half-opened roll. These letters belong to the initial painting, before the vase was fired, and so are fairly well discerned. The first attempt seems to have been made by Maxime Collignon, who published the catalogue of the painted vases of the National Archaeological Museum of Athens.4 His reading was: Θεοί ἠερίων ἐπέων [ἄ]ρχομαι ἄγγ[ελος] ν[έων] ὔ[μ]ν[ων]. Later, H. Heydemann published his own proposal,5 reading on the rolled up ends of the papyrus ΠΤΕΡΟΕΤΙ ΕΠΕΑ, which he corrected to πτερόε<ν>τ(α) ἔπεα, and on its inside area Θεοί· περ(ὶ) τῶν ἐπέων ἄρχομαι ᾄ[δειν. Still later, Domenico Comparetti, assisted by the archaeologist Federico Halbherr,6 read Θεοί· ἠερίων ἐπέων ἄρχομαι ἄλλων, just like P. Kretschmer,7 who also read ἄρχομαι ἄλλ[ων], though with the ending ων supplemented.8
In 1922, J.M. Edmonds, making use of enlarged photographs and the assistance of the archaeologists B. Leonardos and A.J.B. Wace, proposed a somewhat different reading.9 He first noticed that the flanking words, written on the rolled up ends and therefore on the outer side of the roll, must be the title of the book. The title should read not ΠΤΕΡΟΕΤΑ ΕΠΕΑ, as it seems at first look, but reversely ΕΠΕΑ ΠΤΕΡΟΕΤΑ, as it would become apparent if the roll was spread face downwards. He also noticed that “the rolled-up part of the book held by the right hand [of Sappho] is intended to be bulkier than the part held by the left. The latter indeed may be regarded as the first curl beginning the left-hand ‘roll-up’, a curl which will not be completed till the reader proceeds to the next column”. Thus, the text seen on the interior side of the roll must represent the first column, therefore the opening, of the book. The first word read on this first column is ΘΕΟΙ, which Edmonds takes either as an invocation of the poetess at the opening of her book, similarly to the formula found at the opening of numerous official decrees, or as the heading of the first part of Sappho’s book, which might start with hymns to the gods, as was the case with Alcaeus’ Alexandrian edition. He finally inclines to the view that the presence of the invocation ΘΕΟΙ “is due simply to the artist”. What follows, according to Edmonds, are the letters, written in 11 very narrow lines (or 12 with the inclusion of the first line ΘΕΟΙ), which he read as: ΗΕΡΙ | ΩΝ | ΕΠΕ | ΩΝ | ΑΡΧ | ΟΜ | ΑΙΑ | ΛΛΟ | ΝΑ | ΤΩ | Ν. This leads to his final reading:
Ἔπεα πτερόεντα
Θεοί.
ἠερίων ἐπέων ἄρχομαι ἀλλ᾽ ὀνάτων.
The text following θεοί was considered by Edmonds a verse of Sappho, actually the first verse of an introductory poem to the pre-Alexandrian edition of Sappho prepared by the poetess herself. As for the metrical form of the verse, Edmonds found some similarities with several Sapphic and Alcaic fragments, but none exactly similar. Finally, he translated: ‘The words I begin are words of air, but for all that good to hear,’ ‘beneficial’.
In the same year (1922), Edmonds published Sappho’s fragments in the first volume of his Lyra Graeca. There (180–181), he prefixed to her known fragments, as fr. 1a, the verse he read on the vase-painting. He only changed the initial of the first word from η to α (ἀερίων) for adapting the Ionic ἠερίων to the dialect of Lesbos.10 Another proposal, which, however, went unnoticed by the scholarly community, was C.R. Haines’s in Sappho. The Poems and Fragments, London 1926. His drawing of the book roll (p. 53) seems to be the most faithful of all previous transcriptions, but the reconstruction he proposes is rather odd: θεοι, ηεριων επεων αρχομαι, αλλ᾽ <α>ναιτι<ω>ν. In any case, Edmonds’s views, whether of a first verse or of a pre-Alexandrian Sappho edition collected and arranged by the poetess, did not convince the dominant trend in classical scholarship. There is no mention whatsoever of the Attic hydria verse in E. Lobel, Σαπφοῦς Μέλη (1925), in E. Lobel/D.L. Page, Poetarum Lesbiorum fragmenta (1955), in E.-M. Voigt, Sappho et Alcaeus (1971), and in D.A. Campbell, Greek Lyric (1982). Without mentioning even the possibility that the text may come from Sappho, D.L. Page, Poetae Melici Graeci (1962), includes it in the fragmenta adespota, 938 d, where the text is published as
θεοί· ἠερίων ἐπέων ἄρχομαι ......... ,
and the app. cr. as θεοι|ηερι|ων|επε|ων|αρχ|ομ|α.α|τ..|τ.|ν. He obviously followed J.D. Beazley, who had noted in 1928,
Greek Vases in Poland, Oxford, 9 n. 2 (end, continued in p. 10): “the vase shows (besides επεα and πτεροετα) θεοι ηεριων επεων αρχομα.ατ..ν.τ.ν, where my dots represent misshapen and uncertain letters. The last part, where the roll narrows, is evidently meaningless”. Due to an oversight, Page omitted the third last line that contained N. from Beazley’s reading. It was also Beazley’s remark about the meaningless last part of the text that accounts for the numerous dots of Page’s text. Campbell,
Greek Lyric (1993), fr. 938 (d), followed Page (with the oversight) both in the text and the apparatus, only omitting θεοί, which he mentions in a note. He translates: “Airy words I begin” and notes “Lofty? Early?”. Henry Immerwahr in his Online Corpus of Attic Vase Inscriptions includes under no. 764 the Athens hydria. He agrees with the prevalent reading down to ἄρχομαι α-, but his last lines are
He proposes that all this is miswritten for ἀείδειν. The latest treatment of the vase-painting text known to me was by D. Yatromanolakis,
11 who also agrees with the prevalent reading ...