The Cologne Sappho: Its Discovery and Textual Constitution

Jürgen Hammerstaedt

To refer to this article, please cite it in this way:
Jürgen Hammerstaedt, "The Cologne Sappho: Its Discovery and Textual Constitution," Classics@ Volume 4: Ellen Greene and Marilyn Skinner, eds. The Center for Hellenic Studies of Harvard University, online edition of March 11, 2011.


Discovery and Acquisition

In 2002 a group of more than 20 papyri was on the market. They belonged to a private collector outside Egypt. We do not know how and when the papyri became his property. But only after acquisition by the Cologne Papyrus Collection could these ancient documents of inestimable cultural value be finally restored, mounted and conserved properly. It is perhaps even more important that the acquisition of these papyri by a public institution led to their publication and made them accessible not only to the scientific community but also to a broader audience. In this way they at last received the attention they deserve.

First Edition, Partial Overlap with P.Oxy. 1787 fr. 1

Two fragments (inv. 21351) of the Cologne Sappho papyrus were published by Gronewald and Daniel in a first article (2004a). The two fragments do not join physically (Figure 1), but the first editors recognized that the text partially overlapped that of a fragment of Sappho (fr. 58.11–22 Voigt) on P.Oxy. 1787 fr. 1 of the second–third century AD (Figure 2). [1] It was thus clear that the verses contained in the upper part of the column of one of the pieces, which still preserves the top margin, were the immediate continuation of the verses written in the lower part of the columns of the other, which still preserves the bottom margin.

In the second fragment the lines which the Cologne papyrus has in common with P.Oxy. 1787 fr. 1 were followed by a slightly wider interlinear space and the remains of two lines of writing in a different script (Figure 3). These could not be identified with the lines which come after the same poem of Sappho in P.Oxy. 1787 fr. 1. [2]

The new Cologne Sappho confirmed some of the previous restorations which had been proposed for the Oxyrhynchus fragment: [ἰ]⎣ο̣κ⎦[ό]⎣λ̣π⎦ων by Stiebitz (fr. 58.11 Voigt); ⎣ὄρχ⎦ηϲθ’ by Edmonds (fr. 58.16 Voigt); ⎣ἔϲ⎦χατα by Lobel (fr. 58.20 Voigt); ⎣ἀθαν⎦άταν by Stiebitz (fr. 58.22 Voigt). Five verses of Sappho’s poem could now be reconstructed almost completely. [3] They confirmed that the metre of the poem is an acephalous Hipponactean expanded within by two choriambics. This was probably the principal metre of the fourth book of the Alexandrian edition of Sappho. [4]

The Third Cologne Fragment with More Sappho and an Anonymous Lyric Poem

At a later date, the editors recognized a third fragment (Cologne papyrus inv. 21376) and combined it with the second column (Figure 4). Its lower part continues the second text, [5] of which up to then only the remains of two lines written in a different script had survived. Its upper part also contains the missing beginnings of the last four lines of Sappho’s “Tithonus poem.” These were published by Gronewald and Daniel in a second article 2004b (P.Köln inv. no. 21351+21376). The rest of the third fragment was first edited in Gronewald and Daniel 2005 (Figure 5). The remainders of 13 lines run down to the end of the column. There is much uncertainty about the content, extension, metre and form of the poem. But the first editors pointed out one certainty: this text is definitely not by Sappho. None of the forms are specifically Aeolic, and not all of them can be explained as Aeolic. [6] Words like αὐτουργόϲ, ἐπίϐουλοϲ, [ϲυ]νεργόϲ and ἀφέρπω are not found in early poetry. Moreover, the mention of Orpheus, son of Oeagrus, all the beasts of the land (ἑρπετά) and the fine-voiced lyra imply a myth which is not attested earlier than Simonides PMG fr. 567 (Gronewald and Daniel 2007:8). The metrical features of these lines, however, are, in so far as they can be identified, even more cogent. These are written apparently without distinguishing the colometric units. The first line contains a sequence of at least five short syllables. Aeolic metre is therefore excluded. Gronewald and Daniel (2007) analyzed the beginning as lyric anapaests, followed after several gaps by an iambic penthemimeres and further sequences of double and single shorts between the long syllables. Lundon (2007a) points out that the first line could also be interpreted as two dochmiacs. Dochmiacs are the prevalent metre in the Fragmentum Grenfellianum. [7] This lyrical monologue, which contains the lament of a woman abandoned by her companion, had already been indicated by the first editors as a possible parallel for the lyric text in the Cologne Sappho papyrus also on account of other features, like the use of dicola (Figure 6).

In view of this parallel and the fact that the first word, the hapax ψιθυροπλόκε (“whisper-weaving”), recalls the epithet δολόπλοκε in Sappho’s Hymn to Aphrodite (fr. 1 Voigt), Gronewald and Daniel (2007a and 2007b) supposed that the lyric text was erotic in content.

Rawles (2006) proposed a different interpretation, expanding an alternative explanation of Gronewald and Daniel. Taking the dicola in their well-attested function as marks for change of speaker, he suggested that the poem contained an antiphonal exchange between two musicians and focused not on love but on music, specifically the invention of the lyre and its early use by Orpheus. However, as Rawles himself admits, his interpretation does not account for the female character certainly alluded to at the end of the poem.

The most detailed examination of all the relevant evidence has been given by Lundon 2007a, who quotes, besides the Fragmentum Grenfellianum , the lament of Helen in the Hellenistic P.Tebt. I 1 as another parallel. [8] Puglia (2008) accepts and expands Lundon’s conclusions, reconstructing the monologue of a woman who replies to the accusations of the man who had left her and accompanies her laments with the lyre, as Orpheus did.

In spite of differences in their interpretation, both Rawles and Lundon observed several features common to this text and the poems by Sappho that precede it. The most striking aspect is the mention of musical instruments in the three texts. The anonymous poem was apparently added with reference to the preceding poems. [9] But if we want to understand the relationship between Sappho’s two poems and the lyric text properly, we also have to take the hands and other aspects of writing of the Cologne papyrus into consideration.

The Hands and Their Probable Date

Gronewald and Daniel 2005 observed that the letters ε, ζ and θ in the literary hand which wrote Sappho’s two poems [10] in the Cologne papyrus are generally still written in the epigraphic form. There are also some fine examples of the archaic form of ω. [11] One ω (Figure 8, number 2) takes a further step in the direction of the shape in which it normally appears in the papyri, and in most cases in this papyrus too. Lundon (2007a) draws attention to α, whose middle stroke is horizontal, written as a separate stroke after the first diagonal. However, there is one exception (Figure 10), and there are also a few instances of round ε, written in two strokes (Figures 7 and 8).

Generally speaking, the epigraphic features, even if they are not applied throughout, lend a rectangular character to this hand. Nonetheless there are also more curved lines. These are especially noticeable in the right-hand vertical of η, and also in π (underlined in Figure 8), which is rather broad, like that appearing in the oldest dated papyrus document (P. Saqqara inv. 1972 GP 3; Turner/Parsons 1987 no. 79; between 331 and 323 BC; Figure 11). The same rounded features sometimes reappear in υ, which is occasionally written in two strokes (Figure 8). Compare the μ, in three strokes. ϲ always has the lunate form. [12]

As for the other letters, ι extends above and below the line, as the vertical of κ frequently does too, while τ, υ, and ρ descend below the base-line. ω and ϲ are smaller than the other letters.

Gronewald and Daniel (2004a) compared the hand of Sappho’s two poems on the Cologne papyrus with Turner/Parsons 1987 no. 52, of the early third century BC (Figure 12). This manuscript containing anonymous fragments of tragedy (TrGF 625), seems even more archaic, because ϲ is written in four strokes (Figures 13a and 13b), and ω always has the epigraphic form. However, there are also striking similarities between the two hands (Figures 13a and 13b), in regard to not only the archaic letter forms, but also the long verticals of ι and κ and the descenders of τ, υ, and ρ.

I agree that Turner/Parsons 1987 no. 52 is the closest parallel to the first hand of the Cologne papyrus, which the editors dated on this basis to the early third century BC. So it is by far the earliest manuscript of Sappho which we now possess.

The anonymous lyric poem [13] is written in broader pen strokes. As a result the writing is less compact than that of the Sappho poems and the distance between the lines larger (Figure 5). No epigraphic letter-forms can be seen (ε is written in two strokes). For this reason the writing appears rounder than the hand of the Sappho poems (Figures 14a and 14b).

Shortly after the beginning of the anonymous lyrics there are in the second and third lines two major deletions and at least one interlinear correction (Figure 5, number 1). Lundon takes these as possible signs of an autograph, of a text composed and altered by the writer himself. Indeed, the ductus of the writing seems less regular and certainly less formal.

We should also note, with Lundon, that the middle strike of α tends to be inclined, and at least sometimes it seems to be drawn in the same movement as the preceding diagonal stroke (Figure 5, number 2) without the lifting of the pen. But sometimes α seems to be written in three strokes, as it was in the Sappho verses, and the inclination of the middle stroke is not always so accentuated (Figure 5, number 3).

In any case, I strongly believe that these are not just differences in the style of writing, but that the lines are in fact written by a hand altogether different from the one that wrote the Sappho poems. The observation that the τ of the lyric poem never descends below the line (cf. Figure 5, number 4), as it so often does in the hand of the Sappho poems (Figures 7 and 8, passim), seems to confirm this impression.

In view of the apparent similarities of the two hands and the fact that the archaic features of the first hand are not maintained throughout, Lundon concluded that the lyric text might have been added to the papyrus by a second hand in roughly the same period as the poems by Sappho.

I believe that the closest parallel for the hand of the anonymous lyrics is Turner/Parsons 1987 no. 30 (mid-third century BC; Figure 15), although there are of course some differences. The hand of the anthology of lyric passages from Euripides writes a very straight and severe ν and a rather broad δ. It is also more carefully executed. But all in all it looks quite similar (Figure 16). We find the same alternation between α in three movements with horizontal stroke and α in two movements with oblique middle stroke.

At any rate, I am still inclined to date the second hand of the Cologne papyrus some time later in the third century than the first hand. But what is more important: the noticeable difference in formality and execution between the two hands does not favor the idea that the lyric text was added to the papyrus at the same time or in the same situation as the poems by Sappho.

The Division of the Two Poems by Sappho in the First Column of the Cologne Papyrus

It is now generally believed that the verses which the Cologne and Oxyrhynchus papyri preserve before the twelve they share (see Figures 1 and 2) belong to different poems [14] and that the first of the common verses is the beginning of a new composition. [15] Gallavotti (1947) had already suspected the beginning of a new poem in this line of P.Oxy. 1787 fr. 1, while Lobel (1925) had conjectured one two lines later. Gronewald and Daniel recognized three further reasons for distinguishing two poems between lines 8 and 9 of the Cologne papyrus: (1) λιγύραν is repeated in lines 7 and 10; (2) if the constitution of lines 8 and 9 is right (cf. below), κάλα is repeated and line 8 contains an apostrophe to a single Muse, [16] while line 9 seems to mention the “fragrant-bosomed Muses” in the plural; (3) there is a thematic difference between lines 1–8, which compare Sappho’s present existence with her life after death, and lines 9ff, which treat Sappho’s old age.

There are accordingly good reasons for believing that the Cologne Sappho papyrus and P.Oxy. 1787 fr. 1 preserve the ends of two different poems by Sappho in their first lines, followed in both of them by the same poem dealing with old age. All the poems are written in the same metre.

P.Oxy. 1787 fr. 1 and 2: Papyrological and Textual Evidence Concerning Poem-Division

The evidence of the Cologne Sappho Papyrus proved the suspicion that two poems were to be distinguished in P.Oxy. 1787 fr. 1. The last four lines of the Oxyrhynchus papyrus (fr. 58.23–26 Voigt) [17] do not occur in the Cologne fragments either (cf. Figures 1 and 2). But no scholar had previously suspected that these lines might belong to a different poem. Those who assume that Sappho’s poem on old age extended further than fr. 58.22 Voigt need to give convincing reasons to explain why the poem in the Cologne papyrus stops at this point and, of course, to interpret in this sense the remaining parts of lines 23–26. [18]

My contribution tries to set out some of the papyrological and textual evidence concerning Sappho fr. 58.23–26 and fr. 59 Voigt.

The Oxyrhynchus papyrus (text see above) has preserved only the very last words of these lines, but Athenaeus 15.687b gives a partial quotation of the last two lines. [19]

ὑμεῖς δὲ οἴεσθε τὴν ἁϐρότητα χωρὶς ἀρετῆς ἔχειν τι τρυφερὀν;
καίτοι Σαπφώ, γυνὴ μὲν πρὸς ἀλήθειαν οὖσα καὶ ποιήτρια,
ὅμως ᾐδέσθη τὸ καλὸν τῆς ἁϐρότητος ἀφελεῖν λέγουσα ὧδε·
ἐγὼ δὲ φίλημμι ἀϐροσύναν,                    καί μοι
τὸ λαμπρὸν ἔρος ἀελίω καὶ τὸ καλὸν λέλογχε,
φανερὸν ποιοῦσα πᾶσιν ὡς ἡ τοῦ ζῆν ἐπιθυμία τὸ λαμπρὸν
καὶ τὸ καλὸν εἴ‹λη›χεν (Hunt) αὐτῇ· ταῦτα δ’ ἐστὶν οἰκεῖα τῆς

But do you imagine that daintiness can comprehend anything luxurious when divorced from virtue? And yet Sappho, truly a woman, if there ever was one, and a poetess besides, nevertheless was ashamed to separate honor from daintiness when she said: “But I love daintiness …, and for me brightness and honor belong to my yearning for the sun”; thus she makes it plain to all that the desire to live contained for her the idea of brightness with honor; for these are natural properties of virtue. [20]

For metrical reasons, Athenaeus’s second verse needs emendation. The first verse cannot be completely reconstructed, but the Oxyrhynchus papyrus offers us one more word: τοῦτο.

The four letters τολα at the beginning of the last line in Athenaeus’ Sappho quotation return in the first line of P.Oxy. 1787 fr. 2 (Figures 2 and 17). Hunt 1922 was the first to identify this line beginning with the first letters of the verse whose end is preserved in fr. 1.26. He may be right, but his identification is not altogether certain. In this case the other three lines of fr. 2 have to be combined with the remains of the letters at the bottom of fr. 1. [21]

Lobel and subsequent editors up to and including Voigt regarded the last three verses of fr. 2 as the beginning of a new poem (now Sappho fr. 59 Voigt). However, the horizontal stroke beneath the first line of fr. 2 is not necessarily part of a coronis, but can just as easily be interpreted as a mere paragraphos, like the one the editors recognized after the next two lines. The differences in position and appearance between the coronides, used to separate the poems, and the paragraphoi, used to distinguish the strophes of two lines each, can be illustrated very well at P.Oxy. 1787 fr. 3 (Figure 18). In the intercolumnium, before the lines, we see three coronides, which mark the ends, and the beginnings, of different poems (fr. 61 Voigt; fr. 62 Voigt, of 12 lines; fr. 63 Voigt, of 10 lines).

In the same fragment (Figure 18), several paragraphoi are beneath the first letters of the line. If we compare fr. 2 (Figure 17), the first horizontal stroke has a position, well inside the interlinear space, very similar to that of the paragraphoi in fr. 3. [22]

From this difference observed in P.Oxy. 1787 fr. 3 we might well conclude that the trace under fr. 2, line 1, must be a paragraphos and that a new poem did not begin here at all. Di Benedetto saw this in 1985; Luppe (2004) and West (2005) restated this fact.

However, as Lundon (2007a) observes, some coronides in other fragments of the same papyrus extend beneath the line in exactly the same way as the paragraphoi line (clearly in P.Oxy 1787 fr. 21.1–2 and fr. 24.3: figs. 20 and 21). The papyrological evidence does therefore not prove that a new poem begins in P.Oxy. 1787 fr. 2 (Voigt fr. 59), but it does not rule it out completely either.

Some Further Notes on the Text of the Cologne Papyrus

In the places where different readings were proposed I checked the Cologne papyrus on the original. [23]

“New Fragment”: [24]

Cologne papyrus line 3: ] . νῦν θ̣α̣λ̣[ί]α̣ γ̣ε̣[νέϲθω] or γ̣έ̣[νοιτο] Gronewald and Daniel 2004a ([ἔμαιϲιν ἑταίραιϲ’ ἀμ’ ἔμοι] νῦν ϑ̣α̣λ̣[ί]α̣ γε̣[νέϲϑω] Bettarini 2008 exempli gratia): π̣α̣[ρέϲτω] (or π̣ά̣[ρεϲτι] or π̣ὰ̣[ρ ἄμμι]) West 2005. — The top of an upright before νῦν can be reconciled with Bettarini’s proposal ἔμοι̣ νῦν. On the other hand, there is no papyrological evidence to decide between γ and π, but Michael Gronewald excludes West’s supposed α̣ after π (cf. Figure 22).

Cologne papyrus line 4: ] . ν̣έρθε δὲ γᾶϲ γ̣έν̣[εσθ]α̣ι̣ ed. pr., γ̣ε̣[νοίμα]ν̣ Di Benedetto 2005 (who defends Gronewald and Daniel 2004a), π̣ε̣[ρίϲχοι] West 2005, [ἐπεὶ δέ κε γήραιϲα ϑάνω,] νέρϑε δὲ γᾶϲ γέν̣[̣ωμαι] Bettarini 2008 exempli gratia. — No clear papyrological evidence allows a decision between the readings at the line end (cf. Figure 23). The broad horizontal trace of ink at medium height before νέρϑε does not contradict Bettarini’s ([ϑάν]ω̣, νέρϑε).

Cologne papyrus line 5: ] . . ν̣ ἔχο̣ιϲαγ γ̣έραϲ ὠϲ̣ [ἔ]̣ο̣ικε̣ν̣, West 2005: κλέοϲ μέγα Μοίϲει]ο̣ν ἔχο̣ιϲαγ γέραϲ; Hardie 2005: [μολπά μ’ ἔτι Μοίϲε]ι̣ο̣ν̣ ἔχο̣ιϲαγ γ̣έραϲ (Livrea 2007:76n23 would prefer κλέοϲ vel μνάμα instead of μολπά); Di Benedetto 2005: [κῆ μοιϲοπόλων ἔϲ]λον̣ ἔχο̣ιϲαγ γ̣έραϲ, ὠϲ̣ [ἔ]ο̣ικε̣ν (Bettarini 2008 supplies καί instead of κῆ). — Gronewald does not exclude ]ο̣ν̣, but the supposed ο would be unusually large, as θ, and the space between ο and ν seems to be rather long and probably contains further traces of ink (cf. Figure 24).

Cologne papyrus line 6: ] . οιεν, West 2005: πάνται δέ με θαυμά]ζοιεν, Di Benedetto 2005: ψῦχαι κέ με θαυμά]ζοιεν, Bettarini 2008: [οὔ κέν μ’ ἔτι ϑαυμά]ζ̣οιεν. — ζοιεν is quite possible: the surface was once torn upwards in such a way that the lower horizontal was obliquely disposed (cf. Figure 25).

Cologne papyrus beginning of line 8: ] . . . . α . ; West 2005 (after Gronewald and Daniel 2004a app.): [ἤ βάρϐιτον ἤ τάνδε χε]λύ̣ν̣ν̣αν, contested by Di Benedetto 2005. — χε]λύ̣ν̣ν̣αν is ruled out by the evidence. I myself checked α, then τ or π, the rest is not clear (cf. Figure 26).

Cologne papyrus, middle of line 8: κ̣άλα, Μοῖϲ’, ed. pr. (who rules out, for obvious reasons, κ̣αλάμοιϲ’), θ̣αλάμοιϲ’ West. The question cannot be decided (cf. Figure 26).

“Tithonus poem”: [25]

Cologne papyrus line 15: †τα† ϲτεναχίζω θαμέωϲ Gronewald and Daniel 2004a (the metre requires a further syllable); <ταῦ>τα ϲτεναχίζω or <ὄν δὲ> ϲτεναχίζω Gronewald and Daniel 2004a comm., comparing Anacreon PMG 395, at the end of a description of the poet’s ageing body: διὰ ταῦτ’ ἀναϲτυλίζω / θάμα Τάρταρον δεδοικώϲ; West 2005: τὰ <μὲν> ϲτεναχίζω; Janko forthcoming: τὰ <νῦν> ϲτεναχίζω; Führer 2007: <ἦ> τὰ ϲτεναχίζω Burzacchini 2007:100 (defendit Lundon 2007b; reiecit Bettarini 2008:30n53): <ζὰ> τὰ; ϲτεναχίϲδω θαμέωϲ; Burzacchini 2007:100: τά<δε> ϲτεναχίζω.

The most debated single place of the Cologne papyrus is line 18 (Figure 27): Gronewald and Daniel 2004a read ἔρωι δ̣έ̣π̣α̣ϲ̣ εἰϲανϐάμεν̣’ εἰϲ ἔϲχατα γᾶϲ φέροιϲα[ν (l. εἰϲομϐάμεν̣’), and translated: “dass die rosenarmige Eos aus Liebesverlangen den Sonnenbecher bestiegen habe, (ihn = Tithonos) zum Ende der Erde tragend” (the reading was supported with further arguments by Di Benedetto 2005:18–19).

West 2005 proposed: ἔρωι φ . . α̣θ̣ε̣ιϲαν βάμεν̣’. — I am convinced that φ cannot be read.

Magnani 2005: ἔρωι δ̣ί[φρ]ο̣ν̣ εἰϲανϐάμεν̣’, comparing Nonnus Dionysiaca 15.280 (δίφρον ἐὸν ϲτήϲαϲα φαεϲφόροϲ ἥρπαϲεν ‘Ηώϲ). — The ν at the end of δ̣ί[φρ]ο̣ν̣ certainly does not fit the traces.

Austin 2007: Ἔρῳ ἅ̣ρ̣μ’̣ <ἀν>αθεῖϲαν βάμεν̣’, translating: “confia son char à Eros et s’en alla aux extrémités de la terre en emmenant Tithonos”; Bettarini 2007 Ἔρῳ ἄ̣ν̣[ι]’ ἄ̣φ̣ειϲαν βάμεν̣’, taking ἄνι(α) as Aeolic for ἡνία and proposing three different interpretations. — In spite of two examples of “broad” α pointed out by Bettarini the α of ἅ̣ρ̣μ’̣ or ἄ̣ν̣[ι]’ seems highly improbable.

Janko forthcoming: ἔρωι λ̣α̣[λ]ά̣γ̣ειϲαν βάμεν̣’. — The traces do not fit the first λ and the following α well; the γ seems to be ruled out by the traces.

In Figure 28, the two pieces of papyrus are combined more closely using an imaging program. The probability of this recomposition can easily be confirmed by the recomposed letters α and η at the beginning of line 16 in the Cologne papyrus and by the second α of line 17. In line 18, the short trace of a high horizontal line seems to form an ε together with the remaining traces after δ. Before ειϲαν there is the upper left part of a round letter (θ, ϲ, ο), preceded by what seems to be the right end of an α (cf. the first α in line 14, or the second α in line 16).

The textual basis for possible restorations is either ερωιδε̣[.]α̣θ̣ειϲαν or ερωιδε̣[.]α̣ϲ̣ειϲαν. Besides the proposal of Gronewald and Daniel 2004 in the editio princeps, only three of the restorations proposed up to now fit the traces: δέμα θεῖϲαν Danielewicz 2006; δέμαϲ εἶϲαν Austin 2007; δέμαϲ εἰϲαμβάμεν’ Livrea 2007.


Figure 1. Cologne papyrus inv. nr. 21351 (the parts overlapping P. Oxy. 1787 fr. 1 are marked with a box).

Figure 2. P. Oxy. 1787 fr. 1 (lines overlapping the Cologne Sappho papyrus are marked with a box).

Figure 3. Cologne papyrus inv. nr. 21351, bottom of the second fragment.

Figure 4. Cologne papyrus complete (inv. 21351 + 21375); the line divides Sappho's verses from the Lyric Poem.

Figure 5. Cologne papyrus inv. 21351 + 21376, "lyric text"; graphic features: 1. Deletions and corrections in lines 2 and 3 (rectangles); 2. α in two strokes (circles); 3. α apparently in three strokes (squares); 4. τ (arrows).

Figure 6. Two dicola in the Cologne lyric text (inv. 21376).

Figure 7. Graphic features in col. I of the Cologne Sappho papyrus.

Figure 8. Graphic features in col. II of the Cologne Sappho papyrus: 1. θ and ω, in circles; 2. Α particular ω, and some round ε, in squares; 3. Curved π and η, underlined; 4. Rounded υ and μ, marked by arrows.

Figure 9. ω in the Berlin Timotheus papyrus.

Figure 10. A particular α in Sappho's Old Age poem.

Figure 11. Turner/Parsons 1987 nr. 79 (331–23 BC).

Figure 12. Turner/Parsons 1987 nr. 52.

Figure 13. Detail of Turner/Parsons 1987 nr. 52 (left), facing Cologne Sappho poem (right).

Figure 14. Detail of a Cologne Sappho poem (left) facing the lyric text (right).

Figure 15. Turner/Parsons 1987 nr. 30.

Figure 16. Part of Turner/Parsons 1987 nr. 30 (left) facing the Cologne lyric text (right).

Figure 17. P. Oxy. 1787 fr. 2, with diacritic signs: 1. Sappho fr. 58, 26 Voight: τὸ λά[ ̣ ; 2. Sappho fr. 59, 1–3 Voigt: ※ Ἐπιν[---] | Φίλει. | [---] | καιν[---]; 3. Diacritic signs.

Figure 18. P. Oxy. 1787 fr. 3: coronides and paragraphoi.

Figure 19. Paragraphoi in the Cologne Sappho papyrus.

Figure 20. P. Oxy. 1787 fr. 21, 1–2 (coronis).

Figure 21. P. Oxy. 1787 fr. 24, 3 (coronis)

Figure 22. Cologne papyrus, line 3: ] ̣ νῦν θ̣α̣λ̣[ί]α̣ γε̣[νέϲθω] or γέ̣[νοιτο] or π̣ά̣[ρεϲτι] or π̣ὰ̣[ρ ἄμμι].

Figure 23. Cologne papyrus, line 4: ν̣έρθε δὲ γᾶϲ γε̣ν̣[έϲθ]α̣ι or γε̣ν̣[οίμαν] or π̣ε̣[ρίϲχοι].

Figure 24. Cologne papyrus, line 5: ] ̣ ̣ ν̣ ἔχ̣οιϲαγ γέραϲ or Μοίϲει]ο̣ν̣ ἔχ̣οιϲαγ γέραϲ or [κῆ μοιϲοπόλων ἒϲ]λον̣ ἔχ̣οιϲαγ γέραϲ.

Figure 25. Cologne papyrus, line 6: ] ̣ οιεν or πάνται δέ με θαυμά]ζοιεν or ψῦχαι κέ με θαυμά]ζοιεν.

Figure 26. Cologne papyrus, line 8 A + B.

Figure 27. Cologne papyrus, line 18.


Acosta-Hughes, B. Forthcoming. Recalling Lyric in Alexandria. Five Readings. Princeton.

Austin, C. 2007. “Nuits chaudes à Lesbos: buvons avec Alcée, aimons avec Sappho.” In Bastianini and Casanova 2007:115–126.

Bastianini, G., and A. Casanova, eds. 2007. Atti del convegno internazionale di studi “I papiri di Saffo e di Alceo”: Firenze 8–9 giugno 2006. Studi e Testi di Papirologia n.s. 9. Florence.

Bettarini, L. 2007. “Note esegetiche alla nuova Saffo: i versi di Titono (fr. 58,19–22 V.).” Zeitschrift für Papyrologie und Epigraphik 159:1–10.

———. 2008. “Saffo e l’aldilà in P.Köln 21351, 1–8.” Zeitschrift für Papyrologie und Epigraphik 165:21–31.

Burzacchini, G. 2007. “Saffo Frr. 1, 2, 58 V. tra documentazione papiracea e tradizione indiretta.” In Bastianini and Casanova 2007:83–114.

Danielewicz, J. 2006. “Bacchylides fr. 20A, 12 S.-M. and Sappho, P.Köln fr. I–II, 12.” Zeitschrift für Papyrologie und Epigraphik 155:19–21.

Di Benedetto, V. 1985. “Il tema della vecchiaia e il fr. 58 di Saffo.” Quaderni Urbinati di Cultura Classica n.s. 19:145–163.

———. 2005. “La nuova Saffo e dintorni.” Zeitschrift für Papyrologie und Epigraphik 153:7–20.

Esposito, E. 2005. Il Fragmentum Grenfellianum (P.Dryton 50): Introduzione, testo critico, traduzione e commento. Eikasmos studi 12. Bologna.

Ferrari, F. 2005. “Contro Andromeda: recupero di un’ode di Saffo.” Materiali e Discussioni per L’analisi dei Testi Classici 55:13–30.

Führer, R. 2007. “Zum neuen Sappho-Papyrus.” Zeitschrift für Papyrologie und Epigraphik 59:11.

Gallavotti, C. 1947. Saffo e Alceo. Collana di studi greci 10. Naples.

Gronewald, M., and R. W. Daniel, 2004a. “Ein neuer Sappho-Papyrus.” Zeitschrift für Papyrologie und Epigraphik 147:1–8.

———. 2004b. “Nachtrag zum neuen Sappho-Papyrus.” Zeitschrift für Papyrologie und Epigraphik 149:1–4.

———. 2005. “Lyrischer Text (Sappho-Papyrus).” Zeitschrift für Papyrologie und Epigraphik 154:7–12.

———. 2007a. “Griechische Literarische Texte: 429. Sappho.” Kölner Papyri. Band 11. (eds. Ch. Armoni, M . Gronewald, K. Maresch, et al.) 1–11. Papyrologica Coloniensia 7. Paderborn.

———. 2007b. “Griechische Literarische Texte: 430. Lyrischer Text (Sappho-Papyrus)” Kölner Papyri. Band 11. (eds. Ch. Armoni, M . Gronewald, K. Maresch, et al.) 12–19. Papyrologica Coloniensia 7. Paderborn.

Gulick, Ch. B., trans. 1941. Athenaeus. The Deipnosophists. Vol. 7. Loeb Classical Library. Cambridge, MA.

Hardie, A. 2005. “Sappho, the Muses and Life after Death.” Zeitschrift für Papyrologie und Epigraphik 154:13–32.

Hunt, A. S. 1922. “New Classical Fragments. 1787. Sappho, Book IV.” The Oxyrhynchus Papyri Part XV (ed. B. P. Grenfell and A. S. Hunt) 26–46. London.

Janko, R. Forthcoming. “Tithonus and Eos in the New Sappho (fr. 58, 11–23), with a note on Horace Odes I 22.” Symposium Lesbium: Poetry, Wisdom and Politics in Archaic Lesbos: Alcaeus, Sappho, Pittachus (ed. A. L. Pierris). Oxford.

Liberman, G. 2007. “L’édition alexandrine de Sappho.” In Bastianini and Casanova 2007:41–65.

Livrea, E. 2007. “La vecchiaia su papiro: Saffo, Simonide, Callimaco, Cercida.” In Bastianini and Casanova 2007:67–81.

Lobel, E., ed. 1925. ΣΑΠΦΟΥΣ ΜΕΛΗ: The Fragments of the Lyrical Poems of Sappho. Oxford.

Lundon, J. 2007a. “Il nuovo testo lirico nel nuovo papiro di Saffo.” In Bastianini and Casanova 2007:149–166.

———. 2007b. “Die fehlende Silbe im neuen Kölner Sappho-Papyrus” Zeitschrift für Papyrologie und Epigraphik 160:1–3.

Luppe, W. 2004. “Überlegungen zur Gedicht-Anordnung im neuen Sappho-Papyrus.” Zeitschrift für Papyrologie und Epigraphik 149:7–9.

Magnani, M. 2005. “Note alla nuova Saffo.” Eikasmos 16:41–49.

Page, D. L. 1955. Sappho and Alcaeus: An Introduction to the Study of Ancient Lesbian Poetry. Oxford. Corrected reprint ed. 1959.

Puglia, E. 2008. “Appunti sul nuovo testo lirico di Colonia.” Zeitschrift für Papyrologie und Epigraphik 164:11–18.

Rawles, R. 2006. “Musical Notes on the New Anonymous Lyric Poem from Köln.” Zeitschrift für Papyrologie und Epigraphik 157:8–13.

Turner, E., and P. Parsons. 1987. Greek Manuscripts of the Ancient World. 2nd ed. Bulletin of the Institute of Classical Studies Supplement 46. London.

West, M. L. 2005. “The New Sappho.” Zeitschrift für Papyrologie und Epigraphik 151:1–9.


[ back ] 1. MP3 1449 (together with P.Hal. 3 as fr. 44; and P.Oxy. XVIII 2166); LDAB 3899.

[ back ] 2. The text which above (p. 13) is called “Continuation 2”.

[ back ] 3. Lines 5–9 of the “Tithonus poem,” see above (p. 11).

[ back ] 4. For a brief review of the evidence for metrical arrangement, see Lidov, “Notes on the Meter,” this volume. See already Hunt 1922:26 in the editio princeps of P.Oxy. 1787. Ancient references show that the first book of the edition of her poems was written in Sapphic stanzas, and that the second and third book each also consisted of one single metre (for book 2 cf. Hephaestion 7.7 p. 23 Consbruch, for book 3 Hephaestion 10.6 p. 34 C.). Page (1955:144f.), expanding Lobel’s hints (cf. Liberman 2007:48) concluded from the fact that the numerous fragments of P.Oxy. 1787 are all of the same metre that the book which contained them consisted entirely of the same metre. Since the fifth book offered at least two different metres, as probably did the seventh, Page assigned P.Oxy. 1787 and its metre to Book Four, the only book besides the sixth about which nothing else is known. However, Acosta-Hughes (forthcoming:103) points out that Hephaestion does not attribute this metre to a certain book of Sappho’s Alexandrian edition (Hephaestion 11.5, p. 36 C. Consbruch), as he does in yet a third case (regarding book seven: Hephaestion 10.5., p. 34 C.). Liberman (2007:50–52) develops some further ideas about the relationship between the Cologne Sappho and the Alexandrian edition.

[ back ] 5. See above (p. 13), “Continuation 1.”

[ back ] 6. Line 8 πᾶϲ, or πᾶϲ(α) instead of παῖϲ(α). There is a strong Doric coloring, with the only exception of κόροϲ in line 9, which in Doric would be κῶροϲ (Ep.-Ion. κοῦροϲ). Cf. Lundon 2007a:155–157.

[ back ] 7. Ed. Esposito 2005, who reveals further points of contact between the Fragmentum Grenfellianum and the anonymous lyric poem (Esposito 2005:62; cf. Lundon 2007a:162f.).

[ back ] 8. MP3 1606; LDAB 6894.

[ back ] 9. For further cases of imitation of Sappho’s songs in this period, mainly by Theocritus and Apollonius Rhodius and in the Hellenistic epigram cf. Acosta-Hughes (forthcoming:ch.2).

[ back ] 10. “New fragment” and “Tithonus poem”, text see above (pp. 10–11).

[ back ] 11. Comparable forms of ω are known, for example, from the Berlin Timotheus papyrus of the fourth century BC (Figure 9).

[ back ] 12. As it already has in the early document P.Elephantine 1 (311/10 BC) and in P.Vindob. G 1 (UPZ I 1).

[ back ] 13. “Continuation 1” (text, see above, p. 13).

[ back ] 14. “New Fragment” and “Success” poem (texts, see above, p. 10–11).

[ back ] 15. “Tithonus poem” (texts, see above, p. 11).

[ back ] 16. Reading with the first editors κ̣άλα, Μοῖϲ’; for the reading cf. below, p. 26.

[ back ] 17. “Continuation 2” (text, see above, p. 13).

[ back ] 18. I still believe that in fr. 58.23 a new poem begins.

[ back ] 19. The scriptio plena in the Sappho quotation follows the Athenaeus codex A. I plan to prove in a further article that the Sappho quotation does not derive from the third book of Clearchus On Ways of life (fr. 41 Wehrli2).

[ back ] 20. Translation by Gulick (1941:VII 179); his translation of τοῦτο, which is inserted in the Sappho fragment from P.Oxy. 1787, is not reproduced. The interpretation and translation of the second part of the Sappho quotation is extremely doubtful.

[ back ] 21. Cf. the text in Sappho fr. 59.1 Voigt.

[ back ] 22. Half a millennium earlier, they had already served the same purpose in the Cologne Sappho papyrus (Figure 19).

[ back ] 23. Regarding the anonymous lyric poem, the only readings that differ from Gronewald and Daniel’s editio princeps are some corrections and suggestions which were offered by Lundon 2007a:154n23.

[ back ] 24. Cf. the text (above, p. 10), and the constitution of the text by West (p. 14).

[ back ] 25. Text, see above p. 11.