This Italian Sappho anthology, on p. 57-58, has a fusion of several Lobel-Page fragments (cfr this question of mine), among which is 67(a). In l. 3 of that fragment, which is l. 16 of the fusion, the text:

δ]αίμων ὀλοφ . [

is completed to:

δ]αίμων ὀλόφω[ι

This last word, ὀλόφῳ, is untraceable for me. It would seem to be a dative form from *ὄλοφος, but that doesn't seem to exist (Wiktionary and Perseus and my big Rocci dictionary all fail to report it), and Perseus doesn't find it as an inflected form either. The following part of the text:

κ]αὶ τοῦτ' ἐπίκει̣[ρε                  ] θέλ' ὦν τ' ἄπαισαν
δ]αίμων ὀλόφ[ωι                  τέ]λ̣εσον νόημμα

would seem to correspond to the following part of the translation:

e anche questo (mio male) tu voglia (eliminare); perciò tu, che fra tutte (sei) la dea dai pensieri più astuti, dai compimento al mio pensiero.

which in English reads:

and this (pain of mine) may you also want to (eliminate); thus you, who (are) the goddess of the most astute thoughts among all [goddesses], give fulfilment to my thought.

"And this … also" matches καὶ τοῦτ'. "may you … want" might match θέλ(ε), if that is what we take that word to be elided from. "(pain of mine)" goes into the lacuna, hence the brackets. We have to assume "(eliminate)" matches ἐπίκει[ρε, perhaps mis-accented and meant to be ἐπικεί[ρην, an infinitive. "thus you" matches ὦν τ(υ). "among all [goddesses]" would seem to match ἄπαισαν, which however, if genitive plural, should be ἀπαίσαν. δαίμων matches "goddess". "give fulfilment to my thought" is τέλεσον νόημμα. We have left to match "of the most astute thoughts" and the mystery word. Um, problem: the English is too long. I mean, that word can't be a superlative, right? This word would then have to match either "thoughts" or "astute". However, neither give me anything close to this in Woodhouse.

So what is this word?


After thinking it might be a verb, I googled ὀλοφόω, and landed on this, where on p. 172 we find the following:

enter image description here
enter image description here

Let me try to copy those screenshots:

ὀλοφώϊος, ὁ, ἡ, s.v.a. ὀλοὸς, bey Nicander σίνη ὀλοφώϊα θηρῶν, verderblich, tödlich; bey Homer sind ὀλοφώϊα δήνεα Κίρκης, 10, 289, ὀλοφώϊα εἰδὼς, 17, 248. statt ὀλοφώϊος ὤν. Von ὄλω, ὄλλω, ὀλλύω, ὄλλυμι, kommt auch ὀλόω (ὀλοὸς), ὀλόπω, ὀλόπτω, ὀλόφω, ὀλοφόω, wie von ἐλέπω, ἔλπω, ἐλπὶς, ἐλέφω, ἐλεφάω, ἐλεφαίρω. S. auch ὀλοφύρομαι. Andere leiten es von ὄλω u. φὼς ab.

With my shaky German, this should mean:

ὀλοφώϊος, ὁ, ἡ, (s.v.a.=?) ὀλοὸς, apud Nicander σίνη ὀλοφώϊα θηρῶν, pernicious, deadly; apud Homer there are ὀλοφώϊα δήνεα Κίρκης, 10, 289, ὀλοφώϊα εἰδὼς, 17, 248. instead of ὀλοφώϊος ὤν. From ὄλω, ὄλλω, ὀλλύω, ὄλλυμι, comes also ὀλόω (ὀλοὸς), ὀλόπω, ὀλόπτω, ὀλόφω, ὀλοφόω, like from ἐλέπω, ἔλπω, ἐλπὶς, ἐλέφω, ἐλεφάω, ἐλεφαίρω. See also ὀλοφύρομαι. Others derive it from ὄλω and φὼς.

Of all the verbs mentioned with ὀλοφόω, whence my supplement could stem from, I only find ὀλόπτω on [Perseus][3]:

[![enter image description here][4]][4]

But does όει contract to ῳ, so that ὀλοφόω gives ὀλοφόει>ὀλοφῷ which barytonizes to ὀλόφῳ in Aeolic? And assuming it does, how does "pluck out, tear out" fit in that translation?


I just realized I didn't even look in the anthology's notes. Now, footnote 425 on that page refers to l. 14, footnote 427 is about τέλεσον νόημμα, so the only footnote which could be about this supplement is 426:

Integrazione di E. Puglia.

So at least we have a name to our mystery: Enzo Puglia suggested that mysterious supplement. One could think 427 is about both, but it says "Integrazione di Hunt", and Campbell tells me Hunt suggested ὀλοφ[ώϊος there. Moreover, this thing has the exact same combination + translation, and the matching note is referred to "(eliminare)". Ummm… Scouring through the biblio of that thing, we find two E. Puglia references:

  1. E. Puglia, Per la ricomposizione del quarto libro dei canti di Saffo (POxy 1787), «SemRom» X (2007), pp. 17-39
  2. E. Puglia, P. Oxy. 2294 e la tradizione delle odi di Saffo, «ZPE» 166 (2008), pp.1-8

I will now look at those, if I find them, and if one of them contains the answer, I will self-answer. Before I leave you, however, let me remark how the biblio is ordered randomly instead of alphabetically: big boos to that.


For the sake of completeness, here is the full text:

[![enter image description here][5]][5][![enter image description here][6]][6]

Translated as:

... dagli dèi beati avendo ottenuto425... e anche questo (mio male) tu voglia (eliminare)426; perciò tu, che fra tutte (sei) la dea dai pensieri più astuti, dai compimento al mio pensiero427. Tu certo mi amavi ed eri solita (appagare ogni) mia invocazione; per (realizzarla) adesso (aiutami) secondo l'animo (mio428).
(La dea mi rispondeva così):429 "Per te non è (disdicevole) questo motivo e quello che tu vuoi430 (ottenere non) è affatto eccessivo (da avere ... infatti combattere con me431 (non può)432 neppure Andromeda433, (ma per ciò che ha fatto)434 confidando in una (vita) voluttuosa (non) potrà sfuggire435 (agli immortali436); tu infatti sai bene437 che Nemesi (colpisce chi medita infamie e)438 prevale su tutti." O Saffo, poiché ti ama439 (applicò al carro) le ruote e (per te)la veneranda sovrana440 di Cipro (andò a supplicare Zeus e il Cronide le) accordò (di concederti) un grande dono:441 che tutti quanti il (Sole) splendente (circonda con i suoi raggi)442... ovunque siano raggiunti dalla (tua nobile)fama443... e che tu sulle (rive) di Acheronte444 ...

A few comments:

  • Line 14 corrisponds perfectly with «dagli dèi beati avendo ottenuto» (having obtained from the blissful gods);
  • The part we discussed above seems to have to match with the part of the translation I mentioned above, since then «dai compimento al mio pensiero» (give fulfilment to my thought) obviously matches τέλεσον νόημμα;
  • That famous footnote 426 "Integrazione di E. Puglia" actually refers to "eliminare", which would seem to match ἐπικεῖρε as per above; however, since Puglia combined (I believe) the fragments, that article might still give clues to the mystery supplement;
  • «Tu certo mi amavi» (You certainly loved me) matches σὺ μὰν ἐφίλης plus an implied μ(ε);
  • «ed eri solita (appagare ogni) mia invocazione» seems to match virtually nothing, since all we have is ]έτων κάλημμι, and then we end up in the next item;
  • «per (realizzarla) adesso (aiutami) secondo l'animo (mio)» (to (realize it) now (help me) according to (my) soul) matches the last line of the first pic, with πεδὰ θῦμον αἶψα = «secondo l'animo (mio) [subito]» (αἶψα was somehow ignored), and νῦν δ' ἔννεκα = «per (realizzarla) adesso»;
  • «(La dea mi rispondeva così)» is a very intricate translation for ἔφα, "(the goddess) said";
  • «Per te non è (disdicevole) questo motivo e quello che tu vuoi (ottenere non) è affatto eccessivo (da avere» (fixing a few blatant bracketing errors, "(For you) this reason is not unseemly, and what you want to get (is) not at all excessive (to have)") matches the ἔφα line and the following οὔδεν πόλυ; their reason for adding "for you" is mysterious, but whatever;
  • «infatti combattere con me» (indeed fight with me) matches γὰρ ἔμοι μάχεσθαι;
  • «(non può)432 neppure Andromeda, (ma per ciò che ha fatto)» (even Andromeda (can't, but for what she has done)) would seem to match δεῖ σ', Ἀνδρομέδαν, though the "even" is not there and "δεῖ σ'" is ignored, and the lacuna, with its mere two syllables, can hardly contain all that stuff; note 432 reads «οὐ supplemento di Ferrari, γάρ è integrazione di Hunt»; γὰρ I can see, but the only οὐ is in the next line, so the note is misplaced, or it's BS; or, these guys are using the δεῖ σ' Ἀνδρομέδα text but translating the οὐδ' Ἀνδρομέδα one;
  • «confidando in una (vita) voluttuosa» (confiding in a voluptuous (life)) perfectly matches διαίτᾳ χλιδάνᾳ πίθεισα;
  • «(non) potrà sfuggire (agli immortali);» (will (not) be able to escape (the immortals)) matches οὐ λελάθην ἀθανάτοις ἔστι, literally «it is not possible that the immortals have forgotten (sc. Andromeda)»;
  • Isn't that line inmetrical? It looks like –u– –uuu– –uu–u–u… anaclasis, perhaps? But it's a one-off thing due to supplements… very unconvincing; waith though, it's λελᾰ́θειν, not λελᾱ́θειν, so if the privative alpha were long we'd be OK; and it seems it is, so nothing to see here;
  • «tu infatti sai bene» = σὺ δ' εὖ γὰρ οἶσθα;
  • «che Nemesi (colpisce chi medita infamie e) prevale su tutti» is exactly l. 23, and ends the response of the goddess;

The rest, I leave uncommented. So we have more than just one translation problem here.


Reading further into this, I found a mention of this article, which is the beginning of the fusion we're looking at. However, the mystery supplement are from a fragment that isn't yet combined in that article, but appears to be in the text of this poem given in Una mitra per Kleis. Saffo e il suo pubblico, by F. Ferrari, published 2007. Unfortunately, I cannot find it online. Can someone get ahold of a copy of that book and see what is found in it?

[3]: http://www.perseus.tufts.edu/hopper/morph?l=oloptw&la=greek#Perseus:text:1999.04.0057:entry=o)lo/ptw-contents [4]: https://i.sstatic.net/ZtuKq.png [5]: https://i.sstatic.net/cu4su.png [6]: https://i.sstatic.net/B5wPa.png

2 Answers 2


I'm assuming Hanlon's Razor: that they've mangled Hunt's ὀλοφώ[ιος, by removing the inflection, because they didn't want to commit to a case (ὀλοφώ[ι-), and then going back and reaccenting the word, because noone told the proofreader that ὀλοφώ[ι was not a complete word.

As for the meaning "astute", LSJ notes that ὀλοφώϊος in Homer may mean not "destructive" but "deceptive, tricky", derived not from ὀλόπτω but from ἐλεφαίρω. If you ameliorate "tricky" yet further, you get "astute".

I thought of ὁλο-φαος, but I think it impossible that φαος there would contract into φως in Aeolic.

  • I guess that's the best we can do… at least until we get ahold of that book by Ferrari where he presumably operates the combinations and explains the supplements and translation. The thing that still perplexes me is how that becomes a superlative, and where the "thought" are, but maybe a supplement guess + creative translation can explain that away. Oh, and also where the bleep ll. 12-13 come from. That's anothe mysterious puzzle.
    – MickG
    Commented Aug 10, 2019 at 14:22
  • As for the contraction, I guess αο usually contracts to ᾱ in Aeolic (cfr. genitive plural in the first declension), but I wonder if it would at all in this particular instance.
    – MickG
    Commented Aug 10, 2019 at 14:24
  • Not really related, but concerning the word in here, I found out in an article in the Journal Eikasmós that it would be a hápax, that it was proposed by Ferrari in Una mitra per Cleis, the book mentioned in the question here, and that I should compare «Aesch. fr. 47,818 R.2» and «Epich. fr. 198 K.-A.» about it. Where can I find those comparanda? Is the Aeschylus one anywhere here>
    – MickG
    Commented Aug 10, 2019 at 15:20
  • 1
    books.google.com.au/… . It's just the use of θῶσθαι on its own. And it's in a fragment: loebclassics.com/view/aeschylus-attributed_fragments/2009/… Commented Aug 10, 2019 at 16:53

This morning I inspected the English version of Una Mitra per Cleis, the book by F. Ferrari which contains this text.

First off, @NickNicholas was right: the supplement of the question is actually ὀλόφ[ῳ(), with ὀλόφ[ῳα suggested in the app. crit.

I answer here because I want to address the translation. The Italian antholog mangled not only that supplement, but also l. 21, which is supposed to start ο]ὐδ' Ἀ̣[νδ]ρομέ[δα, the ἐπικεῖ[ρε in l. 15, which was supposed to be ἐπίκει[ρ() with the epsilon being supplemented by Puglia and not acceptable as far as the translation goes, and the translation. Here is the translation in the book:

and be willing to take away also this [misfortune ...] then, you who of all
[are] the goddess [who has thoughts the most] astute, fulfill my wish.
You surely were accustomed [to satisfy me in everything] for which I invoked you,
and now for this [purpose come to aid] straightway as my heart desires.
“This reason, [you said,] is surely not [shameful, and] what you wish to obtain
is truly not excessive [to have ... in truth cannot] contend with me
even Andromeda, [but what she has done] trusting in a luxurious
[lifestyle cannot] have escaped [the immortals], and you, since you know well

Based on this, I came up with a few more supplements. Here is the text with those supplements added:

κ]αὶ τοῦτ΄ ἐπικεί̣[ρην, ἄγ΄, ἔμον πῆμα] θέλ' ὦν τ', ἀπαίσαν
δ]αίμων ολοφ[ῴᾱ φρένας· ὦμοι] τέλ̣εσον νόημμα·
σ̣ὺ μὰν ἐφίλης [–uu––u]έτων κάλη<μ>μι,
νῦν δ' ἔννεκα [––uu– μοι] πεδὰ θῦμον αἲψα'.
'τόδ' αἴτιον οὔ τ' [αἶσχρον', ἔφα, 'κὤ]σ̣σα τύχην θελήσῃς
οὐδὲν πόλυ̣ π̣ά̣[σασθαι, ἀλάθως γὰ]ρ ἔμοι μάχεσθαι
ο]ὐδ΄ Ἀ̣[νδ]ρομέ[δ' οἲ', ὄσσα διαίτᾳ χ]λιδάνᾳ πίθεισα
δράσ΄, οὐ] λ̣ελάθ̣[ην ἀθανάτοις ἔστ]ι, σὺ δ', εὖ γὰρ οἶσθα

which I translate as:

Thus, [prithee,] be willing to eliminate also this [misfortune of mine], you of all
The astute goddess [w.r.t. thoughts; oh], give fulfilment to [my] wish:
Verily you liked […] I invoke(d),
Now because of […] straightaway as [my] heart desires".
"This reason is not [shameful] for you, [she said,] and what you wish to obtain
Is not at all too much [to have; for in truth] contend with me
Even Andromeda [can] not, [what she] trusting in a luxurious [lifestyle
Has done, cannot] escape [the immortals;] and you, for well you know

οἶ' = οἶα, since I found οἶος εἶναι means "can", and δράσ(ε) is an augmentless form of ἔδρασε, matching "has done". ἀλάθως = ἀληθῶς, ὦμοι = ὦ ἔμοι. πάσασθαι, in the elided form πάσασθ', seems to be Ferrari's idea, as it is reported as «20 πά[σασθ'?» in the app. crit. Oh, and θέλ' ὦν τ' = θέλε ὦν τὺ = θέλε οὖν σὺ.

If someone has some brilliant ideas to supplement the unsupplemented lines so that they match the translation, please comment. I was unable to come up with anything. Except for πάντα γ]ε, τῶν κάλημμι, which probably is against the accent on the epsilon, and ἔννεκα [τούτω, but no verb for "aid" I found on Woodhouse would fit the meter there.


σ̣ὺ μὰν ἐφίλης [προσδόμεναι πάντα γ]ε, τῶν κάλη<μ>μι,

or πάντ' ἔμε, the accent being justified as the scribe forgetting that it was ἔμε rather than ἐμέ? Scratch the last idea, we'd need ἔμοι instead. What about:

σ̣ὺ μὰν ἐφίλης [μοι παρέχην πάντα γ]ε, τῶν κάλη<μ>μι,

aside from that accent on the epsilon that gets lost? I'm out of ideas.


If you look at the edition's text and Ferrari's text, it looks like Ferrari forgot 4 letters in l. 20. To address this issue, I tried combining the two images of the fragments. I tried to make sure they were resized by the same factor. This was the result:

enter image description here

What this looks like is:

  • Up to πόλ, it's all on one fragment;
  • The upsilon of πόλυ is split between the two fragments, the joint being quite nice, if missing a small portion of the vertical leg;
  • What follows definitely seems to be the two legs of a pi, with a portion of the top, the rest of which is lost to the small gap between the fragments;
  • Then, we seem to have part of an alpha at the bottom, and an acute accent or circumflex accent at the top.

So the transcription should be ΟΥΔ€ΝΠΟΛΥΠÁ[, or at most ΟΥΔ€ΝΠΟΛΥΠÂ[. I'm taking number 1, and the πάσασθαι supplement is vindicated.

Even with just the bottom fragment, the πυφα was hardly vindicated, given we have three vertical legs and part of an alpha, which could at most give ΥΠΑ – as we have seen above, btw.


The raw transcription of P.Halle 3 given in the XV volume of Ox.Pap. doesn't have the accent on the epsilon, so the supplement I gave isn't contra papyrum. However, the construction isn't reported in the dictionary. I'd need a διὰ or similar to govern the genitive τῶν. Metrically, putting ζὰ just before it would work, however, it would supplant the epsilon with its alpha.

I also gave the uncompleted line some thought. ἀρκέω with dative means "assist", so something like:

νῦν δ' ἔννεκα [τούτ' ἄρκεσον ὦμοι] πεδὰ θῦμον αἶψα

could work, although that ὦ in mid-sentence is kind of unjustified. I was trying to conjugate an imperative aorist for *ἄρκημι, formed from the stem ἀρκε- by adding the athematic verbs' suffix, as Aeolic is wont to do, but I'm not sure what it would be. Based on τίθημι-θές, I'd expect the following to be right:

νῦν δ' ἔννεκα [τούτ' ἔλθ', ἔμοι ἄρκες] πεδὰ θῦμον αἶψα

ἄρκημι-ἀρκές/ἄρκες, but I'm not sure.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.