Some Desiderata in the Study of the Derveni Papyrus

Kyriakos Tsantsanoglou
Aristotle University of Thessaloniki
In the present paper, which I hope will be my last on this subject, I do not intend to deal with the reconstruction of the first three columns of the papyrus, the joining of whose fragments has provoked so many rows among scholars. I accept some of the proposals made—naturally not all—and I readily admit that some of our suggestions should be revised. I prefer, however, that any future debate be undertaken between younger scholars, whose patience has not yet been tested as much as mine. Thus, I shall limit myself only to some proposals in two columns indisputably reconstructed, before I proceed to other, more general desiderata.
Throughout this investigation, “we” refers to the three editors of the editio princeps (Kouremenos, Parássoglou, and Tsantsanoglou [= KPT]), while “I” refers only to Tsantsanoglou. The authors of the other proposals are specified by name.

Column IV

3 σ̣ίνεται̣ [..].[: Janko (2008) is right that some traces are visible at a distance of two letters after σ̣ίνεται. He describes these traces as the feet of two vertical strokes. But these feet of verticals appear on the edges of the cuts, left and right, like two thin black strokes. They cannot be relics of letters, or their low ends would be irregularly close to the next line. They are obviously traces of the edge of the carbonized papyrus at the cuts, as is usually discernible in the whole roll. However, there is another trace, which escaped both our and Janko’s notice: a waved horizontal curve somewhat lower than the base line, which cannot but belong to Σ. Following σίνεται, one expects an object, which would usually be preceded by the article. And the only accusative articles that fit the space are τάς (τούς is long) and τὸ σ-, τὰ σ-. Speaking of cosmic justice and of handing over its offenders to the Erinyes, I would venture a synonym of κόσμος: possibly, τὸ σύμπαν (Isocrates 11.12) or τὰ σύμπαντα (Plato Philebus 28d), “the universe.” The Plato passage is especially relevant, since the question in this passage of Philebus concerns former wise men who taught that cosmic order is ordained by νοῦς, king of heaven and earth, and not by the power of chance, or ἀτάκτως. In fact, it is νοῦς that governs τὰ σύμπαντα καὶ τόδε τὸ καλούμενον ὅλον, being in charge of καὶ τῆς ὄψεως τοῦ κόσμου καὶ ἡλίου καὶ σελήνης καὶ ἀστέρων καὶ πάσης τῆς περιφορᾶς.
τ̣ὰ τῆς τύχης γά̣[ρ KPT, τ̣ὰ τῆς τύχης π̣ά̣[θη Janko: We took a tiny trace of ink at the base line as the left-hand foot of alpha. The last two letters can be either ΓΑ̣ or Π, but by no means Π̣Α̣, because the left-hand oblique of the alpha would then overlap the right-hand vertical of Π, no matter how short the latter might be. By enhancing the image, I now see that the end of the horizontal of Γ is visible. It is followed by a high, tiny trace of ink that corresponds to the low trace of ink that we took as belonging to alpha. Apparently, the high and the low traces are the two ends of a slightly curving vertical, whose midpart is cut off at the edge of the piece. Certainly, we should take into account that the letters at the end of the line are usually more crowded than elsewhere in the line. If so, I would read γῆ̣[ν. It would then follow that the god in question did not allow τὰ τῆς τύχης, “fortuitous circumstances,” to seize or take hold of the earth. In other words, the god did not allow earth to be regulated by fortuity, but administered order instead. Socrates, in the Philebus passage, puts it in his usual aporetic fashion:
πότερον … τὰ σύμπαντα καὶ τόδε τὸ καλούμενον ὅλον ἐπιτροπεύειν φῶμεν τὴν τοῦ ἀλόγου καὶ εἰκῇ δύναμιν καὶ τὸ ὅπῃ ἔτυχεν, ἢ τἀναντία, καθάπερ οἱ πρόσθεν ἡμῶν ἔλεγον, νοῦν καὶ φρόνησίν τινα θαυμαστὴν συντάττουσαν διακυβερνᾶν;
In sum, I am making a rough exempli gratia restoration proposal for the text of lines 1–4:
… θ]ε̣ῶν
ὁ κείμ[ενα] μετ̣α̣θ̣[έσθαι οὐκ ἐῶν, οὗ δ’ ἔργον] ἐ̣κ̣δ̣ο̣ῦναι
μᾶλλ[ον ἃ] σ̣ίνεται̣ [τὰ] σ̣[ύμπαντα, οὗτος οὖν] τ̣ὰ τῆς τύχης γῆ̣[ν]
οὐκ εἴ̣[α λα]μβάνει̣ν̣.
… this one of the gods who does not allow what is stable to be altered and whose duty is to hand over (to the Erinyes) rather what are harmful to the universe, this god then did not allow chance to take hold of the earth.
A stronger punctuation after σύμπαντα is also possible, so that οὗτος οὖν vel sim. might start a new sentence.
There can be no doubt that the discussion continues to be about the Erinyes, as in the previous columns and in the ensuing Heraclitus quotation. It is then clear that it is they who constitute the authority to which the offenders are handed over. Why μᾶλλον? Obviously, the distinction is between the universe and the earth. The god’s duty is to hand over to the Erinyes ἃ σίνεται τὰ σύμπαντα, μᾶλλον ἢ ἃ σίνεται τὴν γῆν. In other words, the author corrects the popular concept of Erinyes as avengers only of perjury and homicide, by upgrading their jurisdiction from social to cosmic. We are not told who this god is, but he is in charge of the world’s stability and does not allow fortuity to control earth. Doubtless he is νοῦς. But the author has not yet exposed his system of cosmogony, nor has he made a hint about his henotheistic inclinations and the supremacy of Nous. So he leaves him unnamed among a number of gods in an indefinite partitive genitive.
7 ἥλι̣[ος ̣ ̣ ̣] ̣ου κατὰ φ̣ύσιν KPT (]δ̣ου praeferentes), ἑωυ]τ̣οῦ, ἑαυ]τ̣οῦ vel θνη]τ̣οῦ olim KPT, σκύ]φ̣ου Livrea, μεθό]δ̣ου Janko: I would now prefer ]λ̣ου or ]α̣ου, because a loose bottom end of an oblique is visible in the otherwise muddled area—though I admit that δ has often a similar loose end projecting below its horizontal. If so, I would propose a simple ἥλι̣[ος, ἀλλ]ὰ̣ οὐ κατὰ φ̣ύσιν, with a parenthetic prolepsis on the part of Heraclitus warning the reader that the breadth of the sun he is referring to should not be taken at face value. He has no problem with accepting his ignorance of the real size, inasmuch as his statement about the immutability of the sun’s size holds true with the apparent size as well. I doubt that ἀλλὰ οὐ is a simple case of scriptio plena, as we find sometimes in P.Derv. in verse quotations (VIII.5, XVI.9, XXIV.3). I believe it is a deliberate authorial avoidance of elision for emphasizing two words of substance (ἀλλὰ οὐ); ἀλλὰ οὐ κατὰ φύσιν = NB not in its actual nature. Janko’s μεθό]δ̣ου is too long to fit the size of the gap (M is one of the widest letters of the scribe), and, though later thinkers might well call the system of Heraclitus μέθοδος, I would very much doubt if Heraclitus himself would have used a patently sophistic metaphorical term for describing his epistemology. Our previous ἑωυ]τ̣οῦ, ἑαυ]τ̣οῦ, θνη]τ̣οῦ, and Livrea’s σκύ]φ̣ου are palaeographically impossible.—I do not know why the scribe writes at line 9 the Doric νιν instead of the Ionic μιν, which is attested also by Plutarch 370d and 604a, the other sources of Heraclitus fr. B 94 D-K, but the fact is that he writes very clearly νιν. (Janko erroneously: “μιν imago [...]: νιν perperam ed., Ts.3”)
I restore the whole passage as follows, at times returning to some old proposals of ours:
ἥλι̣[ος, ἀλλ]ὰ̣ οὐ κατὰ φ̣ύσιν, ἀν̣θρω[πηΐου] ε̣ὖρος ποδός [ἐστι,]
τὸ μ̣[έγεθο]ς̣ οὐχ ὑπε̣ρβάλλων· εἰ γά̣[ρ τι εὔ]ρους ἑ[ωυτοῦ
ἐκβήσετα]ι̣, Ἐρινύες̣ νιν ἐξευρήσου̣[σι, Δίκης ἐπίκουροι.
The sun, but not in its actual innate dimensions, is a human foot in width, not exceeding (its) size; for if it steps somewhat out of its width, the Erinyes will discover it.
One might question: “So what if they discover it?” It is in response to this question that I prefer εὔ]ρους to οὔ]ρους, because the first elucidates the word-play that is typical of the style of Heraclitus: “If the sun exceeds its width, the Erinyes will de-widen it.”

Column V

2 ].ο̣ι̣.ε̣[ KPT : ]δοιγ̣ε[ Janko recte. At the end, the left-hand foot of an oblique suggests λ, therefore ]δοιγ̣ελ̣[. Then, since this is the column where the author replies to those disbelieving him, εἰσὶν] δ’ οἳ γ̣ελ̣[ῶσι vel sim. Cf. Plato Euthyphro 3c καὶ ἐμοῦ γάρ τοι, ὅταν τι λέγω ἐν τῇ ἐκκλησίᾳ περὶ τῶν θείων, προλέγων αὐτοῖς τὰ μέλλοντα, καταγελῶσιν ὡς μαινομένου.
3 ]......[..]ι KPT : ].....[....]ι Janko: I would now propose reading τε]τ̣α̣γ̣μ̣έ̣[νο]ι. The result cannot be considered certain, due to the almost complete abrasion of the surface, but some faint traces are compatible with each letter in particular. Of τ only the right-hand tip of the horizontal survives. The other letters are more or less complete, but extremely faint. Νaturally, προστε]τ̣α̣γ̣μ̣έ̣[νο]ι vel sim. is equally possible.
4 ἐπερ̣[ω]τ̣ήσ̣[οντες] KPT, Janko: I write now ἐπερ̣[ω]τήσο̣ν̣τ̣ε̣[ς]. The formerly dotted letters are now better visible, and the rest faintly discernible.
5 εἰ θεμι[...]..η̣δ̣α̣[ KPT : εἰ θέμι[ς ἀπιστ]ῆ̣σ̣α̣[ι] Janko: I believe I can see now εἰ θέμι[ς ταῦτ]α̣ δ̣ρ̣ᾶ̣ν̣[. The letters in G 10 are abraded, and so faint. The first α and the ν are half-preserved, but δρα is very likely.
6 ἆ̣ρ̣᾽ KPT : ἂ̣ν̣ Janko, ἐ̣ξ Ferrari: By eliminating two dark horizontal fibers that obscured the image and had led us in the past to read τὰ | ἐν (suggested by West), alpha and rho are clearly visible, and should perhaps be published undotted.
The new reading of lines 3–6 dictates different punctuation, syntax, and translation for the whole passage. Taking αὐτοῖς as dative of the agent with the perfect participle, deleting the comma at the end of line 4, and placing a full stop after [ἕν]εκεν, we gain a different meaning:
... (προσ)τε]τ̣α̣γ̣μ̣έ̣[νο]ι
αὐ̣τοῖς πάριμεν̣ [εἰς τὸ μα]ν̣τεῖον ἐπερ̣[ω]τήσο̣ν̣τ̣ε̣[ς]
τῶν μαντευομένω̣ν̣ [ἕν]εκεν. εἰ θέμι[ς ταῦτ]α̣ δ̣ρ̣ᾶ̣ν̣,
ἆ̣ρ̣᾽ Ἅιδου δεινὰ τί ἀπ̣ιστοῦσι;
… appointed by them we enter the oracle, in order to inquire on behalf of those seeking a divination. If it is right [for us] to do these things, why do they disbelieve in the horrors of Hades? [in other words, “Why do they trust us in one occasion of supernatural significance, but not in the other?”]
It is true that πάριμεν in Herodotus has a future value, so Janko translates “we will enter,” and Ferrari describes it as an ironic, even sarcastic, future: “we’ll end up entering...”. We are lacking the context, and so we cannot know if the sense was something like: “Whenever we shall enter the oracle, it will always be on their command.” But in the fourth century παρίασι (πάριμεν does not occur) already has a clearly present force (as do εἶμι and its compounds in poetry since Homer): Plato Republic (early fourth century BC) 560c–d οὔτε αὐτὴν τὴν συμμαχίαν παρίασιν, οὔτε πρέσβεις πρεσβυτέρων λόγους ἰδιωτῶν εἰσδέχονται, αὐτοί τε κρατοῦσι μαχόμενοι; Demosthenes 21 (Against Meidias, 348 BC) 213 πλούσιοι πολλοὶ συνεστηκότες, ὦ ἄνδρες Ἀθηναῖοι, τὸ δοκεῖν τινὲς εἶναι δι’ εὐπορίαν προσειληφότες, ὑμῶν παρίασι δεησόμενοι; Aeschines 1 (Against Timarchus, 346 BC) 194 τούτῳ γὰρ παρίασιν ἐκ τριῶν εἰδῶν συνήγοροι.
Other minor points of difference between the text of KPT and that of Janko:
5 μαντευομένω̣ν̣ KPT : μαντευομέν[ων Janko
6 ἀπ̣ιστοῦσι KPT : ἀ[πισ]τοῦσι Janko
7 πρ̣αγμάτων KPT : πρ̣[α]γμάτων Janko.
In all these cases Janko notes “Ts.1 haud recte” and follows “ed.” [Ts.1 = “readings and conjectures by K. Tsantsanoglou in Laks and Most 1997:9–22, with his edition of columns 1–7 (93–128)”; ed. = Parássoglou’s and Tsantsanoglou’s readings as they appeared in the unauthorized edition of the P.Derv. in ZPE 47 (1982)]. What happened is simply that between “ed.” and “Ts.1” (and, naturally, Ts.3 = KPT), in other words some time between 1982 (date of the unauthorized edition) and 1993 (date of the Princeton colloquium), we joined a small piece (F 5a) to the top of F 12, which contained the letters in question: 5 ω̣ν̣, 6 π̣ιc, 7 αγ. The fragment figures in its right place in plate 5 of our edition.—The distance between the group F 5a + 12 + 13 + 11 and G 10, as accepted by KPT and Janko, even if an adjustment of a millimeter might be necessary, does not allow the supplements 7 ἕκαστ[α], 8 ὑπὸ τ̣[ῆς] proposed by Janko, especially if one accepts, as Janko does, the supplement 9 νενικημέν[οι, οὐ] μανθ[άνο]υσιν. In order to attain perfect alignment, we proposed 7 ἕκαστ[ον], 8 ὑπό [τε γὰρ]. Janko is correct that we should have published τ̣[ε and not [τε. Janko is also right that an ι is added supra lineam in 8 ἁ̣μαρτης̣.
Now, leaving aside the contested readings, which will likely only prove to be more and more so, let us turn to another topic: trying to find out who the author of the book is. I have repeatedly made the same attempt, not observing the advice of my good friend, colleague, and co-author Theokritos Kouremenos: “attempting to identify the Derveni author in the light of the available evidence seems to be an exercise of rather low epistemic value” (in KPT 2006:59). No doubt, he is technically right, but I hope it will not constitute a mortal sin to apply guesswork and conjecture in exploring so important a question.
It is this belief that led me to consider Charles Kahn’s suggestion about Euthyphro as a very likely possibility. [1] Whoever reads Plato’s Cratylus and Euthyphro cannot fail to perceive the similarities between the Euthyphro described and presumed in these dialogues and the author of the Derveni book. This does not involve precise affinities in his philosophical theories or his religious creeds, because it is clear that Plato does not take seriously these features of Euthyphro and, when he does not caricature them, he speaks of them with a flippant attitude. The question is rather of the image of the author that comes into sight after the subtle ironies or the grotesque exaggerations are removed. And the impression is of a rather whimsical personage, coming from Prospalta, an Attic deme in Mesogaia, but whose family had connections, possibly as cleruchs, with Naxos; who was a contemporary of but younger than Socrates; who was a religious practitioner professing to be an expert authority in sacrifices and prayers; who distinguished himself from the many and the ignorant; who was a soothsayer, whose fellow citizens did not believe his prophecies and derided him as a lunatic; [2] who was a follower of the then-trendy practice of etymologizing divine names, according to which several thinkers claimed not only to have decoded the deeper sense (ὑπόνοια) of time-honored texts, but also to have figured out fundamental truths about creation, reality, and existence. This description is a faithful reproduction from Plato’s dialogues, with only one exception. It was arbitrary on my part to employ the term “author” for Euthyphro. Yet it was Wilamowitz who observed, long before the discovery of the Derveni Papyrus, that Socrates’ hints about Euthyphro would have been unintelligible in the year of the writing of the Cratylus, some forty years after the incident described in the dialogue, had the latter not put down on paper his idiosyncratic teachings. [3]
Now, it seems that this etymological perversion was widely spread. Philodemus, On Piety (part 1, 19.518–541, in Dirk Obbink’s 1996 edition), in an interesting passage mentioned also by Richard Janko, [4] refers to the attack of Epicurus against those individuals who, by changing some letters in the names of gods, would throw away the divine from the world. They are named: Prodicus, Diagoras, Critias, and others (καὶ ἄλλοι). There follows a special reference to Antisthenes. While there is hardly any evidence about Prodicus, Diagoras, and Critias with regard to etymology, Plato’s Cratylus offers ample information about Euthyphro.
396d–e ΣΩ. Καὶ αἰτιῶμαί γε, ὦ Ἑρμόγενες, μάλιστα αὐτὴν [sc. τὴν σοφίαν] ἀπὸ Εὐθύφρονος τοῦ Προσπαλτίου προσπεπτωκέναι μοι· ἕωθεν γὰρ πολλὰ αὐτῷ συνῆ καὶ παρεῖχον τὰ ὦτα. κινδυνεύει οὖν ἐνθουσιῶν οὐ μόνον τὰ ὦτά μου ἐμπλῆσαι τῆς δαιμονίας σοφίας, ἀλλὰ καὶ τῆς ψυχῆς ἐπειλῆφθαι. δοκεῖ οὖν μοι χρῆναι οὑτωσὶ ἡμᾶς ποιῆσαι· τὸ μὲν τήμερον εἶναι χρήσασθαι αὐτῇ καὶ τὰ λοιπὰ περὶ τῶν ὀνομάτων ἐπισκέψασθαι, αὔριον δέ, ἂν καὶ ὑμῖν συνδοκῇ, ἀποδιοπομπησόμεθά τε αὐτὴν καὶ καθαρούμεθα ἐξευρόντες ὅστις τὰ τοιαῦτα δεινὸς καθαίρειν, εἴτε τῶν ἱερέων τις εἴτε τῶν σοφιστῶν.
Etymology—or rather, weird etymology—is the δαιμονία σοφία that, emanating from Euthyphro of Prospalta, overtook Socrates. The latter decides to contemplate names and/or nouns (περὶ τῶν ὀνομάτων ἐπισκέψασθαι) by making use of this wisdom today, but to conjure it away tomorrow and find some priest or sophist to purify him and his collocutors. Of the thinkers named by Epicurus, it is to Antisthenes that the saying ἀρχὴ παιδεύσεως ὀνομάτων ἐπίσκεψις is attributed (fr. 38 Decleva Caizzi), no matter what meaning modern pedagogy attaches to it. As for the purification that Socrates feels he is in need of, it does not seem unrelated to the description of the etymologists by Epicurus as “deranged,” “madmen,” and “frenzy-stricken” (παρακόπτειν καὶ μαίνεσθαι, καὶ βακχεύουσιν αὐτοὺς εἰκάζει)—as well as to the initial impression of numerous modern scholars, who saw the Derveni author as a raving lunatic.
Even so, however, Euthyphro is but one out of several possible authors who practiced etymology. The difference from the other thinkers, if we continue employing the Philodemus passage, is that the Derveni author (as well as Euthyphro of Plato’s dialogues) does not eliminate the divine element from nature—as we know from other sources too that the aforementioned did (τοῖς τὸ θεῖον ἐκ τῶν ὄντων ἀναιροῦσιν)—but does exactly the opposite: he puts forward arguments that support the role of the divine in cosmogony and cosmology.
Antisthenes—always according to Epicurus—stands also in opposition to this group of thinkers. The sentence that mentions him is important, though considerably obscure. Some trifling changes, without much altering the image as a whole, may possibly illuminate the passage. The text, as published by Dirk Obbink and accepted by Janko, is:
               Column XIX
          κα[ὶ γὰρ] παραγραμ̣μίζ[ουσι]
535     τὰ τ̣[ῶ]ν̣ θεῶν [ὀνόμα-]
          τα, [κα]θάπερ Ἀν̣[τισ-]
          θέ[νης] τὸ κοινό[τατον]
          ὑποτ<ε>ίνων ἀν̣[αφέρει]
          τὰ κατὰ μέρος [τῆι θέ-]
540     σει καὶ διά τι̣[νος ἀπά-]
          της ἔτι πρότ[ερον·]
For indeed they explain the names of the gods by changing letters, just as Antisthenes, substituting the most common, ascribes the particular to imposition and even earlier through some act of deceit.
My proposals, no more than trivial παραγραμματισμοί, are 537 τὸ κοινὸ[ν ὄνομ᾽], 538 ἀν̣[αιρεῖ] (iam Obbink olim), 539 [συνέ]|σει (etiam Obbink olim), 541 ἔτι πρότ[εροι.]:
          κα[ὶ γὰρ] παραγραμ̣μίζ[ουσι]
535     τὰ τ̣[ῶ]ν̣ θεῶν [ὀνόμα-]
          τα, [κα]θάπερ Ἀν̣[τισ-]
          θέ[νης] τὸ κοινὸ[ν ὄνομ᾽]
          ὑποτ<ε>ίνων ἀν̣[αιρεῖ]
          τὰ κατὰ μέρος [συνέ-]
540     σει καὶ διά τι̣[νος ἀπά-]
          της ἔτι πρότ[εροι.]
For they change letters in the names of the gods, just as Antisthenes, who, proposing the common noun, eliminates sagaciously the particular ones, and through some trickery others even earlier.
ἔτι πρότεροι is the subject of the second part of the comparative clause—καὶ (καθάπερ) διά τινος ἀπάτης ἔτι πρότεροι. What Antisthenes proposes by implication (ὑποτείνω) is the employment of the common noun, that is θεός, but rejects and discards τὰ κατὰ μέρος, the particular proper names, say Cronus, Zeus, Hera, Demeter, and so forth. He makes this rejection in a shrewd manner. Others before him had done the same thing through some kind of trickery. Epicurus is not completely hostile to Antisthenes. While he disapproves of his use of etymology and the elimination of the particular gods, he recognizes that his proposal is made prudently. The word συνέσει, if correctly restored, has to do with Antisthenes’ sagacity and not with the craft of those who, in the myth, forced the different gods on humankind. The sentence proceeds by antitheses: τὸ κοινὸν ὄνομα vs. τὰ κατὰ μέρος, ὑποτείνων vs. ἀναιρεῖ, συνέσει vs. διά τινος ἀπάτης.
If then we are to place Euthyphro and/or the Derveni author within the climate described by Epicurus, we must dissociate him or them from the group of outright atheists—Prodicus, Diagoras, Critias, and others—and connect him with Antisthenes, who taught that the oneness of the divine exists by nature (κατὰ φύσιν), but the plurality of gods by convention (κατὰ νόμον) (fr. 39a Decleva Caizzi, again from Philodemus On Piety). And if Antisthenes is partly commended by Epicurus, despite his refusal of the multiplicity of gods and his use of etymology, Euthyphro and/or the Derveni author must belong to the ἔτι πρότεροι, who reached the same conclusion by using the same stratagem, but some more trickery as well. I dare propose that this further trickery is the employment of allegory, as used by the Derveni author. No matter how important his religious or cosmogonic teachings may be, the claim that they derive from an allegoric interpretation of an Orphic poem can only be described as “fraud,” or ἀπάτη.
It is true that the attitude of the Platonic Socrates towards Euthyphro is somewhat different in the homonymous dialogue from what is in the Cratylus. In the latter, the derision is explicit and unambiguous, whereas in the Euthyphro the ironic discussion touches on the subject in which the mantis was supposed to be an authority. Yet the playful attitude and the mockery are plain. Plutarch 580d (De genio Socratis) describes a fictional meeting with Socrates (with most of its elements deriving from the Euthyphro) in which the philosopher is strolling with the mantis ἅμα τι διερωτῶν (asking continually) καὶ διασείων (throwing into confusion) τὸν Εὐθύφρονα μετὰ παιδιᾶς. In both dialogues, though the main subjects are different, at the bottom there is the problem of the multiplicity of the gods. In both, the Platonic Socrates appears to have an implicit critical position on the problem, whereas the mantis, present in the Euthyphro, and implied in the Cratylus, appears as a supporter of the multiplicity. The difference lies in the fact that the “Euthyphro-inspired” etymologies in the Cratylus, which are supposed to lead to an allegorical/physical explanation of the particular multiple divine names, are ridiculed; in the Euthyphro the problems arising from the myths that depend on the multiplicity of the gods, such as the battles among them, are accepted by Euthyphro as undisputed dogma (Plato Euthyphro 7b5 εἴρηται γάρ), though the allegoric interpretation—for instance, of book 20 of the Iliad (θεῶν μάχη)—as the battle of the elements in the process of cosmogony, must already have been widely known. No doubt, in the Euthyphro, the allegoric interpretation would not serve the arguments of the mantis and would lead the discussion to alien paths, whereas etymology is the main topic in the Cratylus. Therefore, it would be futile to look for a significant development in Plato’s thought on this subject between his early (Euthyphro) and middle (Cratylus) periods. [5]
Apart from Plato’s references or hints, no other direct mention of Euthyphro seems to have survived. [6] Similarly, no certain mention of the Derveni book appears to have come down to us. I must say that I strongly doubt if the reference to the hymns of Orpheus by Philochorus or the quotation of an Orphic verse in the Homeric scholia, both quoted also in the Derveni book (XXII.11 and XXIII.11), can be considered citations from the latter. The Orphic hymns and theogonies must have enjoyed some circulation in antiquity, not only as literary works but also as liturgical texts. Certainly, Philochorus, himself a seer and diviner and a prolific writer on religious topics, would have had no need of the Derveni book for citing an Orphic verse.
The only reference that may be associated with the Derveni book with some likelihood is Syncellus Chronography 140c, i.282.19 Dind. = Cedrenus Historia compendaria i.144.16 Bekk. ἑρμηνεύουσι δὲ οἱ Ἀναξαγόρειοι τοὺς μυθώδεις θεοὺς νοῦν μὲν τὸν Δία, τὴν δὲ Ἀθηνᾶν τέχνην, ὅθεν καὶ τὸ ‘χειρῶν ὀλλυμένων ἔρρει πολύμητις Ἀθήνη’. [7] Τhe passage, with the reference to Zeus = Nous omitted and the first two words of the verse inverted (ὀλλυμένων χειρῶν), occurs also in Ioannes Antiochenus, a Byzantine chronicler who is older than Syncellus (FHG iv.1.21 [2]); the verse in Meletius medicus (De natura hominis, Cramer, Anecdota graeca ... Oxoniensium 118) is possibly contemporary with Ioannes. [8] The combination of Anaxagoreans with allegorizing an Orphic poem is really intriguing, even though the particular allegory (πολύμητις, or rather, πολύεργος, Ἀθήνη = τέχνη) seems too transparent for the deliberately obscure author of the Derveni book. [9]
The lack of reference to the book may, of course, be coincidental. But the lack of reference to the author may possibly mean something—for instance, that he was not taken seriously enough by his contemporaries, just as the situation between Socrates and Euthyphro in Plato’s dialogues seems to be. Understandably, it is not my intention to underestimate the Derveni author as regards the invaluable information he conveys about Presocratic philosophy and mystery religion. But I cannot believe that anyone of his contemporary intellectuals, even if he agreed with his theological conclusions, would regard his interpretations of the Orphic hymn as worthy of attention.
Apart from the character of the author, scholars have occupied themselves with detecting the character of the book. Without further ado, I must say that I am convinced that, in spite of the personal touches here and there, the book is intended as a handbook, or rather an instruction book, or even better a vade mecum for prospective μύσται. That word is mentioned only once (VI.8), but the author does employ synonymous expressions. In many ways it is obvious that the sacrificial instructions and the allegorical interpretations are addressed to them. And it is in their τελεταί that the hymn in question was sung. In which mystery cult, however, the initiates were involved, I do not know. I have the inclination to locate it in eastern Attica, at the mysteries of Phlya, at modern Khalandri, where Orphic hymns constituted a part of the λεγόμενα, but also for which, as tradition has it, Orpheus and Musaeus, his pupil, had written their hymns. [10] I also have considered identifying the oracle mentioned in the Derveni book as the Amphiaraeion at Oropus. But I fear that in both instances I am influenced by Euthyphro’s origin, Prospalta in the Mesogaia area.
On the other hand, if the book is really an initiate’s vade mecum, this satisfactorily explains why the roll was burned. The usual passports certifying that the traveling soul was καθαρά and ἄποινος, purified and not owing a punishment, were the well-known Orphic gold leaves. But those leaves, with their text usually corrupt, full of errors and misspellings, attest to a popular production. The relatives of the deceased might easily order one or possibly buy a ready-made one outside the cemetery, thus cheating the immortals—ἡ τῶν θεῶν ὑπ’ ἀνθρώπων παραγωγή, deceit of gods by humans, in the words of Plato (Republic 364d), was an activity as usual in antiquity as it is today. But the personal vade mecum of the deceased, a book circulating in a small and closed circle of initiates, was no doubt the most authoritative passport that, burned together with the holder, could accompany him to the access gates of Hades.
Speaking of the character of the book, we might possibly speculate about its missing portions. In the editio princeps, I conjectured that the surviving papyrus fragments speak for a twenty-sheet roll, the standard size, according to Pliny the Elder: Natural History 13.77 numquam plures scapo quam vicenae, “never more sheets than twenty to a roll.” [11] In any case, the reality is that we possess, damaged or fragmented, a ten-foot-long roll, almost the whole of it. But though the end of the roll is intact, it does not coincide with the end of the book. In 1983 Martin West showed, beyond a doubt, I believe, that “in all probability the text continued in another roll, or several, which perhaps perished on the funeral pyre.” [12] His study, however, concerned the Orphic poems, and so he limited his investigation to restoring the narrative of the Orphic theogony. It was beyond the scope of his book to reconstruct the rest of the physical theory, something that would anyway be much too risky. Because, whereas the narrative of the Orphic theogony is presented in the Derveni book in the order of the verses inside the poem, the unfolding of the cosmological system does not follow a logical sequence—but rather is made with leaps and backward movements, depending on what allegory each quoted Orphic verse would recall to the author.
As regards the constituents of the Derveni book, let me remind you that the first six columns deal with cultic particulars connected only or mainly with souls, Erinyes and Eumenides (who, according to the author, are also souls), Dike, and Hades. A better look shows, however, that we are not dealing simply with cultic instructions, but with a system of eschatology or soteriology that, among other things, details some cultic practices necessary for salvation. Recall that although cultic practices may, in the popular religious conception, be thought to be sufficient, an intellectual preacher of soul salvation cannot content himself with them. The second chapter of soteriology, or the second prerequisite for salvation, is no doubt the “special knowledge.” Special knowledge in the area of soteriology is usually tantamount to mystic knowledge addressed to selected people, the μύσται. Here, however, it is not a mystic object, but a mystic method of approaching and interpreting a religious, probably a liturgical text, the Orphic theogonical hymn. The physical theory of cosmogony, set forth in the second part of the papyrus, would have nothing mystic or mysterious, if it came from the mouth of, say, Anaxagoras or Diogenes of Apollonia. It is the allegorical interpretation that elevates the theory to higher levels than science and philosophy.
Is there a third prerequisite in the same context? As far as I know, every religious teaching on the salvation of souls presupposes a righteous life in this world. The surviving text mentions ἁμαρτίη and ἡδονή, but only en passant, within the context of the other soteriological references. We used to read also ἄνδρες ἄδικοι, but Janko does away with it, perhaps correctly. Nowhere is a lifestyle system described—what would constitute an Ὀρφικὸς βίος—such as is expressly mentioned by Plato, but also alluded to by Herodotus and Euripides; a chapter on pragmatic anthropology or practical ethics is missing. I propose that this chapter, whose size and contents cannot be estimated by any means, is missing from the beginning of the book. If these speculations prove true, the book must thus have consisted of at least three papyrus rolls. Why must it come from the beginning? I imagine that a soteriological teaching must proceed by elimination. A religious functionary should regularly start by demanding of his entire flock that they follow in their life an inviolable ethical code of commandments. A first selection from the whole congregation would be those who perform certain cultic practices, rites, ceremonies, prayers, sacrifices. The last stage of selection would be the clearly mystical selection, the instruction in some sort of special and profound knowledge, the analogue of an Eleusinian epopteia. Here, it would coincide with the allegoric physical interpretation of the Orphic hymn.
I am well aware that all this is much too speculative, and impossible to prove. Further, the placement of the supposedly missing chapter depends, at least, on the meaning of the sentence found in column XXV.10–12: “The god made the sun of such a form and size as is related at the beginning of the λόγος.” If logos is the present treatise, as we suggested based on the common usage of the word in Herodotus, Plato, and Aristotle, the reference may have been to column IV, the Heraclitus column, where there is mention of the stability in the size of the sun as security for the preservation of cosmic order. If, however, logos is the Orphic poem, often referred to as Ἱερὸς λόγος, as Gábor Betegh suggested, [13] the reference may have been to columns VIII and IX, where god or air/Mind is said to have turned a sufficiently large amount of fire into the sun, for enabling the rest of the ἐόντα to condense and form the present world. At first sight, Betegh’s approach seems to allow a further roll before the surviving one, while ours does not. At any rate, λόγος is used technically not only for the whole of a literary composition, but often for a part of it, as usually in Herodotus (1.140.3, 7.137.3 ἐπὶ τὸν πρότερον λόγον; 4.82, 5.62.1 τὸν κατ᾽ ἀρχὰς ἤια λέξων λόγον, and elsewhere). So, whether the beginning of the λόγος is the opening of the Ἱερὸς λόγος or of the chapter that poses the problem to be faced with the physical interpretation of the Ἱερὸς λόγος, a previous chapter may well have existed.
Now, some breaking news. Only while preparing the written version of this paper did I notice that the papyrus seems to provide column numbering. [14] This numbering appears 0.25–0.30 cm above the first line, near the left-hand edge of the column and at uneven distances from it. The problem is not only that the top margin is missing in most of the columns, but also that even where the margin is preserved, it often happens that the particular point is torn away. The number-letters are conspicuously larger than the letters of the text and are written with a different calamus in paler or more dilute ink, a fact that may explain why we missed them before now.
Though the letters surviving are no more than seven, it is easy to discern that the script is somewhat different from that of the rest of the text, as it exhibits a more formal calligraphic style. Gamma, delta, lambda, and what is preserved of nu are more or less common, slightly curvilinear, but not more so than the letters of the text. Distinctive are the epsilon, rounded, almost in a complete circle; the mu, three-stroked, low, and very flat; and the alpha, with a downward curved crossbar. The uncommon alpha and mu are identical to those of the oldest dated Greek papyrus, the Peukestas order (331–323 BC; P.Saqqara inv. 1972 GP 3).
The only letter that may affect the dating proposed for the writing of the Derveni Papyrus (ca. 340–320 BC) is the rounded epsilon. Such epsilons are found in a few papyri dated to the end of the fourth or the beginning of the third century—dated so, perhaps precisely because of the rounded epsilon. From the third century on, rounded epsilons are regular. [15] Examples include the Berlin Scolia and Elegiacs of P.Berol. 13270 and the Epicharmus or Pseudo-Epicharmea of P.Saqqara inv. 5673 GP 6 (both dated by their editors ca. 300 BC). Important here are the papyri that present both the square and the rounded form: the musical Orestes fragment of P.Vind. G 2315 (ca. 300 BC), where the square form is presented in the notes, but the rounded one in the Orestes text, and the accounts of P.Saqqara inv. 5676 GP 9 (late fourth century BC), where the particular entries use the square epsilon, but the contemporary sum totals the rounded one. I doubt if the evidence is strong enough to allow us to date the Derveni Papyrus one or two decades later than what we did, thus approaching the dating of the burial by the archaeologists (end of the fourth or beginning of the third century BC) or the original dating of the papyrus by E. G. Turner (325–275 BC), before he settled for the fourth century.
The numbering appears as follows:
In column V (G 1)               ]λε [16]
In column XIII (I 26)           ]μγ̣[
In column XIV (C 10)          ]μ̣δ
In column ΧΧ (D 4)             ν̣
In column XXI (D 6)            ]ν̣α
In column XXVI (Α 6)          ]ν̣
In column XXIV some random traces conspire to give the impression of letters, H being the most conspicuous; however, not only are the letters more than what are needed for a number, but they are also written lower than the other number-letters; some traces at the usual height are inscrutable to me. At XXVI (Α 6) the lower part of a ν is clear, but I am not sure if the low trace of a curve that follows is compatible with the digamma as written in the fourth/third century (like a c with its high and low ends extended horizontally). If μγ̣ (=43) is correctly read at XIII, being eight columns after λε (=35) at V, this would verify that the letters are really column numbering. The same holds for ]μ̣δ (=44) at XIV, for ν̣ (=50) at XX, for ]ν̣α (=51) at XXI, if ν̣ were confirmed, and for ]ν ̣ at XXVI, if we might read ]νϝ̣. Of these, the number ν̣ of XX is suspicious, as its size is smaller than the rest. Column XXIV does not help.
It now becomes clear that the numbering was continued from roll to roll and that one roll had preceded ours. If our column V is numbered 35 and the numbering started from column I, the preceding roll should contain thirty columns. To have two rolls absolutely equal in length, one should divide fifty-six by two and have two rolls of twenty-eight columns each. This should mean that two more columns (and possibly one papyrus sheet?) are missing from the beginning of the preserved roll—more, that is, than what we had initially calculated. This would actually verify the proposal of Janko. [17] With the evidence available, it is impossible to decide. Even if both rolls were of twenty equal sheets each, the width of the columns and so their number might have differed from roll to roll, if, for instance, no hexameters were cited in the first part of the book, and thus the columns were able to be narrower. In any case, the existence of a further roll and possibly a further chapter that preceded our text is now verified.
Another issue for further investigation is the question of the magi. When we initially detected their presence, we were thrilled at finding Iranian priests involved in Greek religious affairs. But when we ventured to make any proposal about them, we were faced with suspicion. Was the reading correct? Were they Iranians, or rather Greek μάγοι, crooks, quacks, impostors, magicians? But how on earth could the author recommend and advertise a mystic worship that, in his own words, followed the practices of crooks and magicians? More influential voices than ours were needed to reinforce the view that the author really referred to the priestly caste of Persia. Now Janko has detected one more mention of them. The references to the magi have to do with souls: their reception in the Hereafter, the prayers and the offerings that secure a favorable treatment, their relation with certain daimones, who impede their entrance into the domain of eternal bliss.
Now, is it prudent to limit the influence of Iranian religion on the mysteries promoted by the Derveni author only to some concepts about souls and the cultic details accompanying these concepts (hymns, libations, popana)? The last mention of the magi is found in column VI, with the libations and the offerings of the initiates, which we are told are made in the same way the magi do. Column VII starts with a reference to the Orphic theogonic hymn apparently sung by the initiates. I have already attempted to associate the singing of the hymn with the information provided by Herodotus (1.132), that Persian sacrifices had to be accompanied by an ἐπαοιδή sung by a magus, and that this ἐπαοιδή was called “theogony.” [18] The similarity is striking, but again, is it prudent to limit ourselves to the outward resemblance of the cultic elements? From column VII on, the book deals with the allegoric interpretation of the Orphic hymn, and this is the focal point from here on in. In the author’s words, the intention of Orpheus was to say not riddles but rather great things in riddles. What if these “great things” were influenced by the teachings of the magi? There is a vast literature on the possible Eastern philosophical and religious influences primarily in Ionia, the westernmost satrapy of the Persian state, and from there, through Ionian teachers of rhetoric and philosophy to the rest of the Greek world, but mainly to Athens. [19]
The gist of the physical system the Derveni author exhibits, irrespective of whether it is his own or not, is a compromise between materialism and religion. The agent at the basis of this compromise is Nous, the center of the Anaxagorean system, which was popular at the time. Nous in Anaxagoras is corporeal, yet the finest and purest of all things; he is also infinite and self-ruled. In the Derveni book Nous is aerial, like everything in the world; he prevails over all ἐόντα, as he is equivalent with the whole of them, but is also characterized as φρόνησις τοῦ θεοῦ, the thought or the wisdom of god, which is also described as air. Elsewhere he is named “mightiest” and is also compared to a king. It is he who decided and effected the creation—the turn, that is, from the πρὶν ἐόντα to the νῦν ἐόντα—by giving a principal role in the creative process to the sun. Is then this compromise between the Anaxagorean “Mind” or “Wit” and the ancestral “almighty God”—the meeting and fusion of religion and cosmology—unrelated to the Iranian omniscient creator Ahura Mazdā, which means no more than “Lord Wisdom”? I admit that this is much too speculative, because the concept of the creative Mind or Wisdom, a thinking principle in cosmogony and cosmology, extends throughout the history of religions, from Hesiod’s Metis down to the Intelligent Demiurge of modern theoreticians. In between, we may spot lots of stages, from the γνῶσις of the Gnostics, to the λόγος of the Gospel of John, to the Holy Spirit of the Christian Trinity. I pose the question to my philosophically minded colleagues. It is not for me to answer.


Betegh, G. 2004. The Derveni Papyrus: Theology, Cosmology, and Interpretation. Cambridge.
Burkert, W. 1970. “La genèse des choses et des mots: Le papyrus de Derveni entre Anaxagore et Cratyle.” Études Philosophiques 25:443–455.
Ferrari, F. 2007. “Note al testo delle colonne II–VII del papiro di Derveni.” Zeit-schrift für Papyrologie und Epigraphik 162:203–211.
Hussey, E. 1999. “The Enigmas of Derveni: A Review of A. Laks & G. W. Most (eds.), Studies on the Derveni Papyrus.” Oxford Studies in Ancient Philosophy 17:303–324.
Janko, R. 1997. “The Physicist as Hierophant: Aristophanes, Socrates, and the Authorship of the Derveni Papyrus.” Zeitschrift für Papyrologie und Epigraphik 118:61–94.
———. 2008. “Reconstructing (Again) the Opening of the Derveni Papyrus.” Zeit-schrift für Papyrologie und Epigraphik 166:37–51.
Kahn, Ch. 1973. “Language and Ontology in the Cratylus.” In Exegesis and Argument: Studies in Greek Philosophy Presented to G. Vlastos (ed. E. N. Lee, A. P. D. Mourelatos, and R. M. Rorty) 152–176. Phronesis Suppl. 1. Assen.
———. 1997. “Was Euthyphro the Author of the Derveni Papyrus?” In Laks and Most 1997:55–63.
Kouremenos, Th., G. Parássoglou, and K. Tsantsanoglou. 2006. The Derveni Papyrus, Edited with Introduction and Commentary. Florence.
Laks, A., and G. W. Most, eds. 1997. Studies on the Derveni Papyrus. Oxford.
Larfeld, W. 1914. Griechische Epigraphik. 3rd ed. Munich.
Livrea, E. 2008. “Eraclito nel papiro del Derveni.” Zeitschrift für Papyrologie und Epigraphik 164:8–9.
Obbink, D. 1996. Philodemus On Piety, part 1: Critical Text with Commentary. Oxford.
Sider, D. 1997. “Heraclitus in the Derveni Papyrus.” In Laks and Most 1997:129–148.
Tsantsanoglou, K. 1997. “The First Columns of the Derveni Papyrus and Their Religious Significance.” In Laks and Most 1997:93–128.
———. 2008. “Magi in Athens in the Fifth Century BC?” In Ancient Greece and Ancient Iran: Cross-Cultural Encounters. Transactions of the 1st International Conference, Athens, 11–13 November 2006 (ed. S. M. R. Darbandi and A. Zournatzi) 31–39. Athens.
Turner, E. G. 1968. Greek Papyri: An Introduction. Oxford.
———. 1987. Greek Manuscripts of the Ancient World. Bulletin of the Institute of Classical Studies, Suppl. 46. 2nd ed., ed. P. J. Parsons. London.
West, M. L. 1983. The Orphic Poems. Oxford.
Wilamowitz-Moellendorff, U. von. 1920. Platon. 2nd ed. Berlin.


[ back ] 1. Kahn 1973:156nn5–6, 158n9. Extensively, Kahn 1997.
[ back ] 2. Cf. Numenius, fr. 23, who, depending on the Euthyphro, describes the mantis as ἄνδρα ἀλαζόνα (charlatan, quack) καὶ κοάλεμον (stupid fellow, booby).
[ back ] 3. Wilamowitz 1920: i.204f., ii.76f.
[ back ] 4. Janko 1997:89–90.
[ back ] 5. The inconsistency between the Euthyphro and the Cratylus was noticed already by Proclus (Platonic Theology 5.18), who does not recognize the derision in Plato’s etymologies.
[ back ] 6. The title, Πρὸς τὸν Εὐθύφρονα, of a book by Metrodorus the younger is but a shortening of the full title Πρὸς τὸν Πλάτωνος Εὐθύφρονα vel sim.—in other words, a treatise that criticizes Plato’s dialogue or possibly the famous Euthyphro dilemma about the nature of piety; extensively discussed by Obbink (1996) on 25.701–708.
[ back ] 7. Burkert 1970:443n1; Sider 1997:138.
[ back ] 8. The verse is attributed to Orpheus, fr. 856 Bernabé, by Orion Etymologicum 163.23 (where πολύεργος Ἀθήνη is transmitted).
[ back ] 9. In D-K ii.50, the fragment is ascribed to Metrodorus of Lampsacus (61 fr. 6), whose other allegories are, however, referred to Homer, not Orpheus. Hussey (1999:315) specifies the source as Philochorus.
[ back ] 10. Plutarch Themistocles 1.4; Pausanias 1.22.7, 1.31.4, 4.1.6, 9.27.2, 9.30.12.
[ back ] 11. A statement refuted by preserved rolls that consist of more sheets. Turner (1968:4) describes the twenty-sheet roll as a common but not exclusive size.
[ back ] 12. West 1983:94.
[ back ] 13. Betegh 2004:327–329.
[ back ] 14. Parallels are mentioned by Turner (1987:16).
[ back ] 15. Larfeld (1914: Schrifttafel) records two inscriptions, of 378/7 and of 313 BC, with rounded epsilons.
[ back ] 16. The photograph of G 1 taken by Ch. Makaronas when the sheets of the roll were still stuck together shows the ]λε very faintly; it is much clearer after the treatment of Anton Fackelmann, which also produced, however, a large hole in the area.
[ back ] 17. Janko 2008:38.
[ back ] 18. Tsantsanoglou 1997:111.
[ back ] 19. Tsantsanoglou 2008:31–39.