APPENDIX I: THE DOUBLE RECENSION OF THE ARGONAUTICA.1
The fact that there were two recensions or editions of the Argonautica proceeding from Apollonius himself is beyond dispute, but there is considerable difference of opinion as to the time and place at which each of the two appeared. We learn from the lives of Apollonius2 that the first ἐπίδειξις of his work was given at Alexandria. Owing to its unfavourable reception the poet withdrew to Rhodes, and there, after revising the poem, he gave the second ἐπίδειξις. The third ἐπίδειξις took place many years afterwards when he returned to Alexandria at some period after the death of Callimachus. We also learn from our scholia that there was a προέκδοσις or previous edition of the Argonautica distinct from that on which the scholia were written. The difficulty is to determine when and where the προέκδοσις and the ἐπέκδοσις (or later edition) were produced. The scholia throw no light on this point, nor have we any other information to help us. Modern writers on the Argonautica nearly invariably assume that the first edition was published at Alexandria at the time of the first ἐπίδειξις, and that the second edition was published at Rhodes at the time of the second ἐπίδειξις. A modification of this is the view expressed by Linde that the προέκδοσις was published at Alexandria on the occasion of the first ἐπίδειξις, but that there was no second edition given to the public until the occasion of the third ἐπίδειξις at Alexandria. Thus, according to Linde, the successful second ἐπίδειξις at Rhodes was not accompanied by an edition. Merkel,3 however, reasonably urged that if we assume that the first edition was published before the poet left Alexandria for Rhodes we are practically driven into the further assumption that there must have been a third edition as well as a second, for it is scarcely credible that Apolionius could have given a fresh ἐπίδειξις at Alexandria of his work in the form in which it had been published at Rhodes some forty years before: "Eine ἐπίδειξις eines schon edirten Gedichtes ohne neue Umarbeitung ist kaum denkbar; wir hätten also eine dreifache Bearbeitung des Gedichtes, zwei προεκδόσεις, zwei ἐπεκδόσεις." This is possibly true, but there is no actual evidence of two προεκδόσεις; the references in the scholia are invariably to readings contained ἐν τῇ προεκδόσει. Furthermore an ἐπίδειξις did not necessarily involve an ἔκδοσις, and it is in itself improbable that there was any formal edition published on the occasion of the first unsuccessful ἐπίδειξις at Alexandria. It is even improbable, considering the age of the poet, that the whole work had been completed then. In the absence of any definite information, it may be that the προέκδοσις referred to in the scholia was published at Rhodes at the time of the second ἐπίδειξις, and the ἐπέκδοσις at Alexandria when the poem met with such favour on the occasion of the third ἐπίδειξις.
In six passages in the first book we are told in the scholia that there was a different reading in the first edition (ἐν τῇ προεκδόσει):--
(1) I 284 νῦν γε μὲν ἡ τὸ πάροιθεν Ἀχαιιάδεσσιν ἀγητὴ
“δμωὶς ὅπως κενεοῖσι λελείψομαι ἐν μεγάροισιν,
σεῖο πόθῳ μινύθουσα δυσάμμορος κ.τ.λ.
Schol. L δμωὶς ὅπως: ἐν τῇ προεκδόσει κεῖται “βείομαι οὐλομένοισιν ὀιζυρὴ ἀχέεσσιν.
καὶ τὸ ἑξῆς "σεῖο πόθῳ φίλε κοῦρε δυσάμμορος."4
The reading of the earlier edition is rather colourless, and we cannot wonder that the poet altered it. He may also have been influenced by the doubts which were entertained with regard to the form and meaning of βείομαι in Homer. (2) I 512 ῀Ἠ, καὶ ὁ μὲν φόρμιγγα σὺν ἀμβροσίῃ σχέθεν αὐδῇ. “τοὶ δ᾽ ἄμοτον λήξαντος ἔτι προύχοντο κάρηνα
πάν̔τες ὁμῶς ὀρθοῖσιν ἐπ᾽ οὔασιν ἠρεμέοντες
515 κηληθμῷ: τοῖόν τφιν ἐνέλλιπε θέλκτρον ἀσιδῆς. “οὐδ᾽ ἐπὶ δὴν μετέπειτα κερασσάμενοι Διὶ λοιβάς,
ἣ θέμις, ἑστηῶτες ἐπὶ γλώσσῃσι χέοντο
αἰθομέναις, ὕπνου δὲ διὰ κνέφας ἐμνώοντο.
αὐτὰρ ὅτ᾽ αἰγλήεσσα φαεινοῖς ὄμμασιν Ἠὼς
Πηλίου αἰπεινὰς ἴδεν ἄκριας, ἐκ δ᾽ ἀνέμοιο
εὔδιοι ἐκλύζοντο τινασσομένης ἁλὸς ἄκραι,
δὴ τότ᾽ ἀνέγρετο Τῖφυς: ἄφαρ δ᾽ ὀρόθυνεν ἑταίρους
βαινέμεναί τ᾽ ἐπὶ νῆα καὶ ἀρτύνασθαι ἐρετμά.
σμερδαλέον δὲ λιμὴν Παγασήιος ἠδὲ καὶ αὐτὴ
Πηλιὰς ἴαχεν Ἀργὼ ἐπισπέρχουσα νέεσθαι.
Schol. on 515 κηληθμῷ: ἐν δὲ τῇ προεκδόσει μετὰ τοῦτο γέγραπται “ἦμος δὲ τριτάτη φάνη ἠὼς τῇδ᾽ ἐπὶ νύκτα5
βουθύσιον Ἑκάτοιο καταυτόθι δαινυμένοισι,
τῆμος ἂρ ἐκ Διόθεν πνοιὴ πέσεν,6 ὦρτο δὲ Τῖφυς
κεκλόμενος βαίνειν ἐπὶ σέλμασι. τοὶ δ᾽ ὐίοντες.
ἑξῆς δὲ τῶν κειμένων "σμερδαλέον δὲ λιμήν."
Thus, according to the evidence of the schol., in the earlier edition after line 515 came ἦμος δὲ τριτάτη κ.τ.λ., and lines 516-518 were omitted. Dilthey, who is followed by Linde, arbitrarily assumes that there is some mistake in the statement that these three lines were omitted in the earlier text, and thinks that the scholium has been displaced and should in reality be attached to line 518, not to 515. The last line of the passage in the προέκδοσις viz. κεκλόμενος βαίνειν ἐπὶ σέλμασι. τοὶ δ᾽ ἀίοντες cannot have been immediately followed by σμερδαλέον δὲ λιμήν κ.τ.λ., as there would be no construction. In the schol. Par. we find τοὶ δ᾽ ἀίοντες ἰθὺς ἔβαινον, and Gerhard proposed to fill up the lacuna by ἰθὺς ἔβαινον ἄφαρ λελιημένοι εἰρεσιάων, so that the passage would run as follows:-- “κεκλόμενος βαίνειν ἐπὶ σέλμασι. τοὶ δ᾽ ἀίοντες
” [ἰθὺς ἔβαινον ἄφαρ λελιημένοι εἰρεσιάων.] “σμερδαλέον δὲ λιμήν κ.τ.λ. οφ τηε ρεασον οφ τηε ξηανγε ιν τηε σεξονδ εδιτιον λινδε σαψς "φαξιλε ιντελλιγιτυρ ξυρ ιν ηυνξ μοδυμ ποετα ναρρατιονεμ ιν σεξυνδα εδιτιονε ξομμυταριτ: σινε δυβιο ιν εο οφφενδεβατ θυοδ βιδυι σπατιυμ υνο παρτιξιπιο δαινυμένοισι βρεϝισσιμε αττιγερατ, ξυμ εα θυαε ετ πραεξεδυντ ετ σεθυυντυρ ξοπιοσισσιμε εχορνατα εσσεντ. ατθυε ηοξ θυιδεμ ναρρατιονις ϝιτιυμ ξορρεξτιονε ποεταε απτισσιμε τολλιτυρ."
(3) I 542 ἀφρῷ δ᾽ ἔνθα καὶ ἔνθα κελαινὴ κήκιεν ἅλμη “δεινὸν μορμύρουσα ἐρισθενέων μένει ἀνδρῶν.
Schol. μορμύρουσα: δεινῶς ταρασσομένη καὶ κινουμένη φοβερῶς. ἐν δὲ τῇ προεκδόσει “μορμύρουσα τυπῇσιν ἐρισθενέων μένει ἀνδρῶν.
” The reading of the later edition gets rid of the awkward combination of the two datives τυπῇσιν and μένει. Linde adopts Dilthey's view of the reason for the change, "ut onomatopoieticos strepitum navis per reluctantes undas ruentis pondere duorum spondeorum imitaretur." The reason which Merkel suggests is very strange: "videtur κυριότης verbi μορμύρειν quaesita, quam ex Homero observatam exponunt Hesychius et Doederlinus, usurpavit altero loco Apollonius. In priore scriptura ἀφρῷ aegre cum eo verbo iungebatur, prout fit apud Homerum ubique." What bearing the 'κυριότης verbi μορμύρειν' has on the matter is not evident. The note in Hesych. is μορμύρων: ταράσσων, ἀφροὺς ἀποβάλλων: κυρίως δὲ λέγεται ἐπὶ ποταμοῦ μέγα ῥεῦμα ἔχοντος. In our passage ἀφρῷ goes with κήκιεν, and in the other passage where Apollonius uses μορμύρειν (IV 287) ἀφρός does not occur. (4) I 725 τῆς μὲν ῥηιτερόν κεν ἐς ἠέλιον ἀνιόντα “ὄσσε βάλοις, ἣ κεῖνο μεταβλέψειας ἔρευθος,
δὴ γάρ τοι μέσση μὲν ἐρευθήεσσ᾽ ἐτέτυκτο,
ἄκρα δὲ πορφυρέη πάντῃ πέλεν.
Schol. τῆς μὲν ῥηίτερον: ἐν τῇ προεκδόσει τῷ "τῆς μὲν ῥηίτερόν κεν" ἑξῆς ἐστιν "ἄκρα δὲ πορφυρέη." οἱ δὲ μεταξὺ δύο οὐκ εἰσίν. The statement that the verse τῆς μὲν ῥηίτερόν κεν ἐς ἠέλιον ἀνιόντα was in the first edition immediately followed by ἄκρα δὲ πορφυρέη κ.τ.λ. is on he face of it incredible, and so Gerhard proposed τῆς μὲν ῥηίτερόν κεν ἐς ἥλιον ὄσσε βάλοισθα, a reading which is very questionable on account of the trisyllabic form ἥλιος which is never found in the Argonautica, though it occurs once as a proper name in the Odyssey.7 Linde favours Dilthey's suggestion, τῆς μὲν ῥηίτερόν κεν ἐσέδρακες ἠέλιόνδε. The object of the change in the second edition seems to have been to justify the strong expression τῆς μὲν ῥηίτερον κ.τ.λ. by mentioning that the middle of the robe and not merely the border was of a vivid hue. (5) I 788 ἔνθα μιν Ἰφινόη κλισμῷ ἔνι παμφανόωντι “ἐσσυμένως καλῆς διὰ παστάδος εἷσεν ἄγουσα.
Schol. ἔνθα μιν: ἐν τῇ προεκδόσει “ἔνθα μιν Ἰφινόη προδόμου διὰ ποιητοῖο
ἐσσυμένως καλῆς ἐπὶ δίφρακος εἷσεν ἄγουσα.
In the second edition the poet replaced the very unusual δίφραξ (= θρόνος γυναικεῖος Hesych.) by the Homeric κλισμός, and for the Homeric πρόδομος substituted παστάς, a word denoting greater splendour, which is not found in Homer. Merkel says that Apollonius changed πρόδομος because it was used in an un-Homeric sense in the earlier edition; but, even if we assume that this assertion is true, Merkel quite overlooks the fact that we have πρόδομος used in the same way in III 278 of the palace of Aeetes. (6) I 800 δήμου ἀπορνύμενοι λαοὶ πέρθεσκον ἐπαύλους “ἐκ νηῶν, αὐτῇσι δ᾽ ἀπείρονα ληίδα κούραις
δεῦρ᾽ ἄγον: οὐλομένης δὲ θεᾶς πορσύνετο μῆτις
Κύπριδος, ἥ τέ σφιν θυμοφθόρον ἔμβαλεν ἄτην.
Schol. αὐτῇσι δ᾽ ἀπείρονα: ἐν τῇ προεκδόσει “ἐκ νηῶν, ἄμυδις δὲ βόας καὶ μῆλα κόμιζον
αὐτῇσιν κούρῃσιν ἀπείρονα ληίδα δεῦρο.
καὶ τότ᾽ ἔπειτ᾽ ἀνὰ δῆμον ἀάατος ἔμπεσε λύσσα,
οὐκ οἶδ᾽ ἢ θεόθεν ἢ αὐτῶν ἀφροσύνῃσι.
The change was apparently made in order to define more specifically the θεόθεν of the earlier edition, and also to remove the weakness of giving as alternative causes of λύσσα the visitation of the gods and their own ἀφροσύνη. We cannot assume that these six passages, which are all in the one book, by any means comprise all the changes made in the later edition; there may have been many more mentioned in the fuller notes from which our scholia are excerpts. There is one passage in the second book where the scholia in all probability preserve the earlier reading:-- II 963 λεῖπον Ἄλυν ποταμόν, λεῖπον δ᾽ ἀγχίρροον ῏Ιριν, “ἠδὲ καὶ Ἀσσυρίης πρόχυσιν χθονός.
Schol. ἠδὲ καὶ Ἀσσυρίης: πρόχυσιν ἔφη τῆς Ἀσσυρίας, τουτέστι τῆς Λευκοσυρίας, τὴν ἐγκειμένην χώραν αὐτῆς εἰς θάλασσαν, ὡς καὶ ἐν τοῖς ἐπάνω “λεῖπον Ἄλυν ποταμόν, λεῖπον δ᾽ ἁλιμυρέα χώραν
Ἀσσυρίης ἀνέχουσαν ἀπὸ χθονός.
” That these lines, which are quoted for the purposes of explanation, are from the earlier edition is extremely likely. The formula ἐν τοῖς ἐπάνω does not here refer to anything which precedes in our text, and it is acutely explained by Merkel: “"ἐν τοῖς ἐπάνω verba librarii vitium esse reor, qui cum scriptum invenerit ἐν τῇ προτέρᾳ, velut de Aristarchea altera editione est in schol. Il. Σ 182, visus est sibi legere ἐν τοῖς προτέροις, uti aliquotiens vel scripsit, vel scribere potuit, substituitque usitatiorem locutionem ἐν τοῖς ἐπάνω."” It is much more doubtful whether the line which we have cited in the critical note as following II 1116 in some mss. is a survival from the earlier text. The influence of the earlier recension may serve to explain the critical difficulty in IV 538 sqq.-- 538 Υ̓́λλον, ὃν εὐειδὴς Μελίτη τέκεν Ἡρακλῆι
539 δήμῳ Φαιήκων. ὁ γὰρ οἰκία Ναυσιθόοιο
540 Μάκριν τ᾽ εἰσαφίκανε, Διωνύσοιο τιθήνην,
541 νιψόμενος παίδων ὀλοὸν φόνον: ἔνθ᾽ ὅγε κούρην
542 Αἰγαίου ἐδάμασσεν ἐρασσάμενος ποταμοῖο,
543 νηιάδα Μελίτην: ἡ δὲ σθεναρὸν τέκεν Υ̓́λλον.
546 οὐδ᾽ ἄρ᾽ ὅγ᾽ ἡβήσας αὐτῇ ἐνὶ ἔλδετο νήσῳ
547 ναίειν, κοιρανέοντος ὑπ᾽ ὀφρύσι Ναυσιθόοιο.
Here there is no mention of different readings in the scholia, but the confusion in our mss. seems explicable only on the assumption that the readings of the two recensions have been confounded. In L line 540 is marked β᾽, and is followed in the margin by τυτθὸς ἐών ποτ᾽ ἔναιεν: αὐτὰρ λίπε νῆσον ἔπειτα, which is marked α᾽, and then follows 541 which is marked γ᾽. In G after 539 we find next in order τυτθὸς ἐών ποτ᾽ ἔναιεν: ἀτὰρ λίπε νῆσον ἔπειτα. In other mss. the line beginning τυτθὸς ἐών is read after 539, 540, or 547. Brunck, from whose edition the numbering in all subsequent editions is taken, adopted an arrangement of these lines which he found in a work of Angelus Quirinus, Cardinal at Rome, entitled Primordia Corcyrac. This arrangement, according to Brunck's view, Quirinus had discovered in some ms. of Apollonius in the Pontifical Library under his charge. In Brunck's text we find 538-543 as given above, and then come 544 δήμῳ Φαιήκων. ὁ μὲν οἰκία Ναυσιθόοιο
545 τυτθὸς ἐών ποτ᾽ ἔναιεν: ἀτὰρ λίπε νῆσον ἔπειτα.
546 οὐ γὰρ ὅγ᾽ ἡβήσας κ.τ.λ.
The repetition of 539 in this arrangement is objectionable, and no modern editor has followed Brunck. Wellauer suggests that in the earlier edition Apollonius wrote: “Υ̓́λλον, ὃν εὐειδὴς Μελίτη τέκεν Ἡρακλῆι
δήμῳ Φαιήκων. ὁ μὲν οἰκία Ναυσιθόοιο
τυτθὸς ἐών ποτ᾽ ἔναιεν: ἀτὰρ λίπε νῆσον ἔπειτα,
οὐ γὰρ ὅγ᾽ ἡβήσας κ.τ.λ.
” He thinks that in the later recension the poet added what we now have, deleted 544 and 545, and changed οὐ γὰρ to οὐδ᾽ ἄρ᾽. A similar reconstruction of the passage is given by Gerhard. Gerhard and Wellauer, while abandoning as mistaken Beck's view that the mss. of the Argonautica may be divided into two families each containing the reading of one of the two recensions, were still of opinion that the reading of the προέκδοσις could be ascertained through the variety of readings which are found in our mss. and in the occasional citations of grammarians. By a laborious analysis Gerhard thus enumerated eighty-four passages 'in quibus Argonauticorum prior editio detegitur.' In some of these cases, undoubtedly, there is a presumption in favour of this theory, but in the majority it is at least equally probable that we are dealing merely with the vagaries of grammarians and copyists, and that the variations are not so much traces of the earlier recension as perversions of the later. It is noteworthy that in all the six passages where the scholia mention a different reading in the earlier recension there is no such variation in the mss., a fact which must tell against Gerhard's position.8 Ruhnken maintained that Apollonius in revising his poem was especially careful to remove anything which might seem to have been borrowed from the works of Callimachus. Thus he regarded ἁρμοῖ, a variant in the schol. on I 972, as the reading of the earlier recension which was changed to ἶσον in the later 'ne quid Callimachus sibi surreptum queri posset.'9 But, in addition to other resemblances between the writings of the two poets mentioned in the Introduction10 in connexion with the sources of the Argonautica, there is the striking fact that I 1309, καὶ τὰ μὲν ὧς ἤμελλε μετὰ χρόνον ἐκτελέεσθαι, is said by the Schol. to be a line from Callimachus (Καλλιμάχου ὁ στίχος). There is no other evidence for this assertion, but on the strength of it the line is included in the fragments of Callimachus. When we consider the bitter feeling which existed between the poets, it is indeed remarkable that one of the two rivals should take a complete line from the other. It is, of course, possible that the identity may be a mere coincidence, as Weichert suggests. The only thing in any way distinctive about the line is the use of the form ἤμελλε; otherwise it is commonplace enough to have been written by any number of poets independently. Merkel, in his Prolegomena, argues at great length that the second recension was a thorough-going revision of the first in accordance with the progress which had been made in the meantime by the critics, especially Aristophanes of Byzantium, in the knowledge of Homeric usages and the recension of the Homeric text. Be that as it may, it is only reasonable to assume that Apollonius' grammatical studies, continued for many years, must have led to many modifications in his work. How far the completed ἐπέκδοσις differed from the προέκδοσις we have no means of determining. The variations recorded in the scholia certainly do not affect the poem very materially. Linde is strongly of opinion that, with the exception of III 927 sqq.,11 the alterations in the later edition were, generally speaking, of an unimportant character. So too Weichert,12 "Ich bin überzeugt dass diese Verbesserungen von keiner grossen Bedeutung gewesen sind ... Die Verbesserungen, welche er zu Rhodus vornahm, betrafen auf keinen Fall den Plan und die Anlage des ganzen Gedichtes, sondern bezogen sich bloss auf einzelne Stellen, Verse und Worte."