Villa ————— Greek, Roman, and Byzantine Studies 61 (2021) 183–192 Article copyright held by the author(s) and made available under the Creative Commons Attribution License CC-BY https://creativecommons.org/licenses/by/4.0/ A Note on Digenis Akritas G 5.242 and Z 6.1813 Eugenio Villa N 2001 A. KAMBYLIS published an article on the text of Grottaferrata version of Digenes Akritas1 in which he offered a number of corrections, conjectures, and grammatical and orthographical notes on Trapp’s critical edition of the G version.2 As no new edition or critical study of this text has since been published, his emendations and conjectures have remained unadopted, but also unquestioned. Indeed, among many valid corrections—particularly regarding grammar, ac- centuation, and orthography—there are nevertheless some conjectures that do not seem to rest on similarly solid grounds. This paper aims to reject one of these, namely his proposal to insert Z 6.1813 after G 5.241 and to relocate G 5.242 after G 5.244. I will do so by showing that G 5.242 and Z 6.1813 derive from the same line in the common ancestor γ and that, between the two, the former probably retains the text of γ, while the latter innovates on it. I will then suggest a simpler 1 A. Kambylis, “PARASEMEIOMATA. Zum Text der Grottaferrata- Version des Digenes Akrites,” ByzZeit 94 (2001) 29–61 [repr. A. Kambylis, Graeca – Byzantina – Neograeca. Schriften zur griechischen Sprache und Literatur (Berlin 2019) 639–680]. 2 E. Trapp, Digenes Akrites. Synoptische Ausgabe der ältesten Versionen (Vienna 1971). Kambylis also discusses several proposals and notes made by previ- ous scholars. See A. Tsopanaki, “Ἑρµηνευτικὰ καὶ διορθωτικὰ στὸ κείµενο τοῦ Διγενῆ Ἀκρίτα (χειρόγραφο Grottaferrata),” Hellenica 17 (1962) 75–94; N. Eideneier, “Διορθωτικὰ στὸ κείµενο τοῦ Διγενῆ τῆς Κρυπτοφέρρης,” Hellenica 23 (1970) 299–319; L. Politis, “Digenis Akritas. À propos de la nouvelle édition de l’épopée byzantine,” Scriptorium 27 (1973) 327–351. I 184 ON DIGENIS AKRITAS G 5.242 AND Z 6.1813 ————— Greek, Roman, and Byzantine Studies 61 (2021) 183–192 and more effective conjecture to G 5.242 to improve the mean- ing of the line. The scene is one of the most controversial of the poem, because it portrays Digenis encountering a young girl (the nameless daughter of the emir Haplorabdis) abandoned in the desert, promising to take her to her lover, and raping her on the way. According to Trapp’s critical edition, in the G version the description of the rape reads (G 5.240–245): οὐκ εἶχον, ὅ τι γένωµαι, πῦρ ὅλος ἐγενόµην 240 τοῦ ἔρωτος ὁλοσχερῶς ἐν ἐµοὶ αὐξηθέντος· †καταβαλόντες τὸ λοιπὸν χρείαν τάχα ποιῆσαι† ἐν τῷ κάλλει τοὺς ὀφθαλµούς, ἐν τῇ ἁφῇ τὰς χεῖρας, τὸ στόµα τοῖς φιλήµασι καὶ ἀκοὴν τοῖς λόγοις, ἠρξάµην ἅπαντα ποιεῖν πράξεως παρανόµου. 245 244 τὸ στόµα τοῖς φιλήµασι Legrand (ex Z 6.1814) : τῷ στόµατι φιλήµατα G Kambylis (53–54) argues that the problem of these lines is raised not by G 5.242, as implied by Trapp’s daggers, but by the lack of a verb which takes the four accusatives in G 5.243– 244. He therefore suggests to remove the daggers (pointing out that so did E. Jeffreys in her edition),3 to relocate this line after G 5.244, and to insert Z 6.1813 after G 5.241 in order to pro- vide a verb for the following accusatives. Thus the passage, for which he does not provide a translation, should read: τοῦ ἔρωτος ὁλοσχερῶς ἐν ἐµοὶ αὐξηθέντος, <καταλαβόντος πάσας µου µέλεσι τὰς αἰσθήσεις>4 ἐν τῷ κάλλει τοὺς ὀφθαλµούς, ἐν τῇ ἁφῇ τὰς χεῖρας, τὸ στόµα τοῖς φιλήµασι καὶ ἀκοὴν τοῖς λόγοις· καταβαλόντες τὸ (G : an τε?) λοιπὸν χρείαν τάχα ποιῆσαι ἠρξάµην ἅπαντα ποιεῖν πράξεως παρανόµου. 3 E. Jeffreys, Digenis Akritis. The Grottaferrata and Escorial versions (Cambridge 1998) ad loc. 4 Kambylis (54 n.46) says that this line could also be changed to: κατά- λαβόντος ἅπαντα µέλη καὶ τὰς καὶ τὰς αἰσθήσεις. EUGENIO VILLA 185 ————— Greek, Roman, and Byzantine Studies 61 (2021) 183–192 Regarding the daggers in Trapp’s edition, although he does not offer an indication of why he modifies or obelizes the line, he presumably does so because he detects inconsistencies in either the metrical form, the grammar, or the meaning. G 5.242 is a perfect example of decapentasyllabic verse (i.e. it has exactly fifteen syllables, with a break after the eighth and accents on the sixth, eighth, and fourteenth), so the metre could not have been Trapp’s concern. The grammar too does not pose problems, although some pecularities can be detected: in particular, the adverbial τὸ λοιπόν (“so”) has only one other occurence against eleven for λοιπόν without article,5 τάχα (“quickly/rapidly”) has no other occurrence against six for ταχέως and two for ταχύ,6 and the final infinitive (ποιῆσαι) is somewhat uncommon.7 Instead, Trapp must have been puzzled by the meaning of the line and, in particular, of the participle καταβαλόντες: on the one hand, considering that Haplorabdis’ daughter resisted sexual intercourse and that G 5.245 has a singular first person verb (ἠρξάµην), we would expect a singular participle referring only to Digenis instead of plural; on the other hand, this verb has only three other occur- rences, in which it is always transitive and means “overthrow/ defeat” (G 1.139, 4.662) or “throw down [someone from his horse]” (G 8.76),8 neither of which seems possible here. Indeed, as Kambylis points out, E. Jeffreys does not print G 5.242 within cruces; but it must be noted that she retains Trapp’s dag- gers only for G 1.50, so silence does not mean that she regards G 5.242 as unproblematic.9 5 G 1.160, 4.65, 4.90, 4.447, 4.681, 4.868, 4.1026, 6.214, 6.471, 6.571, 6.775. 6 Respectively G 1.294, 3.8, 3.63, 4.362, 6.81, 6.699, and 2.84, 4.753. 7 E.g. G 6.405, 6.632. 8 This exact meaning is not uncommon in Byzantine texts: e.g. Malalas 5.26 ἀπὸ τοῦ ἵππου καταβάλλει; Anna Comn. 1.9.2, 10.4.7, 12.7.3; Niceph. Bryen. 2.13. 9 Her note to G 5.242–244: “The earthiness of G 5.242 has struck com- 186 ON DIGENIS AKRITAS G 5.242 AND Z 6.1813 ————— Greek, Roman, and Byzantine Studies 61 (2021) 183–192 In fact, all translators, including Jeffreys, have struggled with it and no one has offered a fully satisfying and solid trans- lation.10 Mavrogordato renders καταβάλλω with “to rest,” hold- ing that in this occurence the verb is equivalent to καταλύω because κατάβολος (a word which is attested almost excusively in lexicographical works) means “port of call.”11 Odorico assumes τὸ λοιπόν to be the object of καταβαλόντες, even if it appears to be adverbial in all other occurences,12 and renders καταβάλλω with “to drag.”13 Jeffreys and Jouanno stretch the transitive meaning “to throw down [from a horse]” to the intransitive “to dismount.”14 Another minor but not negligible issue is raised by χρείαν. In ___ mentators as extreme bad taste (Trapp obelised the line) … Probably the text has been distorted by a prurient reaction to its sexual content (cf. G 5.247–256 below), though is is unclear when in the poem’s textual history this happened.” In an email exchange (4 Nov. 2020) Professor Jeffreys wrote to me: “one principle I had [scil. in editing the text] was to obelize as little as possible, especially in G. In this case not only is line 242 very awkward but so is the subsequent syntax, with a singular 1st person verb in line 245 rather than a plural. So how much to obelize? At G 8.76 κατέβαλον is applied to Digenis unseating his bride’s brothers with the verb used transi- tively; in interpreting 5.242 I stretched this to an intransitive use … and so ended up with ‘dismounted’. This is not at all satisfactory so perhaps an obelus is the right reaction.” 10 Passing over that of Impelizzeri, who ignores the line: S. Impellizzeri, Il Digenis Akritas. L’epopea di Bisanzio (Florence 1940) 163–164: “e il desiderio dunque cresceva.” 11 J. Mavrogordato, Digenes Akrites (Oxford 1956) 157: “so when we rested as for natural need”: “for καταβάλλω in this intransitive sense, like καταλύω, cf. κατάβολος, a port of call.” 12 See G 1.160, 4.65, 4.90, 4.447, 4.681, 4.868, 4.1026, 6.214, 6.471, 6.571, 6.775. 13 P. Odorico, Digenis Akritas. Poema anonimo bizantino (Florence 1995) 141: “†trascinando a fare il resto come fosse un bisogno†.” 14 Jeffreys, Digenis 149: “dismounting then to perform our natural func- tions.” C. Jouanno, Digénis Akritas, le héros des frontières (Turnhout 1998) 262: “Alors, mettant pied à terre, sous prétexte de satisfaire un besoin naturel.” EUGENIO VILLA 187 ————— Greek, Roman, and Byzantine Studies 61 (2021) 183–192 its three other occurences it means “supplies/things that are necessary [to do something]” (G 2.12, 2.21, 6.444), which would not make sense here as object of ποιῆσαι. Thus, as for καταβαλόντες, each translation is different and no one seems to hit the mark: Mavrogordato renders it “natural need” in con- nection with his assumption that καταβάλλω means “to rest”; Odorico renders it “need/urge [to have sexual intercourse]” and assumes that it is the object complement of τὸ λοιπόν; Jeffreys and Jouanno render it “natural functions,” in accor- dance with LBG s.v. χρεία “(Ort der) Notdurft, Abtritt, Latrine, Abort.” This last does not raise problems per se, but it would imply that Digenis stopped the horse with the excuse of needing to use the bathroom in order to rape Haplorabdis’ daughter, which is not consistent with the character of Digenis and the values of the poem,15 nor with the Serb folk-song that, according to Impellizzeri’s hypothesis, served as model for the episode.16 Now, to emend the passage in G, Kambylis takes a line from Z, the lost common ancestor of T, A, P, and O reconstructed by Trapp on the basis of T and A. As supposed by Kyriakides, Grégoire, and Pertusi and then unequivocally proven by M. Jeffreys,17 Z represented a compilation from E itself or a copy 15 Jouanno, Digénis 262 n.135, notes a similar excuse, albeit in a different context, in Heliod. Aeth. 2.19.6–7. Some reflections of the Aethiopica are found in G, and Jeffreys, Digenis xlvii, suggested that it was one of the sources of *Digenis, but the matter is disputed. 16 S. Impellizzeri, “Un episodio del ‘Digenis Akritas’ e un canto popolare serbo,” AnnPisa SER. II 11 (1942) 221–228. For other hypotheses see E. Trapp, “Hagiographische Elemente im Digenis-Epos,” AnalBoll 94 (1976) 275–287, at 277, 286; A. R. Dyck, “On Digenis Akritas Grottaferrata Version Book 5,” GRBS 24 (1983) 185–192. 17 S. P. Kyriakidis, “Ἀκριτικαὶ µελέται,” Miscellanea Mercati III (Vatican City 1946) 399–430, at 409–420; H. Grégoire, “Notes on the Byzantine Epic. The Greek Folk-Songs and their Importance for the Classification of the Russian Version and of the Greek Manuscripts,” Byzantion 15 (1940/1) 92–103; A. Pertusi, “Alcune note sull’epica bizantina,” Aevum 36 (1962) 14– 45; M. Jeffreys, “Digenis Akritas Manuscript Z,” Dodone(philol) 4 (1975) 163– 188 ON DIGENIS AKRITAS G 5.242 AND Z 6.1813 ————— Greek, Roman, and Byzantine Studies 61 (2021) 183–192 of it and from a manuscript belonging to the γ branch (the same of G) and consequently called g. Since the whole episode is absent in E, the Z compiler relied only on g. The lines cor- responding to G 5.240–244 (there is no line in Z corresponding to G 5.245) are Z 6.1811–1815: οὐκ εἶχον, ὅστις γένωµαι, πῦρ ὅλος ἐφλεγόµην τοῦ ἔρωτος ὁλοσχερῶς ἐµοὶ προσαυξυνθέντος καὶ πάσας µου καταλαβὼν µέλεσι τὰς αἰσθήσεις, ἐν κάλλει τε τοὺς ὀφθαλµούς, ἐν τῇ ἁφῇ τὰς χεῖρας τὸ στόµα τοῖς φιλήµασι καὶ ἀκοὴν τοῖς λόγοις. 1815 As is immediately evident, almost all the lines in Z are basically identical to those in G: Z 6.1811 to G 5.240, Z 6.1812 to G 5.241, Z 6.1814 to G 5.243, and Z 6.1815 to G 5.244 (to which we could add Z 6.1816 to G 5.246). This suggests that both the Z compiler and the diaskeuast of the G version did not depart from their respective antigraphs. Hence, these lines are to be ascribed to the common ancestor γ. But what about G 5.242 and its corresponding Z 6.1813? Kambylis’ proposal would require concurrently that in this passage γ had two lines (one corresponding to Z 6.1813 and one to G 5.242) and that the diaskeuast of the G version omitted the first one and trans- posed the second one, while the Z compiler (or the copyist of g) copied the first one in the right place and omitted the second one. Apart from being somewhat cumbersome, this reconstruc- tion does not solve the problems posed by καταβαλόντες nor, as we shall see, the need for a verb that takes the four accusatives in G 5.243–244. Instead, the possible scenarios are these: the line was missing in γ and both the diaskeuast of the G version and the Z compiler (or the copist of g) inserted one in the same place; the line was present in γ and both the diaskeuast of the G version and the Z compiler (or the copist of g) modified it, hence we would have a case of diffraction; ___ 201 [repr. E. Jeffreys and M. Jeffreys, Popular Literature in Late Byzantium (London 1983) article V]. EUGENIO VILLA 189 ————— Greek, Roman, and Byzantine Studies 61 (2021) 183–192 the line was present in γ and the diaskeuast of the G version copied it while the Z compiler (or the copist of g) modified it, or viceversa. The first option is the least economical and can be easily discarded by considering the verbs: if the two lines were com- posed ex nihilo, how could it be that they share almost the same verb? Between καταβαλόντες of G and καταλαβών of Z, apart from the number, the only difference is the inversion of beta and lambda, which is more likely a scribe’s error rather than a coincidence or a case of multiple emergence. To determine whether the second or the third scenario has more chance of being right, it is crucial to investigate in each of the two lines whether there are elements that suggest a deviation from or the conservation of γ. At first glance Z 6.1813 fits in the passage better than G 5.242: it has a singular verb that can take the accusative and it refers to “limbs” (µέλεσι) and “senses” (αἰσθήσεις) as in the two following lines. However, this line is only apparently less problematic than the one in G. First, καταλαµβάνω is always attested with the meaning “to reach” in its twelve occurrences in Z.18 Therefore the translation of the line should be “and reaching [scil. I, Digenes] all my senses with the limbs,” which is odd since we would expect the subject to be eros, as in the previous line, and the cases to be inverted.19 Moreover, µέλεσι clearly anticipates the datives in Z 6.1814–1815 (ἐν κάλλει, ἐν τῇ ἁφῇ, τοῖς φιλήµασι, τοῖς λόγοις) and αἰσθήσεις the accusatives 18 Z 2.536, 3.909, 3.940, 3.1158, 5.1822, 5.2264, 6.2560, 7.3096, 7.3420, 8.3958, 10.4231, 10.4490. Also in G the verb has always this same meaning: G 1.50, 1.295, 2.4, 3.89, 3.248, 4.405, 4.447, 4.621, 4.764, 4.879, 4.1059, 5.25, 5.111, 6.63, 6.148, 6.244, 6.532, 6.706, 6.797, 7.112, 8.5. 19 See Trapp, Digenes app. ad loc. : “versus ita corrigendus esse videtur: κ. πάσαις µου κατ. αἰσθήσεσι τὰ µέλη.” It is noteworthy that, in their transla- tion of T (which in this passage is identical to Z), Sathas and Legrand felt the need to change the subject and add a verb: C. Sathas and É. Legrand, Digénis Akritas. Épopée byzantine du dixième siècle (Paris 1875) 151: “l’amour ne cessait de croître en moi et se glissait par mes membres dans tous mes sens.” 190 ON DIGENIS AKRITAS G 5.242 AND Z 6.1813 ————— Greek, Roman, and Byzantine Studies 61 (2021) 183–192 (τοὺς ὀφθαλµοὺς, τὰς χεῖρας, τὸ στόµα, ἀκοὴν), but in the first group no word can be referred to one of the senses, while in the second τοὺς ὀφθαλµοὺς, τὰς χεῖρας, and τὸ στόµα are indeed limbs, but ἀκοήν is unequivocally a sense. Trapp’s suggestion to swap the cases and Kambylis’ to have both words in the ac- cusative do not improve the coherence of these three lines. The same difficulties must have been detected by the redactors of A and P, who modified the passage, and by the redactor of O, who got rid of the rape scene altogether: A 2759–276320 δὲν ἤξευρα τὸ τὶ γενῶ πῦρ ὅλος ἐγενόµην, ὁ ἔρως µέ ἑνίκησεν· καὶ αὔξησεν µεγάλως καὶ πάσας µου καταλαβὼν µέλεσι καὶ αἰσθήσεις ἐκάλει τε τοὺς ὀφθαλµοὺς κρατῶντάς την εἰς χεῖρας τὸ στόµα εἰς τὰ φιλήµατα, τὴν ἀκοὴν ’ς τοὺς λόγους. P 372.30–3321 δὲν εἶχα τί νὰ γένω, µόνον ὅλον. Μου τὸ κορµὶ ἐγέµισε φωτία καὶ ὁ ἔρως ἐπλήθυνεν καὶ ἐγέµισεν ὅλας µου τὰς αἰσθήσεις. Διότις ἔβλεπα κάλλος ὀµµάτων. Ἔπιανα τὰ χέρια, ἐφίλουν τὸ στόµα. O 2372–237322 τῆς κόρης δέν της ἄγγιξεν οὐδὲ ἐπείραξέν την, µάλιστα τὸν Εὐδόξιον νὰ ’πάρ’ ἐδίδαξεν την Adding to these inconsistencies the fact that the noun αἴσθησις is never attested in G or E (so probably also not in the *Digenis) and instead occurs two other times in Z (Z 1.142, 3.1200), it seems legitimate to argue that Z 6.1813 does not represent the original line of g, but rather a clumsy attempt of the Z compiler to modify it. Now, it is not uncommon for the Z compiler to modify or rewrite its antigraphs, but the resulting 20 Ed. A. Meliaraki, Βασιλειος Διγενης Ἀκρίτας. Ἐποποίια βυζαντίνη τῆς 10 ης ἐκτατονταετηρίδος (Athens 1881). 21 Ed. D. Paschalis, “Οἱ δέκα λόγοι τοῦ Διγενοῦς Ἀκρίτου,” Λαογραφία 9 (1926) 310–440. 22 Ed. S. Lambros, Collection de romans grecs en langue vulgaire et en vers (Paris 1880) 111–237. EUGENIO VILLA 191 ————— Greek, Roman, and Byzantine Studies 61 (2021) 183–192 line usually fits the text, while in this case it does not. Given this situation, we must conclude that the Z compiler did not grasp or deemed faulty the corresponding line in g, so he tried to ameliorate it. As said above, G 5.242 is problematic both on its own and in its context. Therefore, if the diaskeuast of the G version mod- ified its model, he did not succeed in making an intelligible line. However, while no evidence corroborates this possibility, two clues point to the contrary. On one side, the virtual agreement of G and Z regarding the verb (καταβαλόντες ~ καταλαβών) assures that the diaskeuast of the G version, at the most, in- verted two letters and changed a singular participle to plural. On the other side, the use of ποιεῖν with χρείαν, which has per- plexed editors and translators, has an antecedent, previously unnoticed, in one of the probable sources of the G version, namely the Alexander Romance,23 which describes—as in G 5.242 —a man who forces himself on a woman: Historia Alexandri Magni (recensio β 1.6 = recensio γ 1.6) ὃν εἶδας ὄνειρός ἐστιν· ὅτε δὲ αὐτὸς αὐθεντὶ ἐπέλθῃ σοι, χρείαν σοι ποιήσει. In conclusion, Kambylis’ proposal should be rejected be- cause it does not improve the understanding of the passage and does not abide by what we know about the textual transmission of the Digenis Akritas. As for Z 6.1813, it is certain that it does not reflect the corresponding line of γ and probably also not of g; by contrast, of G 5.242 the opposite is likely. This situation 23 The matter of the influence of the Alexander Romance on the G version is disputed and no extensive study has been developed yet. See T. Korhonen, “Parva Litteraria Byzantina. Digenes Akrites – Byzantin Aleksanteri ja Herakles,” Skholion. The Bulletin of the Finnish Society for Byzantine Studies 2 (2003) 15–17; C. Jouanno, “La réception du Roman d’Alexandre à Byzance,” Ancient Narrative 1 (2000/1) 301–321, at 319–320; U. Moenning, “Digenes – Alexander? The Relationships between Digenes Akrites and the Byzantine Alexander Romance in their Different Versions,” in R. Beaton et al. (eds.), Digenes Akrites. New Approaches to Byzantine Heroic Poetry (Aldershot 1993) 103– 115. 192 ON DIGENIS AKRITAS G 5.242 AND Z 6.1813 ————— Greek, Roman, and Byzantine Studies 61 (2021) 183–192 suggests that the passage was already corrupt in γ, hence either the line corresponding to G 5.242 was faulty or a lacuna oc- curred. In the latter case, the text is irremediable, in the former we should print the line between daggers or try to emend it. If we are to consider the line to be faulty, a simple and economi- cal conjecture, albeit perhaps not flawless, could be to change καταβαλόντες in καταβαλών την. This would require only the confusion of the tachigraphical sign for epsilon-sigma and the one for eta-nu, it would be consistent with the grammar and the vocabulary of the G version, and it would fit what it is happen- ing in the scene. Digenis Akritas G 5.240–245 would thus read: οὐκ εἶχον, ὅ τι γένωµαι, πῦρ ὅλος ἐγενόµην 240 τοῦ ἔρωτος ὁλοσχερῶς ἐν ἐµοὶ αὐξηθέντος· καταβαλών την τὸ λοιπὸν χρείαν τάχα ποιῆσαι ἐν τῷ κάλλει τοὺς ὀφθαλµούς, ἐν τῇ ἁφῇ τὰς χεῖρας, τὸ στόµα τοῖς φιλήµασι καὶ ἀκοὴν τοῖς λόγοις, ἠρξάµην ἅπαντα ποιεῖν πράξεως παρανόµου. 245 242 καταβαλών την ego : καταβαλόντες G I did not know what I was, I became all fire, for the passion had grown utterly in me. So throwing her down [scil. from the horse] to quickly make use of her— the eyes in the beauty, the hands in the touch, the mouth with kisses, and the hearing with words— I began to do everything that is unrighteous action.24 January, 2021 Scuola Normale Superiore Classe di Lettere e Filosofia Piazza dei Cavalieri 7 56126 Pisa, Italy eugenio.villa@sns.it 24 I heartly thank Fabio Vendruscolo for discussing the passage with me, Elizabeth Jeffreys for the insights on her edition, Jacopo Marcon for correct- ing my English, and the anonymous referee for his suggestions and remarks.