Aristophanes of Byzantium’s Text of Homer
2016; University of Chicago Press; Volume: 112; Issue: 1 Linguagem: Inglês
10.1086/689961
ISSN1546-072X
Autores Tópico(s)Classical Antiquity Studies
ResumoPrevious articleNext article FreeAristophanes of Byzantium's Text of Homer†M. L. West†M. L. WestAll Souls College, Oxford Search for more articles by this author PDFPDF PLUSFull Text Add to favoritesDownload CitationTrack CitationsPermissionsReprints Share onFacebookTwitterLinked InRedditEmailQR Code SectionsMoreThe most important pre-Aristarchean text of Homer of which we have any knowledge was that edited by Aristarchus' teacher, Aristophanes of Byzantium. It was edited in the sense that he gave thought to its constitution and furnished it with marginal symbols of critical significance. His ἔκδοσις, as it was called, was not an edition in the sense in which we usually understand it, a particular recension that is reproduced in multiple copies. It probably existed in only a single copy.1 But later scholars were able to say, either from autopsy of that copy or on the basis of indirect testimony, that in such-and-such a passage Aristophanes gave or advocated such-and-such a reading.2 A principal aim of this study is to assess the extent and the nature of his activity as a conjectural critic.Our knowledge of Aristophanes' readings comes largely from the Homeric scholia, occasionally supplemented by other sources. The notices in the scholia derive from Aristonicus and Didymus. Aristophanes' text is cited as ἡ Ἀριστοφάνους (sc. ἔκδοσις, sch. A 91, 124a, 553b, 585a1, Γ 51, ε 393), ἡ Ἀριστοφάνειος (sch. A 423–4, B 192b1), or ἡ κατ᾽ Ἀριστοφάνη (sch. Δ 17a, 142d1, H 436b, P 264b1, α 254a, δ 12d).3 It seems to have been no longer extant in the time of Aristonicus and Didymus, as in several places their knowledge of it is demonstrably indirect, coming from Aristarchus' commentaries or from his contemporary Callistratus.4 Aristophanes himself left no commentary, but some of his Homeric erudition was enshrined in his Λέξεις, and this was a source accessible to later writers.The notices of his readings in the scholia have of course suffered in the transmission from selectivity and corruption. We cannot assume that we have all the information that Aristonicus and Didymus gave, and in several cases scholia deriving from them are clearly corrupt or confused. Nevertheless we have useful information relating to several hundred passages.Inherited ReadingsIt is often recorded that Aristophanes' text agreed with one or more earlier texts, especially that of Zenodotus, but in other places those of Antimachus, Apollonius Rhodius, Rhianus, or (if he antedated Aristophanes) Sosigenes, or with one or more of the "city texts."5 These coincidences were probably already noted by Callistratus and/or Aristarchus. They no doubt resulted at least in part from Aristophanes' use of those earlier texts in constituting his own.In the following places Aristophanes had variants in common with Zenodotus (and in the case of the asterisked passages, only with him): A 91, 598. B 801. Γ 18, 57, 126. Δ 137*. Ζ 121. Θ 290*, 304. Ι 158*. Λ 94*, 219. Μ 59*, 66*, 79*, 127*, 128*, 138*. Ν 2*, 71*, 107*, 237*, 245, 246*, 351*, 551*. Ξ 36, 177*, 208*, 229, 259*, 276, 285, 299*, 310, 340*, 400, 412*, 505*. Ο 139*, 301*. Π 223*. Σ 198*, 400*, 466*, 502. Ψ 461*. α 38*. The increasing density of asterisks as the list proceeds probably reflects scholiasts' diminishing assiduity in reproducing fuller data.Aristophanes shared with Zenodotus the omissions of Κ 497 and O 33.He followed Zenodotus in athetizing the following lines: Η 195–99, 443–64, 475. I 688–92 (Zenodotus only 692). δ 62–64. λ 38–43. He is unlikely to have hit on the same passages independently, so this tends to confirm his use of Zenodotus' text.His other recorded agreements with early texts are as follows.With Antimachus: A 298, 424 (plus three city texts in each case).With Apollonius Rhodius: B 436.With Rhianus (asterisks deployed as above): A 553. Σ 10–11 (omission). Τ 41*. Υ 188. Ψ 81*. δ 12*. ε 296*, 393*. ξ 522*.6With Sosigenes: A 91 (= Zenodotus), 124*, 585. Γ 51. These account for all but one of the mentions of Sosigenes in our sources, the other being at A 435. (Another was conjectured by Heyne at I 453.) It appears that one of the scholars who reported Aristophanes' readings also paid attention to Sosigenes' text for the first few books of the Iliad only.With city texts: A 424 (+ Antimachus), 585 (+ Sosigenes), 598 (+ Antimachus and Zenodotus). O 44. Σ 502 (+ Zenodotus). Τ 86. Υ 188 (+ Rhianus).Except for the atheteses shared with Zenodotus, none of the above material provides evidence for Aristophanes' exercise of his critical faculties, only of his reception of readings already current in the tradition, for which he may or may not have known alternatives. Let us move on to the material that shows him as an active critic.AthetesesFirst, his atheteses, which in most cases we are told were approved by Aristarchus. Indeed they seem usually to be recorded simply as a kind of footnote to Aristarchus' discussions: "already athetized by Aristophanes." Some of them are of lines that were absent in Zenodotus' text, and it may have been those absences that aroused Aristophanes' suspicions and led him to look for reasons why the lines might be spurious. The instances of "om. Zen, ath. Arph" are: Θ 284, 385–87, 557–58. Ι 694. Κ 253. Λ 13–14, 78–83, 179–80 (180 ath. Arph), 356, 515. Μ 175–81, 450. Ξ 376–77. Π 237. The lack of examples from the Odyssey is no doubt due to the relative poverty of the scholia.Since Aristophanes left no commentary, his grounds for atheteses are only known where they were recorded by Aristarchus or another of his pupils. When Aristarchus approved an athetesis of Aristophanes', as he very frequently did, he will generally have approved the reasons for it, and we may suppose that they often appear in the sources as his own reasons.7 But he may often have added further arguments of his own, so we cannot simply assume all the reported reasoning to go back to Aristophanes. However, certain sorts of reason seem to be typical. Many of the lines athetized by both scholars are judged to stand more appositely in another context (Θ 385–87, 557–58. I 23–25, 694. K 387, 397–99. Λ 13–14, 180, 356. Ξ 95. Π 237). Others are condemned as repeating the sense of a neighboring line (K 51–52. Π 261) or being otherwise superfluous (K 253. Λ 515. M 350 + 363. O 56–77, 147–48. Ω 423. β 322); as weakening the force of the passage (Θ 235. I 688–92. Λ 515. M 450. Ξ 213. Ω 6–9); as spoiling the narrative logic (Λ 78–83. M 175–81. Ξ 376–77. Ψ 824–25. Ω 614–17), or being out of harmony with a passage elsewhere (H 443–64. Λ 767–85); as being out of character for the speaker (H 195–99. Θ 164–66); as being inopportune (Θ 284. K 387. Ξ 317–27. O 231–358), or simply risible (Θ 189); as misusing Homeric words or using un-Homeric ones (H 475. K 397–99. Π 261. Ω 6–9, 423).In the Odyssey scholia we sometimes find a different pattern, where the pro-athetesis is mentioned first and then one or more reasons are given. We may assume that here too the reasons are those formulated by Aristarchus:9(1) γ 199–200 καὶ παρὰ Ἀριστοφάνει προηθετοῦντο οὗτοι οἱ δύο στίχοι. ἐκ τοῦ λόγου γὰρ τῆς Ἀθηνᾶς (α 301–302) µετήχθησαν ἐνθάδε.(2) δ 62–64 προηθετοῦντο καὶ παρὰ Ζηνοδότωι καὶ παρὰ Ἀριστοφάνει. τό τε γὰρ "σφῶϊν" οὐχ Ὁµηρικῶς µονοσυλλάβως ἐξηνέχθη, ὅ τε ἔπαινος τῶν νέων οὐκ ἀναγκαῖος. There is unlikely to have been any record of Zenodotus' reasoning. The same applies at:(3) λ 38–43 οἱ ἓξ παρὰ Ζηνοδότωι καὶ Ἀριστοφάνει ἠθετοῦντο ὠς ἀσύµφωνοι πρὸς τὰ ἑξῆς, etc. There is little doubt that Aristarchus agreed with the athetesis; the D scholia give ἀθετοῦνται οὗτοι οἱ ἕξ.(4) λ 435–43 ἀθετοῦνται παρὰ Ἀριστοφάνει. The presence of obeli in the text of the thirteenth-century manuscript M (at 435–42) implies that the athetesis was also Aristarchean.(5) µ 53–54 ἀθετεῖ Ἀριστοφάνης. As a reason for the athetesis is added but no counterargument, we may guess that Aristarchus agreed with it. This is confirmed at the corresponding lines 163–64, where we read καὶ ἐνταῦθα οἱ δύο ὀβελίζονται ὡς ἀδικώτατοι. This implies that 53–54 were obelized, sc. by Aristarchus.(6) At σ 229, where we read simply ἠθέτει καὶ Ἀρίσταρχος, it seems likely that the name is an error for Ἀριστοφάνης.(7) At Ψ 259–61 the report of Aristarchus' athetesis has not been preserved, only the notice that "this was also athetized in Aristophanes." Similarly at α 185–86 and ρ 181. At α 97–98 we just find προηθετοῦντο κατ᾽ ἔνια τῶν ἀντιγράφων, which probably covers Aristophanes.(8) At T 327, after reporting that καὶ Ἀριστοφάνης προηθέτει τὸν στίχον, Didymus cites Callistratus as the source for this information and then gives grounds for the athetesis, which to all appearances are Aristophanes' grounds as recorded by Callistratus (presumably in his work Πρὸς τὰς ἀθετήσεις).The following are atheteses of Aristophanes' that Aristarchus apparently did not follow or that he had doubts about:(9) N 658–59. Aristophanes athetized the lines because they implied the resurrection of Pylaemenes, who was killed in E 576–79. He suggested that the interpolator was motivated by the reference to the man's presence at Troy in 644. Aristarchus explained Aristophanes' thinking but said that instead of athetizing we might take it to be a different man with the same name as the one killed in E.(10) Φ 130–35. Didymus reports that Aristarchus in his commentaries recorded the athetesis by Aristophanes and the grounds for it. He adds, "perhaps Aristarchus agreed with the athetesis, as he offered no arguments against Aristophanes."(11) Ψ 104. Aristophanes claimed that Achilles' denial of φρένες to the souls of the dead was an Odyssean idea at odds with the depiction of Patroclus' ghost as thinking and speaking rationally, unless φρένες here had its physical sense and stood for the body as a whole. Aristarchus resolved the contradiction by saying that Patroclus' ghost retained his mental faculties because he was still unburied, whereas Achilles' statement referred to the dead in Hades.(12) ε 247–48. Aristophanes regarded the two lines as doublets. Aristarchus argued that the second line added something material.(13) ι 253–55. Aristophanes athetized the lines as being repeated from γ 72–74 (sch. γ 71). Aristarchus argued that they were more appropriate in ι.(14) λ 161–62. We are simply told Ἀριστοφάνης ἠθέτει.(15) λ 399–403. Athetized by Aristophanes as generated by Agamemnon's reply in 406–8.(16) o 19. ἀθετεῖ Ἀριστοφάνης ἐπὶ σµικρολογίαι παντελῶς. He must have taken the same view of 91, as the scholion there indicates.(17) o 96. οὐκ ἀναγκαῖον ἀθετεῖν· οὐ γὰρ δοῦλοι οἱ θεράποντες, ὥστε ἐν τῆι αὐτῆι οἰκίαι οἰκεῖν. This is Aristarchus' rebuttal of an athetesis by another scholar, probably Aristophanes, who had thought it inappropriate that Menelaus' herald Eteoneus should not live on the premises.(18) π 49. The evidence comes not from the scholia but from Athenaeus (228d), who quotes the verse and then says, Ἀριστοφάνης ὁ Βυζάντιος νεώτερόν φησιν εἶναι τὸ ἐπὶ πινάκων παρατιθέναι τὰ ὄψα, ἀγνοῶν ὅτι κἀν ἄλλοις εἴρηκεν ὁ ποιητής (α 141) "δαιτρὸς δὲ κρειῶν πίνακας παρέθηκεν ἀείρας."10 It is interesting to see Aristophanes employing the criterion of anachronism. Aristarchus is not associated with the athetesis (which might have extended to 55), and the rebuttal is likely to have been his.(19) σ 281–83. εὐτελὲς τοῦτο, διὸ καὶ κεραύνιον παρέθηκεν Ἀριστοφάνης. This is the only place in Homer where we know a keraunion to have been deployed. According to Isidorus Origines 1.20, ceraunium ponitur quoties multi versus improbantur nec per singulos obelantur, κεραυνὸς enim fulmen dicitur.11 Aristarchus probably avoided the symbol because of its imprecision, preferring to obelize each of the lines he suspected.Peculiar ReadingsI come now to readings cited from Aristophanes and not known to have been found in any earlier source. Many of them must in fact have already been current in manuscripts. Others may have been his conjectures.Oral VariantsIn any Homeric manuscript from Aristophanes' time we would expect to find what may be called oral variants, that is, variants not resulting from ordinary miscopying or from deliberate alteration but from casual misremembering or recomposition by a rhapsode or some other copyist who knew the text well enough to be able to write it out without keeping his eyes fixed on the exemplar before him. Aristophanes will not have produced his own text in this way, but there are likely to have been such "wild" texts among the sources that he relied on. The following instances are evidence of it.(20) K 349. Where Aristarchus and the vulgate have ὣς ἄρα φωνήσαντε, Aristophanes (and some other texts) had ὣς ἔφατ᾽· οὐδ᾽ ἀπίθησε βοὴν ἀγαθὸς Διοµήδης, | ἐλθόντες δ᾽ ἑκάτερθε. The dual participle in the vulgate text is disconcerting, since only Odysseus has spoken. The poet put φωνήσαντε, as if it were "after so speaking to each other," because he was about to use a dual main verb, κλινθήτην (350). It is understandable that some rhapsode found this awkward and avoided it by a different formulation involving an extra line, and that Aristophanes, finding both versions in his exemplars, preferred the one that lacked the oddity.12(21) Ψ 332–33. For these two lines, we are told, Ἀρίσταρχος γράφει "ἠὲ σκῖρος ἔην· νῦν αὖ θέτο τέρµατ᾽ Ἀχιλλεύς." Bolling found it incredible that Aristarchus could have had such a radical variant, and conjectured that his name has replaced that of Aristophanes.13 In any case we must take the variant to be that of a wild text, not one introduced by one of these scholars suo Marte. The replacement of νύσσα by the rare word σκῖρος is noteworthy, but it is not clear what significance is to be attached to it.(22) Ψ 805–6. Aristophanes had a different version of the conditions for the armed duel. According to the version of Aristarchus, three papyri, the testimonia, and the medieval tradition, the victory would go to the man who pierced his adversary's flesh to his inner organs. Aristophanes had a less drastic condition: grazing the skin and drawing blood. This looks like someone's deliberate alteration to mitigate the severity of the contest. I would ascribe the rewriting to a rhapsode rather than to Aristophanes himself.(23) α 424. Instead of δὴ τότε κακκείοντες ἔβαν οἶκόνδε ἕκαστος Aristophanes had δὴ τότε κοιµήσαντο καὶ ὕπνου δῶρον ἕλοντο (= τ 427). This is inferior, as it would suggest that the suitors slept in the palace instead of going to their homes as usual. It is a random formulaic substitution, and in accepting it Aristophanes was surely not making a considered choice. The scholiast adds elliptically ἐν δὲ τῆι Ἀργολικῆι προστέθειται, which Ludwich interpreted to mean that the Argolic text had first δὴ τότε κακκείοντες ἔβαν οἶκόνδε ἕκαστος and then κοιµήσαντο καὶ ὕπνου δῶρον ἕλοντο, on the pattern of I 712–13.(24) β 51ab. Aristophanes here had two plus-lines, adapted from α 245–46. The fact that Telemachus speaks only of the suitors from Ithaca and ignores those from elsewhere had exercised Heraclides Ponticus (frag. 173 Wehrli). But we need not assume that the plus-lines were inserted as an answer to Heraclides;14 even less, that Aristophanes himself took this step.15(25) δ 231–32. The Didyman scholion records a variant version of these two lines and ascribes it to Aristarchus: ἰητρὸς δὲ ἕκαστος, ἐπεί σφισι δῶκεν Ἀπόλλων | ἰᾶσθαι· καὶ γὰρ Παιήονός εἰσι γενέθλης. This is then rejected on the ground that Apollo and Paieon were not the same. The refutation is certainly Aristarchus' (cf. Aristonicus ad loc. and on A 473 and E 899), so the rejected version must have been attributed to an earlier scholar, no doubt Aristophanes, as Cobet and others have conjectured.(26) υ 248. The relevant note in the papyrus commentary P. Oxy. 3710 is plausibly supplemented by M. W. Haslam as ἐλθόντες δ[᾽ ἐς δώµατ᾽· Ἀριστοφάνης] γράφει "αὐτ[ίκ᾽ ἔπειτ᾽ ἀνστάντες (ἔβαν δόµον εἲς Ὀδυσῆος)" (= π 407; υ 247 = π 406).Aristophanes = VulgateIn many places the reading attributed to Aristophanes is also that of the vulgate or a large part of it. The reason is presumably not that the vulgate chose to follow Aristophanes but that the reading was already widely current before him. The passages are:A 108 (οὔτέ τί πω). Γ 35. Δ 17. Ε 289, 555, 638. Ι 73, 203. Κ 334. Ν 348, 358. Ξ 40, 236. Ο 53, 134. Ρ 234. Σ 576. Τ 76–77. Ψ 463. α 254. γ 427. ε 334. ζ 29. ξ 12, 112 (σκύφος), 318. ρ 217. υ 174.(27) Αt π 176, according to the manuscripts of a scholion on Theocritus (1.34b), Aristotle preferred γενειάδες to ἐθειράδες on the ground that ἔθειρα refers to the hair on the head, not on the chin. This looks like an Aristarchean observation (cf. sch. X 315), and Lehrs replaced Ἀριστοτέλης with Ἀρίσταρχος, though he later had doubts about it.16 Wendel in his edition of the Theocritus scholia writes, si quis vero Aristophanem Byzantium restituere malet, non oppugnabo, cum huius nomen cum Aristotelis facillime commutari constet.Casual or Minor VariantsIn the following cases the variant ascribed to Aristophanes seems to have no particular significance and may be nothing more than the kind of accidental variant that may arise anywhere.(28) A 108 (οὔτε …) οὔτ᾽ for οὐδ᾽. Followed by Aristarchus.(29) B 164 δ᾽ omitted (giving asyndeton). So also Aristarchus and αἱ χαριέστεραι.(30) B 192 Ἀτρείωνος for Ἀτρείδαο. Followed by Aristarchus, Dionysius Sidonius, and Ixion.(31) Γ 227 τε καὶ for ἠδ᾽. Followed by Aristarchus.(32) Θ 10 ἀπάτερθε for ἀπάνευθε.(33) N 8 οὐ γὰρ ἔτ᾽ for οὐ γὰρ ὅ γ᾽.(34) N 222 νῦν δ᾽ for νῦν or νῦν γ᾽. δέ is out of place; did Aristophanes intend νῦν δὴ? Nauck emends to νῦν γ᾽.17(35) N 713 σφιν for σφι.(36) N 732 νόον τιθεῖ for τιθεῖ νόον.(37) Π 175 ὃν for τὸν. Followed by Aristarchus.(38) P 178. For ὁτὲ δ᾽ ("and sometimes") Aristophanes had τοτὲ δ᾽. Similar variants are found at Λ 63–64; also ποτέ for τοτέ at ω 447–48.(39) P 751 οὐδέ τι for οὐδέ τε. So also a papyrus and some medieval manuscripts.(40) Σ 14 ἂψ ἐπὶ νῆας ἴµεν for νῆας ἔπ᾽ ἂψ ἰέναι. Followed by Aristarchus and some manuscripts.(41) T 96 ἀλλά νυ for ἀλλ᾽ ἄρα.18(42) T 105 θ᾽ omitted.(43) Φ 249 φόνοιο for πόνοιο. So also a papyrus. Aristophanes is likely to have had the same reading at 137. But the two words are often variants, cf. Z 107, Π 651, P 365.(44) X 51 παιδὶ γὰρ ὤπασε πολλά (?) for πολλὰ γὰρ ὤπασε παιδί.(45) α 208 µὲν for γὰρ. Followed by Aristarchus.(46) γ 486 θεῖον for σεῖον. A nonsensical variant of graphic origin. It is even possible that stray ink caused misreading of Aristophanes' text.(47) ε 83. Aristophanes read στεναχῆισι for στοναχῆισι. στεν- and στον- are often confused, and στεναχή is attested for Zenodotus at Ω 512 (where it is also given by a papyrus), so this will be a casual variant.(48) θ 92 αἶψ᾽ for ἂψ. A common variant.(49) κ 124 εἴροντες for πείροντες. Nonsensical; the π was accidentally omitted or obscured in the manuscript.(50) λ 264 µιν for µὲν.(51) λ 288 οὐδ᾽ ἄρα for οὐδέ τι? The scholion ascribes this to Aristarchus; I do not know why Nitzsch transferred it to Aristophanes.19(52) ν 273, ἤϊα for ληΐδα or λήϊα. Again accidental loss of a letter, as in no. 49.(53) ξ 112 πλησάµενος δ᾽ ἄρα οἱ for καί οἱ πλησάµενος.ArticulationAristophanes' introduction of accentuation testifies to his concern with the correct articulation and interpretation of the text.20 We find evidence of this in the following passages. I include cases where the issue is the presence or absence of mute iota.(54) E 638. Aristophanes, followed by nearly everybody ever since, read ἀλλ᾽ οἷον, not (as Tyrannion did) ἀλλοῖον.(55) Z 148. Aristophanes wrote ὥρηι as a dative; Aristarchus apparently read it as a nominative.(56) Z 285. Aristonicus explains Aristarchus' reading ἄτερ που ὀϊζύος, and notes ἔνιοι δὲ ἀγνοήσαντες γράφουσιν "ἀτέρπου." The edition criticized must have been one in which accents were written, and the probability is that it was Aristophanes'.21(57) Θ 206, Ξ 265, Ω 331. Aristophanes, followed by Aristarchus, interpreted the verse-final Ζῆν as Ζῆνα elided before the vowel with which the following line began, and they made the line-break between Ζῆ and ν᾽.(58) N 12. Aristophanes read ἐπ᾽ ἀκροτάτηις κορυφῆις instead of the genitive singular read by Aristarchus and the main tradition.(59) N 29. Aristophanes read γηθοσύνηι as a dative noun, not γηθοσύνη as a nominative adjective; no doubt also at λ 540.(60) Ξ 148. Aristophanes' ὅσσον δ᾽ for ὅσσον τ᾽ was more than a trivial interchange of particles: it implied a different articulation of the passage, with a new sentence beginning at 148.(61) O 606. Herodian records that Aristophanes contemplated both the accentuation τάρφεσιν (as a noun) and ταρφέσιν (as an adjective).22(62) Π 31. Aristarchus criticized someone who read αἴν᾽, ἀρετῆς instead of αἰναρέτη. This very contrived reading is clearly a scholar's conjecture. Aristophanes may have been the man responsible.(63) T 357. In a fragmentary papyrus commentary, P. Oxy. 4452, there is a possible mention of a proparoxytone accentuation of ταρφειαι and a possibility that Aristophanes was named as the authority for it. But it is very uncertain.23(64) Y 30. Aristophanes read ὑπέρµορον as one proparoxytone word, and no doubt also at α 34–35.(65) Y 53. Aristarchus claimed that the place mentioned was called Θεῶν καλλικολώνη, so that there was no justification for those who accented θέων paroxytone, "running." Aristophanes is the likely target of criticism.(66) Φ 122. A scholion of the h class names Ἀριστοτέλης among authorities who accented ἐνταῦθοι, like ἐνταῦθα; Ἀριστοφάνης (van der Valk) is a likely correction.(67) Φ 252. The scholion criticizes first Aristotle for taking µελανόστου as one word, "having dark bones," and then Aristarchus for reading µέλανός του, as the poet does not use του for τεο/τευ; the true reading is µέλανος τοῦ. The linguistic observation is surely Aristarchus', so the reading µέλανός του was probably Aristophanes'.(68) Ω 84. Aristophanes wrote εἵατο with aspiration, which Aristarchus rejected.(69) β 94, 338. Aristophanes wrote µµεγάροισιν, ννητὸς with doubled nasals; Aristarchus did not.(70) δ 140 ψεύσοµαι ἦ ἐτεὸν ἐρέω· Ἀριστοφάνης οὐκ ἀποφαντικῶς ("as a statement") ἀλλ᾽ ἐν ἤθει (in conversational style, sc. as a question). Herodian continues that there is no need to accent ἦ circumflex because it is not a matter of choosing between two alternatives but equivalent to "whether I am right or wrong." As he has referred to Aristophanes' interpretation, it seems likely that it was Aristophanes who had gone for the circumflex.(71) ε 357. Aristophanes separated ὅ τε, presumably with a diastole, to distinguish it from ὅτε "when."(72) η 317. Aristophanes accented εἰδῆις, and he was generally followed in antiquity (and ever since). Tyrannion (rightly) favored εἴδηις.(73) ν 31. Aristophanes read πὰν ἦµαρ as two words.(74) σ 253. Aristophanes read ἦιεν not ἦεν.MorphologyAristophanes sometimes shows concern over correct morphology or what he takes to be genuinely Homeric.24 I put the following instances under this head.(75) A 122 φιλοκτεανώτατε. Seleucus recorded that Aristophanes read φιλοκτεανέστατε (Eust. 1441.18), as if the compound were formed not from κτέανον but from a neuter s-stem *κτέανος.25 Aristophanes is not likely to have created the irregular superlative form himself, but if he found it transmitted somewhere he might have preferred it as being (supposedly) Attic; cf. sch. β 190e ἀνιηρέστερον· Ἀττικόν, ὡς τὸ πτωχέστερον. Aristarchus and Dionysius Thrax regarded Homer as an Athenian, and I suspect that Aristophanes may have anticipated them; cf. below, no. 83.26(76) A 567 ἀάπτους. Ιn Aristophanes' Γλῶσσαι27 the word was registered as ἀέπτους. At N 318, where the Didyman scholion says Ἀρίσταρχος ἀέπτους, ἄλλοι δὲ ἀάπτους, Lehrs rightly corrected to Ἀριστοφάνης. The word ἄεπτος is used by Aeschylus,28 and Aristophanes may have felt that this form, understood as "unspeakable," made better sense than one that appeared to mean "untouchable."(77) B 447. Aristophanes and Aristarchus read ἀγήρων with contraction, not ἀγήραον. Aristarchus, and perhaps Aristophanes too, justified it by citing P 444 ἀγήρω τ᾽ ἀθανάτω τε.(78) H 238 βῶν. Rhianus' text had βῶ, Aristophanes' βοῦν. Aristarchus retained βῶν, but the Didyman scholia (AT) record the opinion that Rhianus' reading might be acceptable as a contraction of βόα, and also what is evidently a justification for Aristophanes' reading: ἐν τοῖς παλαιοῖς ἐγέγραπτο βον, ὅπερ οὐκ ἐνόησαν οἱ διορθωταί. This invokes the theory that the Homeric text had been transcribed from an old form of alphabet, with some errors arising from the ambiguity of e and o. It implies that βοῦν was not a transmitted reading but an emendation by Aristophanes.(79) Λ 686. Aristophanes had the correct reading χρεῖος ὀφείλετ᾽, probably inherited; cf. 688 χρεῖος ὄφειλον. Aristarchus had χρέως ὠφείλετ᾽. Both appear in a few manuscripts, but the majority have the more familiar χρέος, with either ὀφ- or ὠφ-.(80) M 26 σῡνεχές. Aristophanes, followed by Aristarchus, wrote this with a single ν.29(81) N 92 Πηνέλεων. Aristophanes had Πηνέλεον (and so probably at B 494). I suspect that this was an emendation, supported by the µεταχαρακτηρισµός theory.(82) Ξ 44. Aristophanes gave δείδια in place of δείδω (which is from *δέδϝοα); perhaps a deliberate alteration motivated by suspicion of the form (though it occurs elsewhere), perhaps just a manuscript variant.30(83) Ξ 203. Aristophanes and Aristarchus gave ῾Ρείας as against the Ionic ῾Ρείης given by two papyri and the medieval tradition. They may have been moved by belief in Homer's Athenian origin; cf. above on no. 75.(84) O 10. The transmitted κηραπινυσσων was correctly analyzed by Aristarchus as κῆρ ἀπινύσσων; he recognized the verb as the one found in ε 342 and ζ 258. He was generally followed by later scholars, though some upheld the division κῆρα πινύσσων. Aristophanes had given κῆρ ἀπινύσκων, πινύσκω being a form occurring in post-Homeric poets. We may suspect that this was his own conjecture.(85) O 601. Aristophanes is credited with the unaugmented µέλλε for ἔµελλε. Moritz Schmidt changed the name to that of Aristarchus, who patronized unaugmented forms in other places (B 205, Z 155, Φ 84, δ 39, η 235) and no doubt read µέλλε here. But while his name has replaced Aristophanes' in several places, the converse is unexpected, so perhaps originally both scholars were named. We know that Aristophanes had unaugmented ἔλπετο at P 234, where some copies known to Didymus had ἤλπετο.(86) Σ 576. The reading attributed to Aristophanes, ῥαδηλόν, is corrupt and unmetrical, but it seems that he offered a distinctive variant of the obscure epithet.(87) T 30. Aristophanes read ἀλαλκέµεν for ἀλαλκεῖν. The variant may have been in the tradition, or he may have introduced it on doctrinaire grounds.31(88) Φ 1. The Ammonius commentary (p. 80 Erbse) seems to say that Aristophanes held that ἐϋρρεῖος ought by rights to be ἐϋρρῆος, from nominative *ἐϋρρῆς. This is criticized with a more correct analysis.(89) Ω 425. Aristophanes found διδοῦναι in his sources, as in the paradosis, but took exception to it.32 Possibly he proposed replacing it with παρασχεῖν, as at ν 358 he wrote παρέξοµεν, "δυσχεραίνων τὸ διδώσοµεν." A related form occurs at ω 314 διδώσειν (v.l. διδοῦναι), and we may assume that there too Aristophanes queried it.(90) β 50 ἐπέχραον. Aristophanes read ἐπέχρων. Cf. above, no. 77.(91) ζ 264 εἰσίθµη. Aristophanes gave εἰσίσθµη by association with ἰσθµός. We cannot tell whether this was a deliberate change or an accidental one that had already occurred in the tradition.(92) ρ 221 θλίψεται. Αpollonius Sophista, other lexical sources, and some manuscripts give φλίψεται. Τhe Didyman scholion says διὰ τοῦ θ, ὡς καὶ Ζηνόδοτος: this means that Aristarchus read θλίψεται, and as only Zenodotus is cited as an earlier source, the inference might be that Aristophanes gave φλίψεται. (But it might be simply that Didymus had no information about Aristophanes' reading.) φλίβω is the rarer and more original form of the verb and merits adoption for this reason. Perhaps the poet intended a paronomasia with φλιῆισι in the same line, but to argue for φλίψεται on this ground would be begging the question.Lexical SubstitutionIn many places Aristophanes' text diverged from the mainstream by having a different word. At least some of the instances look like a deliberate change to meet a perceived difficulty.(93) I 551 τόφρα δὲ Κουρήτεσσι κακῶς ἦν, οὐδ᾽ ἐδύναντο | τείχεος ἔκτοσθεν µίµνειν: οὐδ᾽ ἐθέλεσκον Aristophanes. Perhaps a pedantic substitution on the ground that they could have stayed outside the wall if they had chosen to.(94) K 169 ναὶ δὴ ταῦτά γε πάντα, φίλος, κατὰ µοῖραν ἔειπες: Aristophanes had τέκος for φίλος. Perhaps he thought it a more appropriate term for Nestor to use to Diomedes, who has addressed him as γεραιέ; but it may be just an accidental assimilation to Ψ 626.(95) K 391 πολλῆισίν µ᾽ ἄτηισι πάρεκ νόον ἤγαγεν Ἕκτωρ: ἤπαφεν Aristophanes.(96) Λ 26 κυάνεοι δὲ δράκοντες ὀρωρέχατο προτὶ δειρήν: σµερδαλέοι and ἐλ(ε)ιχµῶντο Aristophanes.(97) M 40 ἐµάρνατο ἶσος ἀέλληι: ἐµαίνετο Aristophanes. Cf. O 609, where Aristarchus seems to have had µαινοµένοιο for µαρναµένοιο.(98) N 502 Αἰνείας δὲ πρῶτος ἀκόντισεν Ἰδοµενῆος: πρόσθεν Aristophanes, who may have found fault with πρῶτος for πρότερος. Cf. the comment in Didymus, καὶ µήποτε βέλτιον· "πρότερος" γὰρ ἂν εἶπεν.(99) Ξ 474 αὐτῶι γὰρ γενεὴν ἄγχιστα ἐώικει: γάρ ῥα φυὴν ἄγχιστα ἔοικεν Aristophanes. Ajax conjectures that the man he has just killed is a brother or son of Antenor. If the past tense ἐώικει is right, it is because the man, though still visible, is dead. On the other hand, two lines earlier Ajax said οὐ µέν µοι κακὸς εἴδεται, and Aristophanes may well have felt that ἔοικεν was the right tense to follow that present. He was certainly right to reject γενεήν, as a reference to physical likeness is called for. There is a parallel for φυήν at B
Referência(s)