Proper names, especially Greek ones, are more subject to corruption in Latin manuscripts than other words. The restoration of the correct forms is not always easy. When metre provides no guidance, it is often unclear to what extent they retained Greek morphology or were adapted to Latin inflection. It can also occur that a place or a character is only known from a single Latin source. Likewise, derivative adjectives or patronymics used by Latin poets might not be documented in extant Greek texts. Since all these problems are recurrent in Ovid, it is often possible to call into question the transcription of Greek words.
I shall try to exemplify this with the help of three passages of Ovid's Metamorphoses Book 10. I take Tarrant's OCT textFootnote 1 as a starting point, but I add my own critical apparatus, in which I provide more readings from the recentiores.Footnote 2
(1) 10.76–7
77 (h)(a)emum Ω, edd. plerique: (h)(a)emon B3DrEsV6 A7Ld7(ut uid.) Es2 Bs7Lr26Nr33sV44, Heinsius, rectius
Almost all manuscripts read (H)(a)emum, but a group of recentiores from the twelfth century onwards has (H)(a)emon. Heinsius printed it with the following comment: ‘Haemon pro Haemum tres scripti [for example Ld7 Bs7Lr26Nr33s]. sic & alibi non semel’.Footnote 4 Cf. Met. 2.219,Footnote 5 6.87,Footnote 6 Pont. 4.5.5.Footnote 7 However, HeinsiusFootnote 8 kept Haemum at Her. 2.113. Tarrant says nothing about this in his appendix.Footnote 9 Unfortunately, in all the examples cited the word appears at the end of the line, so metre does not reveal whether Ovid used one form or the other.Footnote 10 None the less, there is no denial that he was especially fond of Greek forms.Footnote 11
To give a couple of examples, Virgil undoubtedly wrote Cyprum, Polyphemum and Menelaum, as the subsequent elisions show:
But Ovid opted for the Greek endings without elision:Footnote 12
Let me list further examples of accusatives in -on guaranteed by the metre in the Metamorphoses: 2.83 Scorpion (cf. Fast. 5.541); 2.415 Maenalon (2.442); 3.539 Tyron; 3.636 Naxon; 4.283 Crocon (cf. Fast. 5.227); 4.466 Sisyphon (13.26); 4.786 Pegason; 7.365 Rhodon; 7.413 Cerberon; 7.466 Siphnon; 7.500 Clyton; 8.261 Daedalon; 8.270 Meleagron; 8.798 Caucason; 9.435 Aeacon (9.440, 13.27); 9.440 Rhadamanthon;Footnote 13 9.646 Cragon; 10.217 Hyacinthon; 10.530 Paphon (cf. Am. 2.17.4; cf. also Ars am. 2.588, Met. 10.297); 11.762 Aesacon (12.1); 12.215 Hymenaeon (cf. Pont. 1.2.131); 12.262 Orion; 12.352 Hippason; 12.378 Phlegraeon; 12.408 Cyllaron; 12.433 Tectaphon (dubium); 13.171 Telephon; 13.257 Coeranon; 13.260 Ennomon; 13.726 Lilybaeon; 14.223–4 Aeolon; 15.61 Samon.
It is noteworthy that none of these forms coexists with a metrically guaranteed form in -um in Ovid's corpus. It is also important to note that at 9.440 the metre demands Rhadamanthon, but that did not prevent the scribe of M from writing R(h)adamant(h)um. Likewise, at 13.27 and 13.772 some manuscripts, including some antiquiores, wrongly give the form in -um.Footnote 14 This strongly suggests that Greek endings are likely to be substituted for Latin ones, as Housman proved for accusatives in -an or -en.Footnote 15 From this, I infer that, whenever there is some evidence for the Greek form in Ovid, we should adopt it.
(2) 10.162
162 amyclide Ω, edd.: am(y/i)cliade Dr Es5, e coni. Unger: amycliades Gf6ac: amyclaide etiam Unger, Merkel, prob. Luck: amiclides Cs BoLr14: amiclede Aac: amiclate Lr7: amiclade So Bo2To2
The patronymic Amyclides is attested only here. It alludes to Hyacinthus, whom elsewhere Ovid calls Taenarides (Met. 10.183), Oebalides (10.196, Ib. 588) or Oebalius (Met. 13.396). According to Bömer,Footnote 16 Amyclides is probably the true patronymic for Hyacinthus.Footnote 17 At the very least, the use of Amyclides shows that Ovid was aware of a tradition in which Amyclas was Hyacinthus’ father (Simmias, fr. 8 Powell; Paus. 3.1.3; Apollod. Bibl. 3.116; Schol. Nic. Ther. 902–3a; Tzetz. Chil. 244–5).
The scansion Amyclīdes would be formed upon an irregular Ἀμυκλείδης created metri gratia, probably by analogy with patronymics in -ειδης of -ευ- stems (for example Met. 1.390 Promethides). This is not impossible: cf. Verg. Aen. 2.82 Belīdae (Cinna, fr. 1.3 Blänsdorf; Ov. Her. 14.73); 7.484 Tyrrhīdae;Footnote 18 Ov. Met. 12.433 Olenīden; 15.624 Coronīden; Prisc. Inst. 2.67.9–13 Hertz.Footnote 19 However, UngerFootnote 20 conjectured Amycliade and Amyclaide. The former would be derived from Ἀμύκλας; the latter from the homonymous city Ἀμύκλαι and would mean ‘Spartan’.Footnote 21 MerkelFootnote 22 printed Amyclaide in his second edition, but he was not followed by subsequent editors. Only LuckFootnote 23 favoured Amyclaide, pointing out the similarity with 4.1 Minyeias (‘M man. 2, Scaliger, Heinsius: minyas vel mineias vel similia Ω’);Footnote 24 14.87 Acheloiadumque (‘U ex corr. W, Constant. Fanensis et Naugerius ex coni.: Acheloidumque Ω’);Footnote 25 15.386 Cythereiadasque (‘b man. 2 k man. 2, Constant. Fanensis et Naugerius ex coni.’).Footnote 26 We could add, among other examples, Met. 6.414 Pelopeiadesque Mycenae; Fast. 2.43 Amphiareiades;Footnote 27 Ib. 295 Amyntiaden; Ib. 345 Dryantiadae;Footnote 28 Ib. 503 Lycurgiaden;Footnote 29 Ib. 631 Cliniadaeue.Footnote 30 The examples just quoted rather stand for Amycliade, which is actually the reading of Dr Es5, while Gf6ac reads Amycliades.Footnote 31 Furthermore, these examples show that omission of letters, often that of -a-, is usual in this kind of words.Footnote 32 Amycliades is indeed a possible patronymic of Ἀμύκλας,Footnote 33 just like we have Amyntiades from Amyntas (Ib. 295), Anchisiades from Anchises (Verg. Aen. 5.407 and elsewhere) or Thyestiades from Thyestes (Ov. Ars am. 2.407). Furthermore, Ovid's liking for alluding to mythographical variants is well known. To my mind, an allusion to a tradition that made Hyacinthus the son of Amyclas makes much more sense than a dull reference to a city or a geographic area.Footnote 34 I think we ought to read Amycliade.
(3) 10.309
309 panc(h)aia GfHLr2M, recc., edd. plerique: panc(h)aica AL3LrLuNTV3, recc., edd. ueteres aliquot, prob. Luck: alii aliter
The manuscripts are split, offering Panchaia and Panchaica.Footnote 35 Heinsius commented: ‘Panchaïa meliores [for example HM EHdLr3P2 P24 Bo3ac]. bene’.Footnote 36 He compared it to Verg. G. 2.139 (totaque turiferis Panchaïa pinguis harenis), but there Panchaia is a noun (as in Lygd. 2.23; Val. Fl. 6.119; Claud. III Cons. Hon. 211, Raptu Pros. 2.81; Plin. HN 7.197, 10.4). He also mentioned App. Verg. Culex 87 Panchaia tura, where Panchaia is an adjective, but the reading is not quite certain, because the best manuscripts offer Pancheia (Γ) or Panchasia (V), while Panchaia is attested in CL,Footnote 37 ‘the least estimable MSS of the poem’.Footnote 38 The only well-attested example of Panchaius seems to be Mela 3.81, who calls the inhabitants of the island Panchai. On the other hand, I have found only one example of Panchaicus: Arn. Adu. nat. 7.27.
Since the evidence for Panchaius/Panchaicus is clearly poor, I suggest that we examine a parallel and well-documented doublet: Achaius/Achaicus. The alternation occurs at Verg. Aen. 2.462, where MPωγ read Achaica castra and Fcy read Achaia castra.Footnote 39 The variation could be explained by the scribal distaste for the cacophonic syllabic combination -ca ca-.Footnote 40 None the less, the same variation also occurs in other contexts where there is no cacemphaton.Footnote 41 In general, adjectives in -icus often loose the letter -c- in manuscripts. For instance, in Ov. Met. 2.727 some manuscripts read Balearia instead of Balearica.Footnote 42 In Ov. Met. 3.518 the adjective Bacchica (Βακχικός, which coexists with Baccheus, Βακχεῖος, and Baccheius, Βακχήϊος) is altered to Bacchia in some manuscripts.Footnote 43
On the other hand, the adjective Achaeus, which is the right transcription of Ἀχαιός, is extensively documented in Latin (cf. TLL 1.384.67–385.60). I doubt that Ἀχαιός could be transcribed both as Achaeus and Achaïus, with a different scansion. Furthermore, apart from Achaïus (and Panchaïus), I can find no other example of any Greek adjective in -αιος (or in -αιϜος)Footnote 44 being transcribed into Latin as -aïus. Therefore, I believe that Diehl (TLL 1.284.8–11) is right when he points out that Achaius is only a scribal misspelling of Achaicus.Footnote 45 The analogy with adjectives in -ius and the noun Achaia, in combination with the scholastic doctrine against cacemphaton, probably smoothed the way for Achaius to substitute Achaicus.
Let me address one final question about the adjectives from Achaia. As often, Ovid also used derivatives in -is (Met. 3.511, 5.306, 15.293; 5.577, 7.504 [nouns]) and in -ias (Her. 3.71). In the Metamorphoses, TarrantFootnote 46 accepted the transmitted form Achais, even if Ovid clearly kept the Homeric form Achaeias at Her. 3.71 (Achaeiadas … matres).Footnote 47 KenneyFootnote 48 suggested that it was very likely that Ovid had also retained the transcription Achaeis in the Metamorphoses (it appears in some old editions). Indeed, Achaeis and Achaeias are the expected transcriptions of Ἀχαιΐς (for example Hom. Il. 1.254) and Ἀχαιΐας (for example Hom. Il. 5.422).Footnote 49
As for Achaicus, Kenney recalled that since Homer the regular Greek form was Ἀχαιϊκός (for example Il. 9.141), and we could expect that Latin poets transcribed Achaeicus rather than Achaicus. But Keeney also knew that this was a complicated issue, since Achaicus relies on the authority of the ancient Virgilian manuscripts (see above; this form is also attested in Hor. Carm. 1.15.35; cf. TLL 1.384.14–42). Perhaps we should consider the possibility that the Latin adaptation had been filtered through the prosaic alternative Ἀχᾱϊκός (and palaeographic rival of Ἀχαιϊκός; cf., for example, Eur. Hec. 521).
All these observations apply, I think, to Panchaius and Panchaicus. If we accept the illegitimacy of Achaius, we must conclude that Panchaius is simply a form attested in manuscripts instead of the true Panchaicus, the transcription of a postulated Παγχᾱϊκός. Thus Panchaeus and Panchaicus would stem from Panchaia, just like Achaeus and Achaicus stem from Achaia.Footnote 50 Other parallels are available: Phocaea produces Phocaeus (*Φωκαῖος; Hor. Epod. 16.17) and Phocӑĭcus (Φωκαϊκός; Ov. Met. 6.9; Luc. 3.172) but not Phocaius; Thebae produces Thebanus, Thebaeus (Θηβαῖος; Claud. Carm. min. 27.91) and Thebӑĭcus (Θηβαϊκός; Stat. Silu. 4.9.26) but not Thebaius.Footnote 51 For all these reasons, I think we should restore Panchaica.Footnote 52 This had been printed by some editors up to Heinsius and was recently favoured by Luck (‘read probably’).Footnote 53
Open access
