Hostname: page-component-74656f9884-npj9w Total loading time: 0 Render date: 2026-11-06T23:45:16.398Z Has data issue: false hasContentIssue false
  • English
  • Français

AN EMENDATION IN APOLLONIUS SOPHISTA'S LEXICON HOMERICUM *

Published online by Cambridge University Press:  03 April 2017

Chiara Meccariello
Chiara Meccariello*
Affiliation:
University of Oxford
*
chiara.meccariello@classics.ox.ac.uk
Rights & Permissions [Opens in a new window]

Extract

The beginning of the second entry of Apollonius Sophista's Lexicon Homericum reads as follows in the codex Coislinianus gr. 345, the only direct witness to this section (1, 14–15 Bekker):

ἀάατος· ὁ ἀβλαβὴς καὶ εὐχερὴς καὶ δι’ οὗ ἄνευ ἄτης. ἐνίοτε δὲ τὸν ἐπιβλαβῆ καὶ δυσχερῆ …

ἀάατος: ‘harmless’ and ‘easy’, and δι’ οὗ ‘without ἄτη’. But sometimes ‘hurtful’ and ‘difficult’ …

Information

Type
Shorter Notes
Information
The Classical Quarterly , Volume 67 , Issue 1 , May 2017 , pp. 332 - 335
Check for updates
Copyright
Copyright © The Classical Association 2017 

The beginning of the second entry of Apollonius Sophista's Lexicon Homericum reads as follows in the codex Coislinianus gr. 345, the only direct witness to this section (1, 14–15 Bekker): Footnote 1

ἀάατος· ὁ ἀβλαβὴς καὶ εὐχερὴς καὶ δι’ οὗ ἄνευ ἄτης. ἐνίοτε δὲ τὸν ἐπιβλαβῆ καὶ δυσχερῆ …Footnote 2

ἀάατος: ‘harmless’ and ‘easy’, and δι’ οὗ ‘without ἄτη’. But sometimes ‘hurtful’ and ‘difficult’ …

The sequence δι’ οὗ in the first clause, which I have intentionally left untranslated, creates difficulties. Villoison, in his editio princeps of the Lexicon, translates the transmitted sequence καὶ δι’ οὗ ἄνευ ἄτης with ‘et per quem sumus ἄνευ ἄτης, id est, sine noxa’, being forced to add a ‘sumus’ that is neither in the Greek text nor deducible from the context.Footnote 3 In its other occurrences in Apollonius δι’ οὗ unproblematically introduces a relative clause with σημαίνει or δηλοῦται as the main verb. In these clauses, which usually convey straightforward semantic explanations, δι’ οὗ simply marks the transition to a more common or more transparent synonym, and can therefore be translated as ‘by which’: see in particular 77, 17–18 ἐρύουσι· ἀντὶ τοῦ ἐρύσωσιν· σημαίνει δὲ τὸ ἕλκουσι, δι’ οὗ δηλοῦται τὸ κατέδονται; 80, 33–4 ζαφελῶς· ὁ μὲν Ἀπίων μεγαλοκότως· ἐπιφέρει δὲ τὸ μενεαίνειν, δι’ οὗ σημαίνει τὸ ἐγκοτεῖν;Footnote 4 122, 34–123, 1 ὀρθοκραιράων· […] ἔνιοι δὲ ὀρθοκεφάλων, δι’ οὗ τὸ ὀρθόπρυμνον σημαίνει.Footnote 5

The meaning ‘by which’ cannot be accommodated in the entry ἀάατος. Not only is a form like σημαίνει or δηλοῦται missing, but ἄνευ ἄτης (whether its sense is ‘without sin’ or ‘without harm’) can hardly be considered a simplifying explanation of what precedes (εὐχερής). More importantly, the connective καί before δι’ οὗ is incompatible with a straightforward relative clause, and rather suggests that an additional meaning is being adduced. Accordingly, Bekker simply excides δι’ οὗ, and his solution has long remained unchallenged.Footnote 6

The excision fixes the text and is supported by a locus parallelus in Hesychius, whose entry ἀάατον (α5) reads τὸ ἀβλαβὲς καὶ εὐχερές, ἢ ἄνευ ἄτης.Footnote 7 However, it undoubtedly entails a trivialization and leaves us with a crucial question: how to account for the presence of δι’ οὗ?

A better solution than Bekker's excision is in my opinion the emendation of δι’ οὗ to ἰδίως, to be understood in the meaning of ‘properly’, ‘literally’. Among the several possible nuances of the adverb, this acceptation is solidly attested in lexicography in the context of literal and etymological explanations of single words.Footnote 8 Apollonius’ Lexicon shows this usage in the entry ὀλοοίτροχος (120, 11), which is explained as follows: ὁ Κομανὸς ἰδίως ἀποδίδωσιν ὅλος τρέχων, ‘Comanus renders literally as ὅλος τρέχων’.Footnote 9

Several other parallels can be found in glossographical works. Examples include Hesychius s.v. τ1404 τρίοδος· καὶ τριόδους πληθυντικῶς· ἰδίως τὰς τρεῖς ὁδούς; Synagoge lex. chres. cod. B s.v. α2618 (= Phot. Lex. s.v. α3456): ἄχρηστος· ὁ ἀνωφελής. καὶ ἰδίως ᾧ οὐδεὶς χρῆται; Phot. Lex. s.v. ε1210: ἐξηγητής· ἰδίως ὁ ἐξηγούμενος τὰ ἱερά; Phot. Lex. s.v. θ157: θηλάστριαν· ἣν θηλάσεταί τις ἰδίως; Phot. Lex. s.v. ε1188 (= EM 386, 13 Kallierges) ἐξεταστής· σημαίνει μὲν καὶ ἄλλα, ἰδίως δὲ τοὺς πεμπομένους ἄρχοντας τοῦ ἐξετάσαι πόσοι εἰσὶν … .

As in these passages, the definition which ἰδίως would accompany in Apollonius’ entry ἀάατος is an etymological one, as the meaning ἄνευ ἄτης is extracted from the structure of the lemma through the typical identification of privative alpha.Footnote 10 This is one of the possible means of building words considered in the Lexicon: see in particular the entry α, 1, 3–4, a general discussion of privative alpha, as well as 36, 12 ἀνωϊστί· […] ἄνευ τοῦ οἴεσθαι; 39, 25 ἀπριάτην· ἄπρατον, ἄνευ τοῦ πρίασθαι.Footnote 11

As to the genesis of the mistake, a misreading can easily have been caused by the palaeographical closeness of καιιδιωσ and καιδιου, particularly if the ending ως was abbreviated in the source manuscript. A confusion between the compendium for ως and that for οῦ is not unlikely, considering the similarity between their respective shapes: a horizontal symbol describable as a reversed tilde can be used for ως (∽: the adverb ἰδίως is so abbreviated, for example, in fol. 29r of the Coislinianus, in the entry λωτόν), while the abbreviation for οῦ resembles a reversed tilde, but with the two curves vertically detached and slightly slid towards one other. One of the possible compendia for ου, a single curve identical to the left-hand half of ∽, is also not dissimilar.Footnote 12 The wrong expansion of the abbreviation for ως would have gone hand in hand with the omission of iota after καί, in either the same copying event or, more likely, subsequently by haplography.Footnote 13

Footnotes

*

This paper stems from ongoing work on a new edition of Apollonius Sophista's Lexicon Homericum, which I am preparing with James Brusuelas and Dirk Obbink on the basis of preliminary work conducted by Michael Haslam. I would like to thank James Brusuelas, Daniela Colomo and Fausto Montana for reading and commenting upon an earlier draft of this article, and the anonymous referee for helpful suggestions.

References

1 The text, here and elsewhere, is quoted according to the latest complete edition available, that of Bekker, I., Apollonii Sophistae Lexicon Homericum (Berlin, 1833)Google Scholar. On the magmatic textual transmission of the Lexicon, the state of the text in the Coislinianus, and the papyrological evidence, see Haslam, M.W., ‘The Homer Lexicon of Apollonius Sophista: I. composition and constituents’, CPh 89 (1994), 1–45 Google Scholar and Id., ‘The Homer Lexicon of Apollonius Sophista: II. identity and transmission’, CPh 89 (1994), 107–19Google Scholar.

2 The remaining part of the entry quotes Od. 21.91 to exemplify the second pair of interpretamenta (τὸν ἐπιβλαβῆ καὶ δυσχερῆ); it then mentions again the meaning ἀβλαβῆ, quoting Od. 22.5. The repetition may seem unwanted, but cf. the entry ἀθέσφατον (13, 5), where πολύν is the first interpretamentum (exemplified by Il. 3.4), and in the second part of the entry ἀντὶ τοῦ πολλαί occurs again to explain ἀθέσφατοι in the further example quoted (Od. 20.211). Moreover, it is probably no coincidence that the entry ἀάατος consists of three groups of definitions (ὁ ἀβλαβὴς καὶ εὐχερής, τὸν ἐπιβλαβῆ καὶ δυσχερῆ, and ἀβλαβῆ), as the adjective occurs precisely three times in the Homeric poems, namely Od. 21.91 and 22.5 (the two passages quoted in our entry as preserved in the Coislinianus) and Il. 14.271. The latter line might well have been quoted, originally, to illustrate the first definition: for the loss of exemplificative quotations in the Coislinianus, see Haslam (n. 1 [1994] [I]), 2 and 29 n. 86, and Id. (n. 1 [1994] [II]), 111 and 117. Note that the adjective is also attested once in Apollonius Rhodius (2.77), and one additional occurrence in the Argonautica is documented by Schol. L Ap. Rhod. 1.801–3a, in a reportedly alternative version of 1.803.

3 Gaspard, J.B. d'Ansse de Villoison, Lexicon Graecum Iliadis et Odysseae (Paris, 1773), 5Google Scholar. Tolle, H., Apollonii Sophistae Lexicon Graecum Iliadis et Odysseae (Leiden, 1788)Google Scholar also prints the paradosis. For other minor textual problems in the last part of the entry, see Tolle (this note), 3 n. 2.

4 As it is clear from this example, when δι’ οὗ σημαίνει is used the implied subject is Homer: see also 25, 6–8 ὅταν δὲ λέγῃ ‘ἄμπυκα κεκρύφαλόν τε’ κόσμον τινὰ ἔοικε περὶ τὴν κεφαλὴν σημαίνειν.

5 Cf. also Hesychius s.v. ε4828 and π2580.

6 It is accepted by Steinike in his re-edition of the entries from alpha to delta: K. Steinicke, Apollonii Sophistae Lexicon Homericum (diss., Gottingen, 1957).

7 Cf. also Synagoge lex. chres. cod. B s.v. α1: ἀάατον· ἀναμάρτητον, ἀβλαβές, ἁγνόν, ἄπονον, ἀπαθές, ἄνευ ἄτης.

8 On ἰδίως in scholia, see for example Meijering, R., Literary and Rhetorical Theories in Greek Scholia (Groningen, 1987), 226–30Google Scholar and Nünlist, R., The Ancient Critic at Work (Cambridge, 2009), 260CrossRefGoogle Scholar, with further bibliography. For the sense required in Apollonius, see in particular Rutherford, W.G., A Chapter in the History of Annotation, being Scholia Aristophanica, vol. 3 (London, 1905), 189–90Google Scholar, who shows that this meaning is attested in the scholia on Aristophanes’ plays and shared with κυρίως; cf. Meijering (this note), 229.

9 Comanus fr. 7a Dyck. Apollonius Sophista quotes the grammarian several times; in one of the verbatim quotations, Comanus refers to the proper usage of a word with the adverb κυρίως (Ap. Soph. 47, 2 = Comanus fr. 6 Dyck).

10 On the modern debate regarding the etymology of ἀάατος, see in particular the entry in the Lexikon des frühgriechischen Epos, Band 1 A (Göttingen, 1979), 2–3, as well as Moorhouse, A.C., ‘ἈΑΑΤΟΣ and some other negative compounds’, CQ 11 (1961), 10–17 CrossRefGoogle Scholar; Olcott, M., ‘The Getty Homer fragment’, in Studia Varia from the J. Paul Getty Museum (Malibu, 1993), 1.93–8Google Scholar, at 94 with nn. 15–17; Id., ‘ΑAAΤΟΣ at Odyssey 22.5: Greek and Indo-European oaths’, Word 44 (1993), 77–90 CrossRefGoogle Scholar, and most recently Krieter-Spiro, M., Homers Ilias. Band X, Vierzehnter Gesang, Faszikel 2: Kommentar (Berlin and Boston, 2015), 130Google Scholar. Traces of the ancient debate can be read, among others, in Eustathius, who suggests double privative α (ἀ + ἀ + ἄτη) and thus the sense βλαβερὸς ἤτοι ἀτηρός: see ad Od. 21.91 (vol. 2.251) and ad Il. 14.271 (vol. 3.639). In both passages, however, he also refers to an alternative explanation, in which only one prothetic alpha is considered, and both the second and the third alpha are traced to the root ἄτη on the grounds of the spelling ἀάτη/ἀάσαι. Although in this case, according to Eustathius, the value of the alpha is intensive, not privative, the interpretation ἀ + ἀάτη is itself sufficient to obtain the sense ἄνευ ἄτης. Eustathius does indeed consider the latter meaning in his commentary ad Od. 22.5 (vol. 2.269), where ἀάατος is explained as δίχα βλάβης and Odysseus’ words οὗτος μὲν δὴ ἄεθλος ἀάατος ἐκτετέλεσται are paraphrased as οὐ πολυβλαβής, ὦ μνηστῆρες, ἐμοὶ ὁ παρὼν ἄεθλος, ὡς ὑμεῖς ἐλέγετε, ἀλλὰ ἐστερημένος ἄτης καί, ὡς εἰπεῖν, ἀβλαβής. However, it is unclear whether Eustathius here presupposes the etymology from ἀ + ἀάτη, as implied in the expressions δίχα βλάβης and ἐστερημένος ἄτης and in the gloss ἀβλαβής, or from ἀ privative + ἀ intensive + ἄτη, as the litotes οὐ πολυβλαβής could perhaps suggest.

11 One may also compare the explanations introduced by the phrase κατὰ στέρησιν found, for example, s.v. ἀβάκησαν (2, 16: the etymology is explicitly traced to Apion), ἀτρεκέως (46, 28), ἀταλόν (46, 2), ἄμνιον (25, 26), ἀμενήνωσεν (27, 3), ἀνήκεστον (35, 9), ἀπάλαμνος (37, 22).

12 See Villoison (n. 3), Tab. V, with reference to actual abbreviations used in the Coislinianus. Cf. also Metzger, B.M., Manuscripts of the Greek Bible: An Introduction to Greek Palaeography (New York, 1981), 30Google Scholar. Note that in the Coislinianus the transmitted δι’ οὗ is written in full (and with diacritics), save for the very common Byzantine ligature ȣ for ου.

13 It must be noted, at any rate, that the manuscript has the abbreviated form of καί at this point.