ἀκολουθοῦντε
(Σ^b α 746, Phot. α 788, Phryn. PS fr. *114)

D. General commentary

The gloss of Σ^b α 746 (= Phot. α 788, A.1), found in the third expansion of the Synagoge and edited by de Borries as Phryn. PS fr. *114, credits Hermippus (C.1, on which see Comentale 2017, 325) with the use of the masculine dual participle ἀκολουθοῦντε instead of the feminine ἀκολουθοῦσαι and states that this preference for the masculine form is customary with dual participles. Cod. Par. Coisl. 345Par. Coisl. 345 actually transmits the feminine dual participial form ἀκολουθούσα, but Cunningham (2003, 579), following Valckenaer, corrects this to the feminine plural participle ἀκολουθοῦσαι found in Photius. As the dual ἀκολουθούσα corresponds exactly to the masculine ἀκολουθοῦντε, the transmitted reading may be correct, but the common contrast made between masculine duals and feminine plural forms in other sources (see B.1, B.2, B.3) argues in favour of ἀκολουθοῦσαι.

Here we focus on the Synagoge gloss and its likely background, refraining from the larger question of dual usage in Homer. The gloss itself does not derive from Homeric exegesis; it never quotes and focuses on a form of the non-Homeric verb ἀκολουθέω. Nevertheless, it demonstrates a knowledge of a grammatical debate typically discussed by ancient grammarians within the scope of Homeric exegesis — that is on the use of the masculine dual participle for feminine referents.

A vestigial feature of the language’s Indo-European roots, the Greek dual was already used only sporadically in Homeric Greek (see Chantraine 1958–1963 vol. 2, 22–9); its preservation in the Attic dialect is both an archaism and a well-known peculiarity (Cassio 2016, 78). Study of the Mycenaean language has shown that feminine words in -ā originally borrowed the endings of the thematic declension -ō since they lacked specific terminations (see Lejeune 1958, 206–7; Bartoněk 2003, 165–6; Cassio 2002, 71–2). This situation recurs in the Homeric language. Noting the absence of feminine duals ending in -ᾱ in Homer, ancient scholars debated the acceptability of dual forms in Homeric passages, ultimately explaining the phenomenon as a σχῆμα — a syntactic anomaly characteristic of Homeric style in which masculine epithets and participles were linked to feminine referents.

Aristarchus’ discussion of dual usage in Homer is particularly noteworthy. Aristarchus considered the presence of the dual in Homer’s idiolect as proof of his Attic origin, arguing that ‘Homer spoke an archaic form of Attic’ (Schironi 2018, 607). This doctrine is expressed in schol. (Ariston.) Hom. Il. 13.197Schol. (Ariston.) Hom. Il. 13.197: {ἴμβριον αὖτ’} Αἴαντε <μεμαότε>· ὅτι συνεχῶς κέχρηται τοῖς δυϊκοῖς. ἡ δὲ ἀναφορὰ πρὸς τὰ περὶ τῆς πατρίδος· Ἀθηναίων γὰρ ἴδιον (‘The two Ajaxes eager [of furious strength]: [The diple is] because he has used the dual continuously. The reference is to the question of [Homer’s] homeland, for [the dual is] typical of the Athenians’: text and translation follow Schironi 2018, 607 with slight adaptations). Aristarchus (rightly) did not accept feminine duals of the first declension. In Il. 8.378 (C.2), he rejected the feminine προφανείσα (referring to Hera and Athena) defended by Ptolemy of Ascalon in favour of the masculine form προφανέντε (B.2), invoking the analogous case of πληγέντε (Il. 8.455, C.3) to support his choice (B.1). On Aristarchus’ position in this debate, see Schironi (2004, 164) and Ascheri (2004, 343–4); Aristarchus’ acceptance of the feminine dual προφανείσα, attested in schol. Hom.Il. 8.378.b2Schol. Hom. Il. 8.378.b2, remains hypothetical (see Ascheri 2004, 343–4 n. 32). According to some sources (e.g. schol. Hom. Il. 3.35 [bT]), Aristarchus’ rejection of such forms led him to suppose a metaplasmMetaplasm at Il. 3.35; refusing to accept παρειά (‘cheeks’) as a feminine dual form (as Demetrius Ixion did), the grammarian instead incorrectly suggested a neuter plural from the unattested *παρειόν. Schironi (2004, 163–5) accepts these claims, arguing that Aristarchus retained the neuter form παρειά to avoid changing the textual tradition, while Ascheri (2004, 343–7) asserts that he preferred the plural παρειάς. Nevertheless, both scholars agree that Aristarchus’ aim was to reject παρειά as a dual — a form that likely came from an Attic interpolation (see Wackernagel 1916, 60). On Aristarchus and the Homeric dual see also Matthaios (1999, 378–82).

The use of masculine forms instead of feminine forms is not restricted to duals, and is described by ancient scholarship as a feature peculiar to Attic and Homeric language. Aristarchus, for instance, preferred the masculine participle ἀκέων (‘silent’) in place of the feminine ἀκέουσα at Il. 4.22 (schol. [Ariston./Apoll.Dysc.] Hom. Il. 4.22, on which see Razzetti 2010, 44–8), and is also quoted by Eustathius, in Il. 1.695.11–2: [ἀκέων] ἀντὶ τοῦ ἀκέουσα εἴρηται Ἀττικῶς (‘[ἀκέων] is said in place of ἀκέουσα in the Attic dialect’). On the masculine in place of the feminine see further [Plu.] Vit.Hom. 2.153–62 (B.5) and Eust. in Il. 2.616.10 (B.6), where parallels are drawn between these usages in Homer, Hesiod, and the tragedians (Eur. Hipp. 387; Soph. El. 977, 980–1). The first occurrence of a feminine dual participle in -σᾱ can be found in Soph. OC 1676Soph. OC 1676, where the feminine παθούσα (‘having suffered’) is coordinated with the masculine ἰδόντε (‘having seen’). Schol. Soph. OC 1676 (B.4), commenting on this passage, underlines the use of the masculine ἰδόντε in place of the feminine as a common σχῆμα, though some scholars suggest emending παθούσα to παθόντε (see Wackernagel 1916, 59; Schironi 2004, 164 n. 29).

Since the replacement of the feminine form with the masculine is explained as a syntactic anomaly, treatises concerning barbarism and solecism (B.3) classify it as a ‘σχῆμα in γένος’ exemplified by the usage of the dual masculine participle πληγέντε at Il. 8.455 (C.3). These treatises proscribe the usage of the σχῆμα to avoid eccentricities in language, since such usage, though justified as a peculiar trait of Homer’s style, would be ridiculous elsewhere. Sextus Empiricus, for example, in his debate on anomaly and analogy sets limits to the imitation of Homer’s language for the sake of appropriateness and coherence, asserting that an exaggerated imitation of Homer’s language by a contemporary would sound out of place and laughable (M. 1.206)S.E. M. 1.206: εἶτα κἀκείνην μάλιστα δεῖ τὴν συνήθειαν μεταδιώκειν ᾗ προσχρώμενοι οὐ γελασθησόμεθα· τῇ δὲ Ὁμηρικῇ κατακολουθοῦντες οὐ χωρὶς γέλωτος ἑλληνιοῦμεν, ‘μάρτυροι’ λέγοντες καὶ ‘σπάρτα λέλυνται’ καὶ ἄλλα τούτων ἀτοπώτερα (‘And we should especially follow that customary usage for which we would not be laughed at when using it: but if we follow the Homeric [usage], we use a Greek that sounds ridiculous, saying μάρτυροι (‘witnesses’) and σπάρτα λέλυνται (‘the ropes are loose’) and other things even more unusual than these’).

Coming back to Σ^b α 746 (A.1), it is one of the four entries in the Synagoge openly address dual forms, together with Σ δ 152, η 127, and τ 111. This last, which concerns the use of the dual τέρπεσθον in place of the plural τέρπονται (‘they enjoy’) in reference to Hera and Athena, derives from the tradition of the D-scholia to the Iliad (see schol. [D] Hom. Il. 4.10). Conversely, as we have seen, Σ^b α 746 (A.1) does not derive directly from Homeric exegesis, although it demonstrates a knowledge of this grammatical debate: it is actually more likely to originate from an Atticist source. As regards Hermippus’ quotation, five comic fragments are preserved under his name in Σ^b (Σ^b α 632 = Hermipp. fr. 74; Σ^b α 746 = fr. 81, C.1; Σ^b α 1354 = fr. 82; Σ^b α 1659 = fr. 66; Σ^b α 2521 = fr. 84). All these glosses come from the expansion Σʹʹʹ, except for Σ^b α 1659, which belongs to Σʹ; among them, Cunningham acknowledges only Σ^b α 632 (which openly quotes Phrynichus) and Σ^b α 746 (= Phot. α 788, A.1) as deriving from Atticist sources. The latter (A.1) has been included among the fragments of Phrynichus’ PS because its doctrine has been supposed to derive from an Atticist source — an attribution supported by both the gloss's concern with the usage of dual forms (a notorious Attic trait) and the usage of a passage by the Athenian comic poet Hermippus, a likely source for the Atticists, to exemplify the phenomenon. The unexpressed subject of the verb κέχρηνται is very likely to be οἱ Ἀττικοί (‘Attic speakers’) since ancient scholars usually describe this σχῆμα as Attic. Moreover, this item is inserted in Σ^b in a sequence of six glosses apparently intended to reproduce a cluster from Σʹʹʹ that preserved materials originating from Atticist discussions (Σ^b 746–51). Attribution of the fragment to Phrynichus’ PS is nevertheless doubtful. Though de Borries recognises Σ^b 746–50 as fragments of the PS due to their connections to the Atticist debate, only Σ^b 751 mentions Phrynichus and has a secure parallel in Phryn. PS 7.13–4Phryn. PS 7.13–4. In at least one case, the identification of Σ^b 747 with Phryn. PS fr. *115Phryn. PS fr. *115, de Borries can be proven wrong: while the gloss describes as Attic the construction of verba sequendi in which a prepositional phrase is substituted for the plain dative, Phrynichus proscribes this very syntax in Ecl. 330Phryn. Ecl. 330 (see entry ἀκολουθεῖν μετ’αὐτοῦ).

Although its provenance from the PS remains doubtful, this gloss (A.1) represents an interesting piece of scholarship on the debate over the masculine pro feminine σχῆμα and shows how Atticism could extend its range of interest by incorporating and reprocessing materials from other fields of ancient scholarship such as those topics traditionally belonging to Homeric exegesis.

E. Byzantine and Modern Greek commentary

N/A

F. Commentary on individual texts and occurrences

N/A