εἰ with neuter adjective and future
(Phryn. PS 3.8–10, Phryn. PS fr. 199)

D. General commentary

The glosses in the epitome of the PS (A.1) and in Photius (A.2) evidently derive from the same materials in the plenior text of the PS. However, the information they provide is quite dissimilar, most likely because they have manipulated the content of their common source differently. On the one hand, in the epitome of the PS (A.1), Eupolis’ line (ἀνόητά γ’ εἰ τοῦτ’ ἦλθες ἐπιτάξων ἐμοί) stands as the lemma. This is first paraphrased (ἀντὶ τοῦ) with the semantically equivalent personal formulation (ἀνόητος εἶ ἐπιτάττων τοῦτο), then an impersonal construction similar to the one used by Eupolis (ἀνόητα εἰ τοῦτο ἐπιτάξεις) is indicated as Attic. A reasonable inference is that it is treated on an equal footing with Eupolis’ fragment. Photius (A.2), on the other hand, posits a sharp opposition between a construction like ἀνόητα εἰ τοῦτο ἐπιτάξεις, which he says was commonly used by people who spoke unsophisticated Greek (οἱ μὲν ἀγοραῖοι καὶ πολλοὶοἱ πολλοί οὕτως), and Eupolis’ ἀνόητά γ’ εἰ τοῦτ’ ἦλθες ἐπιτάξων ἐμοί, which he praises as Attic and artful (Ἀττικῶς δὲ καὶ ἐσχηματισμένως). Photius makes no mention of the personal construction ἀνόητος εἶ ἐπιτάττων τοῦτο.

It is thus hard to reconcile the information provided by the two glosses and reconstruct the content of Phrynichus’ original version. Olson (2014, 98) claims that it is difficult to recognise what Phrynichus actually identified as an Atticism. In his view, the epitome of the PS (A.1) and Photius (A.2) are pursuing mutually irreconcilable aims. He suggests that while the epitome of the PS (A.1) indicates that the impersonal construction of a neuter adjective as an apposition to a hypothetical εἰ-clause (equally ἀνόητα + εἰ + future indicative and Eupolis’ ἀνόητα + εἰ + ἔρχομαιἔρχομαι + future participle) is the proper Attic equivalent of the corresponding personal construction (ἀνόητος εἶ + present participle), Photius (A.2) opposes Eupolis’ ἔρχομαιἔρχομαι + future participle, which is Attic and artful, with the construction that uses only the future indicative.

The construction neuter adjective + εἰ + future indicative, as in ἀνόητα εἰ τοῦτο ἐπιτάξεις, is abundantly documented in Attic prose and poetry (see C.2, C.3, C.4, C.5, C.6). Therefore, while the epitome of the PS appears reliable in its presentation of ἀνόητα εἰ τοῦτο ἐπιτάξεις as correct Attic Greek, Photius’ comment nonetheless requires attention. A further complication, lamented by Olson (2014, 98), could be that Eupolis’ construction ἔρχομαιἔρχομαι + future participle is just standard Greek rather than an Atticism (see C.7, C.8, C.9, but the list of occurrences is potentially infinite), and it is therefore unclear why Photius would present it in opposition to the simple future indicative.

There might be a way out of this complication. For a start, the fact that Photius describes Eupolis’ construction ἔρχομαιἔρχομαι + future participle as an Atticism does not imply that it is exclusively Attic, as Olson seems to believe. Instead, Photius may simply be praising ἔρχομαιἔρχομαι + future participle as a good Attic construction, qua used by Eupolis, and therefore worthy of imitation. Secondly, the fact that Photius indicates ἀνόητα εἰ τοῦτο ἐπιτάξεις as careless and common Greek (οἱ μὲν ἀγοραῖοι καὶ πολλοὶ οὕτως) in opposition to Eupolis’ more artful formulation does not automatically entail that the former is also incorrect. The less elaborate construction ἀνόητα εἰ τοῦτο ἐπιτάξεις may simply have been presented as unsophisticated and banal, and therefore unworthy of special consideration by the aspiring sophist (see F.1). This particular interest in the construction ἔρχομαι + future participle may have a straightforward explanation in light of the gradual disappearance of the future participle in Post-classical Greek and the increased use of the infinitive to indicate the purpose of an action (see Mayser, Gramm. vol. 2,1, 220–3; Blass, Debrunner 1976, §§ 351, 418.4, and 425.4).

Taking these points into consideration, a hypothetical reconstruction of the content of Phrynichus’ gloss might then be along these lines (for the sake of convenience, I follow the structure of the gloss in the epitome of the PS). Eupolis’ impersonal construction ἀνόητά γ’ εἰ τοῦτ’ ἦλθες ἐπιτάξων ἐμοί is first compared to the corresponding personal construction (ἀνόητος εἶ ἐπιτάττων τοῦτο). This use of ἀντὶ τοῦ (‘meaning’, ‘with the sense of’) to introduce the latter construction implies that it is a paraphrase of the former, rather than marking an opposition between the impersonal and the personal constructions as though they are Attic and non-Attic, respectively. Secondly, Eupolis’ line is compared to a similar formulation, namely ἀνόητα εἰ τοῦτο ἐπιτάξεις. Although this too is correct Greek, Phrynichus probably regarded it as unsophisticated and stylistically dull. Consequently, he recommends Eupolis’ formulation as a more masterful choice for the would-be sophist. In condensing this array of information, the epitome of the PS and Photius follow quite different paths. The epitome of the PS (A.1) prefers to omit the section that poses a stylistic opposition between ἀνόητά γ’ εἰ τοῦτ’ ἦλθες ἐπιτάξων ἐμοί and ἀνόητα εἰ τοῦτο ἐπιτάξεις, thus effectively presenting both constructions as though they are of equal merit. Photius (A.2) only devotes his attention to the stylistic opposition between Eupolis’ construction ἀνόητά γ’ εἰ τοῦτ’ ἦλθες ἐπιτάξων ἐμοί, which he recommends as truly Attic and worthy of imitation, and its less sophisticated equivalent ἀνόητα εἰ τοῦτο ἐπιτάξεις, which (though Attic per se) looks like a trivial, common, and unremarkable formulation.

E. Byzantine and Modern Greek commentary

Examples of ἔρχομαιἔρχομαι + future participle in the work of Byzantine writers are not especially rare (C.11, which depends on Menander Protector’s historical writing, C.12). Neuter adjective + εἰ + future indicative is, however, less common (C.10). While these constructions are a sign of a more elaborate style, they do not appear to be of a particularly high register, nor are they limited to the more Atticising writers. There is no exact Byzantine parallel for Eupolis’ construction ἀνόητα εἰ + ἔρχομαι + future participle, nor are there examples of the construction neuter adjective + εἰ + ἔρχομαι + future participle.

F. Commentary on individual texts and occurrences

(1)    Phryn. PS fr. 199 (= Phot. α 2019) (A.2)

While ἀγοραῖος is often used to indicate a mindset or a behaviour (see Phryn. PS fr. *72 [= Σ^b α 296, Phot. α 233, ex Σʹʹʹ; cf. Su. α 308]Phryn. PS fr. *72 (= Σ^b α 296, Phot. α 233, ex Σʹʹʹ),    and Suet. Περὶ βλασφ. 6), less frequently it may also indicate a category of speakers. The examination of some key occurrences of ἀγοραῖος and οἱ ἀγοραῖοι to indicate a sociolinguistic category may clarify the point behind Photius’ remark.

(i) ἀγοραῖοςἀγοραῖος indicates a category of speakers in one passage of Phrynichus: Ecl. 176Phryn. Ecl. 176 ὀπωροπώλης· τοῦθ’ οἱ ἀγοραῖοι λέγουσιν, οἱ δὲ πεπαιδευμένοι ὀπωρώνης ὡς καὶ Δημοσθένης (‘ὀπωροπώλης: This is the form used by those who employ an unsophisticated language, while educated people use ὀπωρώνης [‘fruit-seller’] like Demosthenes (18.262)’: see ὀπωροπώλης). The point of this gloss is that ὀπωροπώλης is the commonly used unmarked form, whereas more educated people prefer the rare form ὀπωρώνης, which is attested in Demosthenes. However, the fact that ὀπωροπώλης is attributed to οἱ ἀγοραῖοιοἱ ἀγοραῖοι does not also mean that it is incorrect Greek. There is no apparent reason why it should be regarded as such, since compound words with a second element -πώλης are perfectly ordinary formations (numerous parallels are attested in Aristophanes, for instance) and are normally accepted by Phrynichus as well (one may compare PS 52.13, PS 107.1–2, and Ecl. 63; see further ὀπωροπώλης). To conclude, the ascription of ὀπωροπώλης to οἱ ἀγοραῖοι indicates that this is a popular word used in everyday speech, as opposed to the cultism ὀπωρώνης, which is a marked form with a clear literary pedigree.

(ii) Two relevant examples of οἱ ἀγοραῖοιοἱ ἀγοραῖοι appear in Theodorus Prodromus’ grammar, Περὶ γραμματικῆς 124.27–9 GöttlingTheodorus Prodromus Περὶ γραμματικῆς 124.27–9 Göttling ὁ δὲ νυκτάλωψ εἶδός ἐστι νοσήματος τὸ παρὰ τοῖς ἀγοραίοις ὀρνιθοτυφλότης ὀνομαζόμενον (‘The νυκτάλωψ is a type of disease which is popularly called ὀρνιθοτυφλότης’) and 129.24–6 GöttlingTheodorus Prodromus Περὶ γραμματικῆς 129.24–6 Göttling λύγξ λυγγός· ἔστι δὲ εἶδος νοσήματος τὸ παρὰ τοῖς ἀγοραίοις κλόξος λεγόμενον (‘λύγξ λυγγός: It is a type of disease which is popularly called κλόξος’) (on this treatise, falsely attributed to the grammarian Thedosius, see Zagklas 2011). In both cases, ἀγοραῖος indicates the vulgar, non-technical name of a disease as opposed to the technical term. Here again a distinction is drawn between every-day and informal language (used by οἱ ἀγοραῖοι) and its marked equivalent (in this case, the Fachsprache).

(iii) An extremely significant parallel for the case of ἀγοραῖος indicating a sociolinguistic category is Phot. Bibl. cod. 114Phot. Bibl. cod. 114 ἡ δὲ φράσις εἰς τὸ παντελὲς ἀνώμαλός τε καὶ παρηλλαγμένη· καὶ συντάξεσι γὰρ καὶ λέξεσι κέχρηται ἐνίοτε μὲν οὐκ ἠμελημέναις, κατὰ δὲ τὸ πλεῖστον ἀγοραίοις καὶ πεπατημέναις, καὶ οὐδὲν τῆς ὁμαλῆς καὶ αὐτοσχεδίου φράσεως καὶ τῆς ἐκεῖθεν ἐμφύτου χάριτος, καθ’ ἣν ὁ εὐαγγελικός τε καὶ ἀποστολικὸς διαμεμόρφωται λόγος, οὐδ’ ἴχνος ἐμφαίνων (‘The style [i.e., Photius is describing a pseudo-apostolic writing called The peregrinations of the Apostles, on which see the comment ad loc. in Bianchi, Schiano 2019 vol. 1, 511 n. 1] is very uneven and diverse. For it uses constructions and terms which occasionally are not careless, but for the most part are trivial and frequently used, and has nothing of the regular and spontaneous style and the natural grace with which the diction of the Gospel and the Apostles is provided, not even showing traces of it’). Photius does not use ἀγοραῖοςἀγοραῖος to indicate strictly incorrect Greek, but rather to refer to common, unsophisticated constructions and vocabulary (notice the contextual use of πατέω) in contrast to more carefully written Greek. This opposition between οὐκ ἠμελημέναις and ἀγοραίοις καὶ πεπατημέναις is strongly reminiscent of the situation in A.2, in that a divide is postulated between more careful and attentive (marked) forms of expression and those that are informal and common (unmarked).

To conclude, the sociolinguistic category identified with ἀγοραῖοςἀγοραῖος and οἱ ἀγοραῖοιοἱ ἀγοραῖοι can be associated with informal, everyday language, the kind of people who would have used at the market, and which can be described as unmarked and unsophisticated. This provides a relatively good explanation as to why, in the glosses discussed above (A.1, A.2), a construction attributed to οἱ ἀγοραῖοιοἱ ἀγοραῖοι is deemed unworthy of imitation by the aspiring sophist, even though it could not be regarded as incorrect Greek. In this sense, ἀγοραῖος and οἱ ἀγοραῖοι are very different from labels such as ἰδιωτικόνἰδιωτικός and ἰδιῶταιοἱ ἰδιῶται, which are rather an indication of incorrect language. One could then argue that ἀγοραῖοςἀγοραῖος and οἱ ἀγοραῖοιοἱ ἀγοραῖοι provide a counterpart in terms of register to οἱ πολλοίοἱ πολλοί, which is in turn more often used to indicate the diffusion of a given expression rather than its precise linguistic connotation (see Matthaios 2013, 104: ‘Dagegen ist insgesamt festzustellen, dass der Markant οἱ πολλοί in seiner quantitativen Dimension zu verstehen ist. Er weist auf die Verbreitung eines Ausdrucks hin, unabhängig davon, ob der betreffende Gebrauch aus der alten Literatursprache stammt oder der synchronen Sprachsituation des Lexikographen zuzuordnen ist’). In a number of cases, these two parameters, οἱ ἀγοραῖοιοἱ ἀγοραῖοι and οἱ πολλοίοἱ πολλοί, clearly go hand in hand, since a word or expression commonly in use (οἱ πολλοίοἱ πολλοί) may also be an element of everyday, informal speech (οἱ ἀγοραῖοιοἱ ἀγοραῖοι). Further, these two categories could provide the opposing polarity to the sociolinguistic category of πολιτικόςπολιτικός, which applies to words and expressions that, though commonly in use, are provided with a degree of linguistic prestige qualifying them as good Attic.