ὀπωροπώλης, ὀπωρώνης
(Phryn. Ecl. 176, Poll. 6.128)

D. General commentary

Phrynichus (A.1) opposes ὀπωροπώλης – which he says was used by οἱ ἀγοραῖοιοἱ ἀγοραῖοι – to ὀπωρώνης, a form attested in Demosthenes, which Phrynichus says was used by οἱ πεπαιδευμένοιοἱ πεπαιδευμένοι. Unlike Phrynichus, Pollux (A.2) puts ὀπωροπώλης and ὀπωρώνης on an equal footing, listing these forms among words for professions that can be used derogatively. Finally, Thomas Magister’s gloss (A.3) is a little puzzling: although he evidently depends on Phrynichus’ gloss, he substitutes ὀπωροπώλης with the periphrasis ὀπωρῶν ὠνήτωρ, which is equally odd given that ὠνήτωρ is a very rare Byzantine alternative to the far more common ὠνητής.

Both ὀπωροπώλης and ὀπωρώνης are unobjectionable compounds. Regarding ὀπωροπώλης, there is abundant evidence in 5th‑century Attic for names of professions that have the second element ‑πώλης (< πωλέω ‘to sell’)Compounds, many of which are primarily attested in comedy (there is a list in DELG s.v. πωλέω; for a thorough investigation of names for professions with the second elements ‑πωλης and ‑πράτης in literary sources, inscriptions, and papyri, from ancient to Byzantine times, see Casarico 1983, Drexhage 1991, Hickey 1996, Drexhage 2002, Ruffing 2002, and Diethart 2005). As for ὀπωρώνης, this form belongs to a group of other compounds with the second element ‑ώνης (< ὠνέομαι ‘to buy’), which indicates resellers and state officials who oversee the buying of goods (a list in DELG s.v. ὠνέομαι). A semantic difference between ὀπωροπώλης and ὀπωρώνης, however, is that while the former simply indicates people who sell fruit, the latter more specifically refers to a reseller (see B.3 and B.4; see DELG s.v. ὠνέομαι on compounds with the second element ‑ώνης: ‘négociants et […] fonctionnaires acheteurs, que seul l’usage attique distingue, parfois, comme revendeurs, du parallèle en ‑πωλης’).

Phrynichus observes that ὀπωροπώλης was used by οἱ ἀγοραῖοιοἱ ἀγοραῖοι, whereas educated people used ὀπωρώνης instead. The point underlying this remark has been deemed elusive by previous scholars, who note that if Phrynichus’ aim had been to say that ὀπωροπώλης was unacceptable, he would have been refuted immediately by innumerable counter‑examples (see Lobeck 1820, 206). Indeed, not only is a compound like ὀπωροπώλης inoffensive as far as its formation is concerned (see above), Atticist lexicographers also regularly approve of compounds with the second element ‑πώλης (see ῥωποπώλης and γελγοπώλης in Ael.Dion. ρ 14 [= Eust. in Il. 3.459.29–460.1], βιβλιοπώλης in Phryn. PS 52.13Phryn. PS 52.13, ῥωποπώλης in Phryn. PS 107.1–2Phryn. PS 107.1–2, βελονοπώλης in Phryn. Ecl. 63Phryn. Ecl. 63, γελγοπώλης in Moer. γ 19Moer. γ 19; many more compounds with a second element -πώλης are recorded by Pollux). To my mind, Phrynichus’ comment that ὀπωροπώλης was used by οἱ ἀγοραῖοι should be taken to mean that ὀπωροπώλης was perceived as unsophisticated Greek and therefore avoided by people who intended to use more distinctive language. In modern terms, the ascription of ὀπωροπώλης to οἱ ἀγοραῖοι is an indication of the sociolinguistic register to which this form belonged, and it does not reflect doubts concerning the correctness of such a compound. This may possibly have something to do with the very profession indicated by this word.

The distribution of ὀπωροπώλης and ὀπωρώνης is extremely uneven. While we are unaware of any attested literary occurrences of ὀπωροπώλης (save for a late and para-literary example, Philogel. 238 ὀζόστομος ἀπελθὼν εἰς ὀπωροπώλην ἠρώτα ‘A man with bad breath went to the fruit-seller and asked’), there are several occurrences in papyri (the earliest being C.2, then P.Oxy. 6.980 [= TM 69486, 3rd century], P.Oxy. 8.1133.7 [= TM 21750, 396 CE], P.Oxy. 64.4441. col. vi.10 [= TM 23666, 315–6 CE]) and inscriptions (SEG 6.435.2 [Konya, South-Central Turkey, undated], MAMA 3.359.2–3 [Korykos, undated, Christian]). All these cases are informal texts (reports, private letters), and ὀπωροπώλης is used as the common, unmarked word for a fruit-seller. The only occurrence of ὀπωρώνης in a classical text is in a passage of Demosthenes (C.1), as recognised by Phrynichus (A.1) too. After an interval of many centuries, it resurfaces again in a letter of Aristaenetus (C.3) and in a late antique papyrus of legal content (C.4). Given the very limited diffusion of ὀπωρώνης, both these occurrences are compatible with the influence of Atticist doctrines, or at least with the use of more elevated language. The re-use of ὀπωρώνης in Aristaenetus (C.3) is consistent with his programmatic imitation of Attic models and language (see Arnott 1968 and Magrini 1981). The papyrus occurrence of ὀπωρώνης (C.4) is found in a partnership contract, a slightly more official type of document in which the use of more formal language is not unknown.

E. Byzantine and Modern Greek commentary

A few documentary occurrences of ὀπωροπώλης are attested in Byzantine texts (SEG 47.2023Α.4–5 [Gadara, approx. 6th century CE] as ὀπερoπώλις [lege ὀπωροπώλης]; SB 18.13599.5 [= TM 36297, provenance unknown, 6th or 7th century CE]). In addition to the masculine, the feminine ὀπωρόπωλις (‘female fruit-seller’) is attested too (see LBG s.v.). While compounds in -πώλης, based on the common verb πωλῶ/ πουλάω 'to sell', are common in Medieval and Modern Greek (see CGMEMG vol. 2, 356), the archaic first element makes ὀπωροπώλης and οπωροπωλείο ‘shop of a fruit-seller’ learned terms with no real life in spoken or everyday Modern Greek. As Ioanna Manolessou informs me, they are found only as official appelations, and appear on shop signs only in order to make them sound more ‘classy’ (the only vernacular word for ‘fruit and vegetable seller’ being the Turkish μανάβης, with the shop being called μανάβικο).

Unlike ὀπωροπώλης, the Atticising equivalent ὀπωρώνης is unattested in Byzantine, Medieval, and Modern Greek (this is because no verb corresponding to ὠνέομαι survives).

Byzantine sources also attest to the existence of a third compound that is the same type as ὀπωροπώλης and ὀπωρώνης, namely ὀπωροπράτης (CIIP vol. 2, no. 978 [Jerusalem, Byzantine times] with the comment ad loc. in SEG 60.1720, Tz. Epist. 57.82.10–4 Leone; the corresponding verb ὀπωροπρατέω is also attested, see Tz. Epist. 57.81.17 Leone). Unlike the other two compounds, ὀπωροπράτης is significantly later. Although πράτης and συμπράτης are attested in classical times (πράτης in Hyp. fr. 163 Jensen, συμπράτης in Lys. fr. 507 Carey and Din. fr. 34 Conomis), these forms were not approved by Atticists (see Orus fr. B 138)[= Phot. π 1142)Phot. π 1142]Orus fr. B 138 (= Phot. π 1142): πρατήρ· οὐχὶ πράτης ὑπὸ τῶν Ἀττικῶν λέγεται ‘πρατήρ: This form, not πράτης, is used by Attic authors’). Poll. 7.8 does include πράτης among derivatives from verbs meaning ‘to sell’, but he calls these forms σκληρά (‘harsh’)σκληρός. More generally, all remaining compounds with the second element -πράτης are regularly (much) later than corresponding compounds with the second element -πώλης (see DELG s.v. πέρνημι).

F. Commentary on individual texts and occurrences

N/A