Philemon
Περὶ Ἀττικῆς ἀντιλογίας τῆς ἐν ταῖς λέξεσιν

In approximately 200 CE, a grammarian and lexicographer called Philemon composed an Atticist lexicon in iambic trimeter entitled Περὶ Ἀττικῆς ἀντιλογίας τῆς ἐν ταῖς λέξεσιν (On Attic Controversy about Words), which has come down to us in two different abridged recensions – L (= Laurentianus 91 sup. 10Laur. 91 sup. 10 [14th century CE], 103v‒104v, on which see Ucciardello 2021, 54–5 n. 6) and V (= Vindobonensis phil. gr. 172Vindob. phil. gr. 172 [15th century CE], 177r‒177v, on which see Ucciardello 2019, 220–4) – each carried by a single manuscript. To the same author is ascribed a work entitled Σύμμικτα (Miscellany), which included discussions of the texts of Homer and Herodotus. The Λεξικὸν τεχνολογικόν ascribed to Philemon and edited by Osann (1821, 1–284), meanwhile, is a 16th-century forgery by Jacobus Diassorinus, as demonstrated by Lehrs (1872) and Cohn (1888). The Atticist Philemon is to be distinguished from the earlier lexicographer Philemon of Aexone (on which see Ucciardello 2007a), a contemporary of Callimachus who composed a treatise Περὶ Ἀττικῶν ὀνομάτων ἢ γλωσσῶν (On Attic Nouns or Obscure Words).

The approximate dating of Philemon rests on the following evidence. Phrynichus, on whom Philemon’s lexicon clearly depends (see B. Transmission, editions, and reference studies), provides a terminus post quem in the mid-2nd century CE. In the 3rd century CE, Porphyry (ad Il. 35.9–10 Sodano) mentions a γραμματικός named Philemon as the author of the Σύμμικτα, from which he quotes a lengthy extract devoted to Herodotean textual criticism (περὶ τοῦ Ἡροδοτείῳ διορθώματος). This fragment discusses the text not only of Hdt. 1.92 but also of two Homeric passages (Il. 15.680–2, 21.126–7), respectfully disputing the opinion of the grammarian Alexander of Cotyaeum. These Σύμμικτα are perhaps to be identified with the Παντοδαπὰ χρηστήρια attributed by Athenaeus (3.114d) to a Philemon (Cohn 1898, 366). If its author is to be identified with the lexicographer, the lifetime of Alexander of Cotyaeum (who was tutor to Marcus Aurelius and died around the middle of the 2nd century CE) provides another chronological reference point. Indirect evidence suggests that Philemon’s lexicon was used by Orus, thus providing a terminus ante quem of the 5th century CE. Cohn (1898, 361–2) identified a fragment of Philemon’s Atticist lexicon in Eustathius’ discussion (in Il. 5.190.14‒6) about the orthography and accent of the noun λη(ι)διον ‘light summer dress’, where Philemon’s doctrine is quoted as ὡς σταμνίον δὲ τὴν προσῳδίαν λέγω (i.e., λῃδίον should have paroxytone accent), an iambic trimeter. Indeed, the Byzantine Etymologica s.v. λῄδιον (Et.Gen. λ 84 ~ EM 563.32) mention Philemon as a source together with Choeroboscus’ Orthography, and the latter – perhaps mediated by the Etymologica – was Eustathius’ source. In turn, Choeroboscus’ source for the entire scholarly discussion surrounding λῄδιον was possibly Orus’ orthographical treatise on the iota subscript, since Orus is quoted by Et.Gen. λ 85 in the entry λῃδάριον, immediately following λῄδιον (Cohn 1898, 363). It is likely, therefore, that Orus had direct access to Philemon’s work and that he used it also in the composition of his Ἀττικῶν λέξεων συναγωγή, Collection of Attic words (Alpers 1981, 112–4; see entry Orus, Ἀττικῶν λέξεων συναγωγή). In the 9th century CE, Choeroboscus himself (in Heph. 183.3 Consbruch) is the source for the title and metre of Philemon’s lexicon as well as for the epithet Ἀττικιστής. Since Ritschl (1832, LXX‒LXXVI), scholars have recognised that a version of Philemon’s lexicon was still circulating in the 14th century, when Thomas MagisterThomas Magister used it for his Ὀνομάτων Ἀττικῶν ἐκλογή (Selection of Attic Words, see entry Thomas Magister, Ὀνομάτων Ἀττικῶν ἐκλογή). The latter’s reliance on Philemon is emphasised by Gaul (2007, 313–4, n. 64; 2008, 189–90), who argues that most entries in Thomas’ lexicon that apparently derived from Phrynichus or Moeris are, in fact, dependent on Philemon.

The codd. of Philemon’s lexicon transmit two rather different abridgements. Cod. L preserves 141 entries covering the letters α‒δ, while cod. V preserves 132 entries, in a more epitomisedEpitome format and with fewer entries per letter but covering the entire alphabet. Only L bears the inscriptio Φιλήμονος. The version of cod. L was first edited by Osann (1821, 285–301) as an appendix to the Λεξικὸν τεχνολογικόν later recognised as spurious. Cod. V was first edited by Reitzenstein (1897, 392‒6), who, however, did not recognise the relationship with L and, consequently, the lexicon’s authorship. This was promptly corrected by Cohn (1898), who identified Philemon as the author. The version used by Thomas Magister appears to have been closer to cod. V in wording but closer to cod. L in extension and to have already lost its original metrical form.

A complete critical edition of Philemon’s lexicon remains a desideratum. Aside from its failure to recognise Philemon's authorship, Reitzenstein’s edition of cod. V suffers from the editor’s frequent attempts to restore the original iambic metre, which sometimes resulted in excessive interventions (see e.g. the entry ἥρως, ἥρωες, and cf. Verdejo Manchado 2020, 265–6 n. 43 for a critique of similarly arbitrary attempts to restore iambic metre in the epitome of Helladius’ Chrestomathy; Isépy 2020, 223–7 offers a different perspective, emphasising the importance of metre as a criterion for reconstructing Helladius’ text). Cohn (1898) was the first to present a synoptic edition of L, V, and Thomas Magister that, albeit limited to the first four letters of the alphabet, offered some textual improvements. Brown’s (2008) unpublished dissertation is the only monograph-length treatment of Philemon’s lexicon to date. It conveniently reports the readings of both L and V as well as that of Thomas Magister, and it offers a commentary on selected entries. While valuable as a collection of useful material, including its discussion of loci classici and lexicographical parallels, it should be used with caution due to the lack of revision. Philemon’s lexicon is discussed also by Tolkiehn (1925); Wendel (1938); Dickey (2007, 98–9); and Ucciardello (2007b).

Brown (2008, 211‒26) conveniently summarises the main concerns of Philemon’s work. Most entries deal with vocabulary, contrasting Attic and non-Attic lexical forms, while some explain the meanings of Attic words or expressions. A favourite subject is verbal morphologyMorphology, verbal: Philemon is keen to prescribe ‘deponent’ presents that had shifted to the active in later Greek (e.g. Laur. 354Philemo (Laur.) 354 ἁλιευόμενος· οὐχ ἁλιεύων), and athematic presents that had been remodelled as thematic forms (e.g. Vindob. 394.14Philemo (Vindob.) 394.14 ζεύγνυσιν· οὐχὶ ζευγνύει). Several entries are concerned with the correct use of prefixesPrefixes in compound verbs (e.g. Laur. 358Philemo (Laur.) 358 διελύσατο· οὐ συνελύσατο). While issues of nominal morphology are also well-represented in this lexicon, it includes little discussion of syntactic constructions. Orthographic and orthoepic discussions, meanwhile, are comparatively frequent: many entries deal with vowel timbre and lengthVowel length, the position and typology of accentAccent, and vowel contractionContraction as well as several consonantal alternationsConsonantal alternations and possibly some cases of syncopeSyncope (see entries ἀριθμός; νεογνός, νεόγονος; ὀψαίτερος, ὀψίτερος). At least one entry (Laur. 355Philemo (Laur.) 355 ἀριθμός) explicitly references pronunciationPronunciation, using the verb ἐκφωνέωἐκφωνέω.

The surviving entries are arranged in alphabetical order, though generally limited to the first letter. As is the case for other lexica, the earliest letters, particularly α, are overrepresented. It is uncertain whether the alphabetical arrangement derives from Philemon’s original work. Several entries appear to have been partially duplicated, even within the same recension, suggesting that they were manipulated in the process of epitomisationEpitome:

• Laur. 356Philemo (Laur.) 356 βάλλ’ ἐς μακαρίαν· ἐς κόρακας ‘βάλλ’ ἐς μακαρίαν (‘Go and be blessed!’): (One could also say) ἐς κόρακας (‘Go and be damned!’)’, Vindob. 393.26Philemo (Vindob.) 393.26 ἐς κόρακας· οὐκ εἰς κόρακας ‘[Say] ἐς κόρακας, not εἰς κόρακας’, Vindob. 393.27Philemo (Vindob.) 393.27 ἐς μακαρίαν· οὐκ εἰς μακαρίαν <λεκτέον> ‘ἐς μακαρίαν: <One should> not <say> εἰς μακαρίαν’;

• Laur. 356Philemo (Laur.) 356 βλαυτίον βέλτιον ‘βλαυτίον (‘slipper’) is better’, Laur. 356 βλαυτίον· σανδάλιον, ὑπόδημα· λέγεται καὶ βλανδίον (sic: perhaps βλαυδίον, as proposed by G. Dindorf in ThGL vol. 2, 278; cf. Hesych. β 686 βλαῦδες· ἐμβάδες, κρηπῖδες, σανδάλια, ὑποδήματος) ‘βλαυτίον: [It means] ‘sandal’, ‘shoe’. One can also say βλαυδίον’;

• Laur. 356Philemo (Laur.) 356 βλήχων· οὐ γλήχων ‘Say βλήχων (‘pennyroyal’), not γλήχων, Laur. 356 βλήχων· βρῶμα διὰ πυρὸς καὶ γάλακτος ἡψημένον ‘βλήχων: A food made from wheat and milk’, Laur. 358Philemo (Laur.) 358 ἡ γλήχων ὡς ἡ κύων ‘ἡ γλήχων [is inflected] like ἡ κύων (‘the dog’, nom. fem. sing.)’;

• Laur. 355Philemo (Laur.) 355 ἀθήρην· οὐκ ἀθάρα<ν> ‘[Say] ἀθήρη (‘porridge’), not ἀθάρα’, Laur. 356Philemo (Laur.) 356 μήποτε δὲ ὤφειλε γραφῆναι ἀθάρη, ὡς προσῆκε. τὴν ἀθάρην περίγραφε. ἢ οὕτω πτισάνη πυρίνη, ἣν καὶ ἀθάρην τινὲς καλοῦσιν ‘One should never write ἀθάρη, as it is fitting. Reject ἀθάρη. Or (it can be used) as follows, gruel made of wheat, which some also call ἀθάρη’;

• Laur. 357Philemo (Laur.) 357 γλαμμᾶν, ἀλλ’ οὐ λημμᾶν ‘[Say] γλαμμᾶν (‘to have bleary eyes’), but not λημμᾶν’, Vindob. 395.10–1Philemo (Vindob.) 395.10–1 λημᾷς σύ γ’, οὐ γλημᾷς <Ἀριστοφάνης λέγει>, λημῶντα δ’ οὐκέτ’, ἀλλὰ <σύ γε> γλαμῶντ’ <ἐρεῖς> ‘<Aristophanes says> ‘you have bleary eyes (λημᾷς)’, not γλημᾷς; however, <you shall> never <say> λημῶντα, but γλαμῶντα (‘bleary-eyed’)’.

As in Moeris, and in contrast to Phrynichus, the approved form is typically mentioned first; occasionally this order is reversed, likely due to metrical constraints. The decision to write a grammatical work in iambics, arguably as a mnemotechnical aid, has a notable parallel in the Chrestomathy of Helladius of Antinooupolis (1st half of the 4th century CE), transmitted only through excerpts in Photius’ Bibliotheca (cod. 279.529b.24–536a.22) and a few fragments in Orion’s Etymologicum (see Heimannsfeld 1911); it may be considered part of a wider tradition of iambic didactic poetryIambic poetry inaugurated by Apollodorus of Athens in the 2nd century BCE (see Fleischer 2020, 90‒3). Traces of the original metre are still recognisable in some of the transmitted entries (Reitzenstein 1897, 388).