βελτίους, βελτίονες
(Moer. β 8)

D. General commentary

Moeris prescribes the s-stem comparative βελτίους (nom. pl. m.) as the correct Attic form, marking βελτίονες as generally ‘Greek’ (see below on the possible contextual meaning of this definition). Indeed, depending on their formation, primary comparatives could be declined as either sigmatic or nasal stems. Regarding s-stems-stems comparatives, they originated from either *-(i)i̯os- and *-īi̯os- (Barber 2013, 185–6) or *-i̯os- (alternations between -ῐ- and -ῑ- in innovative comparatives in -ιων/-ιον being perhaps due to analogical processes, see Nikolaev 2022, 551–4). S-stem comparatives are well attested in the accusative singular of masculine and feminine forms and in the nominative and accusative plural of masculine and neuter forms – that is, ‘in a fossilized and defective paradigm’ (Barber 2013, 146) – as early as Mycenaean (e.g. me-zo-e = /mezohes/, see Bartoněk 2003, 269–70). These s-stem comparatives tend to be replaced by the n-stemNasal stems forms -(ι)ων/-(ι)ον (on which, see Barber 2013, 150, 154, 186; Nikolaev 2022, 551–4; on the whole subject, see also K–B vol. 1, 427; Chantraine 1958–1963 vol. 1, 254–5; Schwyzer 1939, 536–7). However, both s-stem and n-stem comparatives were gradually superseded by -τερ- forms, possibly because the latter attached to the adjectival stem while the other suffixes attached instead to the root (Barber 2013, 146).

Both literary and documentary evidence attests that the s-stem forms are recessive (see e.g. Seiler 1950, 124; CGMEMG vol. 2, 820), as is also suggested by the spread of ‘double comparatives’ (on which, see entry ἀμεινότερος, ῥᾳότερος) and secondary comparatives such as ἀγαθώτερος, μεγαλώτερος, μικρότερος etc. replacing suppletive primary comparatives (see entry ἀγαθώτερος, ἀγαθώτατος). Indeed, as Seiler (1950, 12) pointed out, in Homer, s-stem forms are used primarily for metrical reasons (see, e.g., C.1 where the ‘shorter’ forms κακίους and ἀρείους occur side-by-side with πλέονες) and are far less frequent than other comparative forms (see Chantraine 1958–1963 vol. 1, 255). It is also noteworthy that the occurrences of longer and shorter forms in Aristophanes suggest that, in his time, both forms were used (Willi 2003, 242–3). Attic inscriptionsInscriptions, however, almost exclusively have the s-stem forms, and it is only in the 2nd century BCE that forms in -ον- become common in prose inscriptions; comparatives in -ω, -ους, however, are still used (occurring even in Roman times; see Meisterhans, Schwyzer 1900, 151–2; Threatte 1996, 311). This gradual change partly reflects the development that took place in the koine, where the n-stem forms were progressively disseminated at the expense of the s-stem forms. According to the evidence provided by papyriPapyri and collected by Mayser (Gramm. vol. 1,2, 59–61), in the 3rd century BCE, the ‘shorter’ forms -ω, -ους of ἀμείνων, μείζων, ἐλάσσων, and other high-frequency adjectives were considerably more widely used than their corresponding forms -ονες, -ονας. During the 2nd and the 1st centuries BCE, however, the situation partially changed, although -ους is still the most represented form (the ratio being 10:3, according to Mayser, Gramm. vol. 1,2, 60–1). Among Atticising writers, Aristides and Philostratus regularly use forms in -ους, while Aelian uses -ους only in place of -ονες (see Schmid, Atticismus vol. 3, 23–4; vol. 4, 17–8, 581. On Aelian’s non-purist Atticism, see Rodríguez-Noriega Guillén 2005; cf. also below the data on the distribution of nom. pl. βελτίους and βελτίονες). Papyri of the Roman and Byzantine periods reveal that n-stem forms eventually took over (a detailed list in Gignac 1981, 152–5), although they encountered competition from the secondary comparative suffix -τερ-. Nonetheless, s-stem forms are still attested, largely in stereotyped expressions (see Mayser, Gramm. vol. 1,2, 61; Seiler 1950, 12–4; Gignac 1981, 151 n. 7). In the New Testament, comparatives mostly inflect according to n-stem forms, with exceptions almost exclusively found in the Acts of the Apostles and in John (see Blass, Debrunner 1976, 38–9 and 48; Reinhold 1898, 52).

These data suggest that, when Moeris labels βελτίονες as ‘Greek’, he is generally referring to the usual koine form of the comparative. However, nom. and acc. pl. βελτίους largely prevails in the literary sources, while the nom. pl. βελτίονες occurs fewer than 90 times in the TLG corpus. In the Septuagint, βελτίους has three attestations, while βελτίονες never occurs. Moreover, both nom. pl. βελτίους and βελτίονες virtually never occur in papyri from the imperial age onward.

The distribution of βελτίους and βελτίονες thus calls for further considerations. Although primary comparatives are recessive and forms with disyllabic endings are well attested in the koine, the particular case represented by the distribution of both βελτίους and βελτίονες suggests that in Moeris’ entry, the marker ἝλληνεςἝλληνες is hardly an exact mirror of the linguistic reality of his own times, nor does it indicate the cultivated written koine (for the latter interpretation, see Maidhof 1912, 319–38). In the present case, Ἕλληνες more likely indicates what was perceived to be ‘regular’ according to the criteria of analogyAnalogy (see below re. B.3, B.4, B.5) or simply non-Attic: therefore, Maidhof’s suggestion that Ἕλληνες is a consistent reference to literary koine warrants a review (further discussion in entry Moeris, Ἀττικιστής). Elsewhere, Moeris appears to be very precise in classifying comparative forms: for instance, while ascribing ἀμείνονα to Greek speakers (this time, perhaps, in the sense proposed by Maidhof), he considers ἥσσονα to be ‘common’, i.e., probably pertaining to low-level koine: see entry ἀμείνω, ἥττω.

Other ancient sources also considered s-stem comparatives to be typically Attic: see B.1, B.2, and B.6. Pseudo-Plutarch (B.1, cf. also B.2) regards them as an Attic feature of the Homeric language (see Hillgruber 1994, 117). Some ancient scholars believed Homer to be an Athenian who had written in a form of Old AtticAttic, old: such a theory may have influenced Atticist views (see entries ἥρῳ, οἶσε, and ἱδρῶ, κυκεῶ, Ἀπόλλω, τυφῶ). The other erudite sources dealing with s-stem comparatives (B.3, B.4, B.5) seek to grasp the origin of the -ους ending, indicating that it derived from the ‘regular’ form -ονες through linguistic changes: namely, the loss of -ν and crasis. However, explanations of this nature are not incompatible with the ancient theory according to which the s-stem forms are Attic: ancient scholars who adopted analogy as a guiding principle often regarded dialectal forms as a touchstone of the ‘regular’ ones (see e.g. Probert 2011; Pagani 2015).

E. Byzantine and Modern Greek commentary

The Greek system of comparative adjectives tended towards regularity: in this regard, the suffix -τερ- played a major role (see also D.; further details in CGMEMG vol. 2, 820; interestingly, a ‘double comparative’ β]ελτιότερα is attested in an inscription from Martyropolis, the modern Silvan, dated to the 591 CE or shortly after). Nevertheless, primary comparatives in -ων continue to be attested in Medieval Greek, even if they are rare in non-learned texts and are ‘easily attracted to more common paradigms’ (see CGMEMG vol. 2, 772–3). An ‘irregular’ form like βελτίους (nom. and acc. pl.) is extremely common in the various genres of cultivated Byzantine literature and is considerably more widely used than βελτίονες and βελτίονας, which have only a few occurrences in the TLG corpus (diastratic and stylistic differences are not easily grasped). The idiomatic expression τὰ βελτίω (‘that which is better’) is also well attested.

F. Commentary on individual texts and occurrences

N/A