νεώς, ναός
(Moer. ν 1, Orus fr. A 66)

D. General commentary

Moeris (A.1) and Orus (A.2) deal with the inflectional model of νεώς (‘temple’). This is the Attic equivalent of the standard ναός, derived from the latter by way of what has been called ‘quantitative metathesis’, but which Schwyzer (1939, 245–6), followed by Méndez Dosuna (1993), explained as the synizesis (loss of syllabicity) of a long vowel with simultaneous compensatory lengthening of the following vowel (ναός > νηός > νε̯ώς). The inflection of νεώς, traditionally used to exemplify the so-called Attic declension, is as follows: nom. sing. νεώς; gen. sing. νεώ; dat. sing. νεῴ; acc. sing. νεών; nom. pl. νεῴ; gen. pl. νεών; dat. pl. νεῴς; acc. pl. νεώς; nom. and acc. dual νεώ; gen. and dat. dual νεῴν (see Allen 1871, now outdated; Schwyzer 1939, 514, 557–8; Chantraine 1961, 43–5; Gammage 2019, 44–5; for further information on the ‘Attic’ declension and the Atticists’ attitude towards it, see entry λαγώς, λαγός, λαγωός and AGP vol. 2, Nominal morphology, forthcoming).

Moeris (A.1) prescribes the nominative singular νεώς, specifying that it is oxytone (see below), instead of ναός, a form said to be used by the ‘Greek speakers’Ἕλληνες. By applying the label ‘Greek speakers’ to ναός, Moeris is either pointing to the fact that ναός and similar forms are normally used in koine Greek or classifying ναός as an unmarked alternative to νεώς without any particular diachronic implication (see the discussion of the distribution of both νεώς and ναός below and cf. the entry Moeris, Ἀττικιστής on the evaluative terminology in Moeris’ lexicon). Orus (A.2) explicitly states that the acc. sing. should be written with a final ν: this consideration was probably prompted by the analogical acc. sing. νεώ, attested in Middle and New Comedy (see e.g. C.6) and ‘probably universal after 350 BCE’ in Attic inscriptions (Threatte 1996, 39; further discussion of accusative -ω in AGP vol. 2, Nominal morphology, forthcoming), which by Orus’ time was homophonous with both the gen. sing. νεῶ and the dat. sing. νεῷ (see AGP vol. 1, 260). The need for a distinction between these forms also seems to be attested in the Lexicon Messanense, which is generally considered to be a section of Orus’ Orthography dealing with the iota subscript (B.3; the lexicon is preserved in a 14th-century manuscript in the Santissimo Salvatore monastery in Messina [cod. S. Salvatore 118], see Rabe 1892; Alpers 1981, 80–1; Ippolito 2008; on the iota subscript in the Attic declension, see also EM 616.32–42 [= Hdn. Περὶ ὀρθογραφίας GG 3,2.554.26–555.6]).

Apart from the Atticist sources, various ancient grammatical treatises and lexica deal with νεώς, taking it as an inflectional model exemplifying the Attic declension (the list in section B. is meant to be representative, but by no means exhaustive, since observations on the Attic declension are scattered throughout the corpus of Greek scholarship, often as minimal remarks). These sources are mainly concerned with the inflectional model, accentuation, and phonology of νεώς and other similar nouns and adjectives in comparison with their standard equivalents. Among these, B.1 is particularly noteworthy, since it shows that already the 1st-century BCE grammarian Tryphon (quoted by Athenaeus) observed that νεώς and similar forms take the same accentAccent as that of the nominative of the standard forms from which they derive (cf. Dieu 2022, 468; for alternatives such as λαγώς/λαγῶς, and ὀρφώς/ὀρφῶς discussed by Tryphon and other ancient grammarians, see Dieu 2022, 469, entry λαγώς, λαγός, λαγωός, and AGP vol. 2, Nominal morphology, forthcoming). Tryphon’s view on the accentuation of νεώς corresponds to that of Herodian (as is clear from B.5) and may have been shared by Moeris (A.1), according to whom νεώς is oxytone (however, Vessella 2018, 230 thinks that Moeris’ main concern here is to contrast two different forms, rather than to prescribe the correct accentuation of νεώς).

As regards the distribution of νεώς and ναός, the former is mostly found in Attic authors of the 5th and 4th centuries BCE and is still the norm in Middle and New Comedy (the same is true for other Attic forms such as λεώς and ἵλεως, see AGP vol. 1, 259–60). In Attic inscriptions, νεώς is normal until the late 3rd century BCE (see Threatte 1996, 39–40; López Eire 2002, 82). Significantly, the two occurrences of ναός in Aristophanes (Nu. 306; Lys. 775) are ‘stylistically conditioned’, the former being mentioned ‘alongside ‘Doric’ α’, the latter being part of a ‘‘para-epic’ oracle’ (Willi 2003, 242).

Standard rather than Attic forms begin to be attested in 4th-century Attic prose. An important example is Xenophon (see e.g. C.4 and C.5), ‘who alternates ναός and νεώς even within the same work’ (see AGP vol. 1, 259–60 for further details), but ναός also occurs three times in Plato (Lg. 738c.7, 814b.5; R. 394a.5; otherwise always νεώς) and at least once in Hyperides (Epit. 8.20; see López Eire 2002, 82). ναός eventually became the standard koine form, and the same holds true for other similar forms with an Attic alternative (see e.g. λαός). The standard forms offered a more regular inflection than their Attic counterparts, which must have contributed to the disappearance of the Attic declension (but note that compounds indicating titles such as νεωκόρος ‘warden of a temple’ and νεωποίης ‘official in charge of the temple-fabric’ etc. continue to be attested in documentary sources, often alongside their equivalent in ναο-, see e.g. Gignac 1981, 30–1).

However, this does not mean that νεώς and other forms at some point came to be perceived as a vernacular feature to be avoided in written koine Greek. On the contrary, νεώς is well represented in high-level prose from the Hellenistic age onwards, perhaps because of its literary pedigree. Even if a precise account of the distribution of ναός and νεώς in Hellenistic and later literature is made difficult by possible accidents in the manuscript tradition, it is noteworthy that Polybius apparently used both ναός and νεώς (see de Foucault 1972, 65). The same holds true for many later authors (see AGP vol. 1, 259–61; Lucarini 2017, 7, on Herodian; Gammage 2019, 55–6, on Achilles Tatius). Dionysius of Halicarnassus, Philo of Alexandria, Strabo, Arrian, Appian, and Dio Chrysostomus seem to have had a clear preference for νεώς. Furthermore, although the Septuagint usually has ναός, in LXX 2Ma. νεώς occurs seven times in LXX 2Ma.

On the other hand, documentary texts on papyrusPapyri have almost exclusively ναός (on the papyrus evidence for the Attic declension, see Mayser, Gramm. 1,2, 13–5; Gignac 1981, 29–32; a significant exception is discussed below). This is also the form attested in the New Testament, where the only discernible trace of the Attic declension is ἵλεως in the formulaic expression ἵλεώς σοι (‘[may God] favour you’, see Blass, Debrunner 1976, 36).

To sum up, ναός, which was also sporadically used in 4th-century BCE Attic prose, took over in koine Greek; nevertheless, νεώς often occurs in Hellenistic and imperial prose, perhaps due to the influence of literary models from the 5th and 4th centuries BCE. Within this framework, Moeris’ and Orus’ reference to νεώς as the ‘orthodox’ Attic form is fully understandable. Unsurprisingly, a similar attitude is detectable in Atticising authors of the imperial era, even if some of them seem to be more in line with other representatives of high-level prose in alternating both forms (see already Schmid, Atticismus vol. 2, 18; vol. 4, 582). Indeed, both νεώς (31x) and ναός (10x) are attested in Lucian (see Schmid, Atticismus vol. 1, 226). In contrast, Aelian has only νεώς (in NA 11.3.1 ναός of cod. V is probably a scribal slip, see Schmid, Atticismus vol. 3, 25). In Flavius Philostratus the only form that follows the non-Attic model is apparently the gen. plur. ναῶν in VS 50.12 (Schmid, Atticismus vol. 4, 20). As regards later prose, note that someone like Libanius, who typically adopts literary Attic forms along with high-level koine features, uses both forms (on this point, see López Eire 1991, 98).

From what has been said so far, it is clear that the use of νεώς, although prescribed by the Atticists, need not necessarily be a sign of a tendency towards strict Atticism. Nevertheless, it can sometimes be regarded as such, depending on the context in which it occurs. See, for example, the exceptional attestation of the genitive [ν]εώ in a papyrus dating to the imperial age, P.Giss. 1.99.17 (= TM 27877) [Hermopolis, 2nd–3rd century CE] (C.7). The text, a forensic speech delivered in the context of a trial against a cult association, shows other Atticising features, such as -ττ- instead of -σσ- and the consistent use of the dual (see Luiselli 1999, 135, 147–8, for a detailed discussion). To be sure, νεώς also occurs in another papyrus of roughly the same period, P.Oxy. 43.3094.41 (= TM 15973) [217–218 CE]. Here νεώς is found in the address written on the back of a letter sent by two women. On account of these peculiarities, it is tempting to conclude that the two senders, who were involved in a protracted legal dispute (see Rea 1975, 17), entrusted an educated (professional?) scribe with a letter which must have been of the utmost importance to them (on women’s literacy, see lastly Hübner 2018, with bibliography). The scribe, for his part, may have unintentionally left a trace of his education by choosing to write νεώς instead of the more common koine form ναός (a comparison can be made with an occurrence of νεώς in an Attic inscription of the Roman period, which is ‘probably a learned spelling’, according to Threatte 1996, 40).

E. Byzantine and Modern Greek commentary

As noted above, koine Greek generalises ναός: this is also the standard form in Modern Greek, where ναός usually means ‘church’, i.e. the building in which Christians worship their God. Nevertheless, νεώς survived as a learned form in the imperial age and then in Medieval Greek. More precisely, Byzantine literary texts of various kinds (historiography, hagiography, etc.) often alternate both νεώς and ναός (with the predominance of the latter), as already happened in the literary koine. Something similar may be said of another learned form, λεώς (cf. Thom.Mag. 221.2Thom.Mag. 221.2: λεώς, οὐ λαός, ‘[One should say] λεώς [‘people’], not λαός’): it is attested in cultivated literature (especially history), although much less frequently than λαός, which is also the standard form in Modern Greek (the question of this latter form, however, is further complicated by the alternative λάς, attested in Cypriot Greek: see CGMEMG vol. 1, 454).

F. Commentary on individual texts and occurrences

N/A