nãqi-

This lexeme is found in a variety of inscriptions, most of which are found in Sardis, and datable between the 6^th and 4^th centuries BCE.

[I.Y.]

nom. sg. c.	nãqiš	LW 1.7 (Sardis)
		LW 2.5 (Sardis)
		LW 2.11 (Sardis)
		LW 5.3 (Sardis)
		LW 7.3 (Sardis)
		LW 8,6 (Sardis)
		LW 9.5 (Sardis)
	ṇ[.]q̇iš	LW/N 115 (Koloe)
	ṇ[ãqiš]	LW 1.4 (Sardis)
nom./acc. sg. n.	nãqid	LW 10.12 (Sardis)
		LW 24.3 (Sardis)
		LW 24.18 (Sardis)
nom./acc. pl. n.	nãqida	LW 22.5 (Sardis)

For the grammatical assignment of the form nãqida, see Sideltsev & Yakubovich 2016a:98f and cf. under qi-. For a different interpretation, see Gérard 2005a:96. The sequences to be considered together with this lexeme are LW 10.14 nã=m quid and LW 54.4 nã=mλ qi=g, where the elements nã and qi- are separated by the second-position clitic (ibid., n. 24).

[I.Y.]

Lyd. nãqi- corresponds to Aram. mn zy ‘whoever’ in the Lydian and Aramaic bilingual LW 1 (Littmann 1916a: 35, 71), and the meaning ‘who-, whatever’, accepted in Gusmani 1964a:171, generally suits all the relevant contexts. For the similarity between the use of nãqi- and qi- in the relevant texts, cf. Probert 2018a:365 n. 5. The tentative interpretation ‘if any(one)’ offered in Sideltsev & Yakubovich 2016a:98 n. 24 pertains to diachrony rather than synchrony, as again indicated in Probert 2018a:365 n. 5, and appears to be redundant even as an etymological hypothesis (cf. below). The similarity between the clauses headed by nãqi- and qi- militates against the idea of Haas 1962a:201 to interpret nã-qi- as an equivalent of Lat. ne quis.

[I.Y.]

Lyd. nãqi- obviously represents a combination of the particle nã(ν) and the relative pronoun qi-. As shown by the examples LW 10.14 nã=m quid and LW 54.4 nã=mλ, the fusion of the two elements is not yet complete (cf. above). According to Gusmani 1964a:171 this is the same element as nãν, which in turn was compared with Hitt. nam-ma ‘again’ in Carruba 1959a:38. For a combination of the relative pronoun and additive particle yielding a generalized meaning cf. the case of Lat. quisque ‘every’. An alternative comparison of nã(ν) with Hitt. man ‘if’ and Luw. /man/ ‘whether’, tentatively advanced in Sideltsev & Yakubovich 2016a:98, n.24 brings no obvious semantic gain and is formally more complicated, because it requires the assumption of a distant assimilation *man > *nan and posits two different sources for Lyd. nã(ν).

[I.Y.]