*k^u̯is=h₂o

This lexeme is attested 41 times in inscriptions from Emirgazi, Tabal, Assur, Karkamiš, Hama and Cilicia, dating from the second half of the 13^th to the early 7^th century BCE. 

[A.P.]

The above table offers a representative selection, omitting broken attestations. 

In KARKAMIŠ A23, §7, the order of the signs suggests 〈kwa/i-a-ha〉 rather than 〈kwa/i-ha-a〉, offering enough space for the signs to be rearranged in their respective positions if needed.

Regarding the identification of the dat./loc. sg. forms, see the discussion of the meaning below.

[A.P.]

Because this pronoun is formed using the same strategy known from other Anatolian languages, its identification as an indefinite pronoun was quite straightforward, see the detailed presentation of the /kwa/i-/-based pronouns involving use of the sign *329, read REL and 〈kwa/i〉, by Laroche 1960c:170-174.At the time, the syllabic reading of sign *329 was still somewhat unclear, and Laroche 1960c:170, 173 lists it as 〈hù^?〉. Earlier, Forrer 1932a:41 had suggested 〈ku〉. The grammatical functions of the words spelt with *329 were first discovered by Forrer 1932a:41f., but the actual meaning of the present lexeme only became clear shortly after, see Hrozný 1933a:38, 40, and 87f., who translated /kwatti=ha hantawatti/ as ‘some/any king’ (“un roi quelconque”) in KARKAMIŠ A6, §25 despite a mostly erroneous syllabic reading of the passage and in particular the pronouns. His grammatical interpretation was accepted, though the phonology debated, cf. e.g. Gelb 1942a:54-69, who does however affirm the indefinite meaning of the present lexeme (Gelb 1942a:59).

The indefinite pronoun can occur by itself, or accompany a noun, or be combined with a negation. An example of the latter is found in KARKAMIŠ A30h, §1: ˹(DEUS)ku+˺AVIS (HORREUM)ka-*282-na NEG₃ kwa/i-i-sa-ha su-su-tá, approx. /Kubabas karunan ni kwis=ha susutta/, meaning ‘No-one used to fill (the goddess) Kubaba’s granary’ (Hawkins 2000a:177).

On the basis of its spelling, the dat./loc. sg. form is difficult to distinguish from the adverb HLuw. /kwadi=ha/ ‘somewhere’, only the final plene attested as 〈kwa/i-ti-i-ha〉 in KULULU 1, §14 and KARKAMIŠ A6, §25 may be taken as an indicator of the long /-ī/ expected for that form. However, plene spelling in HLuw. is far from reliable, see Vertegaal 2018a:165, and distinguishing between the forms is therefore a matter of context interpretation.

[AH.B.]

The indefinite pronoun derives from the interrogative and relative pronoun /kwa/i-/ ‘who, which’ in the manner commonly found in the Anatolian languages, viz. by the addition of the particle /=ha/ ‘and’, cf. CLuw. kui-+ḫa ‘some-/anyone’ and Hittite kui-+ki/ka/ku ‘some-/anyone’ (see e.g. HED K:224-226).

The word family around these pronouns is substantial, and the adverb HLuw. /kwadi=ha/ ‘somewhere’ is very likely lexicalised from the paradigm of the indefinite pronoun, while the negative polarity item HLuw. /kwihhan/ ‘(not) at all’ is probably also based on an older form.

[AH.B.]

1. /kwadi=ha/

Amply attested in Luwian rituals (from MS to LNS).

[Zs.S.]

Attestations follow CLL:119 (cf. also Laroche 1959a:56) checked against the photos of the Mainzer Photoarchiv. Restoration in KUB 35.133 iii 23 follows Starke 1985b:282 based on a parallel passage. Starke 1985b:162 reads k[u-i-ḫa] in KUB 32.10, 6’ (followed by Melchert), which is probably correct based on textual parallelism, but the photo shows only the head of a single horizontal.

[Zs.S.]

There is a communis opinio regarding the general meaning of this word as an indefinite pronoun (HED K:231) or similar (Laroche 1959a:56 [‘quiconque’, but only for kuišḫa, since for kuiḫa he assumed the obsolete meaning ‘inscrire, marquer’ (with question mark, entertaining the possibility of a pronoun in KUB 35.79 iv 13)]; HEG A-K:614 [‘wer auch immer’], but ‘manche(r), irgendetwas’ in Tischler 2008a:91 and ‘quelqu’un, quelque chose’ in Tischler 2016a:171; Dunkel 1982-1983a:198 n. 119 [‘quisque’]; Starke 1990a:466 n. 1698 [‘irgendein’]; CLL:119, EDHIL:489 [‘some/any(one)’]; Marcuson 2016a:291 [‘something’]; Ünal 2016a:961 [‘herhangi birisi’]), but since the category of indefinite pronoun includes several different pronouns, a more sophisticated investigation is required, which was only provided by Sideltsev & Yakubovich 2016a:13–22 and Simon 2021e. The former authors argue for the meanings ‘each’ and ‘anything (negative polarity item)’. Although, as Simon’s investigation shows, the passages they quote do not support these meanings, Simon could find support for the meanings ‘each’ and ‘any/someone/thing’ either as a free choice expression (type ‘I can kiss anyone’) or as an existential quantifier (type ‘I kissed someone’).

[Zs.S.]

This pronoun is found in various funerary and dedicatory inscriptions.

[D.S.]

Pace Melchert (DLL:67), tise at TL 5.8 “to place anything on top” and TL 84.6 “to allow anything” are not acc. sg. c., but nom./acc. sg. n. (cf. also Sideltsev & Yakubovich 2016a:106). No animate or indeed human object is present in the phrase.

[D.S.]

For the analysis of tise (a)di in TL 118.6, see Christiansen 2020a:252, contra Neumann 2007a:365.

[D.S.]

Deecke 1888b:264 tried to translate the phrase tise tise prn͂nawati as ‘whoever builds’, comparing it to a Latin phrase of the type quisquis aedificiat (followed by Laroche 1960a:177), although it was later corrected to ‘whatever he built’ (i.e. quidquid…) by Thomsen 1899a:35 n. 2 (further defended by Gusmani 1962a:172-174 and accepted by Houwink ten Cate 1961a:71, Carruba 1978b:170, Melchert in DLL:67, Neumann 2007a: 365 and Sideltsev & Yakubovich 2016a:104-106). It is noteworthy that tise ‘anything’ so far appears on its own, whereas tisn͂ modifies an adjective, i.e. tisn͂ trbbeli ‘anyone inimical’.

Whenever tise tise is iterated, it becomes indefinite, i.e. ‘who-, whatever’.

[D.S.]

The form tise was convincingly connected to CLuw. kuišḫa, which account for the retention of the -s- in Lycian A, i.e. *-sh₂- > Lyc. A -s- (Laroche 1967a:62, Sideltsev & Yakubovich 2016a:107, 109; contra Carruba 1978b:178 “tise = CLuw. kuinza”). The form tisn͂ appears to have been rebuilt within Lycian with the acc. sg. c. ending -n͂ based on the reanalyzed stem form tis- (see Sideltsev & Yakubovich 2016a:110).

[D.S.]

1. tisn͂ke

This lexeme is attested in the inscription of the inscribed pillar of Xanthos, dated contextually to shortly after the ascension of Artaxerxes II in 405-404 BCE and stylistically to around 400 BCE (Borchhardt & Eichner 1997-1999a:19f.).

[D.S.]

The word kize appears to modify the substantive qle (dat./loc. pl.), i.e. ‘to any precinct’, following the comparison by Melchert 1994a:315 (DLL:118) with Lyc. A tise (accepted by Shevoroshkin 2011b:143, Martínez Rodríguez 2019a:224, hesitant Schürr 2016d:162; contra Gehrisch 2018a:83 “rel. pr.”, which would be syntactically awkward). Melchert’s semantic assignment is contextually supported:

kize : qle : pẽniu : tubedi lebi : kudi : pubrati (TL 44c.60-61)

‘I dispatch the lebi with tubedi to any precinct, where he (scil. Gergis) commissions…’

[D.S.]

The status of the -z- vis-à-vis Lyc. A -s-, i.e. kize vs. tise, seems troubling, but ultimately depends on the reconstruction of the pre-forms. Whereas Melchert 1994a:315 sees in Lyc. B a reflex of Pre-Lycian *k^u̯inse, Martínez Rodríguez 2019a:224 regards it instead as a reflex of *k^u̯i-=k̑e; see now under Reconstruction. To account for Lyc. B -z- instead of -s-, we may start from the non-attested form *kizn͂ (= Lyc. A tisn͂), where the -s- would become -z- by regular sound change in Lycian B, i.e. *s > z after a nasal. Thereafter, it spread to the entire paradigm replacing *kise with kize.

[D.S.]

This lexeme is found in the festival text for the Palaic pantheon.

[D.S.]

Checked against the photograph in the Konkordanz (Mainzer Fotoarchiv).

[D.S.]

The relative pronoun kuiša appears at the beginning of the relative sentence and introduces a free-choice relative clause; cf. Sideltsev & Yakubovich 2016a:96-98, Melchert 2016f:291, Carruba 1970a:60.

[D.S.]

HEG I-K:614f.