*sóh₂-/sh₂-´

The 14 attestations of this lexeme are found in inscriptions from Karkamiš and Tabal, but also in Aleppo and Maraş. While the relevant texts mainly date to the 8^th century BCE, a few earlier ones stem from the 9^th and possibly even 10^th century BCE.

[AH.B.]

Against common practice, the attestation in ASSUR letter f+g, f2§12 is analysed as an imperative (following Sasseville, pers. comm.) instead of an indicative, because the clause includes the prohibitive negative particle /ni/ usually combined with imperatives; for further details, see the discussion of the stem below.

The attestations in MARAŞ 4, §§9-10 was thought to represent a form of the verb /as-/ ‘to be’, but Hawkins 2000a:257f. corrected the identification to the present verb based on the morphosyntax of the clause, which includes reflexive pronouns not usually combined with ‘to be’.

Contrary to earlier analyses, the form 〈(“*69”)sa-na-tu〉 from ASSUR letter e, §19 does not belong to the present verb, but to /sann(a)i-/ ‘to remove’; this issue had been pondered by several scholars, cf. Marazzi 1990b:391 and references.

[AH.B.]

The meaning of this verb was debated among scholars for a number of years, cf. e.g. Hawkins 1980b:115f. and Morpurgo Davies 1980b:98 who suggested ‘to miss’, but Marazzi 1990b:391 opted against assigning any meaning.In addition to that, several of the forms from the paradigm were misidentified, e.g. the 1sg. pret. med. was taken to be a 2sg. pres. act. of a putative verb */sah(a)-/ (see Hawkins 2000a:120), and the 1sg. pret. act. /sahha/ in BABYLON 2, §2 was taken to belong to /as-/ ‘to be’, though see the doubts over this expressed by Hawkins 2000a:396, who points out that the form may belong to /sa-/ ‘to release’ instead, a notion reiterated more assuredly for MARAŞ 4, §§9-10 (Hawkins 2000a:257f.). The current meaning ‘to release, let go’ was eventually proposed by Melchert 1989b:32-37 Eichner 1983a:55f. n. 46 had made a similar suggestion already for Lyc. ha-, which he connected to Hittite šai-, as acknowledged by Melchert 1989b:38f. upon examining the verb and its CLuw. cognate šā-⁽ⁱ⁾ both contextually and etymologically and identifying it as the semantic equivalent of Hittite tarn(a)-^(ḫi) ‘to release, let go’; his analysis found immediate acceptance, cf. e.g. Hawkins 2000a:257f., 542. The verb is clearly transitive (thus again Melchert 2000a:181), as nearly every clause it occurs in features an accusative object.

The shape of the logographic determinative *69 has so far escaped interpretation with its horizontal base from whose upper side a dent protrudes upwards; the sign is listed among the hand-shaped symbols in Laroche’s classification, but is markedly less detailed than any of the other hand-based signs.

[AH.B.]

The verb is a root verb and shows the endings of the ḫi-conjugation, see Sasseville 2021a:357f. What appears to be an exception to the pattern, i.e. the form 〈(“*69”)sa-si^-i〉 attested in ASSUR letter f+g, f2§12 and ordinarily parsed as a 2sg. pres. act. indicative, should instead be taken as further confirmation that this verb indeed follows the ḫi-conjugation: the ending /-si/ appears in a context where an imperative is called for (Sasseville 2021a:340 with n. 7), and the expected form would be /sa/, i.e. the bare root. It is not out of the ordinary to see such a short monosyllabic form recharacterised and thus inflected more overtly, in this case with the default ending of the second person singular, i.e. the 2sg. pres. act. indicative from the mi-conjugation. The form should therefore be analysed as done in the table under Forms, i.e. as a 2sg. impv. act.

[AH.B.]

1. /sas(s)a-/

This lexeme is attested in various Luwian festival and ritual texts.

[O.B.]

This verb may be construed with the preverbs āppan (KUB 9.31 ii 34, HT 1 ii 9´, KUB 35.133 ii 10´), ānta (KUB 35.101 obv.^? 6´) and parī (KBo 29.35, 11´).

The broken attestation ša-a-a[t- is immediately preceded by [š]a-a-at-ti-iš ‘you release’, but it is not clear whether a 3sg. pret. or 2sg. pres. ind. act. is intended.

[O.B.]

All forms have been checked against the photographs in the Konkordanz (Mainzer Fotoarchiv).

[O.B.]

Melchert 1989b:33-35 has demonstrated how the usage of CLuw. šā-⁽ⁱ⁾ is equivalent to that of Hitt. lā-^(ḫi) and tarn(a)-^(ḫi), consequently meaning ‘to release, let go’. This meaning is consensual in the literature; see e.g. CLL:183, HEG Š:703, Frotscher 2012a:159, Dempsey 2015a:276, Sasseville 2021a:357.

[O.B.]

CLuw. šā-⁽ⁱ⁾ is a ḫi-conjugating root verb.

[O.B.]

1. šaššā-⁽ⁱ⁾

This lexeme is attested in two grave inscriptions from Limyra that cannot be dated accurately.

[D.S.]

The two forms hati found in a similar curse formula have been assigned to the same paradigm as ha-^(di) and treated as 3^rd plural with denasalization (a sporadic phenomenon) (thus Melchert (DLL: 21f.), tentatively followed by Christiansen 2020a:250). Schürr 2006a:119 suggests seeing a completely different verb ha-^(ti) ‘to have’, which is not independently supported. Sasseville 2021a:60, 546 likewise takes the verbal form hati at face value, i.e. 3sg. pres., but recognizes two different verbal stem formations built on the same root, i.e. ha-^(di) ‘to release’ and ha-^(ti) ‘id.’, the latter being independently supported by CLuw. šā-⁽ⁱ⁾ ‘to release’ and it is thus an original root verb of the ḫi-conjugation.

[D.S.]

For a new collation of TL 106, see Christiansen 2020a:249-251. Outdated Carruba 1970b:34, who takes trbbalahati as one word, see under Lyc. A tawa-.

[D.S.]

The verb hati is found twice in the same curse formula: me=tawa trbbala hati (TL 106.2) / se=we=trbbala hati tawa (TL 131.5). The verb takes the potential offender as subject (contra Christiansen 2020a:251). Consequently, the curse can be best translated as ‘he will release inimical eyes’ (Sasseville 2021a:60; contra Schürr 2006a:119). Outdated Carruba 1970b:34. See also Lyc. ha-^(di) ‘to release’.

[D.S.]

1. elihala-

2. pddẽnehm̃me/i-

1. hata-

2. hha-^(ti)