*Hérk̑-/*Hr̥k̑-´

Attested six times, this verb can be found on inscriptions and letters from Central Anatolia, Tabal, Karkamiš, and Assur. While the Karkamiš evidence dates to the late 10th or early 9th century BCE., the attestations from Tabal, Central Anatolia, and Assur are rather later and belong to the 8th century BCE. The evidence from the latter two locations is recorded on lead strips and may have been brought in from elsewhere.

[AH.B.]

The attestation occurs in a part of the KIRŞEHİR lead strip that is somewhat damaged due to having been rolled up. Therefore the signs are only partly legible and do not constitute fully reliable evidence; the form in question may actually represent a dat./loc. sg. noun.

It is possible that KARKAMIŠ A25a, §5 constitutes a 3sg. impv. act. and should be read /harzatu/ instead, but the fragmented nature of the context prevents any certainty.

Likewise, İSTANBUL 2, A2 appears in a very fragmented inscription and may represent a 3sg. pres. act. /harzadi/.

[AH.B.]

The present meaning for the verb was first suggested by Melchert 1988b:221f., building on earlier work by Morpurgo Davies 1980b:98, who assigned a more general meaning of ‘to get’. Melchert 1988b:222 explains that ‘to collect’ works well when applied to items as in ASSUR letters b and f+g §39, whereas the missing person mentioned in ASSUR letter f+g §46 requires the variation ‘to get hold of’. In the KIRŞEHİR lead strip, the lexeme appears in a rather opaque context with several other words whose meanings have so far escaped interpretation, cf. Akdoğan & Hawkins 2010a:7. Because it is followed, after three intervening words and no explicit clause boundary, by a finite verb, 2sg. impv. act. /sasa/ ‘to release’, it does perhaps not constitute a verb itself but a noun with unclear meaning.

The contexts of İSTANBUL 2 and KARKAMIŠ A25a are too fragmented for a meaningful assessment.

[AH.B.]

The stem of /harza-/ is not wholly clear. Melchert 1988b:222 analyses it as an intensive (previously: iterative) in /-zza-/, which would make it a verb of the hi-conjugation and find support in the form recorded in the KIRŞEHİR lead strip. If, however, the form from İSTANBUL 2, A2 should be analysed as a 3sg. instead of a 3pl. and the attestation from the KIRŞEHİR lead strip turned out to be a noun after all, then this would constitute solid evidence for a mi-conjugation form of a different stem class, perhaps a factitive in /-a-/.

[AH.B.]

Attested in a Luwian ritual fragment (NS).

[Zs.S.]

Attestations following CLL:77 (cf. also Laroche 1959a:43) checked against the photographs of the Konkordanz. Since the first sign has multiple values (ḫar, ḫur, and mur, although [based on the collection of CLL] the last one is not attested yet in Luwian texts) and the meaning of the word is unknown (cf. under Meaning), its reading cannot be safely established (Otten 1953c:114, Laroche 1959a:43, Starke 1985b:363, and Ünal 2016a:957 read ḫar, CLL:77 ḫur [but also mentions ḫar], Tischler 2008a:50, Tischler 2016a:79 considers both ḫar and ḫur). Restoration follows Starke 1985b:363. For possible morphological analyses see under Meaning.

[Zs.S.]

Since the context of the attestations is fragmentary (including the preceding and the following words with one exception, see below; the restoration of a preceding [ḪUR.SAG^ḪI.A-]˹ti˺-in-za in KBo 7.68 rev. iii 14’ by Starke 1985b:363 is not supported by any evidence), the meaning of the word is unknown, it is not even clear if we are dealing with a nominal or with a verbal form (Tischler 2008a:50, Tischler 2016a:79; Ünal 2016a:957). Laroche 1959a:43 proposed acc. sg. (with question mark), CLL:77 a 2nd pl. imp. with the meaning ‘curse?’ (from a stem *ḫurt-/ḫurz-), but classifying it as “far from certain”. He also called attention to the alternative reading ḫarzatan, which could be compared to Hieroglyphic Luwian ha+ra/i-za- ‘to gather, collect’ (see s.v., similarly Yakubovich pers. comm. ‘to collect’).

Both solutions are possible formally. Contextually speaking the only helpful case is KBo 7.68 rev. iii 18’, where ḫar/ḫur/mur-za-ta-an is followed by a-ku-wa-at[-ta (Starke 1985b:363), which may represent the sentence initial clitic chain allowing the possibility that ḫar/ḫur/mur-za-ta-an is the verb of the previous sentence in the regular sentence final position, but this need not be the case.

Accordingly, none of the proposed meanings can currently be supported and the word must remain without assigned meaning until further evidence emerges.

[Zs.S.]

This lexeme is found in a Palaic festival text with Palaic incantations.

[D.S.]

The participial form was previously interpreted as a Hittite 3pl. pres. ind. act. However, due to the join made by D. Sasseville, it has now become clear that KBo 37.150 is a Palaic fragment.

[D.S.]

Checked against the photographs in the Konkordanz.

[D.S.]

The verb has been connected etymologically to either Hitt. ḫar(k)- ‘to hold’ or ḫark- ‘to destroy’ (Kammenhuber 1959a:63, Carruba 1970a:55). Due to the recent join made by Soysal 2017a:84, the connection with Hitt. ḫar(k)- is made much more probable for Pal. ḫark-. If Soysal (ibid.) is right about the lexical identification of (-)k]i^?-i-šar as ‘hand’, the sentence (-)]ki-i-šar i-iḫ-ḫa [ ḫ]ar-kán-du (KUB 35.159 ii 10’) should mean ‘they must hold the iḫḫa in (their) hand’. Therefore, Pal. (-)]ki-i-šar would be an endingless locative. This provides a satisfactory analysis of the sentence.

[D.S.]

HW (2nd ed.) H:280-297