*h2ons-o-
‘offspring’
- Language
- Proto-Anatolian
- Grammar
- subst, c.
- ID
- 3436
1. Individual Anatolian Languages
1.1 Hieroglyphic Luwian
/hams(i)-/, (“INFANS.NEPOS”)ha-ma-si-
‘grandson’1.1.1Transmission
With 28 occurrences, this lexeme is one of the more well-attested in the HLuw. corpus. The attestations cover a large temporal scope with the earliest relevant inscriptions dating to around 1200 BCE and the youngest dating to the 8th century BCE, stretching across nearly the entire HLuw. era. Their geographical distribution is less broad in comparison, but still considerable, and locations range from Malatya in the north-east to Hama in the far south.
[AH.B.]
1.1.2Forms
In BOYBEYPINARI 1, §11, the signs on the block suggest that the reading could also be 〈DOMUS-ni-NEPOS.MI-i-ni-sá〉, cf. Hawkins 2000a:336, 338.
The reading of 〈INFANS.NEPOS-si?-sa?〉 in KARKAMIŠ A27u, l. 2 is uncertain, because the two syllabic signs are barely visible on the stone; Hawkins 2000a:165 transliterates only 〈INFANS.NEPOS〉.
In ÇINEKÖY, §1, the reading should be 〈INFANS.NEPOS-si-sá〉 instead of 〈-sà〉, a typographical error published in the editio princeps; the sign 〈-sá〉 is clearly visible on both the photo and the copy, see Tekoğlu & Lemaire 2000a:969, 983.
[AH.B.]
1.1.3Graphic Features
1.1.3.1Phonological Interpretation
The phonological interpretation of this lexeme is quite clear despite the imprecision of the HLuw. writing system. An exact cognate, CLuw. ḫamš(i)- ‘grandchild’ is spelt unambiguously (cf. CLL:50), and possibly further members of the word family around Hittite ḫašš- ‘to give birth’ (cf. HED H:68f.) lend additional support to the interpretation, although the phonological intricacies with linking /hams°/ to /hass°/ have been pointed out a number of times, see e.g. Melchert 1988b:212f. and Kloekhorst in EDHIL:319-321 and cf. the discussion of the stem below.
The spelling variations obvious in the HLuw. data cover the full range between fully syllabic attestations with or without logographic determinatives and those represented only by logogram(s). The syllabic renditions show very little variation, but the logogram NEPOS comes in two variants: one consists of sign *300 by itself, an elaborate volute-style shape, while the other variant adds sign *488 〈tí〉 underneath or next to it; there is no discernible difference in meaning between the two variants.
[AH.B.]
1.1.4Meaning
The meaning assigned to this word is certain and already listed by Laroche (DLL:39) with reference to Laroche 1958b:188f., who compares it to Hittite ḫanzašša- ‘grandson, great-grandson’. Semantically most instructive is the long introductory clause at the beginning of MARAŞ 1, §1a-i, which uses a number of kinship terms in linear order, and the present lexeme occurs in the slot between /nimuwizza-/ ‘son’ and /hamsukkalla-/ ‘great-grandson’, followed by further descendants such as /nawa-/ ‘great-great-grandson’ etc. More recently, this interpretation has received partial confirmation from another language in ÇINEKÖY, §1, where /hams(i)-/ corresponds to Phoenician ʾšpḥ ‘clan, descendants’ (Tekoğlu & Lemaire 2000a:975f.).
[AH.B.]
1.1.5Stem
As is evident from the different forms attested, the word is a common gender noun and undergoes i-mutation. The communis opinio on the stem is divided, however. Some scholars see the word in some way connected to the HLuw. verb /hassa-/ ‘to give birth’, which has a cognate in Hittite, cf. e.g. Starke 1990a:483 n. 1768, Puhvel in HED H:68f., and Blažek 2003-2004a:5. Other scholars disagree on both semantic and phonological grounds, see e.g. Tischler in HEG A-H:141f., Melchert 1988b:212f. and Hajnal 1995a:235 n. 307, who connect it to Lyc. xahba- ‘descendant’. Carruba 1991a:163f. refuses either of the suggested connections and instead proposes a possible link with /hanta/ ‘before, in front of’, which seems far-fetched.
[AH.B.]
1.1.6Compounds
- DOMUS-ni-NEPOS.MI-i-sá /parni-hamsis/ (BOYBEYPINARI 1, §11)
1.1.7Derivatives
1.2 Cuneiform Luwian
ḫamš(i)-
‘grandchild’1.2.1Transmission
All attestations of ḫamš(i)- known so far come from rituals from the Kuwattalla tradition (from MS onwards).
[E.R.]
1.2.2Forms
| abl./instr. | ⸢ḫa-am⸣-ša-a-ti | KBo 9.143 iii! 13’ | (CTH 763, MS) |
| [ḫa-am-ša-a-t]i | KUB 32.81 i? 14’ | (CTH 761, MS?) | |
| ḫa-am-ša-a-ti | KBo 29.6 obv. 29’ | (CTH 762, NS) |
1.2.3Graphic Features
Checked against the photographs in the Mainzer Fotoarchiv.
[E.R.]
1.2.4Meaning
The lexeme always occurs together with ḫamšukkalla- (on paired kinship terms, see Martínez Rodríguez 2019b). It is found in lists of blessings for the ritual patron (thus in wishes for progeny), and in lists of relatives of the ritual patron who are intended as co-beneficiaries and also to participate in the blessings of their ancestor, e.g.: (28’)... ma-al-ḫa-aš-ša-aš-ši-iš [en-aš] (29’)dumuMEŠ-ti dumu!.munusMEŠ-ti ḫa-am-ša-ti ḫa-am[-šu-uk-kal-la-a-ti] “[the lord] of the ritual with (his) sons, daughters, grandchildren, and great-gra[ndchildren]” (KBo 29.6 Vs. 28’f., modified from Martínez Rodríguez 2019b:184). Laroche’s (Laroche 1958b:188f.) semantic elucidation is based on the contextual analysis of those lists and the comparison with the HLuw. cognates /hams(i)-/ and /hamsukkalla-/. On the assumed match with Hitt. ḫāšša- c. (paired with ḫanzašša-), which also played a role in the recovery of the meaning, see under Reconstruction. The suggested semantic interpretation found general acceptance (DLL:39, HW:424, Oettinger 1979a:439, HEG A-K:196, CLL:50, Melchert 1994a:164f., HED H:68f. and 227, HHw:37, EDHIL:319f., Martínez Rodríguez 2019b:184; for earlier misinterpretations based on a false understanding of Laroche’s suggestion ‘petit-fils’ as ‘small child’, see HW 2nd ed. Ḫ:120).
[E.R.]
1.2.5Stem
Although only an oblique case form is preserved in Cuneiform Luwian, the Hieroglyphic Luwian evidence (see under Lemma) attests to the presence of i-mutation (CLL:50).
[E.R.]
1.3 Luwian in Hittite transmission
ḫammaš(i)-
‘grandchild’1.3.1Transmission
The sole attestation of the word is found in a list of women belonging to the palatial administration.
[E.R.]
1.3.2Forms
| gen. sg. | ḫa-am-ma-ša-aš | KBo 10.10 iv 9 | (CTH 235, NS) |
1.3.3Graphic Features
Checked against the photographs in the Mainzer Fotoarchiv.
[E.R.]
1.3.4Meaning
The word qualifies a palace to which a group of female workers is assigned (ša é.gal ḫa-am-ma-ša-aš). In view of the é(.gal) ḫuḫḫaš, which he translates as ‘house/palace of the ancestors’ (lit. ‘of the grandfather(s)’), and the existence of ḫamš(i)- ‘grandchild’ in Luwian, Laroche 1962a:29 interprets é.gal ḫammašaš as the ‘palace of the grandchildren’ (accepted in HW:424 with a misunderstanding of French ‘petit-fils’, Berman 1972a:38, HEG A-K:196, HED H:68f., Ünal 2007a:156). Differently, del Monte & Tischler 1978a:183 see in ḫammašaš a personal name, hence ‘palace of Ḫammaša’ (followed in HW 2nd ed. Ḫ:120; undecided Weitenberg 1984a:437 n. 384; no translation in HHw:39 ). However, the expected determinative indicating the name is missing, and there is no reason not to take the syntagm at face value.
[E.R.]
1.3.5Stem
The use of the Hittite gen. sg. in -aš is fully compatible with the assumption that the word is a borrowing of the Luwian i-mutated stem ḫamš(i)- (see under Origin).
[E.R.]
1.3.6Origin
Those scholars who accept Laroche’s semantic interpretation as ‘grandchild’ follow him also regarding the Luwian origin on account of the cluster -mš- (for references, see under Meaning).
[E.R.]
1.4 Hittite
ḫāšša-
‘offspring’1.4.1Literature / Comments
Melchert 1973a “offspring of the first generation, child”; cf. HW 2nd ed. Ḫ:397-409 “grandchild” with further literature.
2. Reconstruction
2.1Reconstruction ‘offspring’
On the etymological relationship between Hitt. ḫāšša- c., ḫanzāšša- c. ‘offspring (of the second generation)’ and ḫāš-/ḫašš-(ḫi) ‘to beget’, see Melchert 1973a:64f. with further literature, which has been generally accepted (see Weitenberg 1984a:159, HED H:227, HEG H:191f., EDHIL:323f., contra Melchert 1994a:165). As pointed out by Kloekhorst (EDHIL:323f., see already Melchert 1988b:212-214), the different phonological outcome of the sequence *-VNs- points towards different ablaut grades and/or positions of the accent. Since the root verb of the ḫi-conjugation tells us that *-óns- become -š(š)-, the preform of CLuw. ḫamš(i)- and Hitt. ḫāšša- can be *h2onso-, but with a different position of the accent, i.e. *h2onsó-; CLuw. ḫamš(i)- as a τομός-type and *h2ónso- (with dissimilation of *-ns-´ > -ms-); Hitt. ḫāšša- as a τόμος-type (cf. Oettinger 1979a:439, Starke 1990a:483 n. 1768, Melchert 1994a:164f., 236, 258, Oettinger 2002a:98 “h2óns-o-”; cf. Weitenberg 1984a:159-161 and Melchert 1988b:212-214 “h2m̥só-?”). Considering that Hitt. ḫanzāšša- (< *h2n̥s-ósi̯o-, with secondary t-epenthesis; see Melchert 1988b:212-214, Nikolaev 2021c:396f.) betrays the pre-tonic value of -nz-´, a proterokinetic root noun that was later thematized in Hittite would be formally possible for the Proto-Anatolian word, i.e. *h2éns-/h2n̥s-´ (thus EDHIL:324), but see now for a criticism the PIE lemma under the lemma head; on the sound laws concerning nasal + s that we have adopted here, see also PAnat. *h2éms-u-.
For the often-cited Lydian cognate, see now Lyd. esa- for an alternative suggestion.
[D.S.]
For the lemma head see PAnat. *h2óns-/h2n̥s-´.

