*u̯eth2-r̥/n-
‘word’
- Language
- Proto-Anatolian
- Grammar
- subst
- ID
- 2521
1. Individual Anatolian Languages
1.1 Cuneiform Luwian
ūtar/n-
‘word, spell’1.1.1Transmission
The noun is broadly attested in the Luwian corpus, occurring both in the Puriyanni and Kuwattalla rituals. Moreover, it also occurs in incantations against the Illness of the Child and various fragments (from MS onwards).
[E.R.]
1.1.2Forms
| nom./acc. sg. | ú-tar | KUB 35.101 obv.? 7’ | (CTH 765, MS) |
| ú-tar-š[a] | KUB 35.88 iii 1’ | (CTH 766, NS) | |
| ú-tar-ša | KUB 35.54 ii 13’, 38’ | (CTH 758, MS) | |
| ú[-tar-ša] | KUB 35.54 iii 22’ | (CTH 758, MS) | |
| ˹ú˺-tar-ša | KUB 35.54 iii 38’ | (CTH 758, MS) | |
| ˹ú˺-[tar-ša] | KUB 35.54 ii 4’ | (CTH 758, MS) | |
| ú-tar-˹ša˺ | KUB 35.55, 9’ | (CTH 758, NS) | |
| ú-˹tar˺-ša | KBo 30.190 iii 7’ | (CTH 761, NS) | |
| ú-˹tar˺-ša | KBo 22.137 iii 9’ | (CTH 758, NS) | |
| ú-ta-a[r-ša] | KUB 35.89, 2’ | (CTH 766, MS) | |
| nom./acc. pl. | ú-ut-ra | KUB 35.65 iii 9’ | (CTH 763, MS) |
| gen. adj. acc. sg. c. | [ú-u]t-˹na˺-aš-˹ši˺-in | KBo 9.147, 2’ | (CTH 761, MS) |
| gen. adj. acc. pl. c. | ú-ut-na-aš-ši-in-za | KUB 35.45 ii 19’ | (CTH 760, NS) |
| ú-ut-na-aš-ši-in-zi | KUB 35.48 ii 12’ | (CTH 760, NS) |
1.1.3Graphic Features
Checked against the photographs in the Mainzer Fotoarchiv.
[E.R.]
1.1.3.1Phonological Interpretation
Simon 2010b, who interprets the scriptio plena in CLuw. as both a marker of a glottal stop and a long vowel, tentatively adduces ūtar/n- as a possible example for a scriptio plena as an indicator for this phenomenon. – Any formal correspondence of ūtar/n- with Hitt. uddar/n- must account for the single vs. geminate writing of the dental in CLuw. and Hitt. respectively.
[E.R.]
1.1.4Meaning
In the Puriyanni ritual (KUB 35.54), ūtar/n- is repeated several times in the expression ādduwalza ūtarša, which has been interpreted as the equivalent of the Hitt. expression idālu uddar ‘evil word’ (DLL:153). For this reason, Laroche (DLL:103) gives the meaning ‘chose?’ (Eng. thing), tentatively comparing ūtar/n- in form and meaning with Hitt. uddar-/uddan- ‘word, spell, speech’. This interpretation is commonly accepted today (CLL:247, HHw:212, HEG U:137, EDHIL:932–933, Rieken 1999a:300). However, Starke 1990a:565f. has argued in favor of the meaning ‘water’ based on the context of the Puriyanni ritual, interpreting the expression ādduu̯alza ūtarša as ‘salted water’. In particular, he points out that ūtar/n- functions as the direct object of CLuw. laḫun(a)i- ‘to pour out’ in KUB 35.54 iii 38. The occurrences in KUB 35.88 iii 1’ and KBo 30.190 iii 7’ apparently also support the interpretation of the word as ‘water’ (ibid. 566–567). Rieken 1999a:300 n. 1438 calls attention to the fact, contra Starke 1990a, that the word for ‘water’ used in the same ritual is u̯ārša (KUB 35.54 iii 19).
Outdated interpretation ‘belonging to the land’ by Rosenkranz 1952a:71, followed by Otten 1953a:106 n. 217.
[E.R.]
1.1.5Stem
Both r- and n-stem attested, the available evidence points towards an old heteroclitic stem.
[E.R.]
This lemma was written together with the intern Elia Weber.
1.2 Hittite
uttar/n-
‘word, speech’1.2.1Literature / Comments
HEG U:131-141
2. Reconstruction
2.1Reconstruction ‘word’
While it is clear that a heterocliticon can be reconstructed for Proto-Anatolian, the shape of the root of Hitt. uttar/n- and CLuw. ūdar/n- ‘word’ is highly debated. Melchert 1994a:50, 233, 242, 249, 265, 267 reconstructs a stem *éu̯tr̥-/*utn-´ to a root *eu̯t-, which does not have any correspondences outside the Anatolian branch, whereas Rieken 1999a:301 slightly modifies the root and proposes the reconstruction of *h1eu̯- as a basis for a proterodynamic stem *h1eu-tr̥-/*h1u-tén-s. Another etymology, which is based on Greek αὐδή ‘voice’ and Sanskrit vádati, uditá- < *h2u̯edH1/3- (LIV²:286), has also been suggested. On the other hand, Kloekhorst (EDHIL:933) adduces the connection with the Indo-European root *u̯eth2- ‘to speak’. New evidence vindicating the latter root has been shown by Sasseville forthcomingPaper presented at Arbeitstagung der IG, Ljubljana, June 4th–7th, 2019, who adduces Pal. watḫala- ‘chanter’, which demonstrates the presence of a root-final *h2 (cf. Sasseville 2025b, who opts for a root-final *h3) Taking all evidence together including Hitt. watarnaḫḫ-(i) ‘to report’ (cf. Eichner 1980a:146 n. 69., Kimball 1999a:381, Rieken 1999a:301f., EDHIL:988f), he posits the root *u̯eth2-, whose root-final laryngeal however is dropped without a trace between unaccented syllables; a PIE rule that has already been suggested based on the Indo-Iranian data (Schmidt 1973, Peters apud Mayrhofer 1981a:436). This rule helps explain the singleton -t- in CLuw. ūtar/n- and Hitt. watarnaḫḫ-(i).
Therefore, a heterocliticon of the shape *u̯eth2-r̥/n- ‘word’ is here reconstructed. On the ablaut, see Rieken 1999a:299-302 and Steer 2015a:272-274. For a summary of the literature, see HEG U:138f.
[D.S.]
For a possible root verb built on the same root, see PAnat. *u̯óth2-/uth2-´.