*h1ég(h)-/h1g(h)

‘to drink’

Language
Proto-Anatolian
Class
Base
Grammar
verb
ID
396

1. Individual Anatolian Languages

1.1 Hieroglyphic Luwian

/u-/, u-, BIBERE

‘to drink’

1.1.1Transmission

This lexeme is only found in inscriptions of the 8th century BCE.

[D.S.]

1.1.2Forms

inf. /una/ “BIBERE”-na ÇİFTLİK, §16 (Tabal)
part. nom. sg. c. /uwammis/ u-wa/i-mi-i-sá KULULU 2, §3 (Tabal)

1.1.3Meaning

The meaning of the word -na was deduced by means of the logogram, which pictures a drinking person, and the context, i.e. (“EDERE”)á-ru-na -na=ha ‘to eat and to drink’ (Laroche 1960c:4, Starke 1990a:553, Hawkins 2000a:449). In the context, where the deceased claims that he died by (the grace of) Santa, we find the syntagma -tà-mi-i-sa u-wa/i-mi-i-sa (KULULU 2, §3). Morpurgo Davies (apud Hawkins 1980c:221) translates ‘having eaten (and) drunk’, which would refer to the funeral meal prior to his death (cf. Hawkins 2000a:488). Furthermore, she connects etymologically HLuw. /u-/ with Hitt. eku-/aku-(mi) ‘to drink’ (followed by Starke 1990a:553).

[D.S.]

1.1.4Derivatives

  1. /una-/

1.2 Cuneiform Luwian

ū-

‘to drink’

1.2.1Transmission

Hapax in a Luwian passage of a Hittite festival description (LNS).

[Zs.S.]

1.2.2Forms

2nd sg. pres. ú-ut-ti-iš KUB 35.133 ii 25’ (CTH 665, LNS)

Attestation following CLL:241 checked against the photograph of the Konkordanz. On the grammatical analysis see under Meaning. For a further attestation proposed by Hawkins 1980c:221 see s.v. išta-.

[Zs.S.]

1.2.3Meaning ‘to drink’

The word is attested in the following Luwian sentence (KUB 35.133 ii 24’-25’, transcription following Starke 1985b:280, slightly modified): DU-an a-wa<>za-aš-ti wa-at-ta-ne-i az-za-aš-ti-iš wa-a-šu ú-ut-ti-iš wa-a-šu (not continued). The last four words are rendered basically in two different ways leading to two different meanings of this word:

First, as ‘thou will eat well, thou will drink well’, leading to a verb u- ‘to drink’ (Morpurgo Davies 1980b:106 n. 24 [“may mean”], followed by CLL:241 and EDHIL:236 [with question mark]; similarly Hawkins 1980c:221 [it “may” mean “thou hast eaten well, thou hast drunk well”]; Bossert 1954a:102 wanted to connect it with libation due to the offering context).

Second, as “Die Speise (ist) das Gut / das Gute, der Trank (ist) das Gut / das Gute”, leading to a noun ‘drink’ (Starke 1981a:249, Starke 1990a:465, 565; also Ünal 2016a:974 [who, however, next to utti- ‘drink’ included an uttiš ‘?’ too]; cf. already Meriggi 1957a:218f. [‘Getränk’]; Laroche 1959a:103 [a noun in nom. sg.]). Tischler 2008a:213, Tischler 2016a:406 lists both possibilities.

From a formal point of view both translations are possible (Starke 1981a:250 n. 5 and Starke 1990a:465 n. 1695 rightly points out that the nom.-acc. sg. n. wāšu cannot be an adjective due to the lack of congruence, but it can not only be a noun, as he assumed, but also an adverb, cf. already Bossert 1954a:102). Starke 1981a:251 and Starke 1990a:465 calls attention to the parallel phrase šu?-wa-at-ra wa-a-šu in KBo 4.11 rev. 40 translated by him as ‘Die Fülle (ist) das Gut(e)’. Although this seems to argue for the correctness of his rendering of ú-ut-ti-iš, the parallel passage is otherwise completely obscure and thus his translation is not assured and cannot be used as an argument. Moreover, it also remains obscure what exactly the sense of “Die Speise (ist) das Gut / das Gute, der Trank (ist) das Gut / das Gute” is.

Although the exact meaning of wa-at-ta-ne-i remains debated (see the discussion s.v.), unlike its form (dat.-loc. sg. in congruence with zašti ‘this (very)’), it is clear that we are dealing with a vocative of the Weather God followed by a full sentence (introduced by a=wa), uttered by the priest, who prepared an offering with bread and sweet wine. This sentence must thus either contain at least one inflected verbal form or be a nominal sentence (cf. already Bossert 1954a:102). If this is a nominal sentence, we arrive at senseless sentences, regardless whether wāšu is an adverb (‘*O Weather God! For/in/on/at this very w. the food is well, the drink is well’) or a noun (‘*O Weather God! For/in/on/at this very w. the food is the goodness / goods, the drink is the goodness / goods’). However, if this is a non-nominal sentence, only az-za-aš-ti-iš and ú-ut-ti-iš can be verbs and they would stand in 2nd sg., which perfectly fits the circumstance that before this sentence we have an invocated Weather God and the obtained meaning already makes perfect sense (for this translation cf. already Morpurgo Davies 1980b:106 n. 24): ‘O Weather God! ! In/on/at this very w. thou will eat well, thou will drink well’.

[Zs.S.]

1.3 Palaic

aḫu-

‘to drink’

1.3.1Transmission

This lexeme is mostly found in one mythical passage of a an Old Hittite ritual (CTH 752) and once in the incantations of the festival for the Palaic pantheon (CTH 751).

[D.S.]

1.3.2Forms

3pl. pres. ind. act. a-ḫu-wa-a-an-ti KUB 32.18 i 9’ (CTH 752, OS or MS)
  a-ḫu-wa-an-ti KUB 35.168, 6’ (CTH 752, OS)
  a-ḫu-wa-an-ti KUB 32.18 i 7’ (CTH 752, OS or MS)
  a-ḫu-wa-an-ti KUB 32.18 i 18’ (CTH 752, OS or MS)
  a-˹ḫu˺[-wa-(a)-an-ti] KBo 19.159, 5’ (CTH 752, OS)
inf. a-ḫu-u-na KUB 35.165 rev. 22’ (CTH 751, OS or MS)
  a-ḫu-u-[ KUB 48.53 r.col. 13´ (CTH 753, NS)

The appurtenance of the last attestation to this lexeme is not certain due to the fragmentary state of the context.

[D.S.]

1.3.3Graphic Features

Checked against the Photographs of the Konkordanz (Mainzer Photoarchiv). The attestation a-˹ḫu˺[-wa-(a)-an-ti] (KBo 19.159, 5’) surely belongs here, as the fragment contains the same mythological passage as the other ones, i.e. KUB 32.168, KUB 32.18; cf. Carruba 1970a:49.

[D.S.]

1.3.4Meaning

Otten 1945-1951a:81f. suggests the meaning ‘to drink’ for Pal. aḫu- and brings contextual support for each passage: First, in KUB 35.165 rev. 22’, it is the action to be performed by Ziparwa with the filled cups. Then, he compares the sentence adānti ni=ppa=aš mušānti aḫuwānti ni=ppa=aš ḫašānti with a Hittite passage recurring in various myths, e.g. eder n=e ŪL išpiyer ekuiēr=ma n=e=za ŪL ḫaššikker ‘They ate, but they did not satiate, they drank, but they did not quench their thirst.’ (KUB 17.10+ i 19’-20’). Thus, he equates semantically and etymologically Pal. aḫu- with Hitt. eku-/aku-(mi) ‘to drink’ (followed by Kammenhuber 1959a:37, Carruba 1970a:49).

[D.S.]

1.4 Hittite

eku-/aku-(mi)

‘to drink’

1.4.1Literature / Comments

HW (2nd ed.) E:29-33, HED E, I:261-268, EDHIL:236f.

2. Reconstruction

2.1Reconstruction ‘to drink’

We equate here Hitt. eku-/aku-(mi) with Pal. aḫu- and Luw. /u-/ all with the meaning ‘to drink’; cf. HEG A-H:103-105, HED E,I:261-268, EDHIL:236f.Contra Puhvel (HED E, I:266), CLuw. a-ku-wa-at-ta is a particle chain, not a verbal form (see Oettinger 1979a:91 n. 17, CLL:1).

The exact reconstruction of the root implies a labio-velar, which develops phonologically to /gw/ in Hittite, to /hw/ in Palaic and to /w/ in Luwian. Based on this, Melchert 1994a:236 reconstructs PAnat. *g (Melchert 1993a:247), which is regularly reduced to /w/ in Luwian. For Palaic, he tentatively assumes a special development of intervocalic *g > /hw/. Considering the well-established Proto-Indo-European etymology *h1egu̯h- with its aspirated labio-velar, Melchert concedes that the Palaic sequence /hw/ could be the regular outcome of *gu̯h-, which would support a distinction between aspirates and non-aspirated plosives in Proto-Anatolian. However, he points out that this example alone is not sufficient evidence for concluding the retention of a Proto-Anatolian *gu̯h distinct from *g. Yakubovich 2000a:144 leaves it undecided.Regarding the initial vowel a- in the weak stem, e.g. Hitt. aku-, one can take it as a reflex of *h1 or as restitution analogical to roots in TeT, e.g. Hitt. šeš-/šaš-(mi) ‘to sleep’ (thus Melchert 1994a:66f. with further literature, differently Kloekhorst 2006a:79f., who assumes the sign renders a consonantal glottal stop containing a mute vowel.

Hittite eku-/aku-(mi) ‘to drink’ reflects an amphikinetic root verb with full grade in the singular and zero-grade in the plural (thus Oettinger 1979a:87-91, EDHIL:237). The Luwian and Palaic verbs are not attested in enough forms to assign with certainty a stem class. Therefore, their stem classes are assigned tentatively from the Hittite perspective.

[D.S.]

3. Proto-Indo-European

3.1Semantic Reconstruction

The basic meaning of PIE *h1egu̯h- is ‘to drink’, cf. Hitt. eku-/aku- ‘to drink’, TochAB yok- ‘to drink’.

3.2Morphological Reconstruction

Hitt. eku-/aku-, Pal. aḫu-, Luw. u-, and TochB preterite yāś ‘drank’ seem to reflect an ablauting root verb. In LIV2+, these forms are traced back to a PIE root-aorist *h1égu̯h-/*h1gu̯h-´Differently LIV²:231, where the Anatolian forms and TochAB yok- ‘to drink’ are derived from a reduplicated present stem *h1é-h1ogu̯h-/*h1é-h1gu̯h-. while the present stem TochAB yok- can be derived from a reduplicated present stem *h1é-h1ogu̯h-/*h1é-h1gu̯h-. The reconstruction of a root-final voiced aspirate stop gu̯h (instead of unaspirated g) is suggested by Gk. φ in νᾱ́/ήφω ‘to be sober’ and perhaps also by Pal. aḫu- if /hw/ is the regular outcome of *gu̯h (see PAnat. lemma above).

Kim 2000a, however, comes to the conclusion that a root *h1egu̯(h)- in Tocharian is not tenable for phonological reasons. To explain the attested forms of the different IE languages he proposes to reconstruct the root as *(h1)eh2gu̯h- (with a PIE Narten present *(h1)ḗh2gu̯h-/*(h1)éh2gu̯h-); cf. also LIV2+ s.v. *h1egu̯h- n. 1.

Yet a different approach was presented by Eichner 2015a, who reconstructs the root as *h1ek- ‘to drink (water)’. In this case, the Anatolian lenis root consonant (cf. k in Hitt. eku-/aku-) has to be explained by lenition. Lat. ēbrius ‘drunk’, sōbrius ‘sober’, according to Eichner, do not belong to that root and should be explained differently.As per Eichner, Lat. ēbrius would be based on an adjective “*eks-habro- ‘aus der Haltung (ohne Zurückhaltung, ungezügelt, ungehemmt)’”. Eichner, furthermore, proposes to connect the root *h1ek- with PIE *(H)akah2- ‘water’ (Lat. aqua ‘water’, Goth. aƕa ‘river’). However, the initial a-vowel of the word for ‘water’ seems hardly compatible with a root *h1ek-. This issue as well as the status of Gk. νᾱ́/ήφω ‘to be sober’, whose φ contradicts the reconstruction of a root *h1ek-, require further clarification.

[T.S.]

4. Transmission

Gk. νήφω, Dor. νᾱ́φω ‘to be sober’ (Thgn., Archil. +)The present stem νήφε/o- is likely to be a backformation to the nominal stem νήφοντ- ‘sober’ (← νήφον- < *ń̥-h1gu̯h-(o)n-), cf. GEW II:318f., Weiss 1994bLIV2+ s.v. *h1egu̯h-.

Lat. adj. ēbrius ‘drunk(en)’ (Naev.+), sōbrius ‘sober’ (Pl.+)Cf., e.g., EDLIL:184f., 568f.

TochB yokäṃ ‘drinks’, TochA yoktsi ‘to drink’Ultimately perhaps from reduplicated *h1e-h1gu̯h-, cf. Hilmarsson 1986a:51f., 202, LIV²:231 n. 3, LIV2+ s.v. *h1egu̯h- n. 3.

Cross-references

Referred to from this article