*HEs-(K)-
‘to quench one’s thirst’
- Language
- Proto-Anatolian
- Class
- Base
- Grammar
- verb
- ID
- 334
1. Individual Anatolian Languages
1.2 Cuneiform Luwian
*ḫaš-
‘to quench one’s thirst’1.2.1Derivatives
1.3 Lycian B
*xes-
‘to quench one’s thirst’1.3.1Derivatives
1.4 Palaic
ḫaš-
‘to satiate, quench one’s thirst’1.4.1Transmission
This lexeme is attested in a Palaic text of mythological nature and also once in a Palaic incantation with bread offerings.
[D.S.]
1.4.2Forms
| 2sg. impv. act. | ˹ḫa-ši˺ | KUB 35.165 obv. 1 | (CTH 751, OS or MS) |
| 3pl. pres. ind. act. | ḫa-ša-a-an-ti | KUB 32.18 i 9’ | (CTH 751, MS) |
| ḫa-ša-a-an-˹ti˺ | KUB 32.18 i 18’ | (CTH 751, MS) | |
| ḫa-ša-an-ti | KUB 32.18 i 7’ | (CTH 751, MS) |
See also the probable 2pl. impv. act. ḫ[a]-a-ši-ta-a[n?] on KBo 62.24, 4´.
[D.S.]
1.4.3Graphic Features
Checked against the photographs of the Konkordanz. The word ˹ḫa-ši˺ (KUB 35.165 obv. 1) is missing completely from the transliteration made by Carruba 1970a:14, Carruba 1972a:12. Although Carruba 1970a:14 recognizes traces of signs, he assumes simply a case of erasure, which is not the case. The new readings presented here, i.e. ˹ḫa-ši˺, were given to us generously by the project HFR via Suzanne Görke.
[D.S.]
1.4.4Meaning ‘to satiate, quench one’s thirst’
Otten 1945-1951a:81 compares the Palaic passage “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 suggests that Pal. ḫaš- corresponds semantically speaking to Hitt. ḫaššikk-(mi) ‘to satiate (one’s thirst)’ (followed by Kammenhuber 1955b:355f. n. 1, Kammenhuber 1959a:46f.; Carruba 1970a:55; Tischler 2008a:51). Outdated Bossert 1944a:88.
The instance of ḫaši is interpreted tentatively as 2nd person singular imperative based on the context. The god Ziparwa is summoned to come and eat bread, which has been prepared for him. Therefore, it makes sense for the context, if the speaker invites the deity to satiate himself or (in other words) to rejoice at the offering: nū Ziparwa … ḫaši ḫānda ´Now Ziparwa… satiate! They (scil. the bread) are warm.´Asyndetic relationships between verbs are not uncommon in Hittite incantations; cf. KBo 20.107+ iv 21’-23’.
[D.S.]
1.4.5Stem
Based on the form of the 3rd plural present alone, one cannot properly identify the stem of the verb ḫaš-. It could be a root verb or a ḫi-conjugated verb. With the 2nd singular imperative, one can exclude a root verb, because the form would be endingless; cf. Pal. āš! ‘be!’. Thus, one can tentatively assign a ḫi-verb. For the Palaic verbal stem classes see Oettinger 1979a:558-560.
[D.S.]
1.4.6Derivatives
1.4.6.1Uncertain Derivatives
1.5 Hittite
ḫaššikk-(mi)
‘to satiate (one’s thirst) (also metaphorically)’1.5.1Literature / Comments
HW (2nd ed.):421-423, EDHIL:324f.
2. Reconstruction
2.1Reconstruction ‘to quench one’s thirst’
Based on Hittite ḫaššikk-(mi) and the Palaic verb ḫaš-, it is difficult to provide any certainty regarding the Proto-Anatolian stem formation of this root; cf. HEG A-K:200, HED H:231f., EDHIL:324f. Hittite ḫaššikk-(mi) is a consonantal stem, which contains a k-enlargement. The earlier proposals that it goes back to a formation in *-sk̂e/o- (Otten 1945-1951a:81) have been rejected because of its consonantal inflection (see Eichner 1970a:11, Oettinger 1979a:193f.).
Based on Palaic ḫaš-, an athematic present or aorist is tentatively reconstructed in LIV2:273, which is not probable in light of the new imperative form. A verb of the ḫi-conjugation is more likely; see the discussion under Palaic, Stem.
[D.S.]
3. Proto-Indo-European
3.1Semantic Reconstruction
No non-Anatolian cognates of this root exist, cf. below.
3.1.1Root etymology
According to Klingenschmitt 1994a:242f., the root underlying Pal. ḫaš- ‘to satiate (one’s hunger or thirst)’ etc. may also be attested in Lat. inānis ‘empty (also of the stomach)’. Lat. inānis ‘empty’ would then go back to ‘not satiated, hungry’. While such a semantic development is reasonable, this connection is at least phonologically doubtful (cf. the comment on Lat. inānis above).
The meaning of PIE *h2/3es- (cf. below) is reconstructable as ‘to satiate (one’s hunger or thirst)’.
3.2Morphological Reconstruction
No non-Anatolian cognates of this root exist, cf. below.
3.2.1Root etymology
Pal. ḫaš- ‘to satiate (one’s hunger or thirst)’ etc. shows an anlauting laryngeal which could be either *h2- or *h3-. Lat. inānis ‘empty’ would presuppose *h2- (if < *n̥‑h2es‑ni-), but its affiliation with the root in question is unlikely on phonological grounds (cf. the comment on Lat. inānis above). Thus, the reconstruction of a PIE root is not advisable. On the basis of the available material, the phonological shape of the root may nevertheless be reconstructed as PIE *h2/3es-.LIV²:273 and Klingenschmitt 1994a:242 unanimously reconstruct PIE *h2es- with the second laryngeal.
This root could be further connected with PIE *seh2- ‘to become sated’ (LIV²:520f.). A root extension in *‑eh2‑ (cf., e.g., PAnat. *dmaH-s- ‘to (op)press’) and dissimilatory loss of the anlauting *h2-, viz. *h2es- → *h2s‑eh2- > *seh2‑, is proposed, e.g., by Eichner 2010a:40 n. 3, cf. Ragot 2009a, esp. 235f. following Palmer 1959a:132f.
[A.O.]
4. Transmission
No direct cognates attested.
(?)Latin
Lat. adj. inānis ‘empty (also of the stomach)’ (Naev.+), sbst. f. ināniae (pl.) ‘nothingness’ (Plaut.+)Klingenschmitt 1994a:242f. proposes the preform PIE *n̥‑h2es‑ni- ‘without saturation’ (> *enazni- > *enāni-). However, chronological considerations raise doubts about this etymology: Klingenschmitt (ibid. p. 243 n. 16) assumes that the sound law *-Vzn- > *-V̄n- happened before the weakening of *-a- in medial syllables as allegedly shown by Lat. arāneus ‘spider’ < *araznei̯o- (< *araksnei̯o‑) and profānus ‘outside of the temple’ < *profazno‑. However, profānus is a recent formation because internal -f- cannot be old in Latin. Furthermore, Lat. arāneus, on the other hand, is not necessarily inherited, cf. EDLIL:49f. Besides, there are two good counter-examples substantiating that in Latin weakening of *‑a- to *‑e- predates compensatory lengthening (after loss of preconsonantal *‑z- [and presigmatic *‑n‑]). Lat. anhēlus ‘gasping’ (cf. also an(h)ēlāre ‘to breath hard’) goes back to *ana(n)‑sl- (PIE *h2enh1- ‘to breath’, LIV²:267f.). Lat. ascēnsus ‘ascent’ is reconstructed as *ad‑skansso- (PIE 1.*skend- ‘to leap, jump (forth)’, LIV²:554). Both, Lat. anhēlus and ascēnsus prove that the weakening of *‑a- to *‑e- in closed syllables occured before *‑Vns- > *‑V̄s- and *‑VzD- > *‑V̄D- (including *‑Vzn-), cf. Weiss 2011a:120, 177 n. 11, 193, Opfermann 2016a:127–129. Therefore, Lat. inānis cannot continue *n̥h2esni-.
WH I:688f., EM:314, EDLIL:301, OLD²:945f.