*peh3-s-
‘to swallow’
- Language
- Proto-Anatolian
- Grammar
- verb
- ID
- 3662
1. Individual Anatolian Languages
1.1 Hieroglyphic Luwian
/pass-(ti)/, (EDERE)pa-sà-
‘to swallow’1.1.1Transmission
This lexeme is only attested once in the Hieroglyphic Luwian corpus in a new inscription (TELL AHMAR 6). This inscription was published by Hawkins in 2006 and dated to the reign of Hamiyatas (late 10th early 9th BCE), see Hawkins 2006a:11.
[D.S.]
1.1.2Forms
| 3sg. impv. act. | /pastu/ | (EDERE)pa-sà-tú | TELL AHMAR 6 §32 (Tell Ahmar) |
1.1.3Graphic Features
As Hawkins 2006a:30 points out, the logogram in this attestation cannot be distinguished between *7 (EDERE) and *8 (BIBERE). The choice is made here based on the context. The direct object of the verb is the head of a person, which implies the action of eating rather than drinking.
[D.S.] [I.Y.]
1.1.4Meaning
Based on the representation of the logogram and the phonemic resemblance with CLuw. pašš-(ti) ‘to swallow’, Hawkins 2006a:30 assigns to HLuw. /pass-/ the meaning ‘to swallow’. This verb occurs within a curse formula, where the deity in the shape of a lion is the subject and the head of the offender is the direct object. The verb is accompanied by the preverb INFRA-ta ‘down’, whence the more precise meaning ‘to swallow down’.
[D.S.]
1.1.5Stem
The stem cannot be recovered based on this single attestation. However, by assuming the same stem as the one of CLuw. pašš-(ti), we posit a root verb of the mi-conjugation, i.e. /pass-(ti)/. It is worth noting the noun /patta-/, which was built to the root */pa-/ without the “s-extension”; cf. also HLuw. /pazza-/.
[D.S.]
1.2 Cuneiform Luwian
pašš-(ti)
‘to swallow’1.2.1Transmission
The only Luwian context in which this lemma is attested is in a festival fragment mentioning the ašušatalla-men.
[O.B.]
1.2.2Forms
| 3sg. pret. ind. act. | pa-aš-ta | KUB 35.133 ii 5´ | (CTH 665, LNS) |
1.2.3Graphic Features
The form has been checked against the photograph in the Konkordanz (Mainzer Fotoarchiv).
[O.B.]
1.2.4Meaning
The sole of occurrence of Luw. pašš-(ti) in a Luwian context is fragmentary (the line reads: za-an-da pa-aš-ta DINGIRMEŠ ḪU[L-). Its meaning is therefore inferred mainly on the basis of comparison with the corresponding Hieroglyphic attestation (see above) and Hitt. pāš-(ḫi) ‘to swallow’ (cf. Oettinger 1978a:435, Melchert 1987b:26, Starke 1990a:495, CLL:172, Hajnal 2000a:165, HEG P:514, EDHIL:649, HED PA:182, Dempsey 2015a:243, Vernet 2017a:65, Sasseville 2021a:269f.).
[O.B.]
1.2.5Stem
CLuw. pašš-(ti) is an non-leniting root verb (Starke 1990a:497, Sasseville 2021a:269f., contra Vernet 2017a:65). The geminate -šš- is not visible in the 3sg. pret. ind. act. form on account of the inability of the cuneiform script to render three consecutive consonants, but it is evidenced by the inf. form ⟨pa-aš-šu-u-na⟩ (see the corresponding Hittite transmission lemma).
[O.B.]
1.2.6Derivatives
1.3 Luwian (Glossenkeilwörter)
pašš-(ti)
1.3.1Transmission
All forms of Luw. pašš-(ti) preceded by a Glossenkeil are attested in a snake oracle, whereas the one without is attested in a ritual for the goddess Ištar.
[O.B.] [D.S.]
1.3.2Forms
| 3sg. pret. ind. act. | KUB 49.2 i 14´ | (CTH 575, LNS) | |
| KUB 49.2 i 18´ | (CTH 575, LNS) | ||
| inf. | pa-aš-šu-u-na | KUB 24.7 iii 31 | (CTH 717, NS) |
The form
pa-aš-k[e-ez-zi] on KUB 8.67, 19´ is Hittite (thus Laroche 1959a:80, Starke 1990a:497, CLL:172), because there is no Luwian verbal suffix of the shape -ške/a-.
[O.B.] [D.S.]
1.3.3Graphic Features
All forms have been checked against the photographs in the Konkordanz (Mainzer Fotoarchiv).
[O.B.]
1.3.4Meaning
Friedrich 1947b:18 compares the phrase GAM paššūna, which he recognizes as a Luwian infinitive, with Hitt. katta pašš- ‘to swallow down’ and, thus, assigns the same meaning to the word, which he considers contextually appropriate despite the fragmentary state of the passage (followed by Laroche 1959a:80). The same meaning is also attributed to the Glossenkeilwort
pašta (see van den Hout 1998b:116f., HEG P:514). For more literature on the Luwian lemma and its translation, see CLuw. pašš-(ti) above.
[O.B.] [D.S.]
1.3.5Stem
See under the corresponding Cuneiform Luwian lemma.
[O.B.] [D.S.]
1.3.6Origin
The Luwian origin of the form paššūna is confirmed by the Luwian infinitive suffix -una, whereas pašta is attributed here due to the gloss wedges (see Friedrich 1947b:18, Starke 1990a:497, HEG P:514, CHD P:203).
[O.B.] [D.S.]
1.4 Hittite
pašš-(ḫi)
‘to swallow’1.4.1Literature / Comments
CHD P:203f.
2. Reconstruction
2.1Reconstruction ‘to swallow’
The Hittite verb pašš-(ḫi) ‘to swallow’ (and pašš-(mi))On the double conjugation of the Hittite verb, see Sasseville 2021a:11, 269f. and cf. Oettinger 1979a:435f. and by extension CLuw. pašš-(ti) ‘to swallow’ (and redupl.
pappaša-(i)) have long been connected to PIE *peh3-s- (HEG P:515, EDHIL:649, contra HED PA:183f., all with further literature; see also Oettinger 1979a:435, Sasseville 2021a:546). This verb has also been used to exemplify the different treatment of PIE *h3 as opposed to *h2 beside *-s-; cf. *peh3s- > Hitt. pašš‑ vs. *peh2s- > Hitt. paḫš-. Finally, the bare root is found in HLuw. /patta-/ ‘drink’, which goes back to a formation in *-to-, as well as in HLuw. /pazza-(i)/ going back to a formation in *-sk̑é/ó- (although it is not possible to tell whether the root in the latter case was extended with *-s- or not).
[D.S.]
3. Proto-Indo-European
3.1Semantic Reconstruction
The existence of a root aorist *péh3-t ‘drank’ (Vedic ápāt, Greek Aeol. impv. πῶ(θι)) suggests that the original meaning of the root *peh3- was not ‘drink’, but rather ‘drink up’, ‘take a sip’ (thus García Ramón 1994a:54; Werba 1997a:303) or ‘swallow’. The latter seems to be reflected in the Anatolian verbal formations (Hittite pāš-(ḫi) ‘to swallow’, Luwian pašš-(ti) ‘to swallow’), though it is not unthinkable that the s-extension (of whatever origin; see below and Søborg 2020a:623–625 for a discussion) had a telicizing function (as argued by Friis apud Søborg loc.cit.). However, if we start from a basic meaning ‘take a sip’ or ‘swallow’, the meaning ‘drink’ would have developed predictably in the reduplicated present *pi-ph3-e-ti > *pi-bh3-e-ti ‘swallow repeatedly (vel sim.)’ (see Praust 1996a:185 note 2; on the voicing assimilation caused by *h3 see Mayrhofer 1986a:143f.). For a semantic parallel compare Lithuanian gérti ‘to drink’, Latvian dzer̂t ‘to drink’ from *gu̯erh3- ‘to devour’ (vs. Old Church Slavonic po-žrěti, -žьrǫ ‘to devour, glut’).
3.2Morphological Reconstruction
The variants *peh3- (cf. the root aorist *péh3-t ‘drank’, Hittite pāš-(ḫi) ‘to swallow’, Luwian pašš-(ti) ‘to swallow’, Aeol. and Dor. πώνω ‘drink’, Greek πῶμα n. ‘a drink’, Vedic °-pā́- ‘drinking’, pā́tar-, pātár- m. ‘drinker’, pā́na- n. ‘drinking, a drink’, pā́tra- n. ‘drinking vessel’, Latin pōtus ‘drunk’, pōculum ‘drinking cup’, pōsca f. ‘acid drink of water and vinegar’, Old Irish án ā, f. ‘cup, drinking vessel’, Lithuanian puotà f. ‘feast’), *pih3- (cf. Vedic pītá- ‘drunk’, pītí- f. ‘drinking, a drink’, Greek πῑ́νω, impv. πῖθι, Old Church Slavonic pijǫ (piti) ‘drink’, Old Albanian pi ‘drinks’, përpín ‘devour, swallow’), and *peh3i̯- or *pei̯h3- (cf. Vedic pāy-áyati ‘makes drink’, Old Prussian (impv. 2pl.) poieiti ‘drink!’, Old Church Slavonic pojǫ, pojiti ‘give to drink’) have long puzzled scholars. The currently most popular explanations for these diverse root allomorphs depart from a pre-PIE i-present *péh3-i-, *ph3-i- (cf. Mayrhofer 1986a:174f.; Jasanoff 2003a:117 n. 262; Praust 1996a:184 note 1, with reference to Schindler) or *ph3-(e)i- (cf. EDG:1194f. s.v. πίνω; Lipp 2009a II:205f. “aus einer vor-uridg. intransitiven Aktionsartbildung […] hervorgegangene[n] Wurzelerweiterung”; LIV²:463 s.v. *peh3(i̯)- note 0). The zero-grade allomorph of this stem, viz. *ph3i-, was susceptible to regular laryngeal metathesis before a following consonant, leading to *pih3- (cf. Mayrhofer 1986a:175). The expected distribution of the resulting stem allomorphs (aor. *peh3- vs. pres. *peh3i-/*pih3-) was probably already blurred in the proto-language and *peh3i-/*pih3- could be used as an allomorph of the root. The spread of *pih3- to the aorist (Greek ἔπιον, impv. πῖθι, Old Church Slavonic aor. pitъ ‘drank’) was likely a post-PIE development (cf. Praust 1996a:184 note 1).
Anatolian is the only branch in which the averbo is based on a sigmatic *peh3-s-. The source of the *-s- has been a matter of debate: some have argued for an old s-aorist (e.g., Eichner 1975a:83; Oettinger 1979a:435), an s-extension of indeterminate function (e.g., EDHIL:649 s.v. pāš-i / paš- and 435: “some extension”), or a “telic *-s-element” (Søborg 2020a:640). Any concerted explanation must take the other s-extended verbal stems in Anatolian into account (see Søborg 2020a a for a collection and discussion). It is noteworthy that Hieroglyphic Luwian /patta-/ ‘a drink’ preserves the unextended root.
[S.H.]
4. Transmission
Hittite pāš-(ḫi) ‘to swallow’, Cuneiform Luwian pašš-(ti) ‘to swallow’, Hieroglyphic Luwian /pass-(ti)/ ‘to swallow’, /patta-/ ‘a drink’, /pazza-(i)/ ‘to begin to drink (?)’, Luwian (Glossenkeilwörter)
pappaša-(i) ‘to swallow’, Cuneiform Luwian papašāla ‘esophagus (vel sim.)’, etc.See above and EDHIL:649 s.v. pāš-i / paš-.
Vedic aor. ápāt ‘drank’, pres. píbati ‘drinks’, caus. pāy-áyati ‘makes drink’, perf. papáu, pape, pass. aor. ápāyi ‘was drunk’ (cf. also pres. pānti, s-aor. ápās), pītá- ‘drunk’, °-pá- and °-pā́- ‘drinking’ (madhu-pá- ‘drinking mead’, soma-pā́- ‘drinking soma juice’), papí- ‘drinking’, pā́tar-, pātár- m. ‘drinker’, pā́na- n. ‘drinking, a drink’, pā́tra- n. ‘drinking vessel’, pītí- f. ‘drinking, a drink’, etc.Cf., e.g., EWAia II:113f. s.v. P², Werba 1997a:303 s.v. pā/pī.
Ancient Greek pres. πῑ́νω, Aeol. and Dor. πώνω ‘drink’, fut. (subj.) πίομαι, aor. ἔπιον, impv. πῖθι, Aeol. πῶθι, πῶ, pass. ἐπόθην, perf. πέπωκα, mid. πέπομαι ‘drink’, caus. πιπίσκω, fut. πίσω, aor. πῖσαι, πισθῆναι ‘give to drink’, ποτόν n. ‘a drink’, ποτός ‘potable’, πότος m. ‘drinking, a drink’, ποτήρ m., -ήριον n. ‘drinking cup’, πόσις f. ‘drinking, a drink’, πῶμα n. ‘a drink’, εὔπωνος ὄμβρος· εὔποτος ‘water that is good to drink’ (Hsch. ε 7095 Latte-Cunningham), (?) πῖνον n. ‘liquor made from barley, beer’, etc. Cf., e.g., GEW II:540–542 s.v. πίνω, EDG:1194f. s.v. πίνω.
Latin bibō, -ere ‘drink’ (< *pibō), Faliscan fut. pipafo ‘I will drink’, Sicel πιβε (impv. 2sg.) ‘drink!’, Latin pōtus ‘drunk’, pōtus, -ūs m. ‘a drink’, pōculum ‘drinking cup’, pōtor m. ‘drinker’, pōsca f. ‘acid drink of water and vinegar’; Umbrian punes (gen.sg.), pune (acc.sg.), puni, pone, poni (abl.sg.) ‘sacrificial drink’.Cf., e.g., WH I:103f. s.v. bibō, EDLIL:71f. s.v. bibō, 485 s.v. pōtus; Untermann 2000a:606f. s.v. U.poni.
Armenian əmpem ‘I drink’ (< *en-pibe-).Cf., e.g., Praust 1996a, LIV²:462f. s.v. *peh3(i̯)-, EDAIL:277–279 s.v. əmpem.
Gaulish ibetis (impv.? 2pl.) ‘drink!’, Old Irish ibid, ·ib (< *pibeti) ‘drink’, Middle Welsh 1sg. yfaf, 3sg. yf ‘drink’, Middle Breton 1sg. euaff, 3sg. ef*, Middle Cornish 1sg. evaf; (?) Old Irish án ā, f. ‘cup, drinking vessel’ (< *poh3-neh2-?).Cf., e.g., Schumacher 2004a:516–519, LIV²:462f. s.v. *peh3(i̯)-, LIV2+ s.v. *peh3(i̯)-.
Old Church Slavonic aor. pitъ ‘drank’, pres. pijǫ (piti) ‘drink’ (cf. Russian pit’, Czech pít, Serbo-Croatian pȉti, etc.), pojǫ, pojiti ‘give to drink’ (cf. Russian poít’, Polish poić, Serbo-Croatian pòjiti, etc.), pirъ m. ‘party’ (cf. Russian pir, Macedonian pir, etc.), pivo n. ‘drink; beer’ (cf. Russian pívo, Czech pivo, Serbo-Croatian pȋvo, etc.).Cf., e.g., LIV²:462f. s.v. *peh3(i̯)-, EDSIL:401 s.v. *pìti, 402 s.v. *pȋvo.
Old Albanian pi ‘drinks’, për-pín ‘devour, swallow’ (: Greek πῑ́νω).Cf., e.g., Schumacher & Matzinger 2013a:990.

