*h1és-/h1s-´
‘to be’
- Language
- Proto-Anatolian
- Class
- Base
- Grammar
- verb
- Certainty
- high
- ID
- 198
1. Individual Anatolian Languages
1.1 Hieroglyphic Luwian
/as-(ti)/, á-sa-
‘to be’1.1.1Transmission
This lexeme ‘to be’ is amply attested across the Iron Age period and twice in the Empire period.
[D.S.]
1.1.2Forms
1.1.3Meaning
In the bilingual inscription of KARATEPE 1, the Luwian verb /as-(ti)/ corresponds to Phoen. KN ‘to be’ (Cf. HHG:34f.) and etymologically equates to Hitt. ēš/aš-(mi) ´to be´.
[D.S.]
1.1.4Compounds
- /Astiru(wa)-/ (PN) (cf. Zehnder 2010a:51, 294, Kronasser 1956a:204f.)
- /Astitarhunza-/ (PN) (cf. Zehnder 2010a:51, Kronasser 1956a:204f.)
- /Astiwasu-/ (PN) (cf. Zehnder 2010a:51, Neumann 1983a:146f.)
- /Astuwalamanza-/ (PN) (cf. Yakubovich 2010d:92, Zehnder 2010a:51, Neumann 1983a:146f., Kronasser 1956a:204f.)
1.2 Cuneiform Luwian
āš-/aš-(ti)
‘to be’1.2.1Transmission
This verb is attested from MS until LNS in Hittite rituals and festival texts.
[A.B.]
1.2.2Forms
| 3sg. pres. ind. act. | a-aš-ti | KUB 35.133 rev. iv 10’ | (CTH 665, LNS) |
| 3sg. pret. ind. act. | a-aš-ta | KBo 9.141 rev. iv 4’ | (CTH 761, MS) |
| KUB 35.15 obv. ii! 3’ | (CTH 761, MS) | ||
| KUB 35.65 rev. iii 7’ | (CTH 763, MS) | ||
| KUB 35.89 16’ | (CTH 766, MS) | ||
| KUB 35.103 rev. iii 7 | (CTH 766, MS) | ||
| KUB 25.37 obv. ii 20’ | (CTH 771, NS) | ||
| a-aš-t[a] | KUB 35.101 rev.? 5’ | (CTH 765, MS) | |
| a[-aš-ta | KUB 35.102+ rev. iii 6 | (CTH 766, MS) | |
| 3sg. impv. act. | a-aš-du | KUB 32.9 rev. 31’ | (CTH 761, MS) |
| KUB 35.15 rev. iii! 5’ | (CTH 761, MS) | ||
| KUB 35.54 rev. iii 26 | (CTH 758, MS) | ||
| KUB 35.103 rev. iii 7 | (CTH 766, MS) | ||
| KBo 9.145 r. col. 5’ | (CTH 770, NS) | ||
| KBo 13.260 obv. ii 32 | (CTH 766, NS) | ||
| KBo 29.54 l. col. 13’ | (CTH 765, NS) | ||
| KBo 30.190 obv. ii 3’ | (CTH 761, NS) | ||
| KUB 35.53 rev. iii? 5’ | (CTH 758, NS) | ||
| KUB 35.58 obv. ii 6’ | (CTH 759, NS) | ||
| KUB 35.67 obv. ii 2’ | (CTH 759, NS) | ||
| KUB 35.70 obv. ii 17 | (CTH 762, NS) | ||
| KUB 35.85 4’ | (CTH 759, NS) | ||
| KUB 35.88 rev. iii 15’ | (CTH 766, NS) | ||
| ⸢a⸣-aš-du | KBo 13.264 r.col. 2’ | (CTH 763, NS) | |
| a-⸢aš-du⸣ | KBo 9.145 r.col. 12’ | (CTH 770, NS) | |
| a-aš-⸢du⸣ | KBo 29.31 rev. iv 2’ | (CTH 756, NS) | |
| a-aš-[du] | KUB 35.54 rev. iii 30 | (CTH 758, MS) | |
| KUB 35.101 obv.? 9’ | (CTH 765, MS) | ||
| KUB 35.103 rev. iii 7 | (CTH 766, MS) | ||
| KUB 35.67 obv. ii 3’ | (CTH 759, NS) | ||
| KUB 35.81 4’ | (CTH 759, NS) | ||
| a-aš[-du? | KUB 35.69 9’ | (CTH 761, MS) | |
| ⸢a⸣-[aš-]⸢du⸣ | KUB 35.54 rev. iii 34 | (CTH 758, MS) | |
| a?-a]š-du | KUB 35.54 rev. iii 32 | (CTH 758, MS) | |
| a!-<aš->du | KBo 9.145 r.col. 11’ | (CTH 770, NS) | |
| a-[aš-]du | KUB 35.58 obv. ii 2’ | (CTH 759, NS) | |
| a-a[š-du? | KUB 35.67 obv. ii 6’ | (CTH 759, NS) | |
| 3pl. impv. act. | a-ša-an-du | KBo 29.4 obv. ii? 9’ | (CTH 758, NS) |
| KBo 29.6 rev. 6’ | (CTH 762, NS) | ||
| KUB 32.79 5’ | (CTH 762, NS) | ||
| KUB 35.13 11’ | (CTH 762, NS) | ||
| KUB 35.78 obv. i? 15’ | (CTH 759, NS) | ||
| a-ša-an-d[u] | KBo 7.66 obv. ii? 7’ | (CTH 760, MS) | |
| ⸢a⸣-ša-an-du | KBo 13.264 r.col. 8’ | (CTH 763, NS) | |
| [a?-š(a?-a)]n?-du | KUB 35.14 rev. iv 15’ | (CTH 760, NS) | |
| [a?-š]a?-an-du | KUB 35.15 rev. iii! 13’ | (CTH 761, MS) | |
| [a?-]ša-an-du | KUB 35.92 obv. 28’ | (CTH 440, NS) | |
| a-ša-a-an-du | KUB 32.7 6’ | (CTH 762, NS) |
The attestations follow Laroche 1959a:32f., CLL:33, Starke 1985b, HED E:290, and Yakubovich (personal communication); entirely restored instances have not been included. The transliterations have been checked against the photos of the Mainzer Fotoarchiv and the hand copies.
The form listed in CLL:33 as a-ša-a-an-du and supposedly attested in KUB 35.21 rev. 31’ actually appears on fragment KUB 32.9 rev. 31’ (of the same tablet). While CLL:33 lists the form in KUB 32.7 6’ as a-ša-an-du, the photo of the Mainzer Fotoarchiv shows that it is written with plene a (cf. the hand copy as well as Starke 1985b:132).
An unusual hyperplene spelling a-a-aš-t[u appears in the ritual fragment KBo 27.77 obv. ii 4’ (CTH 754, MS), cf. CLL:33. Although the fragment is in Palaic (at least §1’ and §3’, lines 1’-2’ and 7’ff., cf. Starke 1990a:593), line 5’ in §2 shows a genuine Luwian word with a Luwian ablative ending: ḫa-am-šu-⸢uk⸣-ka4-la-a-ti (KBo 19.152 obv. ii 2’+KBo 27.77 obv. ii 5’; cf. Starke 1990a:593). That may indicate that we are dealing with a longer Luwian passage embedded in the Palaic text, but due to the fragmentary nature of the passage and the presence of less understood words its length cannot be determined (differently Starke 1990a:593, who attributes the entire paragraph to Luwian). Accordingly, the status of this word as either Luwian or Palaic remains uncertain.
[A.B.] & [Zs.S.]
1.2.3Graphic Features
The form in KUB 35.15 rev. iii! 12’ is transliterated as [a-š]a-an-du by Starke 1985b:128. However, the photo of KUB 35.15 shows a sequence x-]QA-du (cf. also the hand copy).
[A.B.]
1.2.4Meaning ‘to be’
The meaning of this word is undisputed and generally accepted as ´to be´, cf. Laroche 1959a:32f., CLL:33, Tischler 2008a:28, Tischler 2016a:35 as well as (discussed under the Hittite lemma eš-/aš- in) HED E:290, HEG A–K:76, HW (2nd ed.) E:97 and EDHIL:250.
With regard to its use in the attested CLuwian texts, most of the forms are imperatives (due to the instructive nature of the ritual texts) that appear with adjectives or participles to describe what ritual components should be like.
When the subject is in the first or second person, sentences with the verb ‘to be’ require a reflexive pronoun, cf. Melchert 2003b:203.
[A.B.]
1.2.5Stem
Athematic root verb (Melchert 2003b:199) with endings of the mi-conjugation.
[A.B.]
1.3 Lycian A
es-(ti)
‘to be’1.3.1Transmission
This lexeme is attested in various types of inscriptions found mostly in Xanthos.
[D.S.]
1.3.2Forms
| 3sg. pres. ind. act. | esi | N 320.12 (Xanthos) |
| (e)si | TL 44b.50 (Xanthos) | |
| 3sg. pret. ind. act. | ese | |
| 3sg. pres. impv. act. | esu |
The form estte in TL 44b.2 does not belong here (contra Neumann 1984b:94f. and DLL:17), see under Lyc. as-(ti). Laroche 1979c:120 and Puhvel (HED E-I:290) interpret hãti (TL 83.12, N 320.21) as a form of the verb ‘to be’, which cannot be sustained.
[D.S.]
1.3.3Graphic Features
1.3.3.1Phonological Interpretation
Due to the Lycian phonological development of *-st- > /s/, the -t- of the endings -ti, -te and -tu can no longer be seen, i.e. esi < *es-ti (Melchert 1994a:288).
[D.S.]
1.3.4Meaning
The first form of the verb ‘to be’ was recognized by comparing Lycian curse formulas with the Greek ones occurring in Greek inscriptions of Lycia: Torp 1898a:25 equates esu httẽmi (TL 91.3) with Gk. ἔστω ἁμαρτωλός (followed by Thomsen 1899a:63, DLL:17, Neumann 2007a:51f.). For the assignment of (e)si in TL 44b.50 to the verb es-(ti) ‘to be’ see Neumann 1984b:95. The two cases of ese (TL 44a.33, TL 44b.36) are attributed to this paradigm as preterit forms instead of to the preverb ese. In TL 44a.33, it is not likely to be a preverb, since the verb a(i)-(di) ‘to make’ never takes the preverb ese. Therefore, it does not belong to the same sentence as the verb ‘to make’, i.e. me uwadraxi : ese : przze “There/it was a bull sacrifice for the first ones”. A translation with the verb ‘to be’ makes both contextually and syntactically more sense. The second example (TL 44b.36) cannot be a preverb, since the sentence would contain no verb and, therefore, the presence of a preverb is unexpected. Moreover, it occurs before the conjunction terñ. These two observations speak for the preterit form of the verb ‘to be’.
[D.S.]
1.3.5Compounds
- Esitm̃mãta- (PN) (Neumann 1983a:147, Neumann 2007a:79)
- Aruwãtijese/i- (PN) (Neumann 1978a:127, Carruba 1979b:88, 90f., Schmitt 1982b:385, Neumann 1983a:146, Neumann 2007a:25f., Zehnder 2010a:50)
- Mullijese/i- (PN) (Neumann 1978a:127f., Zehnder 2010a:50, Neumann 2007a:226)
- Mulese(/i)- (PN) (Neumann 1978a:127f., Neumann 2007a:225, Neumann 2012a:391)
- Ertelijese(/i)- (PN) (Neumann 2012a:394)
1.4 Lydian
el
‘to be(?)’1.4.1Transmission
This word is a hapax legomenon attested in an inscription from Sardis that cannot be dated accurately.
[D.S.]
1.4.2Forms
| 3sg. pret. act. | el | LW 10.13 (Sardis) |
Pace Kloekhorst 2008c:251, the form el cannot be in the present tense, because the verbal ending -l marks the preterit.
The interpretation of the clitic =im as a form of the verb ‘to be’ by Gusmani 1971a has been refuted; see Melchert 1991a:138 n. 14 and Yakubovich 2009a:45.
[D.S.]
1.4.3Meaning
The identification as a verb was first proposed by Brandenstein 1931b:51, basing himself on the preterit ending -l. Based on the syntax of the passage, the meaning ‘to be’ was assigned by Meriggi 1935a:100 and then Gusmani 1961a:189: ak=um nãqid kol fak=m=it el edš wastνunliš qašaas “whatever he revealed(?) to me, was that …” (vel sim.).
[D.S.]
1.4.4Stem
The synchronic stem remains obscure. Gusmani 1964a:100 suggests a root stem and reconstructs the preform *eš- by assuming that the addition of the preterit ending -l would have caused the loss of the preceding -š-. This is not impossible, but it must be considered very tentative until further evidence emerges.
[D.S.]
1.5 Palaic
āš-/aš-(ti)
‘to be’1.5.1Transmission
This lexeme is attested in various Palaic texts.
[D.S.]
1.5.2Forms
| 2sg. impv. act. | a-aš | KUB 35.163 iii 10 | (CTH 753, NS) |
| 3sg. impv. act. | a-aš-du | KUB 35.165 obv. 28 | (CTH 751, OS or MS) |
| 3pl. impv. act. | a-ša-an-du | KUB 35.165 rev. 6’ | (CTH 751, OS or MS) |
| a-ša-an-du | KUB 35.165 rev. 7’ | (CTH 751, OS or MS) | |
| ˹a˺[-š]a-an-du | KUB 35.165 rev. 8’ | (CTH 751, OS or MS) | |
| a-še-en-du | KBo 19.153 iii 13’ | (CTH 751, NS) | |
| a-še-en-d[u] | KBo 62.43 4’ | (CTH 751, NS) | |
| [a-š]e-en-du | KBo 53.301 6’ | (CTH 751, NS) |
The form e-eš-ta[] (KUB 35.163 iii 17) does not belong to the lexeme āš/aš-(ti) ‘to be’; see ēš-(ti). Contra, Kammenhuber 1955b:357 n. 2 interprets this instance as a hittitized form of the verb ‘to be’, which cannot be proven either.
[D.S.]
1.5.3Graphic Features
Checked against the photographs in the Konkordanz.
[D.S.]
1.5.3.1Phonological Interpretation
The e-vocalism of a-še-en-du has not been explained (cf. Melchert 1994a:199), although it is worth noticing that it occurs only in New Script texts.
[D.S.]
1.5.4Meaning
The Palaic verb āš-/aš-(ti) ‘to be’ was equated by Kammenhuber 1955b:356 n. 2 with Hitt. ēš-/aš-(mi) and later on (Kammenhuber 1959a:24) with CLuw. āš-/aš-(ti), HLuw. /as-(ti)/. The lexical assignment and the meaning were taken over by Carruba 1970a:52.
[D.S.]
1.5.5Derivatives
1.6 Hittite
ēš-/aš-(mi)
‘to be’1.6.1Literature / Comments
HEG A-K:109f., HED E-I:285-291, EDHIL:250-252. For a syntactic analysis of this lexeme see Cotticelli Kurras 1991a.
2. Reconstruction
2.1Reconstruction ‘to be’
The reconstruction of the Proto-Anatolian stem alternation is supported by the alternation of the plene-writing in the present singular and by its absence in the present plural forms of Cuneiform-Luwian āš-/aš-(ti) and Palaic āš-/aš-(ti). The vowel *e in the singular forms of the verb ‘to be’ is preserved in Hittite ēš-/aš-(mi). This verb is a root athematic present with accented root in the e-grade in the singular and unaccented root in the zero-grade in the plural. This amphikinetic pattern is preserved in Hittite, Cuneiform Luwian and Palaic (Oettinger 1979a:558, 561, Lehrman 1998a:139, 142). The initial vowel a- in the plural forms cannot represent the vocalization of the laryngeal, which would be expected form a Proto-Indo-European point of view, i.e. 3rd pl. *h1s-énti, due to the lack of independent evidence. Instead, an analogical restoration in -a-, as it happened in šeš-/šaš-(mi) ‘to sleep’, based on other verbs with a phonologically regular outcome have been posited (Kimball 1999a:390 with further literature, Oettinger 1979a:90).
There is also a competing view on this problem: Kloekhorst 2006a:79-81 (and EDHIL:252) argues that in Hittite the first laryngeal is preserved as a glottal stop in prevocalic position. In initial preconsonantal position, *h1 would be regularly lost and, then, in some ablauting verbs, the laryngeal or the glottal stop was reintroduced. This is intertwined with the phonetic realization of *h1 in Anatolian, which is still a debated matter.
[D.S.]
3. Proto-Indo-European
3.1Semantic Reconstruction
PIE *h1es- simultaneously exhibits an existential sense ‘to exist’ and the function of a copulative verb and auxiliary in all Indo-European languages. It is conceivable that the verb’s primary sense was ‘to be there, be present, be available’ (with “adessive semantics”, see West 2016), and that the copulative use developed secondarily (West 2016:487f.).
3.2Morphological Reconstruction
The present-stem inflection of PIE *h1es- is notoriously conservative across all Indo-European languages. The Anatolian branch preserves the present indicative, imperative, the imperfect, and the sk̑‑iterative.
Pal. impv. 2sg. act. āš equals Hitt. ēš, and Lat. ĕs from PIE *h1és! alongside *h1s‑dhí > *h1ǝs‑dhí > Gk. ἴσθι (with initial schwa secundum rather than by vowel assimilation from *ἔσθι; cf. Schwyzer 1939a:677f. n. 11 and Hackstein 2012d [2014]:111 sub “Regel b”), Ved. edhí, OAv. zdī.
3.2.1Anatolian preterite forms
HLuw. pret. 1sg. act. /asha/ ‘I was’ is an Inner-Anatolian replacement of PIE impf. *h1ésm̥ as continued by Hitt. ešun (cf. PIE *éh1esm̥ > Ved. ā́sam, Hom. Gk. ἦα, cf. Watkins 1969:40).
HLuw. pret. 3sg. act. /asta/ < *ast (with final prop vowel -a), Hitt. ešta is from PIE *h1és‑t (cf. PIE *é‑h1es‑t in Ved. ā́s, ā́sīt, Dor. Gk. ἦς).
HLuw. pret. 3pl. act. /asanta/ ‘they were’ goes back to unaugmented PIE *h1s‑ént corresponding to augmented PIE *é‑h1s‑ent in Ved. ā́san, and Hom. Gk. (3pl. > 3sg.) ἦεν.
The Anatolian preterite forms accord with the Indo-European preterite forms of *h1es- as set forth by Hittite ešun, Greek ἦα and Vedic ā́sam (from *ā́sa) except for an analogical extension of ḫi‑conjugation endings. Luwian has generalized the the ḫi‑conjugation endings in the pret. 1sg. (‑ha), and 2sg. act. (‑ta), and Hitt. in the 2sg. act. (‑ta).
There is no unequivocal evidence for a stative h2e‑present ?*h1os‑/h1es‑h2e, or perfect ?(h1e)h1os‑h2e. The lexical atelicity of PIE *h1es- ‘to be (present)’ causes it to be defective (and supplied by other roots) outside the present stem. Hom. Gk. ἦα ‘I was’ shows no trace of an o‑grade; its interpretation as an imperfect is therefore more convincing. Homeric Greek ἦσθα, which has often been suspected to continue an IE perfect (e.g. Schwyzer 1939a:677), is not a secure equation with the HLuw. form, since ‑θα spreads analogically in the athematic imperfect and present; cf. Hom. impf. ἔφησθα ‘you spoke’ alongside ἔφης, and δίδοισθα ‘you are giving’ alongside δίδοις, see Chantraine 1973a:470. For Ved. perf. 2sg. ā́sitha ‘you have been/were’, 3sg. ā́sa ‘has been/was’, and OAv. perf. 3pl. ā̊ŋharǝ̄ ‘they have been/were’, see Kümmel 2000a:618–620, with arguments for considering these forms innovations.
3.3Syntactic Reconstruction and Phraseology
PIE prosodically and syntactically differentiates in the present indicative stem forms the existential and copulative forms of *h1es‑. The substantive verb is orthotonic and covert in affirmative existential clauses, but always overt in negative existential clauses, cf. Hackstein 2012d [2014] on PIE *né=h1esti ‘there is not X’). The copula is clitic, and not obligatorily overt. Germanic (Prokosch 1939a:220, PGmc. *senđí > Goth. sind), Ancient Greek (Schwyzer 1939a:676f.) and Tocharian (stem PIE *h1s-sk̑e/o-, Hackstein 1995a:275f.) provide evidence for clitic copula forms. Likewise the Anatolian branch subdivides the use of ‘to be’ into copula (auxiliary in analytic verb forms) and substantive verb. In accordance with the syntactic type of copulaless nominal clauses (cf. for Hittite, Hoffner & Melchert 2008a:412–414), HLuw. nominal clauses may lack the copula in the present indicative too; cf. e.g. (relative clause) ama=n=za=pa=wa kwa=n=za tanima=n=za ‘all (acc. sg. n.) that is mine’ (BABYLON §4).
[O.H.]
4. Transmission
PIE athematic amphikinetic root pres. 3sg. act. ind. *h1és‑ti/clitic *h1es‑tí/Ø, 3pl. act. ind. *h1s‑énti/clitic *h1s‑entí/Ø, impv. pres. 3sg. act. *h1és‑tu, 3pl. act. *h1s‑éntu
Inflection, all stems: LIV²:241f.; present stem: Weiss 2009a:425–428; present imperative: Forssman 1985a; inflection and derivation: IEW:340–342; derivation: NIL:235–243
Ved. pres. 3sg. act. ásti, 3pl. act. sánti, impv. 2sg. edhí, 3sg. ástu, 3pl. sántu
OAv. pres. 3sg. act. astī, 3pl. act. hǝ̄ṇtī, impv. 2sg. zdī, 3sg. astū, 3pl. həṇtū
EWAia I:144, Gotō 2013a:101, Werba 1997a:159f.; ALF:200f.
Myc. pres. 3sg. act. /esti/*, 3pl. act. /ehensi/, impv. 3pl. act. /ehentō/, Gk. impv. 2sg. ἴσθι
DMic I:203f., Plath 1992a:534
Cl. Arm. pres. 1sg. em, 2sg. es, 3sg. act. ē, 3pl. act. en1sg. *im, analogically replaced by em after 2sg. es
Klingenschmitt 1982a:11, 84, 234
Early Lat. pres. 1sg. act. esom, OLat. 3sg. est, 3pl. sont, Lat. impv. 2sg. es; Osc. 3sg. act. est, 3pl. act. sentExistential verb, e.g. Early Latin esom kom meois sokiois (Garigliano Bowl Inscription, 6/5th c. BCE, Hartmann 2005a:147ff.), and copula/auxiliary verb.
OIr. 3sg. act. is (abs.), 3pl. act. it (abs.)
Schumacher 2004a:295, 308, 310, Schrijver 2020a
Goth. pres. 3sg. act. ist, pres. 3pl. act. sind
Lühr 1984a, Schumacher 2007a:186–196
OLith. 3sg./pl. e͂sti, OCS 3sg. act. jestŭ, 3pl. act. sǫntŭFor the possibility that tu‑imperatives, are continued in Slavic too, see Forssman 1985a:185f.
ALEW 1.1:305f.
Alb. 3sg. act. është (Tosk), âsthë (Gheg), 3pl. act. janë
Demiraj 1997a:207f., Matzinger 2006a:157f., Schumacher Matzinger 2013a:424–427, Neri 2019Alb. është is used as an auxiliary verb to form the analytic verb forms of the nonactive verb (Demiraj 1997a:207).
Toch. B impf. 3sg. act. ṣai, 3pl. act. ṣeyeṃ
Hackstein 1995a:255, Malzahn 2010a:253–256The Tocharian imperfect forms indirectly attest the athematic present stem *h1es‑/*h1s‑, since they descend from the present optative stem *h1s‑i̯eh1‑; see Hackstein 1995a:255, Malzahn 2010a:253–256. The present indicative forms are supplied by the stem PIE *h1s-sk̑e/o- (Toch. B pres. 3sg. act. copula ste, stár‑, 3pl. act. skente, skentar‑, stare), and by the root PIE *nes- (copula and verbum substantivum Toch. B pres. 3sg. = pl. act. nesäṃ).