*h1éd-/h1d-´
‘to eat’
- Language
- Proto-Anatolian
- Grammar
- verb
- ID
- 1018
1. Individual Anatolian Languages
1.1 Hieroglyphic Luwian
/ad-(ti)/, EDERE
‘to eat’1.1.1Transmission
This lexeme is attested twelve times in the HLuw. corpus, with all attestations save one located in Tabal, of which all date the 8th century BCE; only the regional outlier KARKAMIŠ A6, far to the south-east from Tabal, may perhaps date a little earlier and belong to the late 9th century BCE.
[AH.B.]
1.1.2Forms
| 3pl. pres. ind. act. | /adanti/ | (“EDERE”)a+ra/i-x[.. | GEMEREK, D3§3 (Tabal) |
| 3sg. impv. act. | /atstu/ | á-za-tu | KARABURUN, §8 (Tabal) |
| KARABURUN, §10 (Tabal) | |||
| KULULU 1, §12 (Tabal) | |||
| EDERE-t[u] | BULGARMADEN, 1§16 (Tabal) | ||
| 3pl. impv. act. | /adantu/ | á-tà-tu-u | SULTANHAN, §33a (Tabal) |
| á-tà-u | KAYSERİ, §12 (Tabal) | ||
| EDERE-tú | KARKAMIŠ A6, §31 (Karkamiš) | ||
| a+ra/i-tu | KULULU 5, 7§11 (Tabal) | ||
| inf. | /aduna/ | (“EDERE”)á-ru-na | ÇİFTLİK, §16 (Tabal) |
| á-ru?-na | TOPADA, §31 (Tabal) | ||
| [á-tu]-na | KULULU 5, §15 (Tabal) | ||
| part. nom. sg. c. | /adammis/ | “EDERE”-tà-mi-i-sa | KULULU 2, A4B1-2§3 (Tabal) |
1.1.3Graphic Features
1.1.3.1Phonological Interpretation
The attestations of this lexeme offer sufficient evidence for a secure phonological interpretation. Additionally, the word displays 〈tà〉, reflecting the lenis dental (Rieken 2008b:638f.), and shows rhotacism in some forms, as is expected for an intervocalic /d/ in later texts (Melchert 2003b:179-182). The rare occurrence of the sign 〈a〉 instead of 〈á〉 in initial position has caused much discussion in the literature with Kloekhorst 2004a:31f. and Melchert 2010a:156 proposing that it reflects no phonemic difference, although Kloekhorst notes that the HLuw. corpus holds no combination of †〈á+ra〉, i.e. a spelling 〈a+ra〉 could have been forced by orthographic conventions, and Melchert concedes a possible allophone of the vowel usually spelt with 〈á〉 before /r/. Simon 2013a:17 finds that 〈a〉 and 〈á〉 occur in complementary distribution in HLuw. before the 9th century BCE, but cede to do so thereafter, i.e. none of the attestations for /ad-(ti)/ fall into the time when the distribution of the signs was still reliable.
What is striking, albeit not at all unusual in Anatolian, is that the spelling of the root consonant /d/ reliably reflects the epenthetic /s/ that regularly occurs between two dentals in the 3sg. impv. act. form (and similarly built verb forms), i.e. /ad+tu/ is phonetically realised as /atstu/ and spelt accordingly.
[AH.B.]
1.1.4Meaning
One of the most basic verbs, this lexeme is well-attested not only in HLuw., but also in the other Anatolian and IE languages. Its meaning provides the basic level term for the action of eating when undertaken by humans, gods, and divine animals, and possibly (although this is not explicit in the HLuw. texts) by real-world animals. Combined with the pre-verb ARHA /ahha/‘away’, its meaning is altered to ‘to eat up, eat wholly’, while a combination with INFRA /zanta/ ‘down’ is probably to be interpreted as ‘to swallow (food)’ (Hawkins 2000a:481, 523).
[AH.B.]
1.1.5Stem
The verb is a root verb, i.e. endings attach directly to the root, and in the case of two dentals coinciding, s-epenthesis takes place. In contrast to some of the other Anatolian languages, however, the HLuw. forms do not show any ablaut (cf. e.g. EDHIL:262f.).
[AH.B.]
1.1.6Derivatives
1.2 Cuneiform Luwian
ad-/ād-
‘to eat’1.2.1Transmission
Amply attested in Luwian ritual fragments and passages (from MS until LNS).
[Zs.S.]
1.2.2Forms
| 2nd sg. pres. | az-za-aš-ti-iš | KUB 35.133 ii 25’ | (CTH 665, LNS) |
| 2nd pl. pres. mid. | az-tu-u-wa-ri | KUB 9.31 ii 28 | (CTH 757, NS) |
| az-tu-u-wa[- | HT 1 ii 4’ | (CTH 394, NS) | |
| 2nd/3rd sg. pret. | az-za-aš-da | KUB 35.133 ii 8’ | (CTH 665, LNS) |
| 2nd pl. imp. | a-az-za-aš-ta-an | KUB 9.31 ii 26 | (CTH 757, NS) |
| 3rd pl. imp. | a-da-an-du | KBo 13.260 iii 10’ | (CTH 766, NS) |
| KBo 13.260 iii 12’ | (CTH 765, NS) | ||
| infinitive | a-du-na | KBo 7.68 ii 5’ | (CTH 764, NS) |
| participle | a-da-am-mi-in-zi | KUB 35.82 i 5’ | (CTH 761, MS) |
| a-da-am-˹mi˺[-in-zi] | KUB 35.15 iii! 11 | (CTH 761, MS) | |
| [a-]˹ta-am˺[mi- | KUB 35.14 iv 13’ | (CTH 760, NS) |
The attestations follow CLL:39 (cf. also Laroche 1959a:34, 36; HED A &E/I:319; HEG A-K:117; Starke 1985a:253–255) checked against the photos of the Konkordanz (no photo of the relevant part of KUB 9.31 was available).
Restoration of a-da-am-˹mi˺[-in-zi] in KUB 35.15 iii! 11 follows the contextually plausible suggestion of Melchert. Restoration of [a]-˹ta-am˺[mi- in KUB 35.14 iv 13’ follows Starke 1985b:125 based on a duplicate. Melchert suggests furthermore that -]˹mi˺-in-zi in KBo 29.12 iv 4’ belongs to the participle of this verb, which is possible, but cannot be proved due to the fragmentary text.
See s.v. †azzti- that az-za-aš-ti-iš of KUB 35.133 ii 25 belongs here and does not mean ‘food’, contra Starke 1985a:249–253. As correctly pointed out by CLL:39, the interpretation of adamminzi is not assured (thus it is included here only on formal grounds), since the fragmentary contexts do not allow an independent definition of its meaning (cf. also Meriggi 1957b:65, according to whom the assonance with the verb may be fortuitous).
On the spelling variation az(z)- ~ azzaš- see already Otten 1953a:46 and Starke 1990a:44. Since az(zaš)-/ az(zaš)- is a phonologically conditioned regular allomorph of ad-/ād- (Starke 1985a:250 with refs. on the previous discussions), the correct stem form is not the frequent ad-/āz- (contra CLL:39; Tischler 2008a:32, Tischler 2016a:41; Ünal 2016a:949), but ad-/ād- (with Otten 1953a:46 and EDHIL:262).
[Zs.S.]
1.2.3Meaning ‘to eat’
There is a contextually justified general agreement regarding the meaning of this word since its decipherment already by Hrozný 1920a:39 (Bossert 1933-1934a:113; Sommer & Falkenstein 1938a:85; Otten 1953a:46; Laroche 1959a:34; Carruba 1968a:15, Carruba 1982a:44, Carruba 1982b:10, Carruba 1986a:52; HEG A-K:117, Tischler 2008a:32, Tischler 2016a:41; HED A &E/I:319; CLL:39; Rieken 1999a:451; van den Hout 2001a:181 n. 70; EDHIL:262; Maier 2013a:217; Görke 2015a:§18; Ünal 2016a:949).
[Zs.S.]
Last updated: 22 May 2020
1.2.4Derivatives
1.3 Lycian A
*ad-
‘to eat’1.3.1Derivatives
1.5 Palaic
ad-
‘to eat’1.5.1Transmission
This lexeme is found in a festival text for the Palaic pantheon as well as in a mythical passage of an Old Hittite ritual.
[D.S.]
1.5.2Forms
| 3pl. pres. ind. act. | a-ta-a-an-ti | KUB 32.18 i 8’ | (CTH 752, OS or MS) |
| [a-]ta-a-an-ti | KUB 32.18 i 7’ | (CTH 752, OS or MS) | |
| a-da-a-an[-ti] | KBo 19.159 7’ | (CTH 752, OS) |
1.5.3Graphic Features
Checked against the photograph in the Konkordanz.
[D.S.]
1.5.4Meaning
Otten 1945-1951a:81f. suggests the meaning ‘to eat’ for Pal. ad- and brings contextual support: 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. ad- with Hitt. ed-/ad-(mi) ‘to drink’ (followed by Kammenhuber 1959a:19, Meriggi 1963a:3f., Carruba 1970a:52).
[D.S.]
1.5.5Stem
The question whether the Palaic verb ad- belongs to the mi- or ḫi-conjugation cannot be verified with the present attestations.
[D.S.]
1.5.6Derivatives
1.6 Hittite
ēd-/ad-(mi)
‘to eat’1.6.1Literature / Comments
HW 2nd ed. E:128-138, Oettinger 1979a:15f.
2. Reconstruction
2.1Reconstruction ‘to eat’
The comparative evidence of Hitt. ēd-/ad-(mi) ‘to eat’, Pal. ad- ‘to eat’ and CLuw. ad-(ti) ‘to eat’ allows us to reconstruct a root verb of the mi-conjugation with an "amphikinetic" ablaut paradigm, i.e. PAnat. *h1éd-/h1d-´ (EDHIL:262, no ablaut paradigm in given in HED E:315-320). The older view that it represents an acrostatic paradigm as claimed in Oettinger 1979a:89 (and Melchert 1994a:262) is not supported by the bulk of the Hittite attestations nor by the inner-Anatolian comparative data and, therefore, it is considered here to be outdated (EDHIL:262, Sasseville 2021a:260f., 290-295). 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. *h1d-énti, due to the lack of independent evidence. Instead, an analogical restoration in -a-, as it happened in šeš-/šaš- ‘to sleep’, based on other verbs with a phonologically regular outcome have been posited (Kimball 1999a:390 with further literature, Oettinger 1979a:90). For the controversial theory regarding the possible preservation of *h1 as a glottal stop in Hittite and Luwian, see Kloekhorst 2004a, Kloekhorst 2006a:79-81, Simon 2010b, Simon 2013a.
[D.S.]
3. Proto-Indo-European
3.1Semantic Reconstruction
The original meaning of the PIE root *h1ed- was ‘to bite’, whence ‘to eat’ (cf. LIV²:230 with n. 1). According to LIV²:230 with n. 3, the meaning ‘to eat’ originated in the present stem *h1ḗd-/*h1éd-, which had a durative-iterative meaning, i.e. ‘to bite repeatedly’ > ‘to eat’. The meaning of the present stem had then been generalized in PIE already while the original meaning ‘to bite’ was only preserved in derivations (LIV²:230 n. 1) – cf., e.g., PIE *h1(e)d-ont-/*h1d-n̥t- m. ‘tooth’ (< *‘the biting one’), PIE *h1éd-u̯on- f. ‘pain’, and Lith. úodas, Latv. uôds ‘mosquito, gnat’ (< *h1ōd-ó- ‘the biting one’). Cf. Schindler 1975c.
3.2Morphological Reconstruction
As for the verbal formations derived from PIE *h1ed- ‘(to bite →) to eat’, Lat. pres. subj. edim, OIr. ith-, and indirectly Arm. owt- could point to an original ḗ/é-acrostatic present stem *h1ḗd-/*h1éd- ‘to eat’ (cf. LIV²:230 with n. 3). It remains unclear whether the Anatolian material (Hitt. ēd-/ad-(mi) ‘to eat’, Pal. ad- ‘to eat’, CLuw. ad-(ti) ‘to eat’) ultimately goes back to the acrostatic present as well: affirmative LIV²:230 with n. 4, following Oettinger 1979a:89–91 (analogical ablaut ē/ă for original ē/ĕ); negative, e.g., EDHIL:262 and see above sub PAnat. Reconstruction, where the Anatolian forms are explained from a PIE present stem *h1éd-/*h1d-´ instead. – Beside the acrostatic present *h1ḗd-/*h1éd- a root aorist **h1éd-/**h1d-´ meaning ‘to bite’ can be postulated, according to Schindler 1975c (cf. LIV²:230). The aorist participle **h1d-ént-/**h1d-n̥t-´ ‘biting’ was the derivational base of the widespread noun stem *h1(e)d-ont-/*h1d-n̥t- m. *‘the biting one’ > ‘tooth’. There are numerous other noun stems formed to the root *h1ed-; cf. the transmission above and the overview in NIL:208–220.
[S.S.], [T.S.]
4. Transmission
Ved. pres. átti ‘he/she/it eats’, adánti ‘they eat’ (RV+);In Indo-Iranian (or in the prehistory of Vedic), the original ḗ/é-acrostatic root present *h1ḗd-/*h1éd- was remodeled to an amphikinetic root present with generalization of the full-grade root of the weak stem; cf. Narten 1968a:15 n. 44, Oettinger 1979a:89 with n. 14, EWAia I:61, LIV²:230 n. 5. sbst. ádman- n. ‘food, meal’< *h1éd-men-, cf. NIL:209, 215 n. 35. (RV 1.58.2), sbst. dánt- m. ‘tooth’< PIE *h1(e)d-ont-, cf. NIL:210, 216 n. 56f. with lit. (RV+), sbst. ánna- n. ‘food, meal’As per EWAia I:79f., this is a substantivization of the verbal adjective *anná- ‘eaten’. (RV+); cf. also compounds such as Ved. madhuv-ád- ‘eating sweet things, eating honey’< *medhu-h1ē̆d- (whence also OCS medvědĭ m. ‘bear’ < *‘honey-eater’), cf. NIL:208, 212 n. 4 with lit. (RV+)
Gk. pres. ep. ἔδω ‘I eat’ with athem. inf. ἔδμεναι (Hom.+), fut. ἔδομαι ‘I will eat’ (Il.+),The old athematic present which is preserved in the infinitive ἔδμεναι (Hom.+), in the short vocalic subjunctive (> future) ἔδομαι (Il.+) and in the imperative present 2sg. (*)ἔσθι (Od. 17.478?) gradually passed over into thematic inflection (cf. GEW I:444, LIV²:230 n. 6). From the imperative (*)ἔσθι the secondary present formations ἔσθω and ἐσθίω were formed (cf. GEW I:444, LIV²:230 n. 6, LIPP I:54f. [‘deimperativische Ableitung’]). pres. impv. (*)ἔσθι (Od. 17.478?), whence pres. ἐσθίω ‘I eat’Cf., e.g., GEW I:444, LIV²:230 n. 6.; sbst. ep. poet. εἶδαρ n., pl. εἴδατα ‘food’< *h1éd-r̥/*h1ed-n- or *h1éd-u̯r̥/*h1ed-un-, cf. NIL:208, 213 n. 10 with discussion and lit. (Il.+), sbst. ὀδύνη f. ‘pain’< PGk. *edunā (with vowel assimilation, but cf. Aeol. ἐδύνας acc. pl. [Greg. Cor. 597]) < PIE *h1ed-un-ah2-; cf. NIL:209, 214 n. 29 with lit. (Il.+), sbst. ὀδούς, Ion. ὀδών, gen. ὀδόντος m. ‘tooth’< PIE *h1(e)d-ont- ‘tooth’ < *‘biter’, reflecting the original meaning of the root *h1ed-, cf. LIV²:230 with n. 1, NIL:210, 216 n. 56 with lit. For the reconstruction of an amphikinetic stem *h1éd-ont- and the derivational history, cf. Steer 2015a:83–87 with further lit. (Il.+); cf. also compounds such as adj. νῆστις ‘not eating, fasting’< *n̥-h1d-ti- ‘having no food/meal’, cf. NIL:209, 215 n. 41. (Il.+), sbst. ἄριστον (with ᾱ) n. ‘morning meal, breakfast’< PGk. *ai̯eri-sto- < *h2ei̯eri ‘in the morning’ (loc. sg. of PIE *h2éi̯-r̥/*h2(e)i-(e)n- n. ‘day, morning’) + *h1d-tó- ‘eaten’ (verbal adj.); hence literally ‘what is eaten in the morning’, cf. NIL:211, 218 n. 91. (Il.+)
Arm. pres. owtem ‘I eat’;Cf., e.g., LIV²:230f. with n. 17 and lit. sbst. erkn, pl. erkownkʻ ‘labor, pains’;< *h1ed-u̯ō̆n-, cf. Olsen 1999a:139, 270, NIL:209, 214 n. 32. sbst. atamn, -ownkʻ ‘tooth’Ultimately from PIE *h1(e)d-ont-/*h1d-n̥t-; for the phonological development, cf. Olsen 1999a:505, NIL:210, 217 n. 61.
Lat. pres. edō ‘I eat’, inf. ēsse ‘to eat’, pres. subj. edim (Andr.+);Regularly transferred into half-thematic inflection, cf. Meiser 1998a:223f., LIV²:230 n. 7, Weiss 2011a:431. The derivation of inf. ēsse from an original Narten-present is suggested from the pres. subj. edim, edīs, etc. < PIE pres. opt. *h1éd-ih1- with continuous (accentuated) e-grade of the root vowel and zero grade of the optative suffix, cf. Meiser 1998a:201, Weiss 2011a:417. Osc. inf. edum ‘to eat’;Cf. Untermann 2000a:197. Lat. sbst. edō, gen. edōnis m. ‘glutton’< *h1ed-on-, cf. NIL:208. (hapax leg. Varro), dēns, gen. dentis m. ‘tooth’;< PIE *h1(e)d-ont-/*h1d-n̥t- ‘tooth’, cf. NIL:210, 216 n. 56, 217 n. 62. cf. also compounds such as Lat. hērēs, -ēdis m. (f.) ‘heir’Perhaps “komponiert aus *g̑heh1ro- und *h1ed- ‘das Zurückgelassene (?) essend’” (NIL:212 n. 1 with lit. and discussion of a different view); for the first compound member, cf. Gk. χῆρος ‘bereaved’, χήρα f. ‘widow’ (PIE root *g̑heh1- ‘to leave behind’, cf. LIV²:173). (Naev.+)
OIr. pres. ithid ‘he eats’, W ys ‘he eats’;Cf. LIV²:230 n. 8, Schumacher 2004a:377–380, EDPC:113. OIr. sbst. idu f. ‘birth pangs, labor, pain’,< *h1éd-un-, cf. NIL:209, 214 n. 28; differently EDPC:127 s.v. *fedon-. OIr. dét, MW dant, OBret. dant ‘tooth’< PIE *h1(e)d-ont-/*h1d-n̥t- ‘tooth’, cf. NIL:210, 216 n. 56.
Goth. pres. itan, ON eta, OE etan, OHG ëʒʒan, OS etan ‘to eat’;< PGerm. *ete/a- (cf. EDPG:119), which arose from a thematization of the original PIE ḗ/é-acrostatic present. Differently Mottausch 2013a:47, according to whom the Proto-Germanic thematic present is based on the root aorist of PIE *h1ed-. sbst. ON át, OE ǣt, OS āt, etc. ‘food, meal’,< PGerm. *ēta- < pre-PGerm. *h1ēd-o-, cf. NIL:210, 218 n. 74 with lit. sbst. Goth. tunþus ‘tooth’← PGerm. *tunþ- < *h1d-n̥t-, cf. NIL:210, 216 n. 56, 217 n. 64. beside OFris. tōth, OIce. tǫnn, OS tand ‘tooth’,< PGerm. *tanþ- < *h1d-ont-, cf. NIL:210, 216 n. 56, 217 n. 66. sbst. ON jǫtunn m. ‘Riese’, OE eoten ‘id.’,< PGerm. *etuna- < pre-PGerm./PIE *h1ed-un-o-, cf. Lühr 2000a:146f., NIL:209, 214 n. 22. etc.
OLith. pres. inf. ė́sti ‘to eat, guzzle’, 1sg. ė́mi (Modern Lith. ė́du), Latv. pres. inf. êst ‘id.’, 1sg. ē̦mu, ē̦du;Cf. ALEW I:17f., LEW I:124f. The verb shows athematic inflection in the early periods of East Baltic (cf. Stang 1966a:310). In modern dialects, the thematized forms have spread (cf. Zinkevičius 1966a:347f.). The explanation of the long ē-vowel in the present forms of Baltic (and Slavic) as generalized from the strong stem of the original Narten-present PIE *h1ḗd-/*h1éd- (cf. LIV²:230 with n. 3) competes with the explanation as being due to the effect of Winter’s Law (cf., e.g., EDLIL:186, EDHIL:263, ALEW I:18). Arguments against Winter’s Law are presented by Klingenschmitt 2008b:203 n. 203. sbst. Lith. sbst. úodas, Latv. uôds ‘mosquito, gnat’,< *h1ōd-ó- ‘the biting one’, cf. NIL:211, 218 n. 78 with further lit. sbst. Lith. ėdmẽnė, ėdmenė͂ f. ‘mouth; something edible; food, fodder’, Latv. ē̦dmanis m. ‘food, something edible’,< *h1ed-men-ii̯o/ii̯eh2-, cf. NIL:209, 215 n. 37f. with lit., ALEW I:270. etc.