widā(i)-
‘to smash’
- Language
- Cuneiform Luwian
- Class
- Derivative
- Grammar
- verb
- ID
- 791
1. Individual Anatolian Languages
1.1 Cuneiform Luwian
widā(i)-
‘to smash’1.1.1Transmission
There is only one assured example in a Luwian ritual (LNS), perhaps also in an Ištanuwan ritual (NS).
[Zs.S.]
1.1.2Forms
| 3pl. imp. act. | ú-i-da-a-in-du | KUB 9.6 + KUB 35.39 iii 28’’ | (CTH 759, LNS) |
| 3pl. pret. act.? | wi5-ta-an-da | KBo 4.11, 30 | (CTH 772, NS) |
Attestations following CLL:271 (cf. alsoLaroche 1959a:111) checked against the photographs of the Konkordanz. Note that although wi5-ta-an-da in KBo 4.11, 30 (included by Melchert with question mark, by Laroche with double question mark) can formally belong to the same verb and would even fit semantically, this cannot be proven, cf. under Meaning. Contra Laroche 1959a:111 CLL:271 rightly points out that ú-i-ta-at-ta-an in KUB 35.43 ii 11 does not belong here, since the assumption of Laroche (2sg. imp.) is implausible in its context (which suggests an acc. sg., see the discussion s.v.).
[Zs.S.]
1.1.3Meaning ‘to smash’
Although many scholars translate this verb as ‘to smash’ since Watkins 1986b:325f. (Garrett & Kurke 1994a:76; Reichardt 1998a:35, 79; Torri 2003a:60; Yakubovich 2012a:327; Tischler 2016a:443, HEG:535), one can find alternative translations, too: ‘to divide’ (Melchert 1984a:158), but later ‘to strike’ (CLL:271, followed by Tischler 2008a:228 [but cf. above]), and even ‘to bring, to lead’ (Starke 1990a:308, which is false according to CLL:271), if it is not left untranslated (as in Laroche 1959a:111, but also still in EDHIL:838 and Ünal 2016a:976).
The context is, however, exceptionally clear: the passage, a Luwian spell, is introduced by a Hittite description of the ritual act, in which the patient destroys two reed baskets (KUB 9.6+ iii 19’’-23’’). This is followed by the spell (KUB 9.6+ iii 25’’-29’’, Starke 1985b:115, translation adapted from Yakubovich 2012a:327): “Whoever causes evil to the patient, may the gods crush him like reeds, and may they w. him the wītpani-, may they put it under his feet”. Thus we are clearly dealing with sympathetic magic and all three parts of the spell mean the destruction of the evil-doer, just as the baskets were destroyed. Accordingly, the proposals ‘to bring, to lead’, ‘to divide’, and ‘to strike’ are not correct: the first case is obvious and in the latter two, a division and a strike are simply different from the action of smashing and they do not necessarily lead to destruction. Melchert’s first translation, ‘to divide’, was based on the assumed literal parallelism of the Hittite description and the Luwian spell, but such does not exist either in the number of the predicates (the Hittite section has two, the Luwian has three) or in their references (the Hittite verbs are ‘to twist’ and ‘to tear apart’, i.e. a destruction in two steps unlike in the Luwian spell, where the destruction appears already in the first predicate, cf. also Watkins 1986b:326).
Unfortunately, the form wi5-ta-an-da is attested only in a fragmentary context (KBo 4.11, 30), nevertheless, the previous line describes the lifting of vessels and thus if it is a verb, it may refer to their destruction, for instance by smashing them, especially if the preceding word, GUD.MAḪ-li, indeed means something like ‘powerfully’ (meaning proposed by Yakubovich pers. comm.), see the discussion s.v.
[Zs.S.]
2. Reconstruction
2.1Reconstruction
The Cuneiform Luwian ú-i-da-a-in-du ‘they must smash’ has been equated etymologically with Hitt. wezzai ‘he presses, he urges’ and 
wiwidā- ‘id.’ (Watkins 1986b:326, HEG U-Z:582f.). The root can be reconstructed as Proto-Anatolian *u̯ed(h)-. While the Hittite verb reflects a root verb *wid-(mi), in which the root-final alveolar became an affricate before another alveolar (s-epenthesis between two alveolar obstruents rule), i.e. *wizzazzi, before it was remodeled to a stem of the ḫi-conjugation, i.e. wizzai; cf. ezzazzi → ezzai ‘he eats’ (Melchert 1977a:437). The Luwian stem, on the other hand, can continue either a denominative verb in -ā(i)-(di) (< *-éh2-i̯e/o-) or a causative-iterative in -(ā)i-(di) (< *-éi̯e/o-). Since causative-iterative in Luwian show the reflex of an o-grade of the root, the form ú-i-da-a-in-du with the vowel -i- in the root cannot be explained as such. Therefore, it is preferable to assume a denominative verb from a non-attested noun *wida- (PAnat. *u̯ē̆d(h)-aH-). The root vowel could come from an original e-grade with the development of *e > i between a labial glide and a coronal or from an original lengthened e-grade, i.e. *u̯ēd(h)-. Therefore, it is possible that the lengthened grade comes from the nominal formation. Another possibility is that the root verb had an inflection of a Narten present, i.e. PAnat. *u̯ēd(h)/u̯ed(h)-(mi), which would account likewise for the lengthened grade. All possibilities must be kept open until further evidence. The Hittite cognate wizzai does not help either, since both root ablauts discussed are equally possible here. Differently, Melchert 2011a:128 suggests an iterative verb in *-eh2-i̯e/o- with a lengthened e-grade of the root based on Latin verbs such as cēlāre ‘to hide’ (see also Melchert 1994a:231). However, the reconstruction of this type of formation for Proto-Indo-European is very tentative (see Oettinger 1979a:374 n. 239). Finally, regarding the glossenkeil word wiwidā-, whose status as Hittite or Luwian is uncertain, it is a reduplicated stem derived from *u̯ēd(h)/u̯ed(h)-(mi) and inflecting according to the ḫi-conjugation.
[D.S.]
3. Proto-Indo-European
3.1Semantic Reconstruction
The meaning of a possible proto-form of the two hapax legomena Luw. widā(i)- ‘to smash’ and OAv. vādāiia- ‘to crush, break’ may be reconstructed as ‘to destroy’ (< ‘to cause destruction’). For the semantics of the derivational base, cf. Morphological reconstruction below.
3.1.1Root etymology
Based on the fact that PIE *u̯edhh1-García Ramón 1998a:151 regards the root as aniṭ, *u̯edh‑.is attested in Vedic and Greek continuants of a root aorist (Ved. ávadhīt ‘he beat, slew’, Gk. ὦσα ‘I thrusted, pushed’), García Ramón 1998a:151f. reconstructs the telic meaning ‘to hit (once)’.
Judging from the meanings of Ved. vadhá- ‘club, (epithet of Indra’s weapon)’ and others, PIE *u̯edhh1- ‘to hit (once)’ may originally have been restricted to melee weapons.
3.2Morphological Reconstruction
The root vowel of Luw. widā(i)- ‘to smash’ may go back to *‑ē- or *‑ĕ- (cf. above). The root vocalism of OAv. vādāiia- ‘to crush, break’ may continue *‑ē- or *‑ō- (cf. ALF:56–58).Of course, IIr. *ā can always go back to *ŏ in an open syllable (via Brugmann’s Law). In the present case, however, the root syllable is closed (PIE *u̯edh.h1‑).Thus, if the Luwian and Old Avestan forms are equated with each other, the common preform shows lengthened ē‑grade of the root. Regarding the post-radical suffix vowel the evidence points at least to a long vowel (post‑)PIE *‑ā- or *‑ō-.The Old Avestan suffix vowel ‑ā- does not correspond to any suffix vowel reconstructable for o‑grade causative-iteratives like PIE *u̯ṓdhh1‑i̯e/o- ‘to hit repeatedly, attack’ (cf., e.g., Gk. ὠθέω ‘to thrust, push, drive’). This means that an etymology independent from Luwian is less likely.The combination of ē‑grade root and ā/ō‑grade (primary) suffix as well as a following *i̯ suggests denominative status of the formation in question. Therefore, if Luw. widā(i)- ‘to smash’ and OAv. vādāiia- ‘to crush, break’ are regarded as cognates, the most plausible common preform is PIE *u̯ēdhh1‑eh2‑i̯e/o- ‘to make/cause destruction’, a denominative formation in *‑i̯e/o- based on PIE *u̯ēdhh1‑eh2- ‘destruction’.Khot. bādāra- ‘(a weapon)’ is possibly a ro‑derivative of PIran. *u̯ādā- < PIE *u̯ēdhh1eh2. However, the intervocalic ‑d- should regularly have been lost, which is why Bailey 1979a:276 regards the form as “archaic or dialectal”.
3.2.1Root etymology
While the “core” of the root in question is unanimously reconstructed as PIE *(°)u̯edh(°), there are still differing opinions about anlauting and/or auslauting laryngeals.
On the one hand, Kloekhorst (in EDHIL:349–352, followed by EDG:1676f.) proposes an initial laryngeal *h2- to connect Hitt. huett- ‘to pull, draw’ with Gk. ὠθέω ‘to thrust, drive’ (in his view < *au̯othéi̯e/o‑). However, since the retention of the laryngeal would be expected in Luwian (not †hwidā(i)‑), anlauting *h2- is unlikely.
A root-final laryngeal is supported by the Vedic iṣ‑aorist and specifically *‑h1- by Gk. ὠθέω ‘to thrust, drive’ (if < *u̯ṓdhh̥1‑i̯e/o-)For the causative-iterative type in R(ṓ)‑i̯e/o- (LIV²:23), cf., e.g., Gk. πωλέομαι ‘to go to and fro’ (root PIE *ku̯elh1- ‘to turn’, LIV²:386–388) and πωτάομαι ‘to fly about’ (root PIE *peth2- ‘to fly’, LIV²:479).,cf. LIV²:660 s.v. *u̯edhh1- n. 1. Thus, it is advisable to reconstruct the root as PIE *u̯edhh1- ‘to hit (once)’.
[A.O.]
4. Transmission
(?)OAv. 3sg. opt. vādāiiōit̰ ‘to crush, break’ (hapax, Y.29.2)
(?)Khot. bādāra- ‘(a weapon)’
Bartholomae 1904:1410, Bailey 1979a:276
PIE
(?)PIE *u̯ṓdhh1‑i̯e/o- ‘to hit repeatedly, attack’, cf. Gk. ὠθέω ‘to thrust, push, drive’ (Il.+) (GEW II:1144, DELG:1298f., EDG:1676f., LSJ:2031, LSJSuppl:319, LfgrE IV:1323–1328), (?)OCS(+) vaditi ‘to accuse’It may also belong to PIE *h2u̯edH- ‘to talk, sound’ (LIV²:286). (EDSIL:512, Vasmer I:163)
LIV²:660
Indo-Iranian
PIIr. *u̯adhá- ‘mortal, killing’, cf. Ved. vadhá- m. ‘mortal, killer, club, murder, (epithet of Indra’s club)’ (RV+), YAv. acc. sg. vaδǝm m. ‘weapon, wedge’ (hapax, V.14.7)
Ved. aor. ávadhīt ‘he beat, slew’ (RV+), etc.
EWAia II:496f., Grassmann 1996a:1201f., Bartholomae 1904:1344, EDIV:404f.
Greek
(?)Gk. part. pres. nom. sg. m. ἔθων ‘?disturbing, ?confounding’ (hapax, Il.9.540), nom. pl. m. ἐθóντες ‘id.’ (hapax, Il.16.260)This participle may hint at a PIE Narten-present, cf. LIV²:660. However, the meaning as well as the connection with the root in question are still debated, cf., e.g., EDG:378f.
Gk. 3sg. ἔθει ‘(Gk.) φθείρει, to destroy; (Gk.) ἐρεθίζει, to annoy’ (Hsch. ε 636 Latte)
Gk. aor. ἔωσα, ὦσα ‘to thrust, push, drive’ (Il.+)
GEW I:459f., GEW II:1144, DELG:316, 1298f., EDG:378f., 1676f., LSJ:480, 2031, LSJSuppl:104, 319, LfgrE II:418, LfgrE IV:1323–1328According to GEW I:446f., 449, Gk. ἔθειραι f. ‘mane (of a horse), hair’ and ἔθρις ‘(Gk.) τομίας, castrated’ (cf. also GEW III:82) may belong here, too.
Celtic
(?)MIr. fáiscid ‘to press, squeeze’, MW gwascu ‘id.’, OBret. (dem)guescim ‘id.’
OIr. fodb ‘booty (clothing or equipment)’
EDPC:401, 426f., eDIL s.v. fáiscid, eDIL s.v. 1 fodb, fadb
Balto-Slavic
Lith. vedegà f. ‘hatchet, hoe’, Latv. vȩdga f. ‘ice tool, gouge’, OPr. wedigo ‘carpenter’s hatchet’ (E 531)
ALEW II:1204f., LEW II:1211, Karulis 1992a II:499, PKEŽ² s.v. wedigo
Tocharian
TochA wac, TochB weta ‘combat, struggle’
TochA impf. 3sg. wātñā ‘to thrust, stab’ (hapax, A 295 a 3)
TochB wätā?- ‘to fight, struggle’
DTB²:641, 661, Malzahn 2010a:865, 876f.