lukī-
‘to break’
- Language
- Palaic
- Class
- Base
- Grammar
- verb
- ID
- 372
1. Individual Anatolian Languages
1.1 Palaic
lukī-, (redup.) lukīlukī-
‘to break’1.1.1Transmission
This lexeme is found in a Palaic ritual with bread offerings preserved on Old/Middle and New Hittite tablets as well on other Palaic fragments of the New Hittite period.
[D.S.]
1.1.2Forms
| 3sg. pret. ind. act. | ˹lu˺-ki-i-it | KUB 35.165 obv. 2 | (CTH 751, OS or MS) |
| lu-ki-i-it | KUB 35.165 obv. 3 | (CTH 751, OS or MS) | |
| lu-ki-i-it | KUB 35.165 obv. 4 | (CTH 751, OS or MS) | |
| [lu-]˹ki˺-i-it | KUB 35.165 obv. 5 | (CTH 751, OS or MS) | |
| lu-ki-i-it | IBoT 2.35 obv. 8 | (CTH 751, MS) | |
| [l]u-˹ki˺-i-it | IBoT 2.35 obv. 9 | (CTH 751, MS) | |
| lu-u-ki-it | KUB 35.163 iii 21’ | (CTH 753, NS) | |
| lu-ki-it | KBo 13.265, 3’ | (CTH 753, NS) | |
| redupl. | |||
| 3pl. pret. ind. act. | [lu-]ki-i[n-l]u-ki-in-ta | KUB 35.165 obv. 2 | (CTH 751, OS or MS) |
| [l]u-ki-i-lu-ki-˹i-en˺-ta | IBoT 2.35 + KBo 19.150 obv. 8’ | (CTH 751, MS) | |
| [l]u-ki(-)lu-ki-in-ta | IBoT 2.36 obv. 3 | (CTH 751, NS) |
1.1.3Graphic Features
Checked against the photograph of the Konkordanz. The segment a-ar-ra-kat lu-ki-it (KBo 13.265 3’) is actually written a-ar-ra-kat-lu ki-it, i.e. with a space between and instead of between <lu> and <ki>. Its status as mistake is confirmed by the parallel passage; cf. KUB 35.163 iii 21’. It is not certain whether the segment [lu-]ki-i[n-l]u-ki-in-ta (KUB 35.165 obv. 2) is written as one or two words, because the tablet is broken right in the middle of the segment. Judging by the size of the break, it is probable that there is a space between the <in> and <lu>. The two parallel passages, i.e. IBoT 2.35 + KBo 19.150 obv. 8’ and IBoT 2.36 obv. 3 are not uniform regarding the spacing; while the latter leaves a space between lu-ki(-i) and lu-ki-i/en-ta, the former does not.
[D.S.]
1.1.3.1Phonological Interpretation
In KUB 35.165 obv. 2, the reduplicated part of the verb [lu-]ki-i[n-l]u-ki-in-ta ends with an -n as opposed to the two other attestations, where there is no -n. This extra -n is explained here as a case of nasal perseveration; cf. Melchert 1984b:33.On full root intensive reduplication in Hittite see recently Dempsey 2015a:7f., 317-324. However, he does not discuss full stem reduplication.
[D.S.]
1.1.4Meaning ‘to break (into pieces)’
The meaning of the lexeme lukī- and the reduplicated formation lukīlukī- can be elucidated by considering the Hittite passage preceding the Palaic one in IBoT 2.35 + KBo 19.150. Before the incantation addressed to Ziparwa begins, it is said in the Hittite passage that the king repeatedly breaks bread using the Hittite verb paršiyanna-/-i(ya)-(ḫi) ‘to break (imperf.)’ (LUGAL-uš NI[NDA… pár-]ši-ia-an-na-i). At the beginning of the Palaic passage (KUB 35.165 obv. 2), the speaker mentions briefly that they (scil. tabarna and tawannanna) perform lukīlukī-. Afterwards, the speaker describes in details the bread that the tabarna and tawannanna handle one after another (ibid. 4-5) using the verb lukī- for each action in four consecutive sentences. What is clear from the context is that the action of Pal. lukī- is performed repeatedly, which supports the grammatical interpretation of lukīlukī- as an iterative or better imperfective formation (after Kammenhuber 1959a:14, Carruba 1972a:16, contra Melchert 1984b:31-33).
Furthermore, the action of lukī- is performed on bread (ibid. 4-5) that is placed on the tāwa ḫalpiyala (ibid. 3, see under respective Lemma). Therefore, both the imperfective status and the semantics of lukīlukī- can be equated with that of Hitt. paršiyanna-/-i(ya)-(ḫi) ‘to break (into pieces)’ (similarly Carruba 1972a:16f.). The preceding contextual analysis speaks for a meaning other than simply ‘to distribute’ for Pal. lukī- (pace Melchert ibid.), i.e. ‘to break (into pieces)’ = Hitt. paršiye/a-(mi).The previous meaning ‘to set fire to’ suggested by Kammenhuber 1959a:13 based on an etymological connection with Hitt. lukke/a-(mi) ‘to set fire to’ is rejected rightly on contextual grounds (Carruba 1972a:16f., Wallace 1983a:165, Melchert 1984b:32f., Oettinger 1979a:276).
[D.S.]
2. Reconstruction
2.1Reconstruction ‘to break’
Carruba 1972a:17 n. 6 suggests for Pal. lukī- ‘to break (into pieces)’ an etymological connection with the Proto-Indo-European root *leu̯g- ‘to loosen, break’ (likewise Melchert 1984b:33 n. 14, Wallace apud Melchert ibid.). The single spelling of the velar in lu-ki-i-it is in agreement with the posited etymological *g. Wallace 1983a:165 assumes that the accent was on the suffix, which is supported by the constant plene spelling in lu-ki-i-it (differently LIV²:416 n. 4). However, he leaves the question open as to whether the root had a zero-grade or a full grade, i.e. PAnat. *lug-i̯é/ó- (thus Melchert 1994a:193) or perhaps PAnat. *léu̯g-i̯e/o- (with subsequent *eu̯ > u). No cognate of Pal. lukī- has been found so far within the Anatolian branch.The etymological connection with the root *leu̯k- ‘to light’ found in Eichner 1973a:81, Lehrman 1998a:141 is outdated; see above under meaning. Pace Melchert 1984b:33, there is no longer any evidence for an abstract noun luki- and thus the verb lukī- does not have to be taken as a denominative, but can be simply deradical.
[D.S.]
3. Proto-Indo-European
3.1Semantic Reconstruction
Reconstructing Pal. lukī- ‘to break’ as a suffixal present formation (*l(e)u̯g-i̯e/o-, cf. PAnat. Reconstruction above) matches with the fact that PIE *leu̯g̑/g- ‘to break’ is a telic root with a root aorist (cf. Transmission).
3.1.1Root etymology
The Indo-Iranian (e.g. Ved. rok ‘break!’, rujáti ‘to break’ etc.) and the Baltic forms (cf. Lith. láužti ‘to break’ etc.) as well as Gk. ἀ-λυκτο-πέδαι ‘(unbreakable) bonds’ point to a telic root PIE *leu̯g̑/g- ‘to break’.The differing meaning ‘to loosen’ attested in Arm. aor. louci and the nasal present (?)PIE *lunég-/*lung-´ can be explained via ‘to break a bond’ (cf. Gk. ἀλυκτοπέδαι ‘(unbreakable) bonds’).
The Germanic verbs meaning ‘to pluck, weed’ (cf. EDPG:334) can be paralleled by ModHG brechen ‘to break (tr.) (of flowers, twigs, fruits, nuts etc.)’, cf. DWB II:348 s.v. brechen II.8.
3.2Morphological Reconstruction
Since no morphological cognate of the i̯e/o-present Pal. lukī- ‘to break’ can be found in the non-Anatolian languages, only the PAnat. reconstruction above is given.
3.2.1Root etymology
The root underlying Pal. lukī- ‘to break’ is attested in several Core IE branches (Indo-Iranian, Armenian, Greek, Celtic, Germanic, Baltic). Beside the onset *l-, the biradical coda consisted of the semivowel *u̯ and a voiced unaspirated dorsal, viz. *leu̯g̑/g-. The phonological reconstruction is ambiguous, pointing either to PIE *leu̯g- with a plain voiced velar (cf. Ved. aor. rok ‘to break’, YAv. -uruxti- ‘breaking’ etc.) or to *leu̯g̑- with a palatal (cf. Lith. láužti ‘to break’). Neither the remaining non-Anatolian evidence from the satem-language Armenian nor the Palaic form itself permit a decision.On Arm. c < *g̑ and c < *g/u_, cf. Schmitt 2007a:61f., 63. According to LIV²:416 n. 1, PIE *leu̯g̑- is a mere variant of the root *leu̯g-.There are different explanations for the mismatch regarding the root-final dorsal. The auslaut of PIE *leu̯g̑/g- ‘to break (tr.)’ might be regarded as a root extension, viz. *leu̯-g̑/g-.Cf. EWAia II:465 opting for PIE *leu̯- ‘to dissolve’ (= PIE *leu̯H-, LIV²:417) as the underlying root. Cf. also WH I:790, 831 proposing that PIE *leu̯p- (?= *leu̯p- ‘to peel (off)’, LIV²:420) and *leu̯b(h)- ‘to peel off’ (≠ *leu̯bh- ‘to be pleasing’, LIV²:414) belong here, too. Similarly Kümmel (in LIV2+ s.v. Neu: *segwh- ‘ertragen’) on the comparable doublet PIE *segu̯h- ‘to endure’ (ibid.) beside *seg̑h- ‘to overwhelm’ (LIV²:515f.): “lässt an eine alte Variante denken”. Furthermore, cf. KEWA III:65, IEW:686.
[A.O.]
4. Transmission
No direct cognates attested.
PIE
root aor. *léu̯g/g̑-/*lug/g̑-´ ‘to break’, cf. Ved. impv. 2nd sg. aor. rok (VS), ruk (MS) ‘break!’ (EWAia II:465, KEWA III:64f.), Arm. aor. lowci ‘loosened’ (EDAIL:311 s.v. *loyc)
(?)them. pres. *lug/g̑-é/ó- ‘to break’, cf. Ved. rujáti ‘to break, crack’ (EWAia II:465, KEWA III:64f., Grassmann 1996a:1173f.); Goth. us-lukan ‘to pull (a sword)’, OE tō-lūcan ‘to destroy’ etc. (EDPG:334, Holthausen 1934a:207)
(?)nasal pres. *lunég-/*lung-´ ‘(?)to loosen’,If the PIE sequence *-nh2,3V́- developed into PIE *-ngV́- (as per Steer fthc.), the nasal present PIE *lunég-/*lung-´ could be disconnected from the root in question and associated with PIE *leu̯H- ‘to cut off, loosen’ (LIV²:417), hence, PIE *lunéH-ti/*lunH-énti > *lunéH-ti/*lung-énti → *lunég-ti/*lung-énti. However, if the aorist Arm. louci is inherited (as per Klingenschmitt 1982a:268), the assumption made above becomes less likely. EDAIL:311 suggests that both roots were contaminated in Armenian.cf. Arm. lowcanem ‘to loosen’ (Bible+) (EDAIL:311 s.v. *loyc, Klingenschmitt 1982a:159–161, 184); OW guo-llung ‘to let go of; to set free’ etc. (EDPC:249, eDIL s.vv. fo-loing, in-loing, Falileyev 2000a:72)
LIV²:415f., LIV2+ s.v. 1.*leu̯g-, IEW:686
Indo-Iranian
Ved. perf. rurója ‘has broken’ etc.
EWAia II:465, 481 KEWA III:64f., 113, Grassmann 1996a:107, 1173f., 1174, 1177, 1184, 1188
YAv. -uruxti- f. ‘breaking’ (Y.70.3, N.84)< *-luk-ti- < *-lug-ti- (voice assimilation), cf. Martínez & de Vaan 2014a:30f. § 11.16.
Bartholomae 1904:119 s.v. an-ava´uruxtay-, 996 s.v. fra-uruxtay-,
Iran. vb. *rauǰ- ‘to break, burst’, cf. Late Khot. rauśtä ‘to break (intr.)’ etc.
EDIV:318 with refs.
Greek
(?)Gk. ἀ-λυκτο-πέδαι ‘(unbreakable) bonds’ (Hes.+)The affiliation to the root discussed here is called into question by GEW I:80 (followed by DELG:66) preferring Risch’s assumption that it is a recent ad hoc-formation, cf. Risch 1944a:26. EDG:76: “All of this remains highly uncertain”.
GEW I:80, DELG:66, EDG:75f., LSJ:73
(Gk. λευγαλέος, λυγρός ‘wretched, unhappy’ (Il.+))Affiliation with PIE *(s)leu̯g- ‘to swallow’ is preferred here, differently GEW I:108, DELG:632, EDG:850f. The meaning is better explained by a semantic shift from ‘to swallow, (ModHG) schlucken’ to ‘to sob, (ModHG) schluchzen’, cf. also the comment on Lat. lūgeō below.
GEW I:108, DELG:632, EDG:850f., LSJ:1041, 1063f.
Italic
(Lat. lūgeo ‘to grieve’)Kümmel (LIV2+ s.v. 1.*leu̯g-), following Kölligan 2005a, considers this formation belonging to PIE *(s)leu̯g- ‘to swallow’ (LIV²:567f., LIV2+ s.v. *(s)leu̯g-), cf. also EDLIL:351 and the comment on Gk. λευγαλέος, λυγρός above.
WH I:830f., EM:369, EDLIL:351, OLD:1049
Celtic
(?)PCelt. *luxtu- ‘contents; crowd, people’, cf. OIr. lucht m. ‘charge, contents, people’ etc.
EDPC:251, eDIL s.v. lucht, Falileyev 2000a:107
Baltic
Lith. láužti, Latv. laûzt ‘to break (tr.)’ etc., OPr. luse ‘head injury, skull fracture’ (N 97)
ALEW I:563f., 603, ALEW 2.0 s.v. láužti, PKEŽ 2013 s.v. luse, EDBIL:276, 297, Trautmann 1923a:152f., LEW I:347