*ku̯ér-/*ku̯r-´
‘to cut’
- Language
- Proto-Anatolian
- Class
- Base
- Grammar
- verb
- ID
- 596
1. Individual Anatolian Languages
1.1 Hieroglyphic Luwian
/kwar-/, kwa/i+ra/i
‘to cut’1.1.1Transmission
This lexeme is attested in two different texts from Maraş and Cilicia, one from the 9th century, the other from the very end of the HLuw. era in the 7th century BCE.
[AH.B.]
1.1.2Forms
| 1. sg. pret. act. | /kwarha/ | (“MANUS+CULTER”)kwa/i+ra/i-ha-a | MARAŞ 4, §13 (Maraş) |
| inf. | /kwaruna/ | (*486)kwa/i-tu-na | KARATEPE 1 Hu., §48 (Cilicia) |
| (“*486”)kwa/i-tú-na | KARATEPE 1 Ho., §48 (Cilicia) |
In a preliminary analysis, the KARATEPE form (*486)kwa/i-tu-na had been viewed as a verbal noun, not necessarily connected to the verb /kwar-/, but to a putative verb /kwat-/ (Hawkins 2000a:65). Separating the words seemed likely in light of what Melchert 2003a:179f. states about rhotacism in Luwian, viz. that the change in spelling is unidirectional d > r, i.e. intervocalic /d/ is frequently replaced with /r/ but not vice versa. However, this has since been disproved by Rieken & Yakubovich 2010a:216-218, who have shown that the signs used to represent original /lV/, /rV/, and /da/ were virtually interchangeable in late texts because the underlying sounds had then fully merged into the flap [ɾ]; and there is no reason to assume a priori that this would be different before /du/. Indeed, KARATEPE is one of the inscriptions where such interchanges occur (see the examples cited in Rieken & Yakubovich 2010a:206f., 215), and so it seems possible to take the form as a regular infinitive of the verb /kwar-/, not least since CLuw. displays a similar infinitive, /kuruna/, albeit with a zero-grade root no longer maintained in HLuw. (CLL: 116).
[AH.B.]
1.1.3Graphic Features
1.1.3.1Phonological Interpretation
The interpretation as /kwar/ seems assured despite the low number of attestations. Not only is the context in MARAŞ 4 quite clear, but the word is also well connected to the other Anatolian languages by way of numerous cognates (see EDHIL:486f.). The choice of sign *329, 〈kwa/i〉 (previously read REL), in the HLuw. forms shows that HLuw. has lost the ablaut pattern in this verb that is evident in the CLuw. and Hittite forms.
[AH.B.]
1.1.4Meaning ‘to cut, to harvest (?)’
Despite the small number of attestations, the meaning of this verb is clear from the contexts it occurs in and its strong cognate links to CLuw. kuu̯ar-/kur- ‘to cut’ and Hittite kuer-/kur-/kuu̯ar- ‘to cut’. The infinitive is attested in the KARATEPE bilingual and translates, in a syntactically rather opaque context, one of two Phoenician words, viz. qṣr ‘reaping’ or the partially broken off ḥ]rš ‘ploughing’, with the former being the more likely choice (Hawkins 2000a:65). From the context it appears that the infinitive is used as a noun.
As is apparent in the MARAŞ 4 context, the meaning of the verb can be further specified by the use of a preverb, in this case ARHA, /ahha/ ‘off, away’.
[AH.B.]
1.1.5Derivatives
1.1.5.1Uncertain Derivatives
1.2 Cuneiform Luwian
kuwar- / kūr-
‘to cut; to break (?)’1.2.1Transmission
Attested in a ritual with Luwian spells (NS) and in an Ištanuwan song (MS?).
[Zs.S.]
1.2.2Forms
| 3sg. pres. | ku-wa-ar-ti | KUB 35.48 iii 19’ | (CTH 760, NS) |
| ku-wa-ar-[ti] | KUB 35.48 iii 20’ | (CTH 760, NS) | |
| [ku-wa]-˹ar˺-ti | KUB 35.48 iii 20’ | (CTH 760, NS) | |
| ku-wa-al-ti | KUB 35.48 iii 18’ | (CTH 760, NS) | |
| inf. | ku-ú-ru-un-na | KUB 25.38, 11’ | (CTH 773, MS?) |
Attestations following CLL:116 (with slight corrections, cf. Laroche 1959a:58, 59f.; HED K:215) checked against the photos of the Konkordanz, restorations following Starke 1985b:157. Melchert also calls attention to ku-ra-an-t[e- in KUB 35.160, 2’ as a possible further form, which is not impossible, but the tiny fragment does not allow any conclusion and even if it belongs here, it is rather a nominalized form kurant-. Furthermore, he cautiously (“perhaps”) proposes that ku-ú-ri-x[ in KBo 29.29, 4’ may also belong here. Due to the fragmentary context this is impossible to assess and considering the vowel [i], it may rather belong to the verb kūri-/kur(a)i- ‘to cut into slices, pieces’? (see s.v.), if at all.
In accordance with CLL:116, ku-wa-al-ti of KUB 35.48 iii 18’ has not been corrected here to ku-wa-ra!-ti (contra Starke 1985b:157), since an r/l variation in preconsonantal position is not unknown to Luwian, even if this requires further research.
[Zs.S.]
1.2.3Meaning ‘to cut; to break (?)’
There has been a general agreement since Meriggi 1957a:207 n. 40 (cf. also Starke 1981a:149 n. 28; HEG A-K:608, Tischler 2008a:89, Tischler 2016a:186; CLL:116; HED K:215; EDHIL:486) that this word means ‘to cut’ (but without meaning in Laroche 1959a:59f. and in Ünal 2016a:962). Although it definitely fits with the Hieroglyphic Luwian form (see s.v.) as well as the Hittite cognate and it is possible contextually, too, it is worth noting that the context (the two preserved and presumed objects of the verb in KUB 35.48 iii 19’-20’, waḫra- ‘woe, misfortune vel sim.’ and ḫīrūn- ‘oath’, see s.vv.) does not really provide a perfect semantic match. The objects rather point to a meaning ‘to break’, which is supported now by its participle kuramma/i- in Hittite transmission (see s.v.).
[Zs.S.]
1.2.4Derivatives
Derivatives:
- kūramman-
- kurraštarr(i)-
- *kurattar / kuratn-
- gurta-
- kuršammaliya-
- *kuršammašša(/i)-
- kuršawar / kuršawan-
Possible derivatives:
Rejected derivatives:
1.3 Luwian in Hittite transmission
kuramma/i-
‘cut; broken’1.3.1Transmission
Hapax in a cult inventory from Kayalıpınar (LNS).
[Zs.S.]
1.3.2Forms
| nom.-acc. sg. n. | ˹ku˺-ra-am-ma-an | Kp 15/7+ ii 4 | (CTH 530, LNS) |
Attestation following Cammarosano 2018a:406, 414. No hand copy or photo has been published as of today (12 August 2019).
[Zs.S.]
1.3.3Meaning ‘cut; broken’
The word appears in the following context (Kp 15/7+ ii 4-5, text and translation following Cammarosano 2018a:406f.): ALAM LUGAL-aš arḫa 2? kuramman n=at EGIR-pa GAR-ri ‘There are 2 pieces cut off of the statue of the king. They are put back (on)’. Cammarosano 2018a:414 argues that this word is a noun in all other occurrences (as Luwian kuramman-), not a participle, which is supported here by the preceding numeral 2. This is correct, but the word kuramman- means ‘cut(ting)’ in all other occurrences (see s.v.), which does not fit here, because the context including the presence of the preverb arḫa requires a participle. Note that the context allows or even better supports a meaning ‘broken (off)’.
[Zs.S.]
1.3.4Origin
Cammarosano 2018a:414 claims an identity with Luwian kuramman- (kuramma- in his notation), but as it has been shown under Meaning, the two words are not identical (since the text was first published in 2018, there is no earlier discussion). However, since the context requires a participle, only the Luwian participle -mma/i- fits. Luwian also provides a suitable stem kuwar-/kūr- ‘to cut, to break?’ (see s.v.), thus it is indeed a Luwian loanword in Hittite.
[Zs.S.]
1.5 Lycian B
*ker-
‘to cut’1.5.1Derivatives
1.6 Lydian
*qir-, *qr-
‘to cut’1.6.1Derivatives
1.7 Palaic
*kūr-
‘to cut’1.7.1Derivatives
2. Reconstruction
2.1Reconstruction ‘to cut’
The Anatolian root *ku̯er- has numerous cognates in Hittite, Luwian, Lycian and Lydian which have long been known to scholars (see EDHIL:486f., HEG I-K:607-610 with further literature, Giannakis 2001a:149). A root verb of the mi-conjugation is attested both in Hittite as kwēr-/kur-(mi) and in Luwian as CLuw. kuwar- /kūr-(mi) and HLuw. /kwar-/ ‘to cut’. The evidence is clear that this verb goes back to a Proto-Anatolian amphikinetic root verb of the mi-conjugation, i.e. *ku̯ér-/ku̯r-´ (EDHIL:ibid., HED K:215).
Evidence from Cuneiform Luwian derivatives such as the nouns ku(wa)rša- ‘commoners (?)’ and kuršawar/kuršawan- ‘island’ lead us to reconstructing a Proto-Anatolian s-extended verb that co-existed with the root verb, i.e. *ku̯r̥-s- (interpreted as an original sigmatic aorist by Oettinger 1979a:119f., contra EDHIL:ibid., Dempsey 2015a:264f.
One should also consider the newly discovered univerbation zandakwaramma/i-, built from the participial stem with the strong grade of the root.
[D.S.]
3. Proto-Indo-European
3.1Semantic Reconstruction
Since PAnat. attests a root verb, cf. Root etymology below for the semantic reconstruction.
3.1.1Root etymology
Regarding the meaning of the root under investigation, the IE languages are divided in two: only the Anatolian branch (and perhaps Latin)The affiliation of Lat. curtus *‘cut off’, however, is debatable. EM:160f. and, e.g., Meiser 1998a:63, on the one hand, affiliate it to PIE *ku̯er- ‘to cut’, EDLIL:158 and WH I:316f., on the other hand, connect it with PIE 2.*(s)ker- ‘to shear, scratch, cut off’. On this phonological ambiguity, cf. Meiser 1998a:63f.preserves the presumably older meaning ‘to cut’, while the Core IE languages show the further developed semantics ‘to create, form, make’. In order to connect ‘to cut’ (Anat., Lat.?) with ‘to create, form’ (IIr.?, Celt., BSl.), one has to assume a meaning like ‘to carve’ as an intermediary step (as per LIV²:391 n. 1).Cf. also IEW:641 “vielleicht ursprünglich von irgendeiner nicht mehr bestimmbaren Handwerkstätigkeit.”
PAnat. *ku̯ér-/*ku̯r-´ ‘to cut (once), ?break’ denotes a singular verbal action suggesting an aoristic root. The verbal action ‘to carve’, on the other hand, is semantically iterative (‘to cut repeatedly’), and the verbal action ‘to create, form’ (Core IE) is again a perfective event. The disagreement regarding the Aktionsart (telic ‘to cut’) and (iterative ‘to carve’) advises caution in positing a semantic development. Nevertheless, the Anatolian evidence clearly shows the original meaning of the root PIE *ku̯er- ‘to cut’.
3.2Morphological Reconstruction
The PAnat. mi-verb corresponds morphologically to a PIE root aorist *ku̯ér-/*ku̯r-´ ‘to cut’, cf. LIV²:391. However, since the affiliation of the Indo-Iranian material is doubtful (cf. the comment above with ref. to LIV²+ s.v. *ku̯er-), reconstructing a Core-IE root aorist is not advisable solely on the indirect (Lith. kùrti ‘to build, create’)Lith. kùrti ‘to build, create’ can easily be influenced by homophonous Lith. kùrti ‘to light (a fire), heat’ which reduces the probability of it indirectly continuing a root aorist (as per LIV²:392 n. 14).and doubtful (OPr. 〈kūra〉 ‘built’) Baltic evidence. Furthermore, the semantic difficulties mentioned above suggest that the PAnat. formations are independent of the Core IE ones and thus, contra LIV²:391, no PIE root aorist is reconstructed here.
3.2.1Root etymology
The underlying root is reconstructed in LIV²:391f. as PIE *ku̯er-, although we find some instances of anlauting *s- in Indic, cf. Ved. saṁ-skṛtá- ‘prepared’, etc. However, this only case with s mobile may well belong to a different root, PIE *(s)ker- ‘to shear, scratch, cut off’ (LIV²:556f., LIV²+ s.vv.). The first radical is thus not s mobile but a voiceless labiovelar, which is suggested, e.g., by the outcomes in Celtic, viz. OIr. cruth ‘form, shape’ vs. MW pryd ‘id.’ Therefore, the root under investigation is reconstructed as PIE *ku̯er- ‘to cut’.
[A.O.]
4. Transmission
PIIr. root aorist *kar-/*ćar- ‘made, produced’, cf. Ved. ákar ‘id.’ (RV+), OAv. cōrǝt̰ ‘created’
(?)PIIr. *(s)kr̥-tá- ‘(adj.) done; (sbst.) deed’, cf., e.g., Ved. saṁ-skṛtá- ‘prepared (also of the holy language, Sanskrit)’, YAv. kǝrǝta- ‘done’
EWAia I:307–309, Grassmann 1996a:332–345, Bartholomae 1904:444–448, EDIV:236–238According to LIV²+ s.v. *ku̯er-, PIIr. *(s)kar- might belong to PIE 2.*(s)ker- ‘to shear, scratch, cut off’ (LIV²:556f., LIV²+ s.v. 1.*(s)ker-).
(?)Lat. curtus ‘shortened, circumcised, mutilated’ (Lucil.+)
WH I:316f., EM:160f., EDLIL:158, OLD²:524
(?)Osc. petiro-pert ‘four times’, etc.Untermann 2000a:545f. and LIPP II:609 with n. 13 connect it with PIE *per ‘through’.
Untermann 2000a:545f., WH II:283–286
PCelt. *ku̯ri-tu- ‘shape, form’, cf. OIr. cruth m. ‘id.’, MW pryd m. ‘form, shape, time’, etc.
EDPC:182, eDIL s.v. 1.cruth
MW peri ‘to cause, create, make’, per ‘chief, lord’ (*‘creator’), etc.
EDPC:178, Falileyev 2000a:129
Lith. kùrti ‘to build, create’According to LIV²:392 n. 14, this i̯e/o-present is based on the root aorist.
OPr. 3sg. pret. (or pres.) 〈kūra〉 ‘built’This form is phonologically and morphologically opaque, cf. Smoczyński 2005a:208–210.
EDBIL:266f., ALEW I:543, ALEW 2.0 s.v. kùrti I, LEW I:319, PKE޲ s.v. kūra, Smoczyński 2005a:208–210On Lith. kùrti ‘to light (a fire), heat’, Latv. kur̃t ‘id.’ and (denominal) PSlav. *kūriti ‘to smoke’ that might belong to a different root, cf. EDSIL:257, EDBIL:267, ALEW I:543f. s.vv. kùrti I, II.

