*zar-
‘to sway, undulate’
- Language
- Cuneiform Luwian
- Grammar
- u
- ID
- 879
1. Individual Anatolian Languages
1.1 Hieroglyphic Luwian
*/zar/-
‘to sway, undulate’1.1.1Derivatives
1.2 Cuneiform Luwian
*zar-
‘to sway, undulate’1.2.1Derivatives
2. Reconstruction
2.1Reconstruction
All Luwian derivatives listed below contain the root zar- with the semantics of ‘swaying, undulating’, showing different types of derivations. A deradical verb in *-i̯e/o-, i.e. Luw. */zari(ya)-/, can be recognized as the base of the participle *zarimma/i-, from which the denominative verbs zarimma-(ti) ‘to be agitated’ and zarimmi-(ti) ‘id.’ are derived, and of the action noun zariya- c. ‘(a moving waterbody)’. For the Luwian verbs zarimma-(ti) and zarimmi-(ti), no etymology is given by Tischler (HEG W-Z:672).
Considering the semantics posited for the root zar-, a connection to Hitt. zarzūr- n. ‘mixing, concoction (?)’, which Rieken 1999a:359 interprets as a Luwian word borrowed into Hittite, can be adduced. Furthermore, she connects the Luwian root zar- to the Proto-Indo-European root *k̑erh2- ‘to mix’. In order to explain the loss of the laryngeal in the reduplicated syllable, which should have left traces, Saussure’s Law is adduced, i.e. zarzūr < *k̑orh2-kr̥h2, and the second laryngeal would have disappeared in word-final position at any rate (followed by HEG W-Z:675f., doubted by EDHIL:1032). If one connects the root zar- of the other derivatives to this Indo-European root, it becomes difficult to account for the loss of the laryngeal with Saussure’s Law in every word. If the comparison between the two can be upheld within Luwian, one will have to assume an Anatolian root *k̑er- ‘to mix’ without the second laryngeal. Another uncertainty, which is also pointed out by Kloekhorst (ibid.), is that the initial consonant z- does not guarantee an original palatal velar. Pending the discovery of cognates within the other Anatolian languages, which would allow to verify what the Proto-Anatolian initial consonant of the root would be, the reconstruction with *k̑, i.e. *k̑er-, must be considered tentative.
[D.S.] [E.R.]
3. Proto-Indo-European
3.1Semantic Reconstruction
On the semantic reconstruction, cf. the Root etymology below.
3.1.1Root etymology
The root underlying PLuw. *zar- ‘to sway, undulate’ is attested in Vedic and Greek. The Greek formations are used exclusively in the context of mixing or stirring liquids,To denote the mixing of a liquid with a solid or the mingling of two solids, the Greeks used continuants of the root *mei̯k̑- ‘to mix’ (LIV²:428f.).and in Vedic the root is mostly used with soma and milk.The only exception is Ved. yávāśir- ‘mixed with corn (of soma)’ (RV).In view of Luw. zariya- c. ‘(a moving waterbody)’, the Greek (and Khwarezmian) restriction to liquids as well as the comparable tendency in Vedic, the only objects of PIE *k̑erh2- seem to have been liquids. Despite the meaning of Hitt./Luw.? zarzūr- ‘mixing, concoction (?)’, PAnat. *k̑er(H)- does not necessarily mean ‘to mix’ but ‘to undulate’, cf. PLuw. *zar- ‘to sway, undulate’. While the Vedic material is obscured by the preverb ā, the Greek nasal present implies that the original meaning of PIE *k̑erh2- is intransitivePIE nasal present causatives mainly derived from intransitives, cf. Zasada 2019a:47–57.as is the Luwian root. Thus, if the nasal present *k̑r̥-né-h2- ‘to mix’ is regarded as a causative formation of an underlying intransitive root, this verbal root may be approximately reconstructed as PIE *k̑erh2- ‘to undulate’.In Greek, the nasal present was the semantic point of departure. This solution implies that Vedic once also had a nasal present (*śṛṇāti) which provided the causative meaning ‘(to make undulate >) to mix’ for the attested formations, viz. the verbal adjective Ved. (ā́-)śīr-ta- ‘mixed (with milk)’ (cf. the cognate Gk. ἄ-κρᾱ-τος ‘unmixed’) as well as the root noun Ved. (ā‑)śír- ‘(ad-)mixture’.
On the phraseological history and the relationship to Ved. śrayi- ‘to make shine’, cf. Meusel 2018 [2020]a.
3.2Morphological Reconstruction
Since there are no cognates of any of the Anatolian formations attested in the non-Anatolian languages, cf. the Root etymology below for further details.
3.2.1Root etymology
On the basis of Luwian, Indo-Iranian, and Greek,The alleged Germanic (ON hrœra ‘to move, stir’, etc.) and further Vedic (Ved. śrī̆‑) evidence for this root has to be separated, cf. LIV²:328 n. 1 s.v. 2.*k̑erh2- (on Vedic), EDPG:250 (on Germanic).the root in question may be reconstructed with an anlauting voiceless palato-velar *k̑, followed by the thematic vowel and *r. Regarding the coda of the root Indic and Greek clearly point to root-final *h2, thus PIE *k̑erh2‑ whereas the Luwian evidence is in favor of an aniṭ-root PIE *k̑er‑. To maintain the attractive connection, it may be assumed that, similar to other roots like PIE *dei̯(h2)- ‘to shine’, the root-final *h2 of PIE *k̑er(h2)- ‘to undulate’ is some kind of root extension.Cf. on this topic the remarks in the Morphological reconstruction s.v. PAnat. *mérH-/*mr̥H-´ ‘to come as a guest’.
Differently, as presented above sub PAnat. reconstruction, the root PLuw. *zar- (not †zarr- vel sim.) may be interpreted as extracted from formations like Hitt./Luw.? zarzūr- ‘mixing, concoction (?)’ or from the strong stem of an unattested ḫi‑verb with regular loss of the laryngeal as per the de Saussure Effect: *k̑orh2- > *k̑or‑.
However that may be, PLuw. *zar- ‘to sway, undulate’ is ultimately connected with PIE *k̑erh2- ‘to undulate’.
[A.O.]
4. Transmission
No direct cognates are attested.
Ved. āśír- f. ‘(ad)mixture (mostly: milk)’ (RV+, also as second member of compounds), ā́śīrta- ‘mixed (with milk)’ (RV+)
EWAia I:178, Grassmann 1996a:178, Böhtlingk 1855a I:718, Meusel 2018a [2020]According to LIV²+ s.v. 2.*k̑erh2- (2019-04-17), OAv. sār- ‘to be associated, unite’ should rather be connected with (2.)*k̑erH- ‘to bind, plait’ (LIV²+ s.v., 2019-04-17).
Khwar. (caus.) bsʾry- ‘to stir’
EDIV:336