arr-(i)
‘to wash’
- Language
- Palaic
- Grammar
- verb
- ID
- 628
1. Individual Anatolian Languages
1.1 Palaic
arr-(i)
‘to wash’1.1.1Transmission
This lexeme is found in the festival for the Palaic pantheon.
[D.S.]
1.1.2Forms
| 3sg. pres. ind. act. | ar-ri-iº | KBo 40.108 obv. 5’ | (CTH 751, MS?) |
1.1.3Graphic Features
Checked against the photograph in the Konkordanz.
[D.S.]
1.1.4Meaning
The fragment on which the verb arr-(i) is attested was discovered and joined to KUB 32.17 by D. Sasseville. The direct object of the verb arr-(i) is marḫa[n] ‘the guest’. From an etymological perspective, it is cognate with Hitt. ārr-/arr-(ḫi) ‘to wash’, and the same semantics can be applied to the Palaic verb, i.e. arri=an marḫa[n] ‘He washes him, the guest (scil. Zaparwa)’. This is contextually supported by another phrase belonging to the same festival text, i.e. [mā]rḫan ḫuwarnašš[a] ‘Sprinkle the guest with water!’ (IBoT 2.35 iii 4’ + KBo 19.155 iii 24’).
[D.S.]
1.2 Hittite
ārr-/arr-(ḫi)
‘to wash’1.2.1Literature / Comments
HW (2nd ed.) A:224-234, HEG A-H:51, HED A:111-116.
2. Reconstruction
2.1Reconstruction ‘to wash’
Equating Pal. arr-(i) with Hitt. ārr-/arr-(ḫi) ‘to wash’, we can reconstruct a root verb of the ḫi-conjugation. The gemination of the resonant speaks for a root-final laryngeal. Since the verb belongs to the consonantal inflection in Palaic and Old Hittite (instead of the half-thematic inflection), the root-final laryngeal must be reconstructed as *h1 (EDHIL:197f.). Thus, the reconstructed Proto-Anatolian verb is *h1órh1-/h1r̥h1-´ (thus EDHIL ibid.).
For a possible cognate in Luwian, see arramma/i-.
[D.S.]
3. Proto-Indo-European
3.1Semantic Reconstruction
The meaning of the Proto-Anatolian ḫi-verb is ‘to wash’, which should be posited also for PIE *h1erh1- (LIV²:239, IEW:337).
3.1.1Root etymology
Anatolian (‘to wash’) and Tocharian (‘to bathe; to purge [ritually]’) show a very similar verbal action, but the valency of the Tocharian root yär(ā)- is not clear (cf. Malzahn 2010a:799).On intransitive TochA yärnā- ‘to bathe oneself’, cf. the comment below sub Morphological reconstruction.Therefore, based only on the Anatolian evidence, the root is reconstructed as PIE *h1erh1- ‘to wash (tr.)’ (as per LIV²:239).
3.2Morphological Reconstruction
While TochA inf. yä(rn)āssi ‘to bathe oneself’, yärnā[,TochA yärnā is found immediately before the gap in THT 1154 a3 and, thus, no morphological analysis is possible.on the one hand, may continue a nasal present weak stem *h1r̥-n-h2-Cf. LIV²:239, Malzahn 2010a:469 (following Hackstein 1995a:325). However, since “there are so many irregularities [in Prs. X; AO] that the existence of the whole type can be doubted” (Peyrot 2013a:554) and an intransitive meaning (‘to bathe [intr.]’) is rather odd in the context of nasal presents, it should be refrained from projecting the nasal present back to the proto-language.and, on the other hand, TochA abs. yāyruräṣ ‘having taken a bath’ points to a root aorist, the Anatolian data attests a ḫi-verb. In view of the sparse evidence, the antiquity of PAnat. *h1órh1-/*h1r̥h1-´ ‘to wash’ cannot be secured.
3.2.1Root etymology
The underlying root has to be reconstructed with anlauting *h1 because *h2 and *h3 would develop into PAnat. *H and TochA ā̆-, respectively.The only possible anlaut sequence yielding TochA yär- is PIE *h1er° (> PToch. *jär°).As Kloekhorst (EDHIL:198) elucidates, the root-final laryngeal cannot have been *h2 or *h3, either, because in that case one would not expect Hitt. ārri- but †ārrai-.Hackstein 1995a:319 (probably followed by Malzahn 2010a:469) reconstructs PIE *h1erh2-.Thus, the root under investigation has to be reconstructed as PIE *h1erh1- ‘to wash’ (cf. also LIV²:239).
[A.O.]
4. Transmission
TochA yär(ā)- Prs. X ‘to bathe [intr.]; to purge oneself (ritually)’, Kaus. Prs. VIII ‘to bathe [tr.]; purge (ritually)’
Hackstein 1995a:317–319, Malzahn 2010a:798f.
arramma/i-