*(h₁)ass-

Concerning Hitt. āšš-^(mi), Kloekhorst (EDHIL:214f.) suggests reconstructing a pre-form *h₁eNs-, which would have yielded Hitt. āšš- (with geminate -šš-; comparing Hitt. ḫāš-/ḫašš-^(ḫi) ‘to give birth’ < *h₂óms-/h₂ms-´), although he admits that this would be a back-projection without any other cognate. If we adduce Luwian aššatta- ‘peace’ to this root, the assimilation of the nasal would have to be postulated as well. On the other hand, it is also possible that we are dealing with a different type of assimilation, i.e. PAnat. *h₁ah₁s-. Concerning the semantics ‘to remain, stay’, a connection to the root *h₁es- ‘to be’ and *h₁es- ‘to sit’ is attractive, but formally difficult to account for (and rejected by Puhvel in HED A:189 Čop 1969b:196 suggests reconstructing a PIE root *as-, adducing Lat. inānis ‘empty’, although he admits that the geminate -šš­- in Hittite would remain unaccounted for.) because of the a-vocalism of the reduplicated syllable, unless one adduces inner-Anatolian parallels e.g. in Hitt. lālukkiye/a-^(mi), lālukkima- c. and in Pal. tāzzu- ‘to take’ < *dó-dh₃u̯- see Melchert apud Sasseville 2019a:25 n. 13), and because of its mi-conjugation (for expected ḫi-conjugation; cf. Hitt. ēš-/eš­-^(mi) ‘to sit’ → ašāš-/ašeš-⁽ⁱ⁾ ‘to cause to sit’). Nevertheless, a Proto-Anatolian neo-root *(h₁)ass- is here tentatively reconstructed, which could come from an inner-Anatolian o-reduplication, i.e. *h₁ó-h₁s- (← *h₁es- ‘to sit’). Its reinterpretation as a neo-root may in turn account for the mi-conjugation, although this claim might not even be necessary, if one looks at the double conjugational type of papparš-^(ḫi) ‘to sprinkle’ versus papparš-^(mi) (both found in Middle Script texts).

[D.S.]