Cf. Kogan 2015: 189
Level? In spite of the well-known phonological difficulty (Ugr. d presupposes *ḏ in the prototype, as against *z patent in Aramaic and Arabic (One cannot exclude, however, that the root-variant with *ḏ is behind ḥḏw (III) ‘to face, to be in front of)), it is very likely that all the aforementioned forms have the same etymological background (cf. DRS 838‒839, 854). The dialectal distribution of this root with respect to PWS *rˀy is no less puzzling, particularly in EthS where it seems to be completely absent from the languages of the northern branch (unless one compares Tgr. ḥaza ‘to seek, to try’, semantically not improbable, but has become highly prominent almost throughout the southern one
Remarkably, PS *ḥaḳl- is not attested (presumably, lost) in Canaanite.