The latter gives the meaning 'Lunge' only; indeed, there is no context among those quoted in [CAD] which couldn't be interpreted as 'lung'. Note also h_išû 'a part of the human body' [CAD h_ 206]
Originally may have a more specified meaning 'lung(s)'; besides 'lungs' in AKK and 'asthma' in some of the ARB forms, cf. SAB ḥs2y-m 'asthma, whooping (?)' [SD 72] (see the context in [Biella 193]: d_-ʔs1yw ... bn ḥs2y-m w-sʕl-m "what they encountered of bronchial trouble and coughing"). According to Kaufman, a loan from AKK is possible, because the AKK, MND and ARB forms "can hardly be cognate since the Akkadian is almost certainly cognate with the word for "chest": Hbr ḥāze, ARM ḥăd_ê, ARB ḥid_aʔ-" [Kauf 55]. (Note that ARB ḥid_āʔ- /with long -ā-!/ is translated as 'vis-̀-vis, en face de', and not 'chest', in [BK 1 399]; cf. No. .) This is hardly the point. The MND haša 'bowels' [DM 127] does look like a loan, though for quite a different reason: -š- instead of the expected *-s- may point to an Akkadism or Arabism (cf. also h_ašašia do. [ibid. 128], with h_- typical of Arabic borrowings with ḥ). As for the AKK and ARB examples, they are, on the contrary, to be treated as cognates, since no reliable instances of SEM *d_ > AKK š are attested, whereas there are quite a few cases of SEM *ḥ reflected as h_ in AKK (see Introduction). [Holma 88]: AKK, ARB
