*anhtō

Probably from pre-Germanic *h₂ank-t-, ultimately connected with the root *h₂enḱ- reflected in Hittite 𒄭𒅔𒅗 (ḫi-in-ka- /ḫink-/); Ringe explicitly derives *h₂ank-t- > Proto-Germanic *anhtō (Ringe 2017: 117; Rix et al. 2001: 268, s.v. *h₂enḱ-).

Dardano (2020: 41–49) lends further support to this interpretation by analyzing Hittite 𒄭𒅔𒃷 (ḫi-in-kán /ḫe/inkán/) as a deverbal abstract noun derived from, with the semantic development ‘that which is allotted’ → ‘allotted portion, lot’ → ‘fate’ → ‘ill fate, death’ → ‘plague’. Following García Ramón, she favors treating PIE *h₂enḱ- as a Schwebeablaut variant of *h₂neḱ-, while distinguishing this root from PIE *h₂enk-. Old Irish écht < Proto-Celtic *anxtu is probably related, although its deeper analysis remains disputed: Matasović (2009: 37, s.v. anku-) derives it from PIE n̥ḱ-tu- beside *neku-, whereas Irslinger (2002: 96–98) observes that zero grade is unexpected in this type of -tu- noun and instead favors derivation from *h₂enḱ-, supported semantically by Hittite 𒄭𒅔𒅗 (ḫi-in-ka- /ḫink-/) and its derivative 𒄭𒅔𒃷 (ḫi-in-kán /ḫe/inkán/), which demonstrates the development ‘allotted portion, lot’ → ‘fate’ → ‘death’.

If Irslinger’s analysis is correct, Germanic and Celtic preserve parallel inherited dental formations, and no borrowing need be assumed. The alternative derivation from *h₂enk- is much less likely, since LIV explicitly distinguishes that root from *h₂enḱ- and the former offers no convincing semantic route to ‘outlawry, persecution’ (Rix et al. 2001: 268, s.v. *h₂enk-). Orel’s (2003: 20, s.v. *anxtō) association with the Germanic ‘narrow, constricted’ family and thus ultimately with *h₂enǵʰ- remains phonologically possible, but the direct convergence of Germanic *anhtō, Celtic *anxtu, and Hittite 𒄭𒅔𒃷 (ḫi-in-kán /ḫe/inkán/) around ‘fate, death, slaughter, outlawry’ makes the *h₂enḱ- analysis appreciably stronger.