If we accept a semantic development ‘life, lifetime’ --> ‘eternal’ --> ‘that which is eternally valid’ --> ‘law’ (Lühr 2000: 188, Casaretto 2004: 201, Wodtko, Irslinger & Schneider 2008: 281 fn. 10), the possibility of an etymological connection between PGmc. *aiwa-/i-/ō(n)- ‘marriage; law, right’ (cf. OE ǣ, ǣw ‘law, religion, marriage’, OFris. ēwe, ē ‘law’, OS ēo, ēu ‘law’, OHG ēwa, ēa, ēo ‘law, right, will, contract’ and MHG ēwe, ē ‘marriage, alliance’) and Lat. iūs ‘law, right’ is present (Kuhn 1855: 374, de Vaan 2008: 316–317, Kroonen 2013: 16), although rejected by Wodtko, Irslinger & Schneider 2008: 283 fn. 17). We may ultimately derive both legal terms from 1) the acrostatic neuter u-stem noun PIE *h2ói̯-u- ~ *h2éi̯-u- ’life, lifetime’ reflected in Ved. ā́yu- ‘life, lifetime’, Av. āiiu- ‘life, span of a lifetime, time’ and the Greek negation οὐ ’not’ (Wodtko, Irslinger & Schneider 2008: 277, 279) and 2) the proterokinetic u-stem adjective derived from that, i.e., PIE *h2éi̯-u- ~ *h2i̯-éu̯- reflected in, e.g., Ved. yóḥ ‘well-being (gen.sg.), the inflections of which have mixed. Given this etymological connection, the Indo-Iranian forms built on the adjective *h2ei̯-u- ~ *h2i̯-eu̯- seem to provide the missing link between the Germanic and the Latin forms. The Germanic forms would thus most probably constitute a reformation of the original u-stem to a thematic stem, i.e., PGmc. *aju- --> *aiw-a- etc. (Casaretto 2004: 200–201), whereas Old Latin forms such as iouestod ‘justly’ (> Lat. iustō) point in the direction of an s-stem PIt. *jow-os ~ jow-es- ‘oath, law’ > OLat. ious, Lat. iūs ~ iūr-) ‘law, right’ built on the zero-grade form of the adjectival u-stem, i.e., PIE *h2i̯-éu̯- --> *h2i̯éu̯-os ~ *h2i̯éu̯-es- (de Vaan 2008: 316). Alternatively, as pointed out by Sihler (1995: 306), both Ved. yóḥ ‘well-being’ and Lat. iūs (OLat. ious) ‘law, right’ may be regarded as the nom./ acc. of a zero-grade-suffix s-stem, i.e., PIE *h2i̯éu̯-s (weak form: *h2i̯éu̯-es-). Following Sihler’s line of thought, we may wonder, however, why this type of s-stem (with archaic *-s instead of the generalised innovative *-os in the strong forms), which is standard with seṭ-roots (PIE *kreu̯ə2-s > Ved. kraviṣ-, Gr. κρέας ’flesh, meat’), is found with an aniṭ-“root” like PIE *h2i̯éu̯- (Sihler 1995: 307–308). Finally, despite its trivialness, we may also wonder why the semantic development to ‘law, right’ de-scribed above has happened only in Germanic and Latin in two derivationally dissimilar formations. In this paper, I propose a minor modification of the etymologies of both PGmc. *aiwa- and Lat. iūs. I claim that they ultimately hark back to a compounded root noun PIE *h2i̯éu̯-ə1s- ‘which is eternally true’ (i.e., ‘law, right’) consisting of the zero grade of the root *h1es- ‘be’ added to the locative of the u-stem PIE *h2ói̯-u- ~ *h2éi̯-u- ’life, lifetime’. For the secondary meaning of PIE *h1es- ‘be’ as ‘be true’, cf. the participles PIE *h1s-ént- ~ *h1s-n̥t- > Ved. sánt-, Av. hant- ‘being, good, true’ and PIE *h1s-ónt- --> PGmc. *sanþa- ‘true’ (Pokorny 2005 [1959]: 341). The full-grade form PIE *h2i̯éu̯-ə1s- would yield PIt. *jowis-, i.e., a form highly prone to being reinterpreted as an s-stem of the seṭ-root type (PIt. *jowis- --> *jow-(o)s ~ *jow-es- > Lat. iūs ~ iūr-), cf. PIE *kréu̯ə2-s ~ *kréu̯ə2-es- mentioned above, and through a Dybo’s-Law loss of *h1 (Zair 2006a, 2006b, 2012: 145–146), the zero-grade variant PIE *h2i̯u-h1s- could be what has ultimately yielded OLat. iusa (in Paul. Fest., p. 92: iusa iura) and the derived verbs dē-ierāre ‘take an oath’ and pei-ierāre ‘swear falsely’ (< *-iusāre) mentioned by Weiss (2019: 11 fn. 20). In Germanic, the word-internal *ə1 would disappear (PIE *h2i̯éu̯-ə1s- > pre-Gmc. *h2i̯éu̯s-), and the root-final s may have been reinterpreted as the nom.sg. case ending (pre-Gmc. *h2i̯éu̯s- --> *h2i̯éu̯-s), giving rise to a new standard root noun pre-Gmc. *h2i̯éu̯- ~ *h2iu̯-. PGmc. *aiwa- may thus constitute a thematicisation (Thöny 2013: 94–99, 124–127) – maybe partly influenced by PGmc. *aiwa- ‘eternity, age’ or *aiþa- ‘oath’ – of the pre-Gmc. zero-grade variant *h2iu̯-. This would both be in accordance with the system of predicting the radical ablaut grade of an inherited root noun in Germanic from its phonotactics (Hansen 2014: 39–40, 2016: 176) and phonologically regular, since Hansen (2014: 166–167, 170, 2015: 66–69, 71–72) has demonstrated the likelihood of a development PIE #h2iC- > #aiC-.