*h₁u̯eh₂s-d^heh₁-

Although we find solely nominal formations attested in Luwian, it is clear that one has to depart from the Hittite root verb wašta-/wašt-^(ḫi) ´to sin, offend´. On account of its atypical root structure, one must explain it either as a neo-root or as a univerbation. A u-stem is also found as the base of Hittite waštul- (var. uštul-) n., i.e. *-ú-lo- (with post-tonic syncope) and the Luwian root shape wašku- (see below under Derivatives). The Luwian sound change *du̯ > /kw-/ (see kuwayata-⁽ⁱ⁾) implies that a voiced alveolar stop has to be reconstructed. Therefore, it is highly tempting to see a univerbation of waš- (< *u̯os- or *u̯as-) + da-/d-⁽ⁱ⁾ < (*d^hóh₁-/d^hh₁-´; cf. kuwayata-⁽ⁱ⁾); contra EDHIL:986 “*u̯esTh_2/3-”; see PIE lemma below for a justificaiton of the reconstruction as *h₁u̯eh₂s-d^heh₁-.

Obviously, this univerbation led to a neo-root wašt-, which was then nominally derived as a thematic neuter stem, i.e. CLuw. wašt(a)-, a Hittite diphthong stem waštāi- and a u-stem *waštu-, all sharing similar semantics. After the Luwian sound change *du̯ > /kw-/ occurred, the original u-stem wašku- or the newly created thematic stem *waškuwa- served as the base for additional formations, e.g. waškui(t)-, waškuwallimma/i- (see below).

For various Indo-European etymologies that are phonologically or morphologically intolerable, see Oettinger 1979a:504 n. 22, Melchert 1994a:50, HEG W-Z:407f.A root *h₂u̯óms- with Saussure’s Law and loss of nasal is phonologically highly improbable, since the sequence *-ó-ms- would have been expected to simply dissimilate to Hitt. -á-nš-; cf. the reconstruction under Pal. āmša^­-⁽ⁱ⁾.

[D.S.]

Words for ‘sin’ or ‘offense’ can have a variety of semantic origins (see Buck 1949a:1181f.), which makes etymologizing terms such as Hittite wašta-/wašt-^(ḫi) ‘to sin, offend’, waštul-, uštul- n. ‘sin, iniquity, offense’, Luwian wašt(a)- n. ‘sin, offense’, etc. rather difficult. For a critique of earlier proposals, see above. A possible alternative, going back in essence to Hrozný, Sturtevant, Laroche, and Puhvel (see Puhvel 1992a:6) shall be outlined and expanded on here. The starting point for this etymology is the root *h₁u̯eh₂- ‘leave, abandon, lose’ (see LIV²:254; Nussbaum 1998b:15–45, 73–84), attested in verbs such as Vedic vā́yati ‘fades away, becomes empty, lacks, loses’ and Ancient Greek ἐάω ‘abandon, leave behind, leave alone’ (whence ‘let be, allow, permit’; see Nussbaum 1998b:79). An absolute intransitive meaning ‘quit, fail, be deficient’ (see Nussbaum 1998b:81–83) is suggested by Latvian vâjêt ‘become weak’ and a couple of nominal formations, including adjectives in *-(o)no- such as *h₁uh₂-no- > Vedic ūná- ‘wanting, deficient, defective’ (AV+; cf. án-ūna- ‘unfailing, inexhaustible’ RVNussbaum 1998b:83), Avestan ūnā- ‘deficient (offering)’, Armenian ownayn ‘empty’, *h₁u̯eh₂-no- > Latin vānus ‘empty’, *h₁uh₂-ono- > Proto-Germanic *wanaz ‘deficient, lacking (in)’ (Gothic wans, Old Norse vanr, Old English, Old High German wan, etc.). It is not surprising that formations in this semantic range can adopt meanings of negative value judgment (e.g., Latin vānus ‘empty’ and ‘false, lying, deceptive, delusive, untrustworthy’, and further English fail and failure, German fehlen ‘to be missing’ and Verfehlung ‘wrongdoing, transgression’). In this regard, it is important to note that Hittite wašta-/wašt-^(ḫi) also means ‘to miss (a target)’ (cf. Puhvel 1992a:7; HEG W-Z:406), which might preserve the older meaning.

As mentioned above under Reconstruction, the unconventional phonological shape of Hittite wašta-/wašt-^(ḫi) and its congeners suggests that we are dealing not with a primary, but rather with a secondary root or stem of some sort. One famous case of such a secondary root/stem originated in a syntagma *mén-s d^heh₁- ‘to direct one’s mind on sth., to memorize’ (cf. Schindler 1975b:266), with the old acc. sg. *mén-s of the s-stem that is attested as *mén-os n. ‘mind’ in Vedic mánas-, Avestan manah-, Greek μένος. The syntagma is attested as such in Avestan (Young Avestan mąs … dā- ‘to memorize’, univerbated as mązdā- ‘id.’; cf. also Vedic mandhā-tár- ‘pious man’) and is the basis for stem compounds with ablaut in the first element such as Avestan mazdā- ‘knowledgable’ (*mn̥-s-d^heh₁-), mazdā- f. ‘wisdom’, Vedic medhā́- f. ‘id.’. Formations such as the *-ro-adjectives Avestan mązdra-, Vedic medhirá- ‘wise’ (from virtual *mensd^hh₁-ró-) indicate that a secondary root *mensd^hh₁- was abstracted at some point (cf. EWAia II:378 s.v. medhā́- with literature).

The reconstruction of the Proto-Anatolian noun ‘sin, offense’ might be a comparable case. The starting point would be the syntagma of an s-stem *h₁u̯éh₂-s ‘deficiency’ plus *d^heh₁- ‘place; make’, meaning ‘to make a deficiency’ (vel sim.). The root-initial *h₁ would regularly disappear on the way to Anatolian (cf. EDHIL:76); the fate of *h₂ in the position before *s (plus/minus another consonant) is contested, however. Under the assumption that it assimilated to *s (see Höfler 2023a:84 note 7 for a discussion), the resulting *wašš+da-/d-⁽ⁱ⁾ may be seen as the source of Hittite wašta-/wašt-^(ḫi) ‘to sin, offend; to miss (a target)’. The derivation of a u-stem *waš(š)t-u- (cf. Hittite waštul-, uštul- n. ‘sin, iniquity, offense’, Luwian wašku(i)- c. ‘sin, failure’) indicates that *waš(š)t- had been reanalyzed as a root within Anatolian (cf. *mensd^hh₁-ró- from above), perhaps even earlier (see below). Note that geminate /ss/ (i.e., in *waššt°) is not expected to be written before a stop (see Kimball 1999a:444).

While there is no direct evidence for the s-stem *h₁u̯éh₂-(o)s n. ‘deficiency’, it might be indirectly attested in Celtic, Italic, and Germanic, which all exhibit continuants of an adjective *h₁u̯eh₂sto-: Old Irish fás ‘waste, vacant, empty, void’, Latin vāstus ‘empty, unoccupied, waste, desert; vast, immense’, Old English wēste ‘desert, waste’, Old High German wuosti ‘id.’, German wüst. Formally, *h₁u̯eh₂sto- looks like a possessive *-tó-derivative based on the s-stem *h₁u̯éh₂-(o)s (cf. Heidermanns 1993a:688), comparable to *h₂ei̯d^h-s-to- ‘having heat’ (cf. Latin aestās ‘summer’ < *ai̯sto-tāt-; Proto-Germanic *aista- ‘hot’ > Old English āst ‘kiln’). Nussbaum 1998b:80, however, remarks that the adjectives (especially in Latin and Germanic) have a consistently passive meaning ‘deserted, desolate(d)’, rather than simply ‘empty’, which might speak for an origin as a verbal adjective. But rather than resorting to the stem of the sigmatic aorist as the basis for this verbal adjective in *-to- (as per Nussbaum 1998b:81), one could analyze *h₁u̯eh₂sto- as underlyingly *h₁u̯eh₂sd^h(h₁)-to-, i.e., a verbal adjective derived from a secondary root *h₁u̯eh₂sd^h(h₁)- ‘to make deficient, lay waste’, another avatar of the syntagma *h₁u̯éh₂-s ‘deficiency’ + *d^heh₁- ‘place; make’ that was hypothesized above. This would indicate that the reanalysis as a root had already happened in PIE. Note, however, that the coexistence of Avestan mąs … dā- ‘to memorize’, mązdā- ‘id.’, mazdā- ‘knowledgable’, and mązdra- ‘wise’ suggests that different degrees of univerbation and abstraction could exist side by side at any synchronic stage.

[S.H.]

Hittite wašta-/wašt-^(ḫi) ‘to sin, offend; to miss (a target)’, waštul-, uštul- n. ‘sin, iniquity, offense’, waštāi- c. ‘sin, offense’, Luwian wašt(a)- n. ‘sin, offense’, wašku(i)- c. ‘sin, failure’, etc.

Latin vāstus ‘empty, unoccupied, waste, desert; vast, immense’ (Plaut.+).Cf., e.g., EDLIL:655f. s.v. vāstus.

Old Irish fás ‘waste, vacant, empty, void’.Cf., e.g., EDPC:403f. s.v. *wāsto-.

Old English wēste ‘desert, waste’, Old High German wuosti ‘id.’, German wüst ‘id.’.Cf., e.g., EDPG:593 s.v. *wōstu-,Heidermanns 1993a:688.