*g̑^hod-m(e)rs(-i̯é/ó)-

Hitt. kammarš-^(mi) and CLuw. katmarši-^(ti) have been reconstructed as denominal verbs based on an original Indo-European formation *g̑^hod-mer- (see Rieken 1999a:363, HEG A-K:474 and HED K:38, all with further literature). Their assignment to the PIE root *g̑^hed- is hardly debatable, which however requires certain explanations vis-à-vis the unexpected phonology and morphological of the Hittite and Luwian cognates. First, the retention of the voiced velar *g̑^h in Luwian is a priori exceptional, but Melchert 1987a:186 points out that it could be attributed to the “affective” value of the word (if not before back-vowels). Second, the assimilation of the cluster -mm- in Hittite implies that Luwian retains an original cluster *-dm- and is therefore more conservative (cf. EDHIL:432 with doubts). Third, the -s- morpheme remains unexplained as well because the athematic conjugation of the Hitt. kammarš-^(mi) (cf. also the Hittite formation kammaršniye/a-^(mi) with an apparent extension in -niye/a-^(mi)). Unfortunately, since the nominal base is not attested, we cannot verify whether the -s- was already part of the nominal formation. Rieken 1999a:363 (and HEG I-K:474) assumes a denominal suffix -s-, comparing Hitt. ištamašš-^(mi) ‘to hear’ (next to nominal ištāman- n. ‘ear’). However, the -s- has also been taken as some sort of extension (see EDHIL:413) or as a compounded verb šiye/a- (HED K:38). However, in the case of ištamašš-^(mi), we must be looking at a reanalysis of the root followed by a deradical derivation in -šš- (< *-h₁s-). A different approach considering that Hitt. kammarš-^(mi) is a priori an athematic “root” formation is to see a univerbation based on a root noun *g̑^hod- and a root verb *mers- (or *mr̥s-i̯é/ó- for Luwian katmarši-^(ti)), probably the same root as the one of PAnat. *mr̥s-éh₂- ‘false act’. Although this scenario requires more evidence, it is morphologically more credible than to assume that the Hittite athematic formation had been secondarily created after the root had been reanalysed as kammarš- or, if it underlies a formation in *-(h₁)s-, as kammar-.

[D.S.]

In all branches in which verbal continuants of *g̑ʰed- are attested, the verbs belong to a colloquial or vulgar register and denote the process of defecating. For this reason, it is not surprising that Sanskrit hadati, hadate is attested only in Classical Sanskrit, not in Vedic, and that Ancient Greek χέζω was not used in epic poetry but first occurs in Aristophanes’ plays.

For the reasons given above under II. Semantic Reconstruction, it is perhaps not surprising that Hittite kammarš-^(mi), Luwian katmarši-^(ti) do not have morphological parallels attested in other branches. Note that Tocharian B †kenmer “Exkrement” is a ghost word (cf. Malzahn 2014b:197).

[S.H.]

Hittite kammarš-^(mi) ‘to defecate’, Luwian in Hittite Transmission katmarši-^(ti) ‘to defecate, crap’.See EDHIL:432.

Classical Sanskrit hadati, hadate ‘shits’.Cf., e.g., EWAia III:532 s.v. HAD, Werba 1997a:267f. s.v. had,LIV²:172 s.v. *g̑^hed-.

Avestan zadah- m. (in dual) ‘buttocks, rump’, apa-zaδah- ‘with the rump turned upwards’.Cf., e.g., EWAia III:532 s.v. HAD, Stüber 2002a:84–86.

Ancient Greek χέζω ‘ease oneself, shit’, perf. κέχοδα, aor. χέσαι and χεσεῖν, fut. χεσοῦμαι, χέσομαι, perf. ptcpl. pass. κεχεσμένος (Ar., etc.); χόδανος ‘rump’ (Hsch.).Cf., e.g., GEW II:1078f. s.v. χέζω, LIV²:172 s.v. *g̑^hed-.

Armenian jet, -oy ‘tail (of animals), penis’.Cf., e.g., Olsen 1999a:47, EDAIL:432.

Albanian dhjes ‘to shit’, ndot ‘dirty’.Cf., e.g., Schumacher & Matzinger 2013a:236, LIV²:172 s.v. *g̑^hed-.