damaš-ti

Regarding the Anatolian semantics, the meaning of PAnat. *dmaHs- ‘to press (physically)’Cf., e.g., EWAia I:698 s.v. DAM^I, Friedrich 1926a:40f., who cites the instances in KBo 4.2++ obv. i 38–46, where the verb’s object is a purpura- ‘clump’.is primary, and ‘to oppress (psychologically)’ secondary,Cf., e.g., Ved. damāyan in RV.6.47.16 with a human object.cf. Buck 1949a:575. The meaning ‘to tame’ of PIE 2.*demh₂- can be established only for a fairly recent stage of Core IE, while the original meaning has to be approximated as ‘to press (physically)’.

A connection of PIE 2.*demh₂- ‘to press, oppress’ (contra LIV²:116f., cf. above) with PIE 1.*dem- ‘to assemble, build’ is frequently assumed.Cf. IEW:199, LIV²:116 n. 4, de Vries 1977a:581, Orel 2003a:401, Heidermanns 1993a:590, 595, Kluge & Seebold 2011a:1001 etc.A factitive relation seems plausible: ‘to build’, ‘house’ → ‘to make appropriate for living by a house’ > ‘to domesticate, tame’. However, since the primary meaning of PIE 2.*demh₂- is ‘to press (physically)’, this development is impossible.
In order to connect both roots, the original meaning of 1.*dem- ‘to build, assemble’ can hence be approximated as ‘to press on, compress (clay vel sim.)’,For building techniques using clay to fill gaps and plaster to smoothen walls, cf. Gamkrelidze & Ivanov 1995a I:612f. (on “Indo‑European pottery terminology”). An IE parallel is the root PIE *d^hei̯g̑^h- ‘to smear, spread, knead’ (LIV²:140f.), which gave rise to words meaning ‘wall’ etc. (cf. Gk. τεῖχος [n.], τοῖχος [m.] ‘wall’ etc.), on the one hand, and OIr. dingid ‘to oppress’, on the other hand. Hence, the meaning of PIE *d^hei̯g̑^h- should be widened to ‘to apply (a layer of malleable or spreadable material)’ with an already PIE development to ‘to mould, shape’, cf. Meiser 1993a:301–305. Furthermore, cf. OCS gnetǫ ‘to press, knead’ beside Russ. gnestí ‘to press, oppress’ (Buck 1949a:576).thus:
(early) PIE *dem- ‘to press (physically)’

1. > PIE 1.*dem- ‘to press on, compress (clay vel sim.)’ > ‘to build, assemble’ (cf. Lyd. tam- ‘to build’)
2. → PIE 2.*demh₂- ‘to press (physically)’ (cf. PAnat. *dmaH‑s‑) > ‘to oppress (cf. PAnat. *dmaH‑s‑); to tame (cf. Gk. δάμνημι)’

The meanings ‘to build, assemble’ and ‘to (op)press’, might exhibit a polysemy that already existed in PIE. For the root-final *‑h₂- in PIE 2.*demh₂- ‘to press’, cf. below.

Taking into account esp. the Greek (δάμνημι < *dm̥‑ne‑h₂‑) and Latin (domāre < *domh₂‑éi̯e/o‑) evidence,The Vedic reflex of the nasal‑present is rebuilt in a productive way and thus not informative, viz. *dm̥‑n‑h₂- → *dm̥‑n̥h₂‑i̯é/ó‑, cf. LIV²:117 n. 4.the root in question has to be set up as a seṭ‑root with a second laryngeal in the coda. Since PAnat. *dmaHs- cannot be interpreted as the reflex (or preform) of a sigmatic aorist (cf. above), the root-final ‑s- can be aligned with other cases of a root enlargement ‑s-, cf., e.g., *peh₃(s)- ‘to drink’ (cf. [lemma link]), PIE *meu̯(s)- ‘to close oneself’, and maybe PIE *h₂eu̯‑s- ‘to pour, scoop’ (or *(s)h₂eu̯‑s-). PAnat. *dmaHs- is interpreted in this lemma as an extended variant of Core IE 2.*demh₂- ‘to press (physically)’.

While the common ancestor PIE *dem- ‘to press (physically)’ (cf. PIE 1.*dem- ‘to assemble, build’) can be easily reconstructed semantically, the phonological difference in the form of the root-final *‑h₂- still lacks an explanation. Likely parallels can be PIE *men- ‘to think’ vs. *mn‑eh₂- ‘to think, remember’ [lemma link], *h₁ei̯- ‘to go’ vs. *h₁i̯‑eh₂- ‘to move along’ etc., thus PIE *dem- ‘to press’ vs. PIE *dm‑eh₂- ‘to (op)press’ (> PAnat. *dmaH(s)‑,Cf. also OAv. d^ǝmāna‑, OInd. mā́na- ‘house’ < *dméh₂‑no- (Steer 2015a:49f. with similar examples).→ PIE *demh₂- ‘to oppress, tame’).The full grade I (*demh₂‑) can be explained by reanalysis of the Core IE nasal‑present *dm̥-né‑h₂- and other zero‑grade formations of PIE *dmeh₂‑: R(Ø) *dm̥h₂- → Core IE R(e) *demh₂‑.

[A.O.]