Are L. domus and L. domō cognates?

Question

domus

From Proto-Indo-European *dṓm, from root *demh₂- (“to build”). Cognates include Ancient Greek δόμος (dómos), Albanian dhomë (“a chamber, a room”), Sanskrit दम (dáma) and Proto-Slavic *domъ. The same Proto-Indo-European root also gave Old English timber (“building, act of building”) (English timber).

domō

From Proto-Indo-European *dem-h₂- (“to domesticate, tame”). Cognate with Sanskrit दाम्यति (dāmyati), Ancient Greek δαμνάω (damnaō), Old High German zemmen and the Proto-Germanic adjective *tamaz.

The following sentence makes me wonder and have this question: tame is a reflex of PIE *dem-h₂- (“to domesticate, tame”), which has the same formation of PIE *demh₂- (“to build”). Maybe the semantic explanation is "tamed in a house"?

tame

Old English tom, tam "domesticated, docile," from Proto-Germanic *tamaz (cf. Old Norse tamr, Old Saxon, Old Frisian, Middle Low German, Middle Dutch tam, Old High German zam, German zahm "tame," Gothic tamjan "to tame"), from PIE *deme- "to constrain, to force, to break (horses)" (cf. Sanskrit damayati "tames;" Persian dam "a tame animal;" Greek daman "to tame, subdue," dmetos "tame;" Latin domare "to tame, subdue;" Old Irish damnaim "I tie up, fasten, I tame, subdue"). Possible ulterior connection with PIE *dem- "house, household" (see domestic). Meaning "spiritless, weak, dull" is recorded from c.1600.

Answer 1

From the literature, I get the impression that this is a complicated and possibly unsolved issue. Three relevant sources I found but don't have time to analyse now:

• De Vaan, Etymological Dictionary of Latin and the Other Italic Languages (2008):

  • On domo:

    PIE *domh2-eie/o- 'to tame'.

  • On domus:

    PIE *dōm, *dem- 'house'; *dom-o-?

• Ernoud & Meillet (1985) on domo ("domus and domo are not related):

  Sur domō et domus, voir l'important article de M. Benveniste, 'Homonymies radicales en indo-européen', BSL 51, 1955, p. 14 sqq. Il démontre péremptoirement que les deux mots n'ont, à l'origine, rien de commun et que domus, de son côté, doit être séparé de la racine *dem(ə2)- "bâtir".

• Brouillon du texte paru in Françoise Bader (éd.), Langues indo-européennes, Paris, Éditions du CNRS, 1994, 43-64: p. 53:

  Ainsi, Benveniste a-t-il par exemple postulé que gr. dómos "maison" résulte de la confluence tardive d'un dérivé thématisé d'un nom-racine *dom, terme institutionnel désignant une institution politique, un cercle de parenté, avec un dérivé thématique de la racine *dem-(h2)- "construire" (1955, 15-29; l'article contient des explications analogues pour d'autres racines).