Reconstruction talk:Proto-Indo-European/méǵh₂s

Declension table
Is there a reason this is missing the *-s in the masc.nom.sg.? — JohnC5 23:12, 15 May 2016 (UTC)
 * Ah, it's because of the logic that omits the final -s after -h₂. I'll have to think of a way to work around that. —CodeCat 23:47, 15 May 2016 (UTC)
 * I notice that all of the words that lack -s in the nominative are those that have an oblique stem ending in -eh₂-. So that includes the thematic eh₂-nouns, but also feminine adjectives whose oblique stem ends in -yeh₂-, and also the athematic and, while simultaneously explaining why  does have the -s. Is this a rule worth implementing in the module, or is it too far fetched? —CodeCat 23:52, 15 May 2016 (UTC)
 * Hmmm, that would be an odd rule, but root adjectives ending in *h₂ do seem to cause a problem (though I also imagine they are exceedingly rare).
 * As to, which we were discussing before, Ringe's reconstruction is a bit strange to me, especially since it comes just a handful of pages after his discussion of Szemerényi's which definitely should render *h₁widʰéwh₂ as *h₁widʰḗw, which seems even more difficult to propose than *h₁widʰéwh₂. Furthermore, De Vaan would much prefer to reconstruct a thematic adjective based on Italic and Hellenic since he feels that the backformation of and  from a feminine is less likely than a thematic adjective appearing frequently in the feminine because of the prevalence of widows over widowers.
 * I thought Ringe listed as a morphologized exception, and I find mystifying and unmotivated (why the *s). Thus far, I feel the rule should be *s → *∅ / eh₂_# with maybe an exception for *dn̥ǵʰwéh₂s. — JohnC5 03:00, 16 May 2016 (UTC)
 * I've changed the rule to that now. But now the feminines of athematic adjectives are all broken. —CodeCat 15:46, 16 May 2016 (UTC)

Indo-Uralic Root?
See Hungarian nagy.