Thread:User talk:Rua/ἡλίκος (hēlíkos, “which size”), *velikъ and suchlike/reply (4)

I have said that it is confusing. This wording is misleading: “the same suffix as talis …”. Maybe I never followed the link, as it reminded me of cases like ‘’-(t)icus’’, see ‘’-cus’’ below.

Now that I read the entry I see what you mean: helix and talis, and maybe ‘’helikos’’ as well if this is to be inferred, do tentatively agree that *h2el- could be the second element, while *swe-, *to-, and *yos- would give each the respective initial element. It’s a bit messy, and there’s more.


 * Insofar I had implied that the remaining elements would match in *velikъ and helikos, you might object that I haven't read closely: The PSlav entry links *-ikъ without further PIE root, and notes that it were “Probably not related to Sanskrit -ईक (-īká) and Latin -īco”, whereas the AGr entry links *kos-, the same Latin ‘’-ico’’ links to, ultimately, listing e.g.: PSlav *-kъ and *-ъkъ, PGem *-gaz, Latin ‘’-cus’’, AGr -κός, PII *-kas, and *-kasa, *-ćasa … Meanwhile, the /i/ in ‘’helikos’’ et al goes unexplained, and I could not tell the /ks/ in ‘’helix’’, either.


 * ‘’-icus’’ effectively explains PIE “*-ikos” thus: “From i-stem + -cus, occurring in some original case and later used freely” (is there a shorter word for this meme, ‘’o.i.s.o.c.a.l.u.f.’’?). It does not list ‘’-ico’’ as relative, whereas ‘’-ico’’ is: “From -ō suffixed to words with stems ending in -ic (e.g. -icus)” (surely a case of o.i.s.o.c.a.l.u.f.). Lat. ‘’-o-’’ counts eight derivations, it’s a single vowel, I have to ignore this. A bit vexing, AGr. '’hos’’ implicates a diphtong ‘’*ey’’similar to *-ikъ, but in the wrong place, as ’’hos’’ is etymologized “From Proto-Indo-European *yós, *yéh₂, *yód (“who, which”), from the relative stem *yo-, from the anaphoric stem *i-, *ey-.” whereas “the *-i in *-ikъ thus reflects a diphthong *ey”, which is maybe implied by *velьjь, “From vele + jь with tense e becoming ь”? Indeed: *yos links *jь. What’s an i-stem?

This’s what the pages say. It “probably” does not contradict the comparison. I’m first of all focused on the semantics, though, which match so well that no explanation was needed. For a start it might be best to tie up the loose ends.

Dercksen assigns *welh1- to *velikъ indeed, further indexing a doublette *velъkъ with West Slavic variants.

Does that make sense sofar? Wellicht? 109.41.0.191