Reconstruction talk:Proto-West Germanic/oli

PWG?
I don't think this can be of PWG date. Given the stem shape, you'd expect to see geminates in the descendants, but there's none. There's also the vowel *o in an umlauting environment, which wasn't possible until later. I think that points to a later borrowing. —Rua (mew) 13:00, 27 January 2021 (UTC)
 * Maybe considering that the mentions for compounds “oil beam” and “oil berry” (scilicet olive tree and olive I assume) allow for too much phantasy. Did they know the olive tree or not? However it is not much plausible that in  they did not know Roman olives and oil, we can presumably even dig out archaeological findings for it, so on the contrary it could even be Proto-Germanic. But for the umlauting environment – perhaps Leasnam believed that this is how we notate an umlaut, because About Proto-West Germanic does not allow for any notation of umlauts? What does About Proto-West Germanic say about umlauts? So I am not sure this is an argument. Fay Freak (talk) 14:45, 27 January 2021 (UTC)
 * The change PG *u > PWG *o was blocked in an umlauting environment, so in principle PWG would only have had *u in cases like this. Likewise, PG *lj would have become *llj in PWG, which we again see no evidence of in the descendants of this word. That would mean that the word was borrowed after both processes ceased to be productive, but when was that? PWG is only dated by Ringe to a relatively small timespan of the 2nd to 4th centuries, before the first evidence of dialect-specific changes. —Rua (mew) 11:31, 28 January 2021 (UTC)
 * There is no standalone form in Old Saxon, but we do see a plethora of variations in Middle Low German, some of which do show possible gemination of the l, no ? Leasnam (talk) 14:58, 28 January 2021 (UTC)
 * PWG from Latin is a example of *u > *o not occuring. That said, *u ~ *o is somewhat erratic even in native terms, especially before resonants. --  04:01, 23 April 2021 (UTC)
 * PWG *u ~ *o vacillates somewhat only before low vowels; in front of *i and *j, it's consistently *u in other cases, as far as I can see, and earlier loanwords even tend to be adapted accordingly. Rather, PWG *oli, gsg. *oljas (apparently no evidence of gemination, except uncertain evidence in MLG) is definitely an unusual reconstruction, pointing to a late loanword (definitely after the – very late – West Germanic consonant gemination before *j). It's probably a relatively late, not completely adapted loanword, and I'd date it to the 4th century at the earliest. It may still have been loaned into West Germanic at the PWG stage (after the gemination had happened, but before the diversification), but it could also be a later loanword, even independently loaned into the West Germanic dialects. --Florian Blaschke (talk) 19:34, 10 September 2021 (UTC)