Re: [tied] Water, pre/postpositions, somewhat OT

>
> The oldest attested forms of ibai "river" are hibaie-a
> (def.) and hibaia (indef.), pointing to *hibaiV (V=a or e).
>
> >If so, I would consider matching Basque <ipili>, <ipini> "be busy
> >with sth." with Latin <opulentia>, ON afl "harvest"
>
> No such verb. Ipini means "to put", and there is no *ipili.
> Perhaps you mean <ibili> "to walk around", caus. <erabili>
> "to make go, to use, to wear".

Right. Of course.

> The root of the verb <ibili> (*e-bil-i) is clearly *bil
> "round" (e.g. gurpil "wheel" < *gurd-bil "wagon + BIL").
>

Hm. Well IE <ambhi> means around too.

> The root of <ipini> and its variants such as <ibeni>,
> <imini> is more difficult to pin down. My guess is that
> i-pin-i represents *e-b(V)hin-i, while i-ben-i, with i-
> pointing to a high vowel in the root, is *e-beh(i)n-i or
> *e-bain-. <imini> is straightforward from *e-bin-i. If we
> combine all these, we get a possible root *-behin- or
> *-bahin-, where /h/ can be read as the hiatus left by the
> loss of an earlier consonant (not /n/).
>

I recognize that line of reasoning wrt. /e/ > /i/ in the verbal
prefixes from Trask. Could the same rule have applied also on nouns,
so that ibai < *ebai (there's nothing to stop one from assuming a
lost h- in all verbs)? In that case the two roots are getting pretty
close.

Torsten