*h1or+? (was: Renfrew's theory renamed as Vasco-Caucasian)

--- In cybalist@yahoogroups.com, Piotr Gasiorowski <gpiotr@...> wrote:

> There's also 3pl. aru in Old Swedish. In OE, all the dialects have <a>
> or <ea> (i.e., underlying *a), for which there's no analogical model,
> and it's the northern (Mercian, Northumbrian) forms that have final þ,
> which can't be explained as ON-influenced. It's quite clear that 2sg.
> *izi (in all OE dialects) and 3pl. *sindi (in some dialects) were
> replaced by suppletive forms borrowed from a different paradigm. The

/a/

> vocalism points to a preterito-present derived from *(h1e-)h1or-

If truly from the perfect of *h1es+ the 3pl *-e:r could have spread
its -r to other forms in a haphazard way after it became unproductive
and unalyzable as to its previous function. Analogical *oze:r > *arar
could be from haplology, etc.


There are other possibilities, also:

stx > rtx after V (so no fric/stop/fric) with most having analogy


a change of the perfect to middle forms (perhaps in Baltic if

*xYexYsro > *xYixYra > *i:ra

with e>i by CY as sometimes (*ekYs > *is^s.)


and of course I wouldn't say analogy in the past afer z>r couldn't
have happened in Germanic.