On 2008-02-26 14:50, alexandru_mg3 wrote:
> With all my respect for Jens, I think that there is no -t or -nt
> *dHi-dHéh1-t/*dHé-dH(h1)-n.t
>
> See Jassanof:
>
> "
> 1) On the Germanic side, the discovery in 1963 of the Runic 3 sg.
> talgidai (Nøvling, c. 200), which rules out the traditional
> reconstruction of the 3 sg. ending as PGmc. *-de: < *-de:t < PIE *-
> dheh1-t;
> "
Jasanoff is so badly in want of some attestation of perfect passives
that he eagerly interprets a Runic hapax (otherwise completely unknown)
as one, thought the spelling, which is normally <-ide> and only
sporadically <-ida(i)>, may be explained in various other ways, e.g. as
an ad hoc attempt to represent Proto-Norse unstressed *-æ:, for which no
special rune existed. Jasanoff's reconstruction of the weak preterite
also involves reduplicated forms of *dHeh1-, but he uses a hypothetical
middle perfect (*ðeðai < *dHe-dHh1-ói 'factus est' = Ved. dadHé) rather
than an imperfect (*ðiðe: < *ðiðe:ð < *dHi-dHéh1-t). Why he insists that
clearly dialectal constructions are "PIE" is beyond me.
Piotr