On 2006-01-26 21:50, andrew jarrette wrote:
> Is my understanding correct that you're not suggesting here the
> origin of the -ê- of the preterite plural of Germanic Classes IV and
> V? Or are you implying that it traces its origin back to
> reduplication (e.g. of verbs beginning with *e (*h1e))? Or are you
> saying that e-infixation and o-infixation was a Germanic innovation
> in the preterite? That's a new one for me, this is the first I've
> heard of that idea. I will have to read some of those articles you
> recommended to me about this one. I wonder whether it was truly
> infixation or just spontaneous innovative lengthening (for
> contrastive purposes).
In the context of Germanic, I will use *æ: (< PIE *e:, *eh1) for
Germanic *e:1, and *e: for NW Germanic *e:2.
Let's begin with the conjugation of *beran-, pret. *bar, *bæ:run. One
would expect a reflex of the nil grade in the last form, perhaps
something like *burun, as in the past participle. There's no reason why
speakers of (pre-)Proto-Germanic should have felt a need to introduce a
lengthened full vowel here if they tolerated the same vocalism in the
preterite plural and the past participle in the first three declensions.
But if the preterite comes from the PIE perfect de-reduplicated already
in pre-Germanic times, it's imaginalble that a pair *bHebHór-, *bHebHr-'
ended up as *bHor-, *bHe:r- through some kind of haplological reduction
(carried out so as not to leave the stem asyllabic), combined with
compensatory lengthening. Compare *wewórt-, *wewr.t-', where the heavier
root makes it possible to drop the reduplication syllable in both forms
without any undesirable consequences.
We may be dealing with a more straightforward reflex of reduplication
(as in Class VII) in the case of *etan-, pret. *æ:t, *æ:tun, if these
forms go back to *h1e-h1ód-, *h1é-h1d- levelled out as *e:d- > *æ:t-.
The synchronic recipe for the preterite plural *bHe:r- in pre-Germanic
after the changes described above was "take the present stem *bHer- and
lengthen its vowel". The same rule was then employed for the o-grade
presents in Class VI, producing pret.pl. *po:r- as the counterpart of
pres. *por-, etc. Their original preterite must have been similar to
that of Class VII verbs with *oi, *ou, oRC (i.e., indistinguishable from
the perfect of e-grade presents). De-reduplication would have
obliterated the important contrast between the present and the preterite
singular stems, hence the use of the vocalism of the preterite plural
also in the singular.
Piotr