tgpedersen wrote:
> 'Iterative' 'be'?
Did you read carefully? It was an original aorist stem, so the meaning
can't have been the same as that of *h1es-. It's usually glossed as
'grow, become' ('happen' might be a better approximation), and its
suppletive relationship to the athematic present *h1es- is post-PIE. The
*bHeuh2- root does form iteratives, cf. Slavic *byvati 'to happen or
come regularly'.
>>Forms like OHG bim (OE beom), bis(t) (not to mention
>>birum, birut) are not inherited but analogical.
>
>
> Analogical to what?? Afaik 'pim' is the only verb in OHG with that
> athematic ending. And it's 'pim'; 'bim' is taken over from dialects
> further north.
The p/b variation is of course the orthographic effect of the HG shift
and can be ignored here. The word is a hybrid between *im < *ismi <
*h1esmi (Goth. im) and *bijo:.
> One thing that's wrong is your presentation of it: Induction of the
> style Popper didn't like results in an unstated rule from which you
> deduce the rule you desired. That's not proof. It shows possibility,
> not necessity.
The "labial-labial ban" on /fw, vw, pw, bw, mw/ is still there in Modern
English, although it's no longer absolute: the clusters are marginally
acceptable since they occur in rare loans (pueblo, bwana, Buenos Aires,
puissant), often with alternative pronunciations showing that
English-speakers have not yet come to terms with them. Such clusters
don't show up in Germanic where we would expect them for etymological
reasons, as in Goth. fo:n (which we _know_ to be related to *pah2wr/n-,
and there's no better explanation of the absence of the etymological
*w). Compare with that the loss of root-initial *v after the prefix *ob-
in Slavic, where *tv-, *dv-, *kv- and *gv- are permissible while *pv-
and *bv- are not. I don't claim to have _proved_ that *bijo: lost its *w
in early Germanic, but if it had had one after the *b, it would have
lost it anyway, so you can't use this form to postulate *bHe-y/w- -- it
can be explained without recourse to exotic alternations.
Piotr