On 05-02-17 11:30, tgpedersen wrote:
> Or inversely, that -o:n in *tek^รพ-o:n etc is the Hoffmann suffix (but
> then length is unexplained), and the present participle is the
> extension of that with the demonstrative particle nom.m.f.sg. *so,
> oblique *to (+ further suffixes).
No, the pres.part. ending shows no trace of a laryngeal. There are also
some subtle differences concerning the quality of the vowel (*e ~ *o
variation in the pres.part. versus stable *o in the Hoffmann suffix) and
the compound-like behaviour of Hoffmann stems. Also, *-e/on(t)- is
attached to verb roots, while the Hoffmann suffix forms nouns and
adjectives exclusively from nouns. To be sure, we also have the
"individualising" suffix *-on(t)- (the one which accounts for a large
number of Germanic weak nouns, including hypocoristically abbreviated
names like Odo and Hugo). Opinions vary as to whether it's
etymologically related to the pres.part. suffix. It's also formally
distinct from the Hoffmann suffix, but confusible with it: in some cases
it's hard to decide which one of the two was originally involved.
> How come the nominative is always seen as derived from the oblique
> stem by reduction?
Not always. Some people think that pres.part. *-e/ont- = *-e/on- plus an
extension. It's still a moot question.