On 08.02.2006, at 11:23, Patrick Wynne wrote (off-list):
> What we do NOT ever see is a pa.t. formed by insertion of a nasal
> infix between a verb stem and a pronominal ending, as you propose
> in **_ub + n + ni_.
That's certainly a strong point. Just to clarify, there are two
assumptions that led me to this reconstruction:
1) The infixed past tense forms derive from original suffixed forms
by regular sound change, as in Q. _lemba_ < _*lebnâ_ (Etym s.v.LEB-).
2) that the apparent past tense marker *_-ê_ was originally a marker
of the 3. sg.
Ad 1) It's true that Tolkien's own wording suggests that there was an
original distinction between nasal infixion and suffixation of _-ne_
with subsequent metathesis.
[One such passage making this distinction occurs in the Early Qenya
Grammar (ms.), where Tolkien writes that the past stem was formed
by addition of the suffixes _-ye_, _-ie_, or _-ne_, but that the most
common of these, _-ie_, "is normally accompanied by stem strengthening
consisting of (1) a-infixion, (2) n-infixion, (3) vowel lengthening"
(PE14:56). -- PHW]
I could, and probably should have written **_umb-ni_. Which brings
Ad 2) That, I'll admit is a very weak point, as Tolkien's writings
suggest otherwise. It was an entirely ad hoc assumption, in order to
explain one particular form. It seemed to me (somewhat) admissable
because Tolkien's languages (unlike 'real-world' ones) are subject to
re-formation/ interpretation without further notice. Also, re-
formation of case endings (especially in past tense) based on the 3.
sg. are frequent in the world's languages.
Yet *_ê_ or (or, at some stage *_ie_, cf. Helios' post) is indeed
indicative of past tense in particular and so throughout the corpus.