Fournet on PIE Morphology and Root Structure (was: S mobile)

From: Richard Wordingham
Message: 53558
Date: 2008-02-17

--- In cybalist@yahoogroups.com, "fournet.arnaud" <fournet.arnaud@...>
wrote:
>
> I have published a part of my own theory
> of PIE morphology and root structure at :
>
> http://w3.ens-lsh.fr/llma/sommaires/LLMA_7_02_Fournet.pdf
>
> You can have a look.

Is this a frozen text or a work in progress?

I have now looked through it, and I have a number of comments on the
purely IE part of it:

1. (p6) You say of the four reconstructions *pitar, *p&ter, *p&2ter
and *ph2ter (normalising to the usual Cybalist spellings), 'Ces
différants formats ne sont pas plus vrais les uns que les autres, ni
mutuellement exclusifs'. But *pitar is wrong for European languages,
and *p&ter implicitly fails to address the differing Greek reflexes of
the laryngeals.

2. (p7) Voiced aspirates without voiceless aspirates exist in Kelabit
(spoken in Sarawak).

3. (p8) Note that there is a big difference between pre-glottalisation
and post-glottalisation. While post-glottalised consonants are
similar to ejectives, preglottalised consonants are similar to
implosives, and it is then the back consonants which are frequently
missing. (For example, Proto-Tai lacks preglottalised velars and
palatals, though the place of the latter may be taken by the
pre-glottalised semivowel *?j. The gap is continued in Modern Thai.)

4. (p9) As has already been pointed out, you overlooked the PIE a ~ a:
ablaut series.

5. (p10) In the PIE verb, /o/ is not a mark of the past - it is a mark
of the *perfect*, and I think of the indicative singular (i.e. strong
forms) only.

6. (p10) English _got_, _bore_ and _born_ are zero grade, not o-grade.
The history here is that (1) r., l., m., n. > ur, ul, um, un in
Germanic, and that the /u/ was generalised before all /r/, /l/, /m/
and /n/ in the verb, and further extended in Middle English, and (2)
the simple past in English often uses the vowel of the past
participle. The original o-grades of _get_ and _bear_ survive in the
archaic _begat_ and _bare_.

7. (p10) The infinitive of Latin do: _give_ is _dare_, with both
vowels short. The vowel after the 'd' of the stem are long only in
the present subjunctive, _do:_, _da:_, _da:s_ and _da:ns_, and nowhere
else.

8. (p23) Slavonic *od, *ed 'one' is very doubtful. It is usually the
following part that is seen as going back to PIE, to *oinos to by precise.

9. (p28) '#n' has been mistyped for '#s'.

10. (p30) Using English _skip_ as an example of a s-mobile is not
sound - the word cannot derive from Old English, and no North Germanic
cognate is documented.

11. (p30) The implication of your analysis of Latin _simplex_ is
striking - that the -plex element does not relate to 'fold'!

12. (p35) The fa``al analogue exemplified by _momordi_ is the perfect,
not the 'past'. I think it a little misleading not to note that the
reduplicated presents, such as your example Greek _pimple:mi_, are
also common.

13. (p35) (See point 6 above.) English _to bear_ is not Germanic
Class VII. OE _beran_ was Class IV (beran, bær, bæ:ron, -boren), and
it makes sense to describe Modern English as such. Modern English
examples of Class VII are fall, fell, fallen; hold, held, held; know,
knew, known. Of course, this then makes nonsense of the comparison
with Form III.

I also have some comments on the Hebrew examples:

1. (p9) Hebrew 'give' is <na:tha:n> - the /t/ is not the emphatic. I
believe this point has already been made.

2. (p23) <'ans^> (s^ = shin) is not a possible Hebrew word. The
singular <'iysh> would require explanation, so I suggest you just give
the plural <'a^na:s^iym> 'men'. Are you claiming that the third
consonants (PIE *r v. a Semitic sibilant) are cognate?

Richard.