Re: [tied] Unreality...

From: enlil@...
Message: 33115
Date: 2004-06-06

Me:
> You wish to equate patterns seen in IE and those seen in Sanskrit
> together as if they are equal

Jens:
> No. Never did that, never said I would.

Answer does not compute. Warning. Warning. Blood pressure at dangerous
levels. Warning. Trying to understand.


>> In IE, things are a little different and that minute difference
>> matters.
>
> Certainly, and it varies with the definition of the cover-term "IE".

At least we're agreeing here.


> Indications strongly advising against the identification of *e and
> *o as **a and **a:. I have never called PIE /o/ a lengthened version
> of /e/. Others have, but I have not followed them.

But *o is to be considered some function of *e in your theory, no?
That's the whole idea of monovocalism, that one of the vowels is
merely a particular reflection of the other.


>> but that leaves out *e:. Oh maybe we need a triple-long vowel too.
>
> Maybe we do. For the final stage PIE we do need an
> opposition /e/ : /e:/ : /e::/, but the third degree could be noted
> as hiatic /ee/. I have not been able to find evidence of an
> opposition of length in hiatus, [...]
> [...]
> Some say that [it is dumb/illogical/crazy], but I disagree. I may
> perhaps be allowed to express my opinion and state my reasons.

The whole problem and why this discussion has gone on for such a long
time is that you apparently don't have any adequate reasons for
integrating all these anti-universals and oddities into a protolanguage
you dare call Pre-IE. The language is robotic in your formulation of it.


>> We don't run into this problem in Sanskrit, and this
>> is what makes it tough to argue against. The point is, taking away
>> /i/ and /u/, Sanskrit really DOES have a single vowel /a/ with a
>> long counterpart /a:/ and it's a very elegant solution indeed.
>
> How civil.

Yes, I'm not just a caveman :)
And now comes the long assumptive paragraph that I can't agree with,
nor can anyone because it's not sufficiently proven...


> Not for the many cases of change of /e/ to /o/ which are all /e/ if
> projected back to the day before the change occurred. At that time
> some other o's were not vowels at all.

Based on a random thought that can also be replaced by any number of
alternative solutions.


> Apart from a few isolated cases of apparently fundamental /i/ as the
> root vowel, the only pre-apophonic /o/ I know is that of reduplicated
> verbal stems, as the perfect, the intensive and the reduplicated
> aorist (and, some say, the reduplicated present).

Which can also be replaced with other solutions that are equally
satisfying or better since they may not need double-long vowels and
hypothetical *R.


> If that reflects a sound law it may be projected back [...]

Only "if". That's the problem. You don't try to prove it by
eliminating the other possibilities. Still these possibilities linger
for every one of your assertions.


> That would potentially leave a two-vowel system for that stage of IE
> morphophonemics. However, we were talking about lexically given root
> vowels, and that's a different matter altogether.

Alright, so you accept that a former phonemic difference between *e and
*a could potentially be used for the purposes of morphological processes
like perfects, statives, etc. which would "potentially leave a two-vowel
system" for this stage of pre-IE but your issue remains the "preponderance
of *e" in verb stems that forces you to analyse beyond what I'd do to
reduce everything to one vowel like we can Sanskrit. Is that it in a
nutshell?


> I do not feel ashamed to tell anybody what I believe I find.
> My "unicorns" do not occur simultaneously, by the way.

It doesn't matter. My theory has less unicorns at any stage of preIE
by far. The only one you've identified is Final Voicing which might
be explained in a more pleasing manner anyways. We'll see.

Double-long vowels can't be explained away no matter what. They are
simply exceedingly rare and that's that. That is a problem for you
which isn't a problem for me because I leave the two-vowel system
as is and don't meddle with it further.


> But what we can know about it may amount to just that. The
> accusation for dodging is not at all fair. In my original
> presentation of my IE morphophonemics in 1978 (reprinted 1999) I
> added an appendix of 166 wordforms for which I specified each change
> they underwent when processed by the phonetic rules I had
> formulated.

Alright. Well my problem involves my hectic life right now, having
moved back to Winnipeg. So I need to get myself to the UofM library
and look at a few things but that's a long bus trip and I can't find
the time right now even when I desperately want to. Mea culpa then.
We'll see if I'm in the gutter a couple of monthes from now and if
not, I'll be nosing a few books.


= gLeN