Re: [tied] Re: Unreality of One-Vowel Systems (was: Bader's article

On Tue, 25 May 2004 enlil@... wrote:

> Jens:
> > No, that's a complete distortion of your own conduct in message
> > 32876: You expressly excluded the *possibility* of monovocalic
> > systems.
>
> They don't exist. I refuse to discuss vampires and werewolves with you.

I insist on their right to be considered at least to the extent and in the
sense that one such exists in Sanskrit. I also insist on the possibility
that the past may have seen things we do not have in the present world.
Many things are found only once; when the lone witness dies, its testimony
is stricken from the realm of theoretical possibilities. That seems
unwise.

>
> > I have never made such an assumption,
>
> Then what the hell is the basis of your pre-IE theory? Do you even know?

Something close to what you say yourself inthe following.

> In reconstructing ancestral forms of IE, we both depart from differing
> initial conclusions.

No, I don't believe I have any. I test many, but I do not assume they are
right until that appears to be the case.

> An examination of these 'initial conclusions' is
> important in evaluating which competing theory is the best. Obviously,
> the best one has the strongest foundation (ie: the soundest observation),
> the conclusion that eliminates the most unknowns to answer our question
> most efficiently and thoroughly.

Basically yes, but there is a qualitative side to this also. Some choices
may be so pitiful as to be valueless. If one theory can explain two forms
out of a thousand, and another one three, and nobody has anything better
yet, even the three-hits theory is not likely to be anywhere near the
truth. Also if one theory can explain all the facts by elaborate ad-hoc
fairy-tales and another one can only explain half, the latter could still
be preferable.

> The basis of my theories should be understood by now as deriving from a
> single common-sense deduction that unintuitive accent alternations
> between
> stem and suffix were once regular at some point in "pre-IE". There is
> simply no denying that this MUST be so.

I agree it is worth testing the possibility and I believe it proves
rewarding. Still, a priori it could have been a wrong assumption: The
linguistic sign is arbitrary.

> It doesn't tell us when such a
> stage occured but it MUST be so, even if it's a million years ago. This
> is my initial conclusion and it is an inevitable one. From this
> conclusion,
> I arrived at the Penultimate Accent Rule which later turned out to be
> a predictable quasi-penultimate accent. Others may arrive at something
> other than this to explain the unpredictable accent, but there's no
> denying the soundness of the conclusion.

If we extract different conclusions from it, both procedures cannot be
sound. They could both be wrong, however. Or does "conclusion" in the last
sentence mean "prerequisite"?

> You've made clear, in contrast, that the foundation of your views rely
> on your 'abstract analysis' of the IE vowel system.

We both do abstract analysis of the interface of phonological and
morphological alternations. Your analysis is not different from mine in
kind, only in the implementation.

> However, your
> analysis, just like the same analysis on Sanskrit, is not guaranteed
> to have any bearing on the vowel system of its prestage.

That goes for yours as well.

> Therefore, your
> 'initial conclusion' is immediately and obviously flawed. The conclusion
> that the analysis _does_ have bearing is disproven by Sanskrit and
> our knowledge of IE. As a result, your methodology demonstrably involves
> picking a solution by random without eliminating any unknowns. It is
> fair then to immediately dismiss your views altogether from the onset
> and choose the competing hypothesis that _is_ based on a sound
> conclusion.

One cannot sensibly begin with a conclusion. You are construing a daevic
world where all the good words have come to mean bad things. I leave open
a possibility which I believe I can see, but do not find occasion to use
it as the matter turns out, how can that disqualify me all that much? But
you are free to dismiss anything you want to dismiss. The witness is
excused.

Jens