Re: [tied] How to prepare **udon soup (was: PIE rhotacism)

On Thu, 22 Nov 2001 01:27:09, "Glen Gordon" <glengordon01@...>
wrote:

>>Original system:
>>
>>*i *u *i: *u:
>> *a *a:
>
>It starts off as a perfectly balanced system alright...
>
>
>>I. Loss of high vowels: [...]
>>
>> *a *a: *ai *au *a:i *a:u
>
>Now here's where the bullshit starts. I defy you to justify your
>ideas using mainstream linguistics.
>
> 1) One-vowel systems don't exist anywhere, period. Biggest
> problem of all.

Good, as it's not a one-vowel system.

> 2) With only *a and *a:, *a: would most certainly dissolve
> into another short vowel in order to greatly improve
> speech economy.

Evidence?

> 3) Since *a: is as (or more) predominant than *a, as shown
> by your lala-land reconstructions(1), your long vowels must
> be typologically reinterpreted as unmarked (short) and
> your short vowels, what little exist, as extra-short or
> non-existant. Of course, the latter solution runs into
> even more problems concerning syllabicity.

The paradigms I gave for "water" are acrostatic and mesostatic,
respectively, precisely because of the long vowels they originally
contained in the root, unlike the normal proterodynamic and
hysterodynamic paradigms. There is no problem: *a: is *a:, and *a is
*a (where Pa:n.ini's last sutra can always be invoked).

> 4) Since *a: is in reality "short" by general typological
> constraints, the assertion that {*a > *e AND *o} DOUBLY
> defies the strong tendency for such vowels to LOWER, not
> rise.

If you have /a/ and /a:/, the tendency is for one to front, and the
other to back. Cf. Germanic, Baltic, Slavic, Hungarian, etc. etc.


=======================
Miguel Carrasquer Vidal
mcv@...