Re: SUPERL and comments...

["Arthur W. Protin Jr." (GC-ACCURATE) <cbmvax!uunet!PICA.ARMY.MIL!protin>]

Folks,
    I wondered when I read what Ron Hale-Evans wrote:

> *Why* can't we create a language where grammatical correctness implies
> factual correctness?

if he really did not understand the answer to his own question since
he did cite a reference to the answer, "Godel, Escher, Bach" by
Hofstadter.  And I decided not to comment since if he could not
understand the answer when he read it there, he would not understand
it when I repeated it here.  BUT, I can not assume that all of
you read GEB, or any other explanation of Godel's Theorm.  So I
must try to explain it as best I can.
    Godel's Theorm has to do with number theory and the application
of number theory to numerical representations of proofs, and thereby
to proofs in general.  Simply put it claims to prove that any formal
representation/manipulation scheme is:

    A) "incomplete" - there are truths that can not be represented/
                arived-at; or
    B) "inaccurate" - there are falsities that can be represented/
                concluded; or
    C) "both" - all of the above.

While it maybe the case that Godel's Theorm is flawed within the formalism
used or the formalism is flawed such that a falsehood has been proved true,
I would not spend much time or effort looking for this language (the
existance of which would constitute a contradiction of Godel's Theorm).

    The arguments against such a compact language are many and varied!
I wish I had my research into this matter formalized such that I
could present it or a reference to where it had been presented.
the combined arguments could easily fill a monograph if not a full
text, but I doubt that any simple concise statement will every be
available.
    Some of the issues are:
    1) signal to noise ratios - the trade off of error correction
                against data density; and
    2) vocabulary - the expanse of an already intractable collection; and
    3) information theory - there are orders of magnitutdes more data
                in our language than we are aware of.
On the issue of vocabulary, I refer you to the posting
from  Coranth <gryphon@dino.ulowell.edu> :

> Ron Hale-Evans writes:
...
>> Another question: Why can't we create such a compact language?
>> Again, Hofstadter has an interesting comment on this in *Metamagical
>> Themas*; he devises the word "cohatalat", which is an acronym for "come
>> over here and take a look at this", a phrase most of us use often enough.
>> Why hasn't a word like "cohatalat" evolved to fill this niche? Actually, I
>> tend to think we *can* create more compact language in just this way. I
>> don't think I would have any problem using "cohatalat" in everyday speech
>> if anyone actually knew what it meant.
>
> The group I hang around with has many such, including faip (for all intents
> and purposes), wysiwyg (pronounced wisy-wig, what you see is what you get)
> and w'kw'bfy (pronounced wik-wib-fee, we know what's best for you)

These example are adding words to English which already has the world's
largest vocabulary.  Yet the size of the average person's vocabulary does
not vary from language to language as much as the size of the available
vocabulary varies.  Thus I expect that we are nearer the limits of our
ability to use language than we are to the limits of languages.

    thank you,
    Art


Arthur Protin <protin@pica.army.mil>
These are my personal views and do not reflect those of my boss
or this installation.