Re: Hypersimple & Dreadfully Unnatural Grammars

I tried using this method on sentences with increasing complexity and soon
came to the conclusion that the only way to make understandable utterances
is to "cheat" on the strict RPN rule.  If I find it necessary to break the
rules I made then why did I make them in the first place?

I jumped into this project to find an answer the question "can a usable
language be designed around RPN notation?"  I'm beginning to think that the
answer is "no".  Natlangs tend to be delightful mixtures of order and chaos,
and perhaps that kind of "living on the edge" is actualy necessary for a
language to have sufficient expresive power.

With a maximally chaotic grammar you can say anything you want, but nobody
will understand you.  With a maximally ordered grammar everyone will
understand you, but there's only a handful of things you can say, none of
them original.  :)

--gary.


-----Original Message-----
From: charles <catty@...>
To: Multiple recipients of list CONLANG <CONLANG@...>
Date: Thursday, March 18, 1999 10:07 AM
Subject: Re: Hypersimple & Dreadfully Unnatural Grammars


>On Thu, 18 Mar 1999, Gary Shannon wrote:
>
>> Suppose it were declared that no "functional" word may have more than two
>> arguments?  The trivalent verbs would be seperated into two "functional"
>> words.
>
>> By keeping the stack of pending concepts to a maximum of two it becomes
>> easier to parse the sentence on the fly in your head as you hear it
uttered.
>>
>> Are their instances where this approach fails?
>
>I have tried this, breaking up 3-argument predicates into 2 interlinked
>2-argument predicates, and while it does work, natural laziness (which
>is an advanced kind of efficiency) soon drops the 2nd predicate.
>Another (better?) way to do it might be the trigger-language approach
>of using only 1 core argument + lots of oblique args.
>I've also been wondering whether a small verb can be incorporated
>into each noun, like a case tag actually, and eliminate the main verb.
>It seems that native languages of British Columbia do so.
>