Re: Non-polairty

Hi!

Sai Emrys <saizai@...> writes:
> (Perhaps this will be a series - non-[insert kludgy but omnipresent
> language feature here]. Eris would be proud.)
>
> So in this case: non-duality / non-polarity / etc. Or another way to
> put it - analogy / continuous scales.
Yippie!  One of my favorite research areas in conlangs. :-)

> The problem: there are too many words that reference what are actually
> (multi-dimensional?) (infinite?) spectra as if they were two (or more,
> but finite) points.
>
> E.g.: Good/bad. Old/young. Rich/poor. Etc.
These are not the same kind of scales in my view.  I'll explain that
in a second.

> I object. I would like to eradicate all of these - and the associated
> (cringily kludgy) Esperantoesque practice of having some word be the
> base and then antonyming it (e.g. malbono [sp?] et al). Likewise with
> referring to the spectrum, or the unkown quantity, with one of the
> endpoint words - e.g. "how old are you?".
Very good! :-)

> More, I would like to have some sort of grammatical and elegant way of
> having a *continuous* spectrum. I don't object to having reference
> points along the way (e.g. for color or temperature), just to having
> voids inbetween.
Ah, fun!  My conlang Qthyn|gai does not have a continuous scale, but a
generic way of either subdividing into three or into nine steps.  The
former for a vague scale, the latter for somewhat more precide scale.
And then, it has some additional affixes for several other things.
See my grammar page for details.

So first of all, Qthyn|gai comes with generic modificators for scales:
you use 'high' + scale_affix to indicate the level of height.  This
makes 'low', too, and even 'flat'.

I found this fair enough for human communication, but if you need
more, just use a system similar as for


> The simplest idea I can think of for auditory mode is to use the two
> major continuous factors of speech - vertical & horizontal tongue
> position. This would work OK for finite spectra - i.e. 0-1 or (0,0) to
> (1,1) - but not so much for the infinite. Not sure how to work that.
That I find funny but not practical.  I'd use a stackable system
instead, similar to the north, east, south, west system where you can
have south south west etc.   That's what my conlang does, too.

> For that matter, as a starter, do any of you actually understand what
> I'm talking about and why it annoys me? (Does this sound familiar?
> :-P)
YES! :-)))

Here's more about Qthyn|gai scales and how I try to make them unbiased:

  1) 0...infty scales:

     If a scale has one prominent end point, usually labelled 0, and
     if that scale is infinite (at least for human intuitive
     judgement), then I'd collapse lexicon entries into only one
     entry.  Example: young vs. old.  I'd not have young, just old or
     better 'age'.  Young is a small amount of age, old is a large
     amount of age.  Then there is zero age, the end point on one
     side, but there is no endpoint in the other direction of
     the scale.

     Therefore, I only have one word.

  2) All other one-dimensional scales get two completely unrelated
     lexicon entries for each direction, so that no bias arises.
     There are still different types of scales here:

     2a) inftyA...0...inftyB scales:

         Those with a natural continuum from one infinite to the
         opposite infinite, with a 0 in between.  These are
         'height' vs. 'depth':
            'high'    = 'height' large_degree
            'low'     = 'height' small_degree
            'flat'    = 'height' zero_degree
         and
            'flat'    = 'depth' zero_degree (other point of view!)
            'shallow' = 'depth' small_degree
            'deep'    = 'depth' large_degree

     2b) constA..constB scales:

         These get two unrelated lexicon entries, too.  Examples:
            probable...improbable   (obviously biased in English :-))
            empty...full

         Saturated scales of any kind fall into this category.  Note
         that the constants are from a human's point of view -- sometimes
         there is a physical infinity on one side that a human just
         does not see.  E.g.
            black...white

         The reason for giving these two entries in the lexicon is
         that the selection of one of the two endpoints for the lexicon
         while the other one would be the opposite, is bias!
         I had some discussions about Lojban here, which, as I
         understand it, sometimes (always?) only has one lexicon entry
         for these scales.

     2c) Non-opposites:

         Well, the standard example is
            good vs. bad

         Who wants to decide whether these are collinear?  If there is doubt,
         scales get two lexicon entries, again unrelated.

**Henrik

PS: Qthyn|gai (but not much about scales, I'm affraid):
    http://www.theiling.de/conlang/s7/

Replies