[YG Conlang Archives] > [jboske group] > messages [Date Index] [Thread Index] >

[Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: [jboske] more true (was: RE: Re: ka ka (was: Context Leapers)

In a message dated 10/4/2002 6:09:48 AM Central Daylight Time, a.rosta@hidden.email writes:

<<
You misunderstand. I mean that the scale of True is unbounded; it
does not stop at 1 (the minimum threshold for wholly True). I
see the scale thus:

<-------|-----------|------>
   false | 'sort-of' | true

'sort-of' = "neither wholly true nor wholly false"

"Kareem is tall" would be within the 'true' part of the scale,
but "The taller guy is tall" would be even further along it.

For {pi PA} truth values ({pi PA va'e}, {ja'a xi pi PA}),
{pi no} = wholly false, {pi ro} = wholly true.

>>
Hmmmm.  I'm not sure how to deal with this, being so entrenched in [0,1] truth.  I suspect that this could be projected back onto the usual interval (or rather (0,1)) in a variety of ways, but essentially by introducing the notion of designated/acceptable values and calling all those "true" (and calling all the antidesignated values "false").
On the other hand, allowing the values to expand outward (as I know from multivalued cases) makes the situation easier to see sometimes -- though it gives weird laws (i.e., non-Boolean).  This kind of modification is probably not going to cause too much trouble, once its conventions are thoroughly laid out.