Re: Lean Lujvo and fat gismu

[Art Protin <protin@USL.COM>]

In a message that seems to have addressed to only me, lojbab
speaks to the list:

> Not having seen John's posting on heap paradox, etc. since
> he forgot to cc it to us...
>
> Art has taken the scalar "not bigger than" relation and not
> the contradictory one.   Using the contradictory version
> x1 is (not) bigger than "something-unspecified" means
> It is not the case that x1 is bigger than "something-unspecified".
>
> But unless x1 is the smallest thing in the universe it IS
> bigger than something and this is therefore misleading (if
> not wrong, since you can always argue that the
> "something-unspecified" is some kind of universal standard.

I am sorry, but the "something-unspecified" is simply unspecified.
It's selection is my, the speakers, choice.  And I can easily
select it such that "it is not the case that X1 is bigger than
the something-I-did-not-specify".  This works fine.

Lojbab continues:

> But in the color realm, things get really messy.  Asuuming that
> white light consists of the entire frequency spectrum of light,
> one must argue that "white" is bluer than something-unspecified
> which is a legitimate standard for blue, because you can find
> some white light that has more blue in it (through instensity)
> than the blue standard you are using, whatever that standard is.
> White is bluer than some things we would call blue, and is also
> bluer than brown.  This is the reality of physics but has nothing
> to do with language.  White is not blue in any language that I know of.

Of course white is not blue.  And the physics mentioned by lojbab
is proof that white can not be more blue since it is mixed with all
the other frequencies!

> We don;t perceive colors as being more of that color than a standard,
> but rather as being more-similar to some ideal image of the color than
> something else - aboundary line that is not well-fixed and is
> dependent on many other factors than the thing we are actually
> evaluating for blueness.  Thus a comparative blue as a gismu would
> have to have all of those places specified, because they are
> relevant to the question of whether it is blue (and they would
> have to be part of the place structure - none of the hand-waving
> we did to eliminate "standard" and "background" and "ambient light"
> and "observer" which we did in the non-comparative blanu that we
> adopted over JCB's choice.

Things are blue because I see them as blue.  Why must we encumber
the language with so much arbitrary data from one model of perception?
Something is blue because the speaker experiences optical stmulation
of sufficient quantity and of sufficent quality to subjectively
claim that it is blue.  Maybe it is really white but illuminated
by a blue source.  The speaker should not have to make claims
about the real nature of things that are beyond the speaker's
ability to know.  I talk about things the way I see them!
(With regard to color, size, or the logic of this discussion.)

> The other problem with comparatives used in TLI Loglan is that
> JCB was extremely inconsistent as to when they were used.  He
> used them with some, but not all adjectives.  (A detailed list
> will shortly be posted for ftp, as I have just finsished my mapping
> file from Lojban to TLI gismu). But if you look in your trusty
> L4/L5 you will find many listings of words as adjectives in
> E-Trans that do not have comparative places.  To be philosophically
> consistent, all such must have a comparative, and indeed, since all
> gismu are at once adjectives along with everything else, every
> gismu needs a comparative place (and every lujvo, as well - and
> this comparative is independent of any places derived from the
> source gismu).

NOW THIS IS A GOOD REASON!!!

The consistancy of the language is more than sufficient reason
to justify the change.  I strongly recommend that the other
aguments not be mentioned anymore as answers to this question.
I have yet to accept any of those others as even being valid
let alone sufficient.

> Finally you have the problem of multi-place gismu.  If you
> take klama, you have trouble deciding how to formulate the
> comparative version, because there are 5 places.  Is the
> adjectival version of klama more klama-ish in the x1 dimension,
> or in the x2 dimension, etc.  (you will easily answer that
> x1 is the correct answer based on your English bias).
> Is the adjectival form of "selklama" still more-so in the
> (now) se "go-er" place?  You've gone and turned Loglan from
> a langauge with a strong bias towards the x1 place in
> semantics, and this breaks down in places where there must
> be no such bias.  Predicate logic is one of them.
>
> But then you take "nelci"  which is the implicit comparative
> here - the x1 or the x2.  English usage attributes both
> comparatives:  x1 likes x2 more thanx3 is completely ambiguous,
> because you don;t know whether x3 is being compared with x1 or x2.
>
> Comparatives have way too many problems for them to be the
> philosophical basis of Loglan/Lojban place structures and
> semantics, and those problems stem from our English bias that
> insists on dividing predicates into different "kinds", which
> is unacceptable in a logical language.
>
> lojbab

[All of lojbab's message is included here and in the order
in which I received it.]

    thank you all
    Art Protin


Arthur Protin <protin@usl.com>
STANDARD DISCLAIMER: The views expressed are strictly those of the author and
are in no way indictative of his employer, customers, or this installation.