Re: more epistemic perversity

[Logical Language Group <lojbab@ACCESS.DIGEX.NET>]

I'm going to bow out, agreeing with Rick that we are going around in circles
and not getting anywhere, and the Lojban-relevat points have already been
raised, whether or not anyone has been convinced.

I am sometimes using sloppy wording as in
>>In English we do not have these implied extra places, but instead have the
>>stated or elided assumption of a single absolute reality.  In Lojban there
>>is no absolute reality and we have the elided expression of a variety of
>>x4 epistemologies that allow different truths to coexist or to collide.
>
>I can't make any sense of that. Neither in English nor in Lojban need
>one make any assumptions of absolute reality. As for the x4, I thought
>we had agreed that it was for how x1 acquired knowledge x2, and it had
>--More--
>nothing to do with the truth of x2.

where perhaps I should have said "different knowledges to coexist or to
collide".  But that does seem odd in English.  People can know different things,
and can hold different or even opposing truths, but I don't think I have
ever heard of plural use of knowledge.

And the reason for this likely is that knowledge presumes some kind of
absolute truth or fact in English.  We don't usually consider in English that
any realm of information where two peopl can justifiably hold opposing
truths is a realm of "knowledge" as opposed to "opinion".  Thus the sense
of "know" that applies to "knowledge" DOES assume some kind of real-world
absolute truth, which among other things would be observer independent.

But having noted in making the fatci jetnu distinction (which pc pointed
out to me once in a phonecon), that by some philosophies there are NO
absolute truths, it seemed to me not to be culturally neutral to make
"know" pertain only to facts as opposed to "truths" whioch are relative.
The x4 was added to make this distinction evident, and originally the
same words were used for jetnu as for djuno.  But soemwhere along the way
people pointed out that "truth" is based on "metaphysics" whereas
"knowledge" is based on "epistemology", and that the x4 of djuno should be
tied to the x1 of djuno and not to the x2.  i can accept this, though I
suspect that a metaphysics probably is enough tied to particular
epistemologies such that the two are largely interchangeable,

Having removed the absoluteness of the truth of le se djuno, much of the
tie to the standard English sense of "know" is lost.  I thus interpret the
"know" more in the sense of a superset of propositions that one can be
attacjed to.  The x4 of djuno can refer to opinion, to belief, to logic,
to observation, to dream states, to cultural customs.  And for these to
all be acceptable, it CANNOT be required that the x2 necessarily be something
"true" in some speaker independent sense.

The other words relevant to this discussion are not as broadly encompassing.
krici pertains to believe without necessarily having justification or
evidence.  birti has to do with the emotional response of certitude.  jinvi
explicit refers to propositions which are subjective by there nature and
in which there is generall some acknolwedged disagreement - we don't call it
an "opinion" if everyone agrees.

I feel that Lojban needs a general word usable to report a proposition that
someone holds/is convinced/knows which says nothing about what the speaker
feels/knows/holds/presumes as truth.  I intended that to be djuno, and
its use elsewhere in the language design is consistent with that meaning

du'o, the BAI cmavo based on djuno, means "according to" - thus saying
that the x1 of djuno is the one who is "presuming truth" if anyone is.

We also have ju'o, used to express "certainty" as contrasted with "ia"
whcih expresses belief.  I see djuno and birti as being closely related,
with birti being an emotional claim, whereas djuno is a justified claim.

We have the evidentials, which each pertain to epistemological means of
knowing, some of which are peculiar to the knower.  I believe that if
I make an assertion, using an evidential to indicate my epistemology,
that anyone else should be able to report my claim using "djuno" and
the corresponding epistemology to the evidential I used, regardless of
whether that speaker is necessarily willing to presume my knowledge is
true.  All that should matter reagrding MY knowledge is whether >I< think
it is true, and not whether I can convince someone else of that truth.

I think that the internal consistentcy of Lojban requires this
understanding of "knowledge".  The speaker should not be the judge of
truth of se djuno, only le djuno, unless we somehow introduce the speaker
as a place into this.  (I cannot easily articulate why I think that And
is arguing something else when he compares mlatu and djuno.  But asserting
a bridi to me implies tgat I claim that the bridi relationship holds, and
my tie to that relationshipo is only as asserter.  To say that with djuno,
that not only am I asserting the relationship of le djuno and le se djuno
et. al., that I am also asserting that >I< presume le se djuno is true,
seems to be an added level of involving the speaker into the meaning of
the bridi that I feel is inappropruate to Lojban.  But if this doesn't
make sense to others, I cannot do any better).

I have been arguing this rather more strongly than perhaps I should because
discussions based on "djuno" have come up again and again, each time driving
the definition of the language in some new direction.  "djuno" has been a
key factor in the distinction of "du'u" and "sedu'u", the addition of "tu'a"
for sumti raising, the addition of "kau" for indrect questions, and probably
others that I am not calling to mind at the moment.  EACH of these additions
has hinged on a distinction that Lojban makes which English does not clearly
make.  That alone should make it clear that all the nuances of English "know"
should not transfer to Lojban "djuno".

I think it is sufficient to leave djuno somewhat vaguely defined, so long as
it is remembered that its meaning is constrained by all the decisions on the
language design that have stemmed from that meaning.  Very seldom will a
usage that I might maike of djuno using my interpretaton conflict with one
that Jorge would make - I merely state that such conflict is possible if
he deduces from my assertion about someone else's knowledge that I also
presume/know that knowledge to be true.  Usage will decide, but the difference
here is subtle enough that usage will not decide very soon - there aren't that
many places where someone will use djuno about a fact theyt dispute.

But I feel rather strongly that I cannot give in to a definition creep which
seems justified primarily by my choice of a word as the keyword when we have
repeatedly said that the keywords are not definitional, and when in particular
we are groping at dsintictions that English does not have clear words for.
And I cannot do this especially when I know that djuno will come up yet again
in some other semantic controversy, and this debate will by implication
affect that other controversy.  I don't like Lojban words to be defined by
debate, especially in English.  It is bad for the language, and I think it is
bad for the community.

I will do my best to resist further discussion of the meaning of djuno,
and will especially stay away from discussion of English "know".

lojbab
----
lojbab                                                lojbab@access.digex.net
Bob LeChevalier, President, The Logical Language Group, Inc.
2904 Beau Lane, Fairfax VA 22031-1303 USA                        703-385-0273
Artificial language Loglan/Lojban: ftp.access.digex.net /pub/access/lojbab
    or see Lojban WWW Server: href="http://xiron.pc.helsinki.fi/lojban/";
    Order _The Complete Lojban Language_ - see our Web pages or ask me.