Re: Summary of summaries on DJUNO

[And Rosta <a.rosta@UCLAN.AC.UK>]

Lojbab:
> >This is not the problem, though it is quite likely that Lojbab
> >believes it somehow is. Whether or not you believe there are absolute
> >facts, it remains the case that whenever you assert p in Lojban your
> >sentence is asserting p to be an absolute fact.
>
> No you are asserting it to be TRUE.  Absolute means independent of all other
> considerations.

I am currently unable to see any relevant distinction between "true"
and "fact". (I do see an irrelevant distinction, which is that
propositions are true, while possible states-of-affairs are facts.)

> But even given this, asserting that "John knows X" should assert only that
> John knows X, and not necessarily assert X itself, which is what I see your
> definition as requiring.

Asserting "John knows X" DOES assert only that John knows X. However,
for all speakers of English with the implausibly self-proclaimed exception
of you, the relationship "know" denoted by the word _know_ is such
that John cannot know X if X is false. It therefore follows
inescapably that asserting only that John knows X includes an
assertion that X is true.

This is said for the 100th time. I simply cannot comprehend that you
still don't see what I and others are saying.

> >Language and
> >communication requires that we at least pretend that there are
> >absolute facts.
>
> I think we need a real postmodernist, and not myself, to argue the contrary.
> The postmodernists most clearly are NOT willing to make such pretensions.
> I've been involved in theo other side of various debates with postmodernists,
> and I understand their argument even if I disagree with it.

Even if we had a postmodernist to argue the contrary - and thank god
we don't - their contribution would be irrelevant to the present
discussion, since what is at issue is the definition of one specific
gismu, and not any general philosophy of communication.

Even if you had your way and {djuno} were defined as you would wish,
the definition that others are adovcating for {djuno} could be
assigned to some other brivla with exactly the same place structure.
Remember that.

> >Lojbab has no response to 2.1.
>
> No, I have a different paradigm in mind, and I see the differences implicit
> in krici/jinvi/djuno, including in their place structures, to render this
> particular contrast invalid.

Could you (re-)spell this out more clearly?

> krici and jinvi do not have epistemology places,

This may be so. What do you think the difference is between jinvi's
x4, the "grounds" for the opinion, and djuno's x4, the "epistemology"
for the opinion?

> and hence there is a difference between djuno and those two above and
> beyond whether x2 is true (if the truth of x2 is even necessary).  Thus
>
> >                 jinvi = justified belief
> >                 djuno = justified true belief
>
> is an inadequatre contrast since it does not account for the epistemology

Certainly this paradigm assumes a comparability of djuno's x4 and
jinvi's.

> Furthermore, krici is defined quite clearly to not merely apply to any old
> belief, but to beliefs that are held without evidence.  Thus opinions which
> are based on evidence are not se krici, and this difference is not the
> same as "justification", since there are other ways to justify a belief other
> than by evidence (logical deduction for example)
>
> The krici/jinvi/djuno contrast thus appears to be a good paradigm, but it
> fails based onthe Lojbanic definitions.

Well here you are indeed (at last?) making what seems to me a
reasonable response to the arguments of others.

Certainly I had been thinking that {krici} was belief for which there
is not necessarily any rationale, rather than belief for which there
is no evidence. That is, I thought it included things taken on faith,
unlike jinvi, but I didn't notice that it excluded things not taken
on faith.

So I think you have succeeded in showing that the 2.1 paradigm is not
as compelling as it at first seems.

> >His response to 2.2 is that keywords
> >are not defining. My response to that is that other things being
> >equal it is better to have a keyword that fits the actual definition
> >rather than a completely misleading keyword.
>
> I agree, but that was not the primary consideration in choosing the keywords,
> and jhence is an improper argument for the semantics of the Lojban.

I don't see how you make this conclusion. What does it matter how the
keywords are chosen? We know they're not defining, and are merely
indicative clues.

> >His response to 2.3
> >involves counterintuitive definitions of {birti}, etc.
>
> All the Lojban words may have "counterintuitive" definitions to the English
> speaker, absed on the English words used to define them.  I have attempted to
> arguie that the very abbreviated definitions given in the gismu list are
> not adequate to define the words semantically enough to make the decisions
> people seem to want to make now, suing English-language arguments.  They in
> partcuilar seem to lead people to false paradigms of the 2.1 sort, and I am
> trying to couynter that by mentioning the actual paradigms that I used when
> coming up with the definitions.

Good. That is progress.

When someone uses "birti" I pretty much imagine they mean "certain
about". If you instead want it to mean "emotionally wedded to" or something like
that, then it is most unlikely to get that meaning unless you tell us
about it.

(Funnily enough, though, I had always assumed that {djuno} meant
"believe", until this debate began! (But that's for reasons to do
with my linguistics background) - And I could never see any
difference between {jinvi} and {djuno}.)

> The probably bottom line is that the Lojban word will end up having a vague
> meaning that includes your definition as well as mine.  Since yours
> is largely a restricted version of mine (all things that you could use djuno
> for, I could also, but not the reverse), my definition is probably better
> for the gismu, since (and Colin Fine has argued this better than I have)
> it is better to have broad gismu and use lujvo to restrict and narrow their
> definitions.

I agree, with the proviso that the restricted version is preferable
if the more inclusive version is duplicated by another gismu/brivla.

I would find your version of {djuno} (= my originally assumed one)
more palatable if I had a clear idea of how it differes from {jinvi}.

--And