Re: [lojban] An approach to attitudinals

[Richard Todd <richardt@flash.net>]

Jorge Llambias wrote:
> Saying {ui le truralju cu stace} does not affect the assertive
> force of the bare bridi, but {a'o}, {e'o} or {xu} do. With these,
> the proposition is no longer used to make an assertion. Its truth
> value is not affected, but now the speaker is not claiming to know
> what that truth value is.

I see how you can get at the most useful versions of various
attitudinals by defaulting them on an ad-hoc basis to one form or the
other.  I don't see why this is logical, desireable, or culturaly
neutral.

What is the justification for {a'o} affecting the assertive force of a
bridi, while {ui} does not?  From other responses (both public and
private), I get the impression that not everyone agrees with your
either/or stance.  Some feel that all could be either depending on
context, which I agree is workable but not anywhere close to ideal.

I wish we could just be explicit about the effect of the attitudinal on
the truth value of the sentence it's attached to.  Then there'd be less
confusion and more possible shades of meaning.  Everyone wins.

It bothers me that the only completely safe interpretation of:

.a'o do dunda le karce mi

is

{hope emotion} associated in an unspecified way with {you give me a
car}, which may or not be true depending on the nature of the
association.

With a suffix, there's still context involved, but at least you know up
front whether the speaker is asserting a true statement.  This could go
a long way towards clarity.

Using the placement of the attitudinal, as I've seen suggested, would
force us to pick a word to which to attach a truth-altering
attitudinal.  That keeps the placement from doing what it's intended to
do, which is emphasize that the emotion has more to do with that word
than the whole sentence.  Despite that objection, I do like the idea it
better than leaving things alone.  In my mind, this part of the language
has a bug and we should fix it.

> >Then, when someone misunderstands me, I can point to the
> >list and say, "Nope, attitudinal number 206 doesn't do that.  Study
> >harder."
> 
> Is that what you do when someone doesn't understand that a bridi
> marked with {xu} is not an assertion of the bridi? The effect
> of {a'o} is similar to the effect of {xu}.

When I think of {xu} I don't classify it with the attitudinals.  In my
mind it's:
{ma}=provide a sumpti, 
{mo}=provide a brivla, 
{xu}=provide the truth value.
(the goal of all of these being to come up with a true assertion)

So I don't know what the answer to your question is.  If it is an
attitudinal, then I admit the non-propositional sense of it would be a
bit of a stretch.

> >ko'a: .au mi na speni do
> >
> >ko'e: .o'onaisai ba'e mi cliva .i mi birti
> >       le du'u mi speni do
> 
> What did ko'e understand? Apparently not the wish that ko'a tried
> to express (which would have been a good reason for ko'e to
> get angry). Did ko'e get angry just because (according to ko'e)
> ko'a asserted a false proposition?

It was just a silly exaggeration of what I was getting at, where {ko'a}
always assumes the propositional form, and {ko'e} always assumes that a
sentence is an assertion.  So, yeah, that's why {ko'e} is angry; he's
fed up with {ko'a} constantly spouting what he believe to be
nonsense/lies.

In {ko'e}'s mind, {ko'a} says:
Consistent with my hopes, I'm leaving.
Just as I desire, it is not true that we are married.

To {ko'a}, it's more like:
I hope I will leave.
I wish it weren't true that we're married.

I think I've made my point the best I can make it.  I'm not trying to
start an argument.  All in all, so far I think lojban is a very
enjoyable language to learn.  Perhaps someone else will think of a
better resolution.  I do think we need some sort of 'ruling' on these
things, which will be written down and accessible.  Perhaps updating the
reference grammar would be a good idea, or a bug-tracking database (like
is used in software).

Richard