Re: [lojban] non-ka properties

[John E Clifford <kali9putra@yahoo.com>]

{kau} began life as a meaningless marker to distinguish between two meanings of 
things like 

ko'a cusku le du'u ma mrogau:
He said who the murderer was.  and
Who did he say the murderer was?

(I think they got the usage exactly backward: that the {ma} is in the 
subordinate clause is obvious; that it functions outside that clause is useful 
additional information -- and not just for questions, since we sometimes want to 
raise expressions out of subordinate clauses, but cannot in general do so).

There was also the enduring question of what an indirect question rally was 
which was here solved by just noting that it was an indirect question, to be 
treated however that finally would be.

As such, it can occur just about anywhere that a proposition can occur, 
basically the array of psychological verbs (I don't think anybody thinks it just 
goes with {djuno}, though that is the stock example).  But not elsewhere, since 
its "meaning" is completely tied up with indirect questions.  If, perhaps, you 
have a theory about what indirect questions are, it may be that you can come to 
attach some meaning to {kau} that plays a role in this interpretation, but I 
don't know of any such interpretation nor what that interpretation might be (the 
most likely seems to me to be that {kau} is a choice function that picks out the 
right thing from the list of possibilities, but that still doesn't seem to have 
any utility outside indirect questions: answering {ma kau} to {ma mrogau} would 
probably earn you a kick, and deservedly so).




----- Original Message ----
From: Jorge Llambías <jjllambias@gmail.com>
To: lojban@googlegroups.com
Sent: Sat, June 25, 2011 4:08:30 PM
Subject: Re: [lojban] non-ka properties

On Sat, Jun 25, 2011 at 3:10 PM, John E Clifford <kali9putra@yahoo.com> wrote:
> {kau} is an indirect question marker, according to CLL.  It is hard to see how
> that can expand out from there to anything else (though it may easily be
> expanded to contexts other than {lo du'u), other sorts of abstractions, in
> comparable senses: I know what fits in here, and the like.  But what can it
> possibly mean in a main clause?

"I know what fits in here" is not a good description of "kau". "kau"
is not related to the speaker's knowledge. The idea that it is comes
probably from the common indirect question examples like "mi djuno lo
du'u ... kau ...". But in such examples the "I know" part comes
strictly from "mi djuno", "kau" only contributes the "what fits in
here" part. Any indirect question that does not involve "mi djuno"
shows that "kau" has nothing to do with the speaker's knowledge:

la djan djuno lo du'u lo ma kau nenri lo tanxe .i mi na go'i
"John knows what's inside the box. I don't."

mi ba troci lo nu smadi lo du'u ma kau nenri lo tanxe
"I will try to guess what's inside the box."

In those examples "I know what fits in here" makes no sense, it's just
"what fits in here".

In a main clause, "kau" creates a tautology. Tautologies are not very
informative things to say, they don't communicate anything. Since they
are true by definition, they don't really say anything about the
world, but they can be colourful:

- do klama ma
- ma kau

"Where are you going?"
"Wherever."

Of course that's necessarily true, but not very informative. That's
how I interpret "kau" in any case, "the value that makes the bridi
true".

There are open problems with that definition: What happens when no
value makes the bridi true? (My tentative answer at this point is that
this violates a presupposition, but I took a different position in the
past.) What happens when more than one value make the bridi true? (My
tentative answer here is that then makau refers to all the values.)
And then there is the issue of scope, which is a frequent problem in
Lojban. I think "kau" should be within the scope of preceding
operators and have scope over what follows, as usual, but I may not
have always been fully consistent with that in my own usage.
.
> The nearest thing I've seen to making functions out of predicates seems to be
> {na'u}, which makes an operator (close to a function apparently) out of a
> selbri.  I suppose this would require some specification somewhere of domain 
>and
> range at the introduction, but (in spite of not being mekso) {na'u cmene ce'u}
> seems to work in theory.

But making a function (or an operator) out of a selbri is not what
Felipe wants. He wants to be able to refer to a function, not make
one.

"na'u cmene ce'u" is not grammatical, you would need to convert "ce'u"
into an operand, and you can do that with "mo'e". But the problem of
"na'u cmene mo'e ce'u" is exactly the same one of "lo cmene be ce'u".
If "na'u cmene mo'e mi" is the value that the operator "na'u cmene"
assigns to the operand "mo'e mi", then you are using the the form
usually used for the values to refer to the function itself, simply by
filling the argument place with "mo'e ce'u". If that's what "ce'u"
will do in MEX, there's no reason for it not to do it also outside of
MEX.

Of course mathematical notation does that all the time, using f for
the function and f(x) for the value of the function at x, but Lojban
tends to (or attempts to) be more punctilious than that.

mu'o mi'e xorxes

-- 
You received this message because you are subscribed to the Google Groups 
"lojban" group.
To post to this group, send email to lojban@googlegroups.com.
To unsubscribe from this group, send email to 
lojban+unsubscribe@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/lojban?hl=en.

-- 
You received this message because you are subscribed to the Google Groups "lojban" group.
To post to this group, send email to lojban@googlegroups.com.
To unsubscribe from this group, send email to lojban+unsubscribe@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/lojban?hl=en.