Re: lei xau-dja-sei

["Jorge J. Llambias" <jorge@INTERMEDIA.COM.AR>]

And:
> kb> le prenu pa'a tcidu le cukta
> kb> (The cmaste happens to use this very example; look under pa'aku.
> kb> It means "respectively")
> I don't have an up to date ma'oste with this in. And I'm
> rather irked to find that my time away from Lojban has led
> me to forget how to construe BAI before selbri

It's simple: "BAI broda" means "broda BAI zo'e".

> [O hurry up
> codical refgrammar!].

ie au

> But even so, I find your report pretty
> mystifying: I can't see how it would work.

It could be called a lojbabism, I guess. The normal working of pa'a,
as I understand it, would be something as in:

   le prenu cu tcidu le cukta pa'a le nu le smani cu ciska cy
   The people read the books as the monkeys wrote them.


> xj>        le prenu cu tcidu lo frica cukta
>
> Your version means "book having something to do with differers".
> I want "each person read a book not read by the others".

You could use "lo frica je cukta" for more precision. But you're right
you still have to make assumptions. I know it's not what you want.

> xj>        le remei poi ge pa ke'a fraso gi pa ke'a dotco
> xj> The slight difference is that the Lojban here allows for one
> xj> member being franco-german and nothing is said of the other,
> xj> but I think pragmatics should take care of that.
>
> Is "pa ke'a" legit?

Yes.

> What does it mean?

I think that's an open issue. "ke'a" here stands for a mass, and it
is not completelly agreed, as far as I know, what does a number in
front of a mass do. My preference is that it demassifies the mass
and refers to individual components. If you don't agree with that,
then you can use "pa lu'a ke'a" instead, which gives you one of the
elements of ke'a, i.e. one of the pair in this case.

> Something like "one member of
> which"?

Right.

>Anyway, I'm seeking something corresponding to exactly what I
> was after.

I know. How about:

le remei po'u pa fraso ku joi pa dotco
The couple which is a french and a german.

> xj>        le mamta be la xorxes a la and
> xj> Notice that this does _not_ expand to:
> xj>    le mamta be la xorxes ku a le mamta be la and
> But it does expand to
>
>            le ga mamta be la xorxes gi mamta be la and

Well, it would if that were grammatical, but it isn't.

> - this would work for "ro mamta be la xorxes a la and"

which is what "le" gives you, more or less.

> but not
> for "lo mamta be la xorxes a la and", which could refer to just
> your mum.

Right, that would be at least one of the mothers of Jorge and And.


> xj>        le mamta be su'o le nanmu
> xj>        Each of those that are a mother of at least one of the men.
>
> Okay, "le mamta be su'o le nanmu" is good. Let's change the
> example to:
>
> 3a   the sisters of the men   [each of the men have sisters who are
>                                being referred to] (same as 2c)

Same thing:

        le mensi be su'o le nanmu
        Each of those that are a sister of at least one of the men.

Ah, I see, here we are not saying that each man _has_ a sister.
I think a prenex is in order then:

        ro da voi nanmu zo'u le mensi be da
        For each man, x: Each of x's sisters.

("le mensi" is "ro le su'o mensi", so I am saying that there is
at least one mensi for each da.)


> 3b   the mothers of some men

Same deal:

        da poi nanmu zo'u le mamta be da

It's not so bad. Kind of a topicalization thing, which is right,
because you are putting emphasis on the quantifier of the modifier.

> > 2c the children of the women
> xj>        le panzi be su'o le ninmu
> xj>        Each of those that are a child of at least one of the
> women.
> The latter is better, but how to get the meaning where each of the
> women has children?

Let's see, we can be even more succint:

        le ninmu zo'u le panzi be ny

Or:
        le ninmu zo'u le ny panzi
        Each of the women: Their children.

Yes, definitely works.

> (Same goes for "children of xorxes and and".)

This one's a bit tougher:

        roda po'u lu'a la xorxes ce la and zo'u le panzi be da
        For all x which is a member of {Jorge, And}: The children of x.

Can't think of a nice and short form. We could say:

        la xorxes e la and zo'u le panzi be da
        Jorge and And: Their children.

But you would have to believe me that da runs over the prenexed even
when not explicitly bound to them. It would be within acceptble
convention, though, what do you think?

> OK, but not enough of a solution for me. Change example to:
>
> 3c we are french and german [but not necessarily dual nationality]

Yes, I see, "mi'a fraso ja dotco" does not guarantee at least one
French and at least one German. Actually, this is not such a good
example because "mi'a" is a mass, so "mi'a fraso je dotco" would
probably give you what you want. Let's say you want "the people
are French and German", and you insist on using "le prenu" rather
than "lei prenu" for "the people". Then I don't know.

> The examples in 1 are respectives.

For respectives, I think that the answer has to be in using
prosumti, something like:

        le prenu cu tcidu le py cukta
        Each person reads their book.

where the "py" will force "le cukta" to get a different referent
for each prenu.

> The examples in 2 are different.

Yes, I think what's going on there is that we need external scope for
quantifiers of things which are sintactically internal. Maybe you can
do something with your old scope jumpers, which I forgot what they
were.

> ki'e markl, xorxes, kris
> i'o sono vostro pendo and

je'e nostro pendo
i'o sono pendo cudu so'i pendo
co'o mi'e xorxes