Re: ka/ni kama

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

>2- {ni} as an indirect question.
>
>2a.                 mi nelci le ni la meris cu ninmu
>                       I like to what extent Mary is a woman.
>
>2b.                do mi zmadu le ni ce'u nelci le nu ri dansu
>                      You exceed me in the amount we like to dance.
>
>An example from the refgram:
>
>2c.                 le pixra cu cenba le ni ce'u blanu
>                       the picture varies in-the amount-of (X is blue)
>                      The picture varies in how blue it is.
>
>In these cases, we can't replace {le ni...} with a number. Suppose
>that {le ni la meris cu ninmu cu du li piso'i} "the extent to which Mary
>is a woman is a lot", then 2a. would be saying {mi nelci li piso'i},
>which, to me, is nonsense, and it is certainly not what is meant.
>
>If {le ni ...} is indeed a number, as the definition and the examples (1)
>indicate, then the examples (2) would be cases of sumti raising.

2a does seem to be sumti raising, in that if you replaced "le ni ...
by its numerical value, you would be saying that you are fond of a number.
Probably it is a sumti-raising from
mi nelci le ka/ni le ni la meris cu ninmu cu zmadu leni lo'e ninmu cu ninmu"
or something like that.  I am fond of the degree to which Mary's femininity
exceed the norm for women.  (On translation, it seems clear that it
should be lo'eni lo ninmu cu ninmu).  All of this presupposes that two women
can have different amounts of womanliness - an assumption implicit in using
"ni".

I do not see why 2b and 2c cannot be replaced by a number, provided that one
can figure out an appropriate scale.  Actually in the case of 2c, color
measurements usually use two or 3 numbers to measure amount of blueness
(saturation, hue, and maybe some other number) and it might be true that in
collapsing these 3 numbers to a single scale you might have a change in
ka blanu that is not a change in ni blanu.  But the e-trans of 2c says
specifically that this is a change in "how blue it is" which would rule
out a change in quality of blue that was not also a variation in quantity.

>But the problem is that very often {ni} is used in the indirect question
>mode without tu'a marking, even in refgram examples.

I think you are uncomfortable with the idea of something having a quantitative
measurement when no scale for such quantification has been defined.  One
could consider this a sound scientific prejudice %^).

>I don't agree they are parallel cases. If something is not long by a
>given standard, why would the amount of it being long by that standard
>be different from zero.

I think we are confusing standards and standards %^).

One way of defining a standard for truth is by similarity to some ideal.
With such a standard, something is "not broda" if its delta from that ideal
exceeds some amount (usually its delta from some ideal of na'e broda).

The other kind of standard often used is a measurement on an absolute scale
from 0.

We know how to use quantities of the second sort.  The whole concept of the
fuirst kind of quantity is rather new in semantics (it doesn't seem to have
occurred to the color-testers in the 50s who were testing Sapir-Whorf using
cross-cultural color comparisons, but then I have n't read the lieterature of
those experiments).

It is possible that ni broda for some broda will use a scale of degree of
delta from an ideal. in which case "zero" might be a high degree of meeting
the ideal and tus better than "one" whatever that value might mean.  It is
all in the x3 of ni.


>I don't think it applies to ka, since ka abstractions are never
>full bridi, they can't claim anything by themselves. For example:
>
>        le jubme cu claxu le ka ce'u cukla
>        The table lacks the property of being round.

"never"???

le ka lemi speni cu fetsi cu se flalu  fi ti
so I can say
ka lemi speni cu fetsi
just as much as I can
lemi speni cu fetsi
Both claim relationships - different ones.  The first claims that the second
relationship can be chacracterized.

This may be trite, but it does make a claim.

I will leave it to someone to come up with an example of "na ka" though %^)

>In the case of nu, I'm not sure whether there is agreement that
>talking of {le nu broda} makes the implicit claim that it happens.

I am sure it does not.  The whole point of abstractions is to allow someone
to talk about relationships without assuming that they exist.

Maybe that solves the previous problem.  Surely
na ka lemi speni cu nakni
just as
lemi speni cu na nakni

>As for {jei}, assuming it is the truth value and not the indirect
>question, then it obviously doesn't claim its bridi. If the bridi is
>false, then {le jei <bridi>} will exist and be the truth value "FALSE".

No, it will be a truth value other-than-1.  Most people assume "false"
means "zero".

>>It doesn't
>>relate at all with si'o, su'o, or li'i, so the "rule" that ni is bending
>>when it assigns values to non-broda is not really a rule of abstractors at
>>all, but an incorrect generalization based on the behavior of ka.
>
>I don't claim any such rule for abstractors. If ni has the number
>meaning it can't really be compared with ka. If it has the indirect
>question meaning then yes, sometimes it acts just like a ka. In
>those cases {le ni <bridi>} is very similar to {le ka <bridi> la'u
>makau}.

I don't know why the ka is necessary.  Why not "makau poi bridi la'u ke'a"?

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.