curmi

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

>> >Why does {curmi} have "under conditions" but not {gasnu}?
>>
>> curmi: like xanka, the conditions tend to relate to the x1's state of mind,
>> while conditions might also be found inside the sumti clause being permitted,
>> that are unrelated to the state of mind.
>>
>> Why should gasnu?  You either ARE an agent, or you are NOT.  I don't see
>> much basis for this comparison.
>
>You could just as well say that you either LET something happen, or you
>do NOT. Why is your state of mind important when you let something fall,
>but not when you make it fall?

Because if your state of mind is important (at least as people normally
see states of mind being related to actions), you can use zukte.  "let"
implies a mental decision on the part of a person that can affect the
occurance of x2. curmi requires such a mental state. gasnu does not.  A
mindless object can be a gasnu, but not a curmi.

One does not say that A "lets B happen", unless B's potential happening
imposes some kind of mental state on A that relates to a potential
capability of A to interfere with B. I can't truthfully say "I let the
sun rise", because I can't help it or hinder it.  I can say "I let my
son go outside" because I have the potential to interfere, and "let"
implies a mental decision to NOT interfere.  It is uncertain whether "I
let my son think about going outside" makes sense since I don't know how
I could stop him, but I could say "I let myself think about going
outside."

I suppose it is possible to stretch curmi to include amental x1 - does a
brake pedal not depressed "let" the car keep moving?  Or is it only
possible to say thhat with reference to the point of view of a person
who could decide to use the brake pedal.  Note that "curmi" is not
necessarily "fail to stop/interfere" - the brake failing to stop the car
did not "curmi" the car going off the cliff.

The gismu list is an attempt to divide up semantic space as it is used
by people, not to try to impose some kind of artificial parallelism on
all concepts.  The division need not be optimal, so long as it is
adequate.  I would suspect that if I had an amental example parallelling
curmi, I would be able to find some phrasing using some other gismu that
does not imply a mental state.  The gismu list is adequate, if we have a
way to do so for all actual thoughts that occur to Lojban speakers.  I
don't care about potential thoughts - anything is possible.

lojbab