Re: [lojban] Predicate logic and childhood.

[And Rosta <arosta@uclan.ac.uk>]

Rob:
#On Wed, May 02, 2001 at 04:21:22PM +0100, And Rosta wrote:
#> #People seem to be implying that as soon as there is cause and effect involved,
#> #you are not allowed to use logical connectives. Not that you can choose not to
#> #use them in favor of a cause-and-effect statement, but that you just can't use
#> #them. I have yet to see an answer to why there should not be a choice of
#> #sentence structure.
#> 
#> I have lost the thread, I'm afraid, and can't reconstruct what are the issues
#> under debate. Would you be willing to take the trouble to recapitulate?
#
#The sentence that started all of this is "If you clean your room, I'll let you
#go to the park."
#
#I suggested {ko (do bazi) nicygau ledo kumfa .ijo mi curmi lenu do klama le
#panka}.

Thanks. I have been reading the thread, but I still don't see how you reach
the conclusion that 
  "People seem to be implying that as soon as there is cause and effect 
  involved, you are not allowed to use logical connectives. Not that you 
  can choose not to use them in favor of a cause-and-effect statement, 
  but that you just can't use them. 

I see nothing at all wrong with translating the sentence with "ijo".

Complications are added by your suggestion of "ko". The English is
not an imperative, and the translation does not require "ko". Of course
there are English sentences like:

   Clean your room and I'll let you go to the park.
   Take one more step and you're dead.

   Clean your room or I won't let you go to the park.
   Stop or I'll shoot.

and these present some semantic issues much debated in linguistics.

#xorxes thinks that this sentence means something different than "If you
#clean your room, I'll let you go to the park" and should not be used to
#translate it, instead suggesting either using the x3 of curmi (spawning a
#side-debate about what the x3 of curmi really means) or some sort of
#"conditional" which expresses it in terms of cause and effect. (I have no idea
#how this would be formed in Lojban, and I don't believe an example has been
#provided, though it might be 'rinka' with a 'nu' on both sides.)

I would translate your English sentence with ijo but changing "ko" to "do".

Your Lojban sentence means to me either (a) "Iff you go to the park, I hereby
order you to clean your room", or (b) "I hereby order that iff you go to the
park you clean your room", which is close to what you want, but allows
that the addressee may neither go to the park nor clean their room.

(a) seems silly, but seems right for "Queue here if you want to see the film"
= "Iff you want to see the film I hereby instruct you to queue here".

These problems arise because {ko} conflates both the command and
the reference to the addressee. These are logically separate and if
they're not linguistically separable then problems ensue.

#lojbab posted a confusing message where he suggested that logical connectives
#would be used for things like "If wishes were horses, then beggars would ride",
#but it seems that, for one thing, that wouldn't work at all, and also that
#seems to be another issue which I don't think was satisfactorily resolved.
#
#I believe that both .ijo and the conditional would give a good enough
#approximation of the English meaning of the sentence. English is fuzzy enough
#that we don't need to argue about how to translate the most detailed
#implications of a sentence.

I don't think implications need be translated at all. Just translate the meaning
the sentence encodes, and what the original implicates the translation will
also implicate. For example, your English example implicates a command
without encoding it, and hence the Lojban can perfectly well do the
same.

#Incidentally, I'd like to see how this "conditional" would be constructed. I'd
#rather not hunt through the archives of the list.

You mean how to translate the English sentence? Well, to repeat, mine
is:

   do nicygau ledo kumfa .ijo mi curmi le nu do klama le panka

--And.