Re: [lojban] Re: Are Natlang the best case for entropy in communication ?

[Escape Landsome <escaaape@gmail.com>]

> No evidence of any kind. It's beginning to sound like a mindless chant.
>
> As those in the scientific community are quite fond of saying, PROVE IT.


Entropy is defined as proportional to sum of p.ln p where p are the
probabities of the given choices.   That means if you begin by
admitting equiprobability for the sake of simplicity, the more choices
you have the more confusion you get.

A paradigm being a natural set of choices that arises in natural
context, it follows that the complexity of two different paradigms
cooccurring in the same utterance is less than one single paradigm
which has product cardinal of the two former paradigms, because the
context specifies clues that helps to desambiguate each of the two
former paradigms by the other, --- which cannot be done with one
single big paradigm.

This is well-known and widely acknowledge in Linguistics.

Thus, the example you give with
[paradigm A with 2 choices] with [paradigm B with 2 choices]
is much less likely to confuse than the one with
[paradigm C with 2x2 = 4 choices]

the negative difference is the amount of information given by the cross-context.

Thus, any natlang example of 4 choices given as Aa, Ab, Ba, Bb, will
be less harmful than the lojban example of so'V where you have Xa, Xe,
Xi, Xo.

This general reasoning must be amended by the fact that some paradigm
are much more used than others.   It is a reasonable argument that the
QUANTITY paradigm is much more useful than another very particular
paradigm, such as RACE OF HORSE, for instance.

-- 
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.