Hi,
So far as I know the creators of Lojban never
claimed nor intended it to be ISOMORPHIC to logic.
Rather, the goal as I understand it was to make a
language that provided the best possible compromise between
* logical soundness
* practical usability
A language that was simply a verbalizable form of
predicate logic would fail the "practical usability" criterion.
As Cycorp has found, carefully and correctly
encoding an ordinary English sentence in predicate logic takes a trained
individual at least 10-15 minutes. This kind of time-requirement is not
viable for a spoken or written language.
What is really cool about Lojban, IMO, is that it
shows how far one can go in the direction of logicality, without making huge
sacrifices in terms of the time required to express commonsensically simple
things.
Regarding your comment about the existence of
different orderings of sentences that are semantically equivalent, I don't think
emphasis is the main point here, but rather cognitive naturalness. If you
tried to impose a fixed ordering on all Lojban sentences, I think you'd be
making the process of sentence-formulation too cognitively unnatural, which
means that speaking and understanding would be made to take a significantly
longer time than with the current version of Lojban. So I view this as an
example of the necessary compromise between logical precision and practical
usability by human beings.
Of course, the Lojbanic system of
precision/usability compromises is not the only possible such system, and there
may be a better one. But it is clear to me that over the history of
Loglan/Lojban a lot of clever folks have put a lot of thought into "tuning" the
Lojbanic system of compromises. Even if one found another
logical-language-structure that was fundamentally better than Lojban, I still
suspect it would take a lot of effort to "tune" it into a really workable
language (which so far as I can tell, Lojban *just barely* is, in spite of all
the work that's gone into it...)
- Ben Goertzel (also a Lojban novice)
|