Re: Gender (was: LANGUAGE LAWS)

At 1:57 am -0400 23/10/98, Nik Taylor wrote:
>Mathias M. Lassailly wrote:
>> So those 8 categories do not classify items in the world into big
>> families through specific logics now forgotten ? :-)
>
>They do have general meanings, for example, in Bantu (and I'm assuming
>Xhosa is the same way, being Bantu and all), gender 1 is human.
>However, they are not fully consistent.
That's correct on both accounts.  And although the categorization must have
followed some 'specific logics now forgotten', it was hardly scientific in
the modern sense.

[snip]
>females.  But, if the culture changes, those reasons may be lost, so
>perhaps the exceptions in many languages may have been rational in
>earlier stages, but who knows?  It's like Sally Cave's metaphor of the
>old city, you may have ancient buildings, built thousands of years ago,
>next to buildings built just a couple of years ago, built according to
>different architectures and for different technologies
Yes, and I've come across a theory that the familiar IE three gender system
is the result of two 'different architectures' - an animate/inanimate
division & a male/female division - but I don't recall which is thought to
be the older system.

It strikes me that this is also the problem with a_priori IALs that try to
categorize all entities - they can do this categorization only in the light
of the understanding of the culture & science of their times.  As knowledge
advances, their "philosophical" categories are going to start to show
inconsistencies also.

But I must abandon this discussion for a few days - I'm off to la Belle France.

Ray.