Message-Id: <199511080121.UAA26155@locke.ccil.org>
Date:         Tue, 7 Nov 1995 20:01:32 +0000
Reply-To: ucleaar <ucleaar@UCL.AC.UK>
Sender: Lojban list <LOJBAN@CUVMB.BITNET>
From: ucleaar <ucleaar@UCL.AC.UK>
Subject:      Re: Incredible!
To: John Cowan <cowan@LOCKE.CCIL.ORG>
Status: OR

la djan cusku die
> > There are 2 kinds of syllable, C(@) and CV. @ is schwa and can be
> > omitted between certain consonant pairs. Cmavo are all of form CV
> > or CVCV or CVCVCV, etc. Gismu are all of form C(@)CV (with 17 C
> > and 5 V, that gives 1445 possible gismu; Lojban actually has 7 V
> > phonemes and 22 C phonemes, so that gives 2904 possible gismu).
> This involves throwing overboard the current gismu-assignment system
> with its effort to be mnemonic in six languages proportional to their
> number of speakers, of course.  That principle is actually far older
> in the Project than the rafsi principle.

The policy needn't be thrown overboard; it just gets applied less
effectively. The traditional etymological method is not that useful
- after all, very few gismu bear any mnemonically useful resemblance
to their English counterparts - and since the algorithm itself was
only ever rough-and-ready, and never perfected, its role should not
be overplayed. As I said, patterns like "culture gismu end in O" are
more useful.
----
And