Re: [lojban] Consonant cluster clarification

[Richard Curnow <richard@rrbcurnow.freeuk.com>]

On Wed, Jan 31, 2001 at 05:59:09PM -0500, Pierre Abbat wrote:
> 
> It appears to me that all stage 3 fu'ivla formed from a CVCCV gismu violate
> this rule, for instance "jinmrtitani" has the four-consonant cluster "nmrt" in
> it. So is the rule actually that consonant clusters can have up to three
> consonants in a row between syllabic consonants?
> 

To get consistency with the rules for stage 3 fu'ivla, it certainly
seems true that a syllabic consonant (l, m, n or r) is required so that
no cluster of >3 other letters remains.

I was wondering if anyone has a view on what happens in an initial
cluster of a (stage 4) fu'ivla.  The rule is that all pairs in the
cluster must be valid initial pairs.  Does a syllabic consonant within
the cluster (if any) cause the end of what is considered the initial
cluster, or not?  i.e., do the consonants after a syllabic consonant
only have to be valid pairs, i.e. not necessarily valid _initial_ pairs?
My interpretation of the book is that this is *not* the case, i.e. all
pairs within the initial cluster must be valid initial pairs.  Under
this interpretation, the longest initial cluster would be 5 consonants
in length, of the form CCCml or CCCmr (where the CC pairs are valid
initial pairs, as is the final C+m.), to be interpreted as CCC,mr etc.
This exploits the fact that m is the only syllabic consonant that can
begin a valid initial pair.

Thoughts anybody?

-- 
----------------------------------------------------------------------
Richard P. Curnow                                       rpc@myself.com
Weston-super-Mare
United Kingdom                            http://go.to/richard.curnow/