TECH: more on morphology problem - some opionions

[Logical Language Group <lojbab@GREBYN.COM>]

Well, no discussion of the problem on the net. But Nora and John Cowan
have given some verbal feedback.  They seem to agree that:

1) The type 3 borrowings that cause the problem are prevalent these days,
and must continue to be easy to make.

2) le'avla/borrowing morphological space is sufficiently large and rich
enough that, when we get to the point where people want to make them,
type 4 (pure borrowings) le'avla have plenty of options, even if we add
some constraints.

Therefore, both favor solving the problem by constraining the pemritted
forms of le'avla enough to ensure that no morphological problems exist,
and Cowan further wants to take some preventative action by defining some
of the losse ends more rigidly.

The consensus thus far (which I can accept as well) might include

1) that close-commas (forced syllable breaks) which are 'significant'
would be banned from le'avla.  In other words, no two le'avla could differ
merely by whether there is a close-comma forcing a syllable break at
some point.  This eliminates some strange vowel glide posibilities and
presumably would allow us to  have "io" in a le'avla in contrast to "i'o",
but not requiring opposition of "io,io" with "i,oi,o" etc. sommething that
we had already forbidden in the case of UI cmavo in succession due to
possible resolution problems.

Close commas could still be used in non-significant pronunciation-aiding
ways, and to make the components of the lea'vla clear, as in
"bangr,speranto", where you know that the classifier rafsi is "bang-"
for "language", and can also see that the 'r' at the juncture is intended
to normally be pronounced syllabically.  (But given the restriction, would not
cause problems if you instead pronounced the 's'  syllabically per the
syllabization "ban,grs,per,an,to", though the comma would encourage against
this).

Close commas could also still bne used in names.  Parsers and other computationl
 products might not be requires to support them, though.

2. A new phonological category of "permissible initial cluster" be defined,
to apply to le'avla.  Clusters at the beginning of words would be required
to  be such that all pairs of consonants within the cluster be permissible
initial pairs, per the standard set already defined.

If vowel-initial borrowing roots had a suitable consonant preposed (and Nora
suggests 'x' as the norm, as being an easily invisible sound, and not too
hard after a liquid), then the use of only 1 of the three {r,n,l} as glue
syllabic consonants to attach a rafsi on the front of the root as a
classifier, should prevent ever having that first consonant cluster being
a permissible initial cluster as just defined.  This is because you would
always have r/n/l followed by another consonant, and r/n/l are never the
first letters of a permissible initial pair.

3. A self-consistent definition of permissible medial cluster would be
devised that covered these messy combiantions that can arise in le'avla.
the current definition in the Synopsis was written only with gismu and
lujvo in mind, and hence is vague and/or forbidding of the norm for lujvo.

4. Cowan will investagate, and may propose some norms for final clusters,
which would set a non-mandatory standard for use in names.

How does this sound?  Does anyone care???

lojbab