Re: PHONO: Nasal assimilation (was: An incongruent orthography:

Matthew Kehrt writes:
 > Roger Mills wrote:
 > >
 > > John Cowan wrote:
 > >
 > > >Michael Poxon scripsit:
 > > >
 > > >> Surely all these instances are just assimilation of a nasal to a
 > > following
 > > >> homorganic stop, which explains why the versions with -m don't work;
 > > >
 > > >But they *do* work in Italian; the nasal is forced to be homorganic to
 > > >the stop no matter what the nasal's nominal place of articulation is.
 > >
 > > That's true, but then, the only normal final nasal in Italian is /n/ (unless
 > > memory is failing again).  Same in Spanish, despite a few loan in /-m/,
 > > which give some speakers problems, i.e. "álbum ~ álbun", or dialectally even
 > > "álbung".
 > >
 > > Buginese, with only /-N/ assimilates it to a following vd. stop.
 > >
 > > And Engl., like Indonesian, has contrastive final nasals (more functional
 > > load?), so we have to be more careful about assimilating them......(?)
 > >
 > > Nonetheless, Engl. fast speech rules seem clearly to allow assimilation,
 > > though /m/ does seem to be a little resistant, with a tendency to insert a
 > > [p] rather than assimilate--
 > > comfort ['cVmpfR=t] rather than ['cV__f@rt] (I disremember X-Sampa lab-dent.
 > > m)
 >
 > If anyone can tell me, what the HECK does all that mean?
 > I'm normally pretty good at following advanced conversations on the
 > list, but that's just gibberish.
 > -M
 >
Uhh, I thought I understood it, and I've no training in linguistics.
What part don't you understand?

If you assimilate a final nasal, then you can't tell what nasal it was
originally, yes?  So if it's possible to have words in the language
that differ only by their final nasal, like ham and hang in English,
you'd expect there to be some resistance to assimilation, as otherwise
you couldn't tell what the word was.  In some languages there's only
one permissable final nasal, and in these cases it's okay to
assimilate it as it can't cause an ambiguity.

My own conlang (which is unnamed, but I'll refer to it from now on by
the provisional codename LC-01) only allows /-N/ as a final nasal on
roots, precisely so that assimilation can take place in compounds.