Re: [lojban] stress, capitalization & audiovisual isomorphism

[pycyn@aol.com]

In a message dated 11/26/2001 11:44:00 AM Central Standard Time, arosta@uclan.ac.uk writes:

#Indeed, it would be: specifically, it would cease to be an isomorphism,
#since <y> would sometimes mean [@] and sometimes [j].

but <y>-adjacent-to-V would always mean [@] and <y>-not-adjacent
to-V would always mean [j]. And vice versa from sound to spelling.
So why is this less of an isomorphism than <V>-in-penult-syllable
= stressed-[V] and <V>-not-in-penult = unstressed-[V]?

Hard to argue the point, except to say that the isomorphism would likely fail at the phonemic level, since the y-glide is an allophone of /i/, not /y/.  And there is the added glyph w for something that is only an allophone of /u/.  On the other hand, it gets rid of ', an unsightly object, which nonetheless represents a real sound and not an allophone of anything (except non-juncture).  So, as Chao would say, we could reanalyse the sound system, making the ' be an automatic reflex of vowel-vowel contact when glottal stop was absent.  But then we would always have to write in glottal stops (or maybe just be more careful about spaces).  While we're at it, why not make the glide an allophone of /y/  -- and make w also apply to the buffering vowel yer, allowing one to write out dialects?