Re: Self-segregating morphology again - in simpler terms, with list of methods

Jim Henry wrote:

> Maybe it would make sense to collect a list of methods
> for self-segregation.  I'm thinking about this a lot lately
> because I want my next conlang to have such a feature.
Jaghri marks the end of a word component by an open syllable. So all
word components are of the form V, CV, CVCCV, CVCCVCCV, and so on. A
syllable consisting of a single vowel without a consonant marks the end
of a word and specifies its role in the syntax.

panngajriiyidraghiatilkuriutimbiibalriingurpai:
panngajrii: panngajri - i
yidraghia: yidra - ghi - a
tilkuriu: tilku - ri - u
timbii: timbi - i
balrii: balri - i
ngurpai: ngurpa - i

But Jaghri isn't entirely self-segregating, since some word components
are formed of shorter morphemes joined together:

pertikra (bluebird): per- + tikra
kagvifta (drumstick): kag- + vifta
milticni (thirty): milti + -cni
vilnimpa (lemon): vilni + -mpa

There's no way of knowing whether pertikra is per- + tikra or perti +
-kra without looking it up. This could be fixed by placing a stress on
the first syllable of a two-syllable morpheme: pertíkra, kagvífta,
mílticni, vílnimpa, or on the second syllable: pertikrá, kagviftá,
miltícni, vilnímpa.