--- On Sat, 2/28/09, Matthew Barnett <[log in to unmask]> wrote:
> steve rice wrote:
> > A major problem arises with the diphthongs. Bislama
> would render these as <ae>, <ao>, <oe>,
> which might work better than my more typically Euro-American
> assignments. The problem arises from the pseudo-ending
> <in>/<en>, which helps distinguish certain
> > laik (similar) vs laikin (enjoying, pleasure)
> > no (no[t]) vs ?noin (knowing, knowledge)
> > The problem is that ?noin should be a disyllable. So
> how do I force the break? I considered no'in and
> nowin--the second may be slightly better--but it seemed
> easier to change -in to -en: noen. This also means that
> while -en may remind users of -in[g], it also suggests the
> participial -en:
> > pruven. (Though the usual meaning of such words is
> active, in context I usually don't see the need to mark
> such things: da raiten lanwij "written language,"
> not "writing language." I can imagine a fantasy
> story where a certain language always inscribes itself on an
> available surface when used, but that's atypical.)
> > Any suggestions?
> I'd suggest a single rule as to whether two adjacent
> vowels are a single diphthong or two syllables.
I should've mentioned that I tried using <ay>, <oy>, <aw>, but the last one in particular seemed to cause confusion, so I switched back. The current "rule" is simply to define three diphthongs and say that in other cases the vowels are separate. Even using <w> and <y> to form diphthongs wouldn't resolve all problems: if you have a sequence such as ayV, does it group as (ay)V or a(yV)? On the other hand, under the current system, aiV groups as (aI)V and ayV as a(yV), so "aian" groups unambiguously as aI.an ("iron"). (I like consistent rules as much as anyone, but consistency is a means, not an end.)
> I think I prefer the diphthong, with an apostrophe to
> represent, say, a glottal stop if you want them to be
> separate syllables.
I hadn't thought about it, but I don't use a glottal stop myself. I prefer the use of <w>, but it sets up an awkward special rule. You're probably right about the apostrophe to avoid diphthongs, but since I dislike proliferating apostrophes, I'll probably stay with -en to reduce the number of cases where it might occur. Someone may eventually import ?oke'ik (I should've mentioned that -er -> a and ar -> o).