| My (highly fallible) memory has it that there were some studies at UCLA in
the 70 or 80's maybe by Vicky Fromkin and/or Peter Ladefoged on both various things that counted as r's and on sequisyllabic words like Carl and Earl. The archives should have some informative spectrograms, if so. The only postvocalic r's that seem to be problematic in Lojban are the Broad Brits (including those of the upper US), who try to get by with lengthening and various odd congestions in the back of their throats (which they regularly deny happens and sometimes does not), and really thorough uvular French r's. The first ends up being just the vowel to many ears, the second falls dangerously close to /x/ for people not too used to the real thing (and even some who are). Trills, flaps and retroflex fricatives work best -- and the last two make for easy syllabics. The uvular also works if extended (though I always need to spit shortly thereafter). In passing, I should note another syllabic consonant (or two) in some Lojban idiolects. These are /s/ and /z/ (I have not heard /c/ nor /j/ but they are surely possible and even likely) They turn up in Chinese names and correspond to Chinese sounds : {s,ma} (Ssu-ma) and {laud,z} (Lao-tse) and other like that. The first case can be treated as a buffered consonant cluster, maybe the second, too. But the "buffering" is quite different from the usual ones, being a buzz with not particular vowel articulation. |