On Thu, 14 Oct 2004 17:06:06 +0930, Adrian Morgan (aka Flesh-eating Dragon) <[log in to unmask]> wrote: >I would like to introduce a new consonant into Gzarondan phonology, >namely a phonemic distinction between /x/ and /x_w/ (unlabialised >versus labialised velar fricatives). > >I'm trying to decide what other changes to make to the phonology in >order to make the new phoneme set look reasonably naturalistic. I >would appreciate some advice. I've forgotten who or what speaks Gzarondan. You said naturalistic, so I'm assuming ordinary Earth-type humans. >Intuitively, I feel that if there is a /x/ vs /x_w/ distinction then >there ought to be a similar distinction for (at the very least) all >fricatives and approximates at or behind the velar position. This >would mean adding a /h/ vs /h_w/ distinction. It seems to me that /k_w/ and /g_w/ are more likely with /x_w/ than /h_w/ is, but I haven't exactly studied a lot of phonologies -- there are certainly much more unusual things in Natlangs. >Alternatively, I could do away with /h/ (but how many natural >languages don't have /h/ ?) and use the /x/ vs /x_w/ distinction in >place of the /x/ vs /h/ distinction. This conserves the total number >of phonemes. > >Building on the above approach, I could rewrite history to say that >all unvoiced fricatives once distinguished labialised and >non-labialised forms, but that /T_w/ later became /f/ and /s_w/ later >became /S/, whereas /x_w/ remained as it was. Your sound changes sound OK. The question I have is: why were only voiceless fricatives labialized? You might want to widen your scope and look at the whole recent *history*, including phonotactics and the vowel system. Hope this helps. Jeff >This sounds reasonable to me, but perhaps there's a better option. > >The current consonant phoneme inventory is listed below. > >Stops, Nasals & Trills: > >/p/ /b/ /m/ >/t/ /d/ /n/ /r/ >/k/ /g/ > >Fricatives, Laterals & Approximants: > >/f/ >/T/ /D/ >/s/ /z/ /K/ /l/ >/S/ > /j/ >/x/ >/h/ > >Adrian.