[YG Conlang Archives] > [ceqli group] > messages [Date Index] [Thread Index] >
--- In ceqli@yahoogroups.com, Rex May - Baloo <rmay@m...> wrote: > on 2/21/04 9:39 PM, HandyDad at lsulky@r... wrote: > > > --- In ceqli@yahoogroups.com, Rex May - Baloo <rmay@m...> wrote: --SNIP-- > Tho the ceqli wordshape rule calls for one or more > consonants at the beginning, it seems to in practice be one or two > consonants. Can we use Italian as a guide? If it's permitted in Italian ? > initial clusters, I mean ? it's permitted in ceqli? > > Looking at the vocabulary, we have as initial clusters: > > sf > sk > sp > st > ts > xp > xt > zb > > nothing for xk yet, but that'll be permitted. What about the voiced equivalents of those in the list? "zv, zg, zd, dz, zhb, zhd, zhg, zhv..." Plus "xf". > > That's not counting lrmn as consonants, of course. > > The only thing I'm really crazy about that I think you don't like is the ts. My concern is that a significant number of major world languages -- Asian and middle-Eastern ones, mainly -- have difficulty with consonant clusters, so I think we ought to permit schwa buffering. But if schwa buffering occurs between the two initial consonants of a word, then it aurally separates into a Ce-class word followed by a word with a single initial consonant: "zbano" => "z'bano" => "ze bano" Solutions are: 1) Prohibit initial consonant clusters (therefore, all consonant clusters). 2) Require full [e] pronunciation of letter "e" everywhere. 3) Require full [e] pronunciation in the words "se", "te", "xe", and "ze". 4) Prohibit the words "se", "te", "xe", and "ze". 5) Prohibit schwa buffering (i.e., demand accurate pronunciation of word-initial consonant clusters) 6) Not worry about it, figuring that we'll know when someone means "ze bano" not "zbano". I also looked at the Spanish model of schwa buffering, where they prefix 'sC-' words with a schwa before the 's'. But then the schwa runs into the tail end of the preceding word. > but if you think about it, it's analogous to the c sound > x is to c as s is to ts. They're both actually affricates, as is j, not > clusters. Quite so. Remember way back when I was urging that 'zh' be represented by "j" and 'j' by "dj"? And if we're going to have "ts", then why not "dz"? :-) > If we had an extra Roman letter, we'd use it for ts. And it's > very useful for German loanwords. as well as many Mandarin ones. And if we > can agree on this set, I don't think we'll be needing any more. I don't > want any ps or bz or anything like that. > -- --Krawn