Re: Consonent Clusters at the End of Words Saul Epstein Thu, 13 Nov 1997 12:34:52 -0600 From: Rob Zook Date: Wednesday, November 12, 1997 4:42 PM > At 02:53 PM 11/12/97 -0600, Saul wrote: > > > >a palatal fricative. Also, since Vulcan doesn't otherwise have a > >dental fricative, we'd probably end up with a phonetic realization of > >[ds] for /dth/. > > Hmmm. Actually I was thinking of the opposite, a phonemic /dth/ but > realized as the [dth] where the was the fricative. That doesn't sound opposite. Your proposed rule transforms aspirates to fricatives when following a voiced stop at the same place of articulation. Since Vulcan otherwise has no dental fricative, but does have an alveolar fricative nearby, it is likely that the latter would be substituted. Which is not to say that a dental fricative couldn't show up as an allophone of the aspirate. It could go either way. I'm just reminded of a situation in various Indo-European languages at different stage, in which a voicing rule would have transformed [s] into [z], but in some languages transformed it into [r], because they had no [z]. The past participle of "lose" was "loren" before that verb weakened, which still shows up in "forlorn." > >>Unfortunately, this does not seem quite consisant either, because > >>p -> f and k -> ch does not have the same relationship as t -> > >>th. > > > >k -> x is the same relationship, though. As is p -> wh. Oh, wait. > >Do you mean the aspirate or the fricative th? > > I mean the fricative. For example the IPA chart shows no dental > stops, and t is a voiceless alveolar stop. That's a somewhat ambiguous part of the chart. There are dental, alveolar, and postalveolar stops, all of which are typically represented by , , . This works because it is rare for a language to have separate stop phonemes in each of these places. When a language does, there are diacritics to distinguish them. > So with t -> th not only does the > amount of closure change as in p -> f, and k -> ch, but the point > of articulation changes as well, because th as realized in English is > a dental fricative not a alveolar fricative. Technically, the point of articulation is changing in each of your examples -- except for t -> th, which might or might not. > >> appears in k'wawl, and in this case the aw could be some > >>kind of vowel dipthong maybe /ao/? If not, perhaps you can have an > >>[approximate]+[lateral] at the end of a word: yl wl. > > > >It winds up being much the same thing. I tend to restrict > >approximates to occuring before vowels, but that's just me. > > Well, lets look at an English example, "bowl". In that case not > only do you have a diphthong, but in the dialect I speak you have an > ending [wl] sound. Fascinating. I'd probably have to hear you talk to know what you mean. > >>Or if /w/ > >>does indeed represent a voiced bilabial fricative as Saul > >>suggests it could mean you can have [fricative]+[labial] at the end of a > >>word: fl vl sl zl cl jl xl hl. > > > >Or we could change _this_ w to u. But the clusters above would be > >OK as well. We could expand it to fricative+approximant. French- > >flavored with l, Russian-flavored with y. They wouldn't phonetically > >be clusters, probably, but lateralized and palatalized consonants. > > Hmmm. Maybe you can explain the difference? Well, if you listen to the right sort of native French speaker say the word "table," they can do it in one syllable. The final sound is a [b] but as they release it, if not before, they get their tongue into an [l] position. Russian has a semi-letter, whose name I've forgotten (myakiznak?) which palatalizes the preceding consonant. Basically, you make the usual sound represented by the preceding consonant, but you push the middle of your tongue up towards the roof of your mouth as you do so, lending the consonant an accompanying "sh" sort of sound. > >>Did I understand you aright Saul? In your last message you > >>suggested /w/ was also a fricative. Since /wh/ is already the voiceless, > >>that would seem to make /w/ voiced. Frankly, I'd rather see a voiced > >>approximate with a voiceless allophone. > > > >Mm. The difference between a voiced approximate and a voiced > >fricative is tiny. The only thing I'm trying to do here is secure an > >orderly role for wh/w in the phonology. Even if /w/ is phonemically a > >fricative, it might almost always be pronounced as an approximant. > > So you would say /wh/ and /w/ both fricatives with approximate > allophones? That's how I'd like to describe them, yes. > >>khp khb kht khd khk khg khq > >>th thb tht thd thk thg thq > >>dh dhb dht dhd dhk dhg dhq > >> > >>I kind of like these simply because of their exoticness. That > >>could give us words like qyoodhk or kakhg which look nice and tricky to > >>pronounce ;-) > > > >That almost makes it look like h is a voiceless form of <'>. > > Similar. I think /h/ is a voiceless glottal fricative. Yes. And if I'm right, <'> between two consonants, neither of which can be made syllabic, is pronounced as a minimal vowel -- which is a sort of voiced [h]. -- from Saul Epstein liberty*uit,net http://www,johnco,cc,ks,us/~sepstein "Surak ow'phaaper thes'hi thes'tca'; thes'phaadjar thes'hi suraketca'." -- K'dvarin Urswhl'at