Theiling Online    Sitemap    Conlang Mailing List HQ   

IPA/CXS questions

From:Tristan McLeay <conlang@...>
Date:Thursday, December 8, 2005, 5:22
On Wed, 2005-12-07 at 23:04 -0500, Mark J. Reed wrote:
> On 12/7/05, Roger Mills <rfmilly@...> wrote: > There's also "reversed epsilon" (CXS [3]) which I at least > modify with > circumflex (IPA hook) for the retroflexed _stressed_ vowel of > bird, hurt et > al. [3^]. But I'm not sure this is proper CXS. Your [r\=] gets > the point > across too. > > I thought the CXS for rhotic hook was `, not ^?
Yes, that's true. ` is a rhotic hook on vowels and a retroflexion diacritic on consonants. I can't think of what ^ means though---I think some people might use it to denote superscripts, thus t^h instead of t_h
> The unstressed final version (as in the -er suffix) would then > be [@^]; the > two vowels are phonetically distinct IMO. > > Not IMD; "bird" and "encumbered" and "burred" are a three-way perfect > rhyme. But that way lies YAEPT.
I don't mean to take you up on the YAEPT, but I am slightly curious---is it really a perfect rhyme, and the difference in stress doesn't get in the way? I thought most dialects had the last syllable of "encumbered" unstressed, and rhyming is usually considered to occur from the last stressed syllable unwards---thus "encumbered" can normally only rhyme with two-or-more syllable words. (At this point, Roger wrote and Mark snipped:)
> > One advantage of this system is > > that for non-rhotic dialects, you simply drop the diacritic ^.
Not quite true---most transcriptions of non-rhotic dialects mark length, whereas American English at least is typically written without length marked, thus /3`/ -> /3:/ and /@`/ -> /@/. Still, in the normal way of writing RP you can always work out from the symbols whether a length mark should follow or not. (One common way of writing Australian English doesn't mark length used by frex. the Macquarie Dictionary; it strikes me as pretty silly because then the only phonetic difference between /fVs/=[fa_"s] and /fAs/=[fa_":s] is not even mentioned.) -- Tristan


Roger Mills <rfmilly@...>