Re: CHAT: XS vs. Kirshenbaum vs. Who-knows-what
| From: | Tristan McLeay <zsau@...> |
|---|
| Date: | Friday, January 30, 2004, 6:58 |
|---|
On Fri, 30 Jan 2004, Morgan Palaeo Associates wrote:
> Working Title: "FITS - Frivolous IPA Transcription Scheme"
>
> bil lad dap ret pal vel uv pha gl
> +-----+-----+-------+-------+-------+-----+-------+------ +-----+
...
> | P B | f v |TsS DzZ| s` z` | K~ G~ | K G | K` G` | h~ H~ | h H | fric
> | | | x X | | | | | | | lat fric
... (etc.)
~ is so strongly associated with nasalisation I don't see the point in
trying to usurp it for palatisation/pharyngealisation. Also with things
like /x/ for a lateral fricative? Seems silly.
> [{] begin double articulation or affricate (when necessary)
Is that prefix, a postfix, a roundfix or an infix? If it's a postfix, I'd
suggested [}] (i.e. [tS}]), otherwise it looks like it's grabbing what
comes after, and take something else for a voiced implosive; as an infix,
it increases the space between the two sounds, the opposite of what should
happen.
> Now, as for the vowels, let us first note that people tend to use
> x-sampa [3] for the vowel in "bird" regardless of whether it's really
> [3], [8], [3`], [8`], or anything in that vicinity (for my part, I
> still need rhoticity to be unambiguously explained to me, as it's
> something I have unresolved questions about). In general, [3] gets
> used to mean, "a roughly mid-central vowel, but with a different
> quality from schwa". Since in practise people rarely bother with
> higher resolution than that when discussing vowels in the mid central
> region, I have not bothered to implement such resolution, and have
> designated [R] to serve the above role
This is only true for English, TMK, and only because it doesn't have much
of a reason to change (whereas in AuE, at least one of [V] and [A:] have
to change to show the long/short, so you might as well change both). I
think you really should be distinguishing @\ from 8 from 3 from 3\. Or at
the very least rounded unrounded pairs.
Furthermore, seeing something like [bRd] will make someone think of a
rhotic vowel rather than a central vowel. If this system gains any degree
of currency, people will get a passive knowledge of it, but I doubt that
it'll be enough to remember that [R] is a proper vowel, not [r\=]...
> (not wholly unlike Kirsch).
It's actually 'Kirsh'.
> Vowel diagram
>
> +---------------------------+
> | | also [R] to represent the vowel in 'heard'
> | i y i> u< | u | i.e. roughly mid-central but distinct from schwa
...
> | e c q o |
> | @ |
> | E C Q O |
...
> | a ( A ) |
Roger Mills has alread shown the problem here. While ( and ) arranged like
that have certain mnemonic properties that X-Sampa { and } lack, there's
still going to be confusion and ( and ) are still grouping pairs. Grouping
pairs should only ever be used for groupings, because that's what people
see them as (> and < are sometimes groupers, but their original use was as
less than/greater than, so they're excepted, and anyway, they have less
grouping strength than (), {} or [] have, being smaller).
I also really dislike your use of c and q as vowels. I think it's insanity
to try to use consonants as vowels and vowels as consonants, excepting
when talking about a syllabic consonant. It'll do more harm than good.
| doesn't seem big enough to act as a consonant/vowel, especially one as
wide as an upsidedown m.
Also, there is an argument for not using @ as a letter in newer Ascii IPAs
because things tend to say 'ho! an email address!', which is not at all
useful when they follow this up with 'I'd better mask it to stop spammage'
as in Yahoo! Groups.
> [\~] palatised [\@] mid centralised
As complained about above and by Roger.
> [\f] voiceless ["] begin primary stress [:] long
> [\v] voiced ['] begin secondary stress [;] half long
I'd prefer inversion of primary and secondary; ['] looks enough like the
primary IPA mark that confusion would ensue unless you had strong reasons
(i.e. based on Sampa) to do it otherwise. Also, the fewer times you have
to reach for shift the better :)
I also agree with
> [\F] breathy voiced [,] syllabic break [.] extra short
A <.> is a longer break than a <,>; I'd suggest <.> for syllabic break
a la IPA and ^ (reminding us of the upsidedown ^ in IPA) for extra short.
> Numbers 0-9 are not used at all.
> Numbers 0 and 1 should be avoided lest confusion arise with 0 and l.
Like you've just done :)
--
Tristan
Mecht most toreck, ånd absolut mecht most toreck absolutelik. Gehalchte
menn vaore nichæ ever uevel menn, jetsvao svao indfluedels evaore
nemmfremm, ånd ne othoritet.
--- Jochn Emerich Edvard Dalberg, Herr Acton
/maiS mQ:tSaitS, Qn afsluS maiS mQ:tSaitS afslulaitS. jQ:f m&: wE Sa jev
ivu m&:, ZEsi fo efluduS evuJE:f, Qn (mwipet_ho/nu:t_horit_heS)/