X-SAMPA confusion (was: Introduction)

On Wednesday, November 17, 2004, at 11:00 , Rodlox wrote:

> ----- Original Message -----
> From: Geoff Horswood <geoffhorswood@...>
> To: <CONLANG@...>
> Sent: Wednesday, November 17, 2004 11:05 PM
> Subject: Re: Introduction
>
>
>>>
>>> I hate to be the bearer of bad news...but some folks here  are trying to
>>> replace X-SAMPA.
>
>> Aaarggh!!  I just try to learn a system and they go and replace it!! ;P!
>
>   that was pretty much my reaction.
No one on Conlang is in a position to replace X-SAMPA (Extended Speech
Assessment Methods Phonetic Alphabet).

SAMPA itself was drawn up between 1988 and 1991 by the SAM consortium to
form the basis of a machine-readable phonetic alphabet for international
purposes. X-SAMPA is an extension of SAMPA suggested by J.C. Wells of
University College. London, in order that all IPA symbols may be mapped
into 7-bit ASCII. Neither SAMPA nor X-SAMPA are have ever been norms on
Conlang and they cannot, therefore, be "replaced".

>
>> With what?  Reasons??
>
>  if I understand, so they can call it "conlang x-sampa".
With respect, that is a silly, cynical remark.
===================================================


On Wednesday, November 17, 2004, at 09:18 , Andreas Johansson wrote:

> Quoting Geoff Horswood <geoffhorswood@...>:
>
>> Aaarggh!!  I just try to learn a system and they go and replace it!! ;P!
>> With what?  Reasons??
>
> The reason is that certain features of X-SAMPA are widely perceived as
> very
> annoying. The system mostly used here is called "CXS", and is in most
> respects
> very close to X-SAMPA.
Partly so, but also for historic reasons. Once upon a time there were many
systems for "IPA ASCII" (there probably still are). When long years ago I
joined Conlang, the prevailing system here was one known as _Kirshenbaum_
[sic] but it had two drawbacks:
i. it was designed principally for dealing with English and used a system
of tags between angled brackets to denote more 'exotic' sounds;
ii. Evan Kirshenbaum did not develop the system further.

Gradually over the years, especially as more non-anglophones joined the
list, X-SAMPA became more widely used. However, some Kirshenbaum features
tended to survive, especially:

> Actually, I thought you were using it, since you transcribed Kazakh 'a'
> as [&],
> which represents the the vowel of English "cat" in CXS. In X-SAMPA it is
> a low
> front rounded vowel.
Yes, in Kirshenbaum [&] is the RP English 'a' in _cat_, in both SAMPA and
X-SAMPA the symbol for that sound [{] thus in SAMPA/ X-SAMPA English _cat_
is written /k{t/.

But in X-SAMPA [&] is the IPA _capital_ OE ligature, a low front rounded
vowel, lower than the sound of French 'eu' in _neuf_ which is lower-case
oe ligature and represented in both X-SAMPA and CXS as [9]. It is the
_rounded_ version of X-SAMPA  and CXS [a] and occurs in Danish _drømme_.
In in CXS the sound is written [&\].

The result was that *in pactice* a mixed system was being used and
occasionally misunderstanding did arise, particularly over the symbol [&]
which some read as [æ] and other read as [Œ]. This was clearly
unsatisfactory. CXS (Conlang X-SAMPA) was codified in order to regularize
the position and to *avoid confusion*. IMHO those who are responsible for
CXS deserve praise, not sneers.

> If that really is what you meant, Kazakh phonology and
> spelling is rather weirder than I thought ...
Indeed. I had assumed _s&ljemjets1z bje_ was using CXS rather than X-SAMPA.

But now I am confused. Is Geoff really using X-SAMPA and the first vowel
of _s&ljemjets1z_ *rounded*? That would be weird vowel harmony.

Ray
===============================================
http://home.freeuk.com/ray.brown
ray.brown@freeuk.com
===============================================
Anything is possible in the fabulous Celtic twilight,
which is not so much a twilight of the gods
as of the reason."      [JRRT, "English and Welsh" ]

Reply