Re: CXS suggestions

--- In conlang@yahoogroups.com, Isaac Penzev wrote:
>First of all, as one of the contributors to the present version of CXS, I'd
>like to state that the exclusive advantage of CXS is its stability that may
>accept it as a de facto standard. It originated from X-SAMPA that should be
>machine-readable, and made ***A FEW*** adjustments to make it more
>human-friendly.
Define "stability", please. You already said yourself that it *may* change,
if a new feature, like /;/ for palatalization, is gaining enough
popularity...


>Also, caeruleancentaur wrote:
>
> > With respect to my own conlang's phonetics, I've always been puzzled
> > by the pairs /c/ /J\/ and /j\/ /C/, the voiceless and voiced palatal
> > plosives, and the voiced and voiceless palatal fricatives,
> > respectively.  The present symbols offend my sense of order!  :-)
>
>This issue was discussed some time ago. The stability of the standard was
>the main argument against changing it.
This part makes perfect sense to me. *All* the ASCII-IPA constructions I've
seen agree on the values of the small Latin letters ... /C/ (like some
others) also seems to be fairly universal. The palatal voiced fricative can
then be considered a close variant of /j/, and the voiced plosive a more
distant one.


>CXS came into being from usage, not from discussion or agreements.
Of course, any such creations tend to be more or less filled with
illogicalities. Person A derives symbols by logic X, person B by logic Y,
etc - and if you group symbols from many such systems together, it'll soon
seem like there's no logic at all.


>So, if somebody wants changes (I
>don't) - use them (with explanations!), and  things may change one day.
I'll certainly be using at least /a\/ from now on ... you can consider my
first message an initial explanation then, rather than a vote for a new
system.


> > Most of you apparently use X-SAMPA or Henrik Theiling's CXS
>
>It's not Henrik's. It's a kinda freeware. For details contact Tristan
>McLeay, but I think he is nomail now.
Aha. At least Henrik has the only online description I've seen, so I assumed
it was his doing (well, compiling - it seemed obvious from the start that it
had borne out of usage)


> > - /e\ 2\ 7\ o\/ for /e_o 2_o 7_o o_o/ (middle vowels)
>
>I am not aware of a natlang that distinguishes five levels of vowels. So
>normally if the middle vowels are merely middle (without opposition
>close-mid :: open-mid), ppl usually use symbols for close-mid with a
>footnote about their real nature.
Yes, actually I think these symbols might not get that much phonemic use.
Myself, I went through a lot of confusion before realising that many
languages which only have /e o/ (and optionally /2 7/) use those phonemes
for *middle* vowels, not *mid-high* - but indeed, this might be an issue of
the sources just not bothering to mention the realization, rather than not
having more exact phonemic symbols available...


> > Currently, /F J/ do not really suggest nasality, ... And /K K\/ suggests
>neither laterality
> > nor fricativity.
>
>CXS is not mnemonic.
>
>-- Yitzik
That's not what it isn't; it's not *consistent*. Or do you really mean that
eg. the /&/-normalization was for any other than mnemonic reasons??


--- In conlang@yahoogroups.com, Benct Philip Jonsson wrote:
>Carsten Becker skrev:
> > I'd furthermore suggest to partly adopting Z-Sampa (search
> > kutjara.com/wiki), since it contains all the symbols from
> > the Disordered Speech chart. AFAIK, there is no
> > representation of these symbols in (C)XS yet.
>
>Interesting, but as Yitzik said CXS grows by usage,
>not by discussion and voting!
*adds Z-Sampa to his things-to-check-out list*


>I've already used [w\] for this sound when transcribing Tibetan.
>IMHO this makes a better parallel to [v\].  Also since [p\] is
>voiceless bilabial fricative [w\] is less misleading for bilabial
>approximant.
I guess that makes sense, too... though I have a fairly strong "velarized"
association with the "w" symbol.


>IPA small capitals are all capital+backslash in X-SAMPA.
I noticed the same one day, when I started mapping how CXS uses each Latin
letter. It's a pretty good mnemonic, maybe it ought to be actually MENTIONED
somewhere...

---

Tho I should mentioned that I don't really like the way all ASCII phonemic
alphabets are just translitterations of the IPA. Especially when it comes to
conlanging: it's too easy to just pick phonemes which are already
well-known, instead of inventing some of your own. If there were actual
SYMBOLS mentioned somewhere for phones like a uvular tap or a velar ejective
lateral or non-sibilant (post/palato)alveolar fricatives, I bet we'd be
seeing them in conlangs a lot more.

John Vertical

_________________________________________________________________
3 vrk:n sääennuste http://www.msn.fi/uutiset/saa/

Replies