Re: X-SAMPA { and }

> That's basically what I'm doing right now. I'm going directly from the IPA
> (because we have to start somewhere) and try to design a scheme
> with two main
> features: redundancy and controlled ambiguity. Redundancy means
> that for many
> phones, there will be more than one way to transcribe them. This
> will be due to
> the fact that I will use diacritics (in fact signs after a
> letter) to show for
> instance the "corresponding fricative". So imagine that - marks
> "corresponding
> fricative" (it probably won't, but I didn't choose which signs
> will correspond
> to which diacritic). Then the voiced bilabial fricative will be
> marked /b-/,
> even if maybe it will also be standardly marked as /B/. Just
> choose which one
> you want, depending on your needs.
I see a remarkable similarity between the various IPA-ASCII transcriptions
and the difference between inflecting and agglutinating languages. In both
cases you have the choice of representing each and every phone (or word)
with a single character (inflected root word), or by marking all the
characteristics of the phone (or word) using a specific symbol (morpheme)
for each characteristic such as frication (tense).

Basic IPA needs many characters, although certain features of the glyphs
identify certain features of the sounds, such as the curly tails used with
many palatal sounds, or the ? and h character of most pharyngal and glottal
consonants.IPA thus resembles an inflecting language in some ways.

Cristophe's proposal resembles the agglutinating languages, where each
quality of the phone is expressed by a characteristic additional glyph
(well, not for each quality, so it is only partially agglutinating). So we
have, say, "_" for frication, "!" for click, and the root glyphs stand for
places of articulation and on themselves represent unvoiced stops.

Phonetic representation by writing {vcd alv stp}{vcd alv frc} is then the
model for the isolating languages.

Ok, this was just a silly observation, but I had to make it anyway.

Maarten

Reply