Re: How to Make Chicken Cacciatore (was: phonetics by guesswork)

On Thu, 22 Jul 2004 05:49:02 -0700, Philippe Caquant
<herodote92@...> wrote:
> Ah, great. In fact, I didn't know of that word "phone"
> meaning just something like "utterable sound", that's
> why I was using "phoneme" instead. Now I see better.
Ah, OK. Yes, phones (marked with [square brackets]) and phonemes
(marked with /slashes/) are different.

Have you also understood the meaning of "allophone"?

> So let's reconsider the X-Sampa problem:
>
> - X-Sampa refers to many and various languages, so
> clearly one cannot say that X-Sampa handles about
> "phonemes", but about "phones".
Yes, exactly!

> - nevertheless, every single X-Sampa character should
> (at least in theory) refer to one real phoneme at
> least in one particular natlang, meaning that this
> phone is a distinctive phoneme in at least one known
> language.
>
> Is this right ? I hope so.
That is my understanding, too.

There are, after all, many more in-between sounds that humans *can*
make, but the reason they do not receive a separate IPA (and, hence,
X-SAMPA) symbol is nearly always because the distinction between such
an in-between sound and another, very close, sound is not phonemic in
any language, so both sounds are marked with the same IPA (X-SAMPA)
symbol. (Possibly with diacritics such as "retracted tongue root" or
"open" or "unvoiced" or "no audible release" or whatever, if you want
to be slightly more precise.)

IPA and X-SAMPA are also sometimes used for phonemic notation, but
there's no particular reason why they are especially suited to this.
(SAMPA, i.e. "non-extended X-SAMPA", on the other hand, was intended
for a phonemic notation of a particular language; for example, there's
a SAMPA for English, one for German, etc.) IPA is more intended to
mark up phones.

So when using it to mark phonemes, it's customary to pick a "typical"
allophone of that phoneme and use that as the symbol for that phoneme
in that language.

This can mean that the same phone may be represented by different IPA
symbols when representing different languages, if they are considered
different phonemes.

(To pick an example from what you said - [e] and [E] are different
phonemes in French but not in Spanish, so they might both be marked
with /e/ in Spanish. Then you might have a French word and a Spanish
word which are pronounced (nearly) identically, but one has /e/ in
phonemic notation and the other has /E/ -- but both would have [E] in
phonetic notation. Remember, phonemic = to do with phonemes; phonetic
= to do with phones.)

> To verify this, let's take
> two contrastive examples about French (sorry that I
> cannot remember the scientific names):
>
> - in French, there is an "open o" and a "closed o",
> although this very much depends of the area, of the
> social and cultural level, etc. But I cannot see any
> example where this distinction is relevant in meaning
> (I may be wrong) : so these are simply phones, as far
> as French is concerned. But in some other language,
> these could be phonemes.
Yes.

In French, they are allophones of the same phoneme; in other
languages, they can be separate phonemes.

I don't think it's useful to talk about sounds being phonemes, since
phonemes are a bit more abstract than sounds -- so I'd say, for
example, that "open o" is a *realisation* of the French phoneme /o/,
or "one of several allophones" of the phoneme /o/, or that "/o/ and
/u/ are different phonemes in French", but I'm not sure I'd say, for
example, that "[u] is phonemic in language X" the way I'd say that
"the difference between [u] and [U] is phonemic in language X".

> So I have to come back to my first idea: since X-Sampa
> concerns very many languages, including languages that
> someone will never listen to in his whole life,
Very likely. For example, I don't know whether I'll hear click sounds
by native speakers any time soon; the languages I know of that contain
them are in southern Africa AFAIK. Or ejective sounds such as are
found (I believe) in the Caucasus.

> So we could see at once that, ex, "mild th" exists in
> English and in Icelandic, but not in French;
also in some dialects of Spanish in some positions. I don't think it's
a separate phoneme though, but simply an allophone of /d/.

Cheers,
--
Philip Newton <philip.newton@...>

Reply