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

On Fri, 23 Jul 2004 00:01:21 -0700, Philippe Caquant <herodote92@...>
wrote:

>phonematic distinction concerns
>meaning, while allophonic distinction does not
Exactly, that's the very center of the definition.

>So it seems that a huge part of the discussions hold
>on this list concerns allophones, which is clearly one
>of the most peripheral part of linguistics, the very
>surface of the orange (probably at the same level as
>orthograph). When I write for ex "when I was in
>Alsace, I used to pronounce "Nord" this way, but when
>I moved to Paris, I modified it that way", I'm talking
>about allophones.
No, that's not true. Phonemic distinctions may vary from dialect to dialect
(as others pointed out recently and repeatedly). If one distinguishes
between _fausse_ /fos/ and _foss_ /fOs/ or between _maître_ /mE:tr/ and
_mettre_ /mEtr/ but another doesn't, then this doesn't mean that these
sounds are allophones of each other. The phonological/phonemical analysis
(and thus the notion of allophones) doesn't make sense but within a very
homogeneous group of speakers.

I think that this might be the difference between allophones and
archiphonemes (a term I didn't know before): archiphonemes are a notion
that tries to reconciliate the different phonemic systems of the different
dialects of a language, saving thus the unity of the language.

>Topics like allophones are something very concrete,
>while syntactic topics are harder, and semantic or
>cognitive topics, even more abstract and difficult,
>thus discouraging, especially after a hard work day.
The notion of allophones doesn't make sense without the notion of phonemes,
which is an abstrahation: cognitive, if you will. And meaning isn't
possible without phonemes that make it distinct from other meanings. But
how can there a uniform meaning in a language that doesn't have uniform
phonological/phonemical distinctions?

==========================================================

On Thu, 22 Jul 2004 05:49:02 -0700, Philippe Caquant <herodote92@...>
wrote:

>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, it
>seems vital that after the scientific definition
>(alveolar, epiglottal, etc, etc), and the X-Sampa
>character, and the computer code, there will be
>detailed examples for several (I suggested about 12)
>natlangs; so that one could see immediately whether he
>is concerned or not for his own purpose, and what is
>the phone like. And more, phones like, for ex, "s",
>"z", "th", "sh", etc, should be grouped in a family
>(archiphonetic) and then further detailed.
(I don't think this is an adequate use of the term 'archiphoneme', since
that term, same as phoneme, is always language specific.)

As I see it, the groupings you have in mind do exist in the IPA system:
They're just the signs next to each other on the phonetic tables, since the
tables aren't arbitrary but based on principles of pronunciation.

g_0ry@_^s:
j. 'mach' wust

Reply