Re: Allophones Question

Hey.

I'm responding to this separately from the Spanish discussion, since it doesn't
deal with Spanish at all; but it might be helpful to read the two replies
together to triangulate on what I think is an appropriate way of thinking about
the question of allophony.

At 9:44 AM +0100 2/19/03, Christophe Grandsire wrote:
>(Occam's
>razor here. Choose the simplest way to explain all the data).
>
>This is exactly the same reasoning that brings the fact that French has the
>phonemes /2/ and /9/ (X-SAMPA for IPA o-slash and oe-ligature) while in the
>lexicon they happen to be in complementary distribution. The fact is, the
>French people do differentiate them, and although they don't have [9] in open
>syllables or [2] in closed ones, they can pronounce them easily if asked to do
>so (any French person - except the inhabitants of St-Etienne and around there
>who don't have the distinction between mid-low and mid-high vowels at all, and
>pronounce them always mid-high - will be able to explain you the difference
>between [p2] and [p9] nearly as well as a linguist - they will probably say
>it's the same as between [pe] and [pE] -), even if [p9] doesn't exist in the
>French lexicon). And if you want to argue that it's an artificial distinction
>brought by French education, you'd be wrong, because even if it's true that we
>are still taught that /2/ and /9/ are different sounds, we're also taught
>that "a" in "patte" is a different from "â" in "pâte", and yet nobody makes the
>distinction anymore (the two sounds used to be [a] and [A], but have merged in
>a single [a] /a/. I myself still made the distinction 20 years ago, but don't
>anymore, and had to relearn it when I learned Dutch) or understand what the
>distinction is.
Speakers of languages with long written traditions are notoriously unreliable sources
of intuitions concerning the phonemic status of sounds. French speakers may
insist that there is a "phonemic" distinction between [2] and [9], but the fact
that they are in complementary distribution belies this intuition. There is a
large psycholinguistic literature which shows that speakers attend to and can
reproduce all sorts of details of pronunciation which aren't necessary under
the assumption that all and only the distinctive properties of sound structure
are stored in the mental lexicon. Just because a difference is perceived,
remembered and reproduced it doesn't follow that it must therefore be
underlying.

So structuralists would be opposed to the phonemic status of [2] and [9] on the
basis of their distribution; [2] occurs in open syllables and [9] occurs in
closed syllables (remember, they aren't concerned with speakers' intuitions).
Generativists would not see French speakers' ability to differentiate these
sounds to be evidence of their underlying status. Since Ockham was already
invoked, I shall call on him again to pronounce that there is no underlying
distinction between [9] and [2] in French; to say otherwise would clearly be to
needlessly multiply entities.

Dirk
--
Dirk Elzinga                                               Dirk_Elzinga@byu.edu

"It is important not to let one's aesthetics interfere with the appreciation of
fact." - Stephen Anderson