LINGUIST List 7.581

Fri Apr 19 1996

Sum: [u] for [y] substitution in English/French bilinguals

Editor for this issue: T. Daniel Seely <dseelyemunix.emich.edu>

Directory

1. Bill Holowacz, [u] for [y] substitution in English/French bilinguals

	A while ago I posed a question regarding the substitution of [u] 
for [y] in english speakers of french. I'd like to thank everyone who 
replied to me. So, here's the summary of the answers.

	One of the questions concerned whether 'naturally' bilingual children
(those who learn French by interacting with native french speakers)
differentiate [u] and [y] correctly. In general the answer is yes.The two
main factors in qualifying this answer are: the relative amount of
french/english input, and the age of exposure to the second language. A
completely bilingual child will initially have a mixed morphological system.
Subsequently, sounds will not belong to either individual language. After
the age of approximatley two or three years, the two languages separate into
two distinct systems.
	The problem arises in adults who try (consciously or unconsciously) 
to map the vowel system of the first language onto the second. In languages 
whose two systems are very similar, there is no problem. The english system,
however, is more simple that the french. With regard to rounded vowels, the
feature of backness is redundant. All rounded vowels are back vowels, and
vice-versa. This is not the case in French. When the english system tries
to classify an incoming [y], the most salient feature, the rounding from an
acoustic and visual point of view, is retained. The only high rounded vowel
in english is a back high rounded vowel, [u] and thus in english speakers,
the connection between french[y] and english [u] is made.
 It was proposed that this mechanism should be the same for all the
front rounded vowels coming into english - that they be classed as back
rounded vowels and not front unrounded vowels. After a little bit of
unscientific observation, I'd say that the vowels [o/] and [oe] are never
pronounced as [e] or [E] (ascii keyboards are the pits for linguists, huh?)
but rather as [schwa], [open o], or [upside-down omega], more or less
confirming this hypothesis.
 However, as many of you were quick to point out, in Scottish and
northern english dialects, [u] is non-existant and is replaced by [y]. (not
necessarily the same as the french [y]). Again, after a little unscientific
observation, I propose that speakers with northern british accents will
substitue northern [y] for french [u]. Thus, where an american/canadian
would say [bu] for [by] "bu", the northern british speaker would say
 [by] for [bu] "boue".
I always thought this was simple overgeneralisation, but seeing as this
seems to happen mainly to british speakers, it could be a variant of the
process described above. The high back rounded vowel [u] is mapped onto the
only high round vowel in the speakers native system:[y], where rounding is
the most marked feature. I didn't actually see what northern british
speakers did with [oe] or [o/].
 Noone really bought my spelling argument except in the case of
cognate words, where english speakers will pronounce a word with english
pronunciation and intonation in the middle of a French sentence, if the
meaning of the french and english words are sufficiently close: ex.
"suspend". (notice <u> does not have the value [u] in english) Whether this
is due to laziness, or another larger phenomenon, noone really said.
 The only other comments were general critiques of the way in which
french teachers try to teach the [y]: by making [i] with the lips rounded.
While this makes a good approximation for beginners, it doesn't relly
describe what a french speaker does in pronouncing the sound.


References:

 John Kelly and John Local. "Doing Phonology" (Manchester University Press,
1989)

 J. E. Flege. Journal of Phonetics 15, (1987), pp. 47-65.

 " Effects of equivalence classification on the production of
foreign language speech sounds. In Allan James &
Jonathan Leather (Eds.), Sound Patterns in Second
Language Acquisition. Providence: Foris Publications, 1987.


If you have any other references to articles wich refute or confirm what
i've said please pass them along. Thanks again.

Bill Holowacz
billsoftdom.com or w.holowaczhermes.humana.univ-nantes.fr