nl-kr-request@CS.ROCHESTER.EDU (NL-KR Moderator Brad Miller) (12/08/87)
NL-KR Digest (12/08/87 01:02:10) Volume 3 Number 58
Today's Topics:
Re: Language Learning (acquiring native accents)
More anecdotal stuff on language learning
Re: Language Learning (a Turing test)
Re: Language Learning (anecdotes)
Re: Language Learning
Multiple regional dialects and acquiring "native" accent
----------------------------------------------------------------------
Date: Tue, 1 Dec 87 18:53 EST
From: wales@locus.ucla.edu
Subject: Re: Language Learning (acquiring native accents)
In article <2360@tut.cis.ohio-state.edu> paul@tut.cis.ohio-state.edu
(Paul W. Placeway) writes:
> Of those adults who have been in the new culture, certainly some do
> not ever learn all of the subtleties that comprise the local accent,
> and some do learn to perceive the differences, but do not learn how
> to produce them well.
> As far as phonemic discrimination, the study I cited does not show
> perfect results, but then again, 95% as good as natives in 3-4 weeks
> of training isn't all that bad, either. I do not know if a similar
> study has been done for people who have had years of experience.
> Years of practice probably cover the remaining 5%, however...
Keep in mind that the standard of perfection is quite high. An adult
trying to acquire a local accent well enough to "pass for a native"
needs to do a virtually flawless job, or else he will be found out. No
matter how "good as native" that 95% of someone's speech may be, the 5%
that deviates from the local norm will stick out like a sore thumb in
the listener's ears.
A friend of mine -- a young man from Manchester, England -- recently
spent a couple of years in Kentucky and southern Ohio. Upon his return,
he still spoke with the same Manchester dialect as he did before he left
-- but now with occasional slight traces of a Midwestern US accent. Yet
he told me that, when he went home to England for a brief visit, every-
one back in Manchester claimed he sounded like an American! (I assured
him, by the way, that he most definitely did not sound at all like an
American, and he was much relieved. :-})
What presumably happened was that his friends back home simply didn't
notice the "Manchester" components of his speech (since these were no
different from the norm as far as they were concerned), but the occa-
sional features from the American Midwest stood out very clearly. If
my friend had visited another part of England, I suspect they would have
had no trouble identifying him as being from Manchester.
-- Rich Wales // UCLA Computer Science Department // +1 (213) 825-5683
3531 Boelter Hall // Los Angeles, California 90024-1596 // USA
wales@CS.UCLA.EDU ...!(ucbvax,rutgers)!ucla-cs!wales
"Sir, there is a multilegged creature crawling on your shoulder."
------------------------------
Date: Tue, 1 Dec 87 23:19 EST
From: "Martin Minow THUNDR::MINOW ML3-5/U26 223-9922" <minow%thundr.DEC@decwrl.dec.com>
Subject: More anecdotal stuff on language learning
>From: portal!cup.portal.com!Angelique_N_Wahlstedt@uunet.uu.net
>can [adults have trouble learning foreign languages] because
>many adults weren't exposed heavily enough to a new language in order to
>learn it? ...
>All hearing kids learn languages mainly
>by OVERHEARING in addition to people talking to them.
Also by *speaking* languages. I don't think you can learn a language
solely by listening. (I'm learning one now, and find writing and speaking
essential.) Why else do people move their lips when they read?
>From: Rick Wojcik <ssc-vax!bcsaic!rwojcik@beaver.cs.washington.edu>
>There is a standing challenge in the L2 community to find a single
>person who has acquired native fluency in a foreign language in
>adulthood. The judge of this can only be native speakers of the
>language in question. ...
My ex-boss (a native speaker of Swedish) learned Finnish as an adult.
He *has* convinced native speakers that he, too, was native.
----
Having learned several foreign languages as an adult (French in college,
Swedish, Danish, and Norwegian in Sweden, and I'm currently struggling
with Yiddish in night school), I discovered that:
1. You have to speak the language -- after I was in Sweden for three months,
I refused to speak or listen to English. Two weeks of chaos ensued, and
when I was able to communicate again, I had a *real* grip on the language.
Just listening or just reading are not sufficient. This is the real
problem with school language courses.
2. Reading is a good way to build vocabulary -- I read the daily newspaper
(written with the vocabulary of a New York Times/New York Review of
Books) from cover to cover daily. This usually took 2-3 hours per day.
3. You learn what you need to know. You *only* learn what you need to
know. Even after speaking Swedish for over 20 years, I haven't lost
my American accent. More importantly, I lack understanding of subtle
distinctions (was the meal "quite good" or "rather good"?)
Martin Minow
minow%thundr.dec@decwrl.dec.com
------------------------------
Date: Wed, 2 Dec 87 11:16 EST
From: Rick Wojcik <rwojcik@bcsaic.UUCP>
Subject: Re: Language Learning (a Turing test)
In article <2314@homxc.UUCP> lewisd@homxc.UUCP (David Lewis) writes:
>Has anyone heard about a test like this?
>Take the pool of native speakers and subject them to this
>test: to tell whether an arbitrary speaker of their language is a native
>speaker or not.
>The problem: sure, maybe they
>can say that your wife is not a native speaker of Spanish or French. But
>perhaps they'll also say that of other native speakers.
>Only if they can tell native/nonnative with high accuracy should the group
>be considered a valid test of fluency.
>--
>David B. Lewis {ihnp4,allegra,ulysses}!homxc!lewisd
You are probably right up to a point. Many valid dialects of a language
do sound like foreign accents. For the sake of argument, let us require
that 'native speakers' and examiners be only those with command of the
standard dialect of a language.
But I should clarify the nature of the Turing test. I never did like it
as a test for artificial intelligence, but it does seem ideally suited for
language mastery. Turing originally called the test the "Imitation Game",
which was to involve 3 players. The object was for one of the players--the
questioner--to figure out the identities of the other two. The other two
players might be a man and a women, both of whom would be hidden or disguised
appropriately. So the questioner would have to figure out which was the man
and which was the woman. Turing suggested that the game be played with a
computer trying to imitate a human. I suggest that it be played with a mature
learner of a foreign language trying to imitate a native speaker. This means
that the questioner would have a real speaker and a "ringer" to
distinguish between. It would be interesting to try this with "ringers"
who had learned the language in question before and after puberty. I
suggest that post-puberty learners will almost always be exposed.
===========
Rick Wojcik rwojcik@boeing.com
------------------------------
Date: Wed, 2 Dec 87 11:52 EST
From: Rick Wojcik <ssc-vax!bcsaic!rwojcik@beaver.cs.washington.edu>
Subject: Re: Language Learning (anecdotes)
In article <2360@tut.cis.ohio-state.edu> paul@tut.cis.ohio-state.edu (Paul W. Placeway) writes:
>
>A seven year old child has just spent 7 years, of 365 days/year, 12-18
>hours/day of language practice; most adults do not spend anywhere near
>this amount of effort learning a new language unless they have spent
>years of time in the new culture. Thus most adult language learners
>do not 'count' in such a comparison. Of those adults who have been
Let us limit ourselves to cases of adult immersion in a foreign
language. Even trained phoneticians can't seem to rid themselves of an
accent. A case in point is Dr. Lehiste, whom you mentioned in a
previous note. She has an extremely sharp mind, a good memory, and more
knoweldge of phonetic detail than 99.999(etc.)% of the human population.
She has lived in the US for several decades. She may be a busy woman,
but her business is language. Why does she still speak with an accent?
Another case in point is the late Roman Jakobson, one of this century's
finest linguists. The standard joke in the linguistic community is that
he spoke five languages in fluent Russian. The fact remains that no
amount of practice or immersian can make an adult the equal of a child.
===========
Rick Wojcik rwojcik@boeing.com
------------------------------
Date: Thu, 3 Dec 87 20:45 EST
From: Stephen Smoliar <trwrb!aero!venera.isi.edu!smoliar@ucbvax.Berkeley.EDU>
Subject: Re: Language Learning
This account of learning the "beetle" sound reminds me of one of the
sillier remarks in Jerry Fodor's THE LANGUAGE OF THOUGHT. While he
was discussing the role of visual exemplars in understanding, I would
think his remarks are as applicable to audible exemplars:
The present point is that the process by which one becomes
acquainted with the exemplar is not itself a process of
hypothesis formation and testing; it is, rather, the
process of opening one's eyes [ears] and looking [listening].
The accounts of both Alen Shaprio and Robert Stanley would indicate
that phoneme learning is not just a matter of hearing good exemplars.
Attempts by the learner to reproduce the phoneme clearly seem to be
an indication of hypothesis formation on the learner's part (hypothesizing
how to shape the mouth and control the breath) and testing based on both
what the learner hears and how others react to it.
------------------------------
Date: Thu, 3 Dec 87 23:13 EST
From: Paul Placeway <paul@ptero.cis.ohio-state.edu>
Subject: Re: Language Learning (a Turing test)
In article <2966@bcsaic.UUCP> rwojcik@bcsaic.UUCP (Rick Wojcik) writes:
< ... When
< did she learn German, anyway? That's an essential point here. (Don't
< forget that the Estonia of her childhood had close ties to Germany.) It
< is also worth noting that, despite her many years of residence in
< America and her linguistic sophistication, she retains a noticeable
< foreign accent in English. Her control of English is about as good as
< it can get in adult language learners.
Good point. Yes, I suspect that she learned German very much earler
than English, thus providing evidence for your theory.
<
< >The similarity of dialect does not allways hold either. Elizabeth
< >Zwicky does not speak the same regional dialect of SAE that I do, even
< >though the two of us spent the majority of our lives growing up within
< >10 miles of each other, in the same side of the same city. Our
<
< You miss the point. I never said anything about the social and ethnic
< factors that shape dialects. The Columbus neighborhood that you and she
< grew up in contains a mixture of Northern and Midland dialects.
< Elizabeth's dialect (Northern) and yours (Midland?) are recognizably
< American.
Actually, I think that you missed mine also. I am *quite* aware of
the dialect boundry that falls through Columbus, but in the example I
gave, it isn't a problem: My accent is also definitely Northern, much
closer to the SAE spoken in Cleveland or Ann Arbor, MI. Mr. Lewis (in
article <2314@homxc.UUCP>) expressed basically the same point as mine:
when you are looking close enough at the accent of a native, fluent
individual vs. the associated dialect (which is the "average" of the
community), the level of variation of the norm of that individual vs.
the dialect gets so high as that a test such as yours might not be
statistically meaningful, and is probably fuzzy, at best. In other
words, I think that such a test would have to be ammended, so that
"native fluency" would be some minimum _percentage_ of a large sample
of actual native speakers saying "yes, this person speaks the same
dialect as I".
The main thing I object to is the "no adult" part. "Very very few",
perhaps, but I am _very_ wary of universal statements such as this.
Many universals have been shown to be false (my favorite is "No
language has a syntax with unlimited cross dependencies", the counter
example to which is Swiss German).
-- Paul
------------------------------
Date: Fri, 4 Dec 87 13:39 EST
From: mark edwards <edwards@unix.macc.wisc.edu>
Subject: Re: Language Learning (anecdotes)
In article <2999@bcsaic.UUCP> rwojcik@bcsaic.UUCP (Rick Wojcik) writes:
:In article <2360@tut.cis.ohio-state.edu> paul@tut.cis.ohio-state.edu (Paul W. Placeway) writes:
:>
:>A seven year old child has just spent 7 years, of 365 days/year, 12-18
:>hours/day of language practice; most adults do not spend anywhere near
:>this amount of effort learning a new language unless they have spent
:>years of time in the new culture. Thus most adult language learners
:>do not 'count' in such a comparison. Of those adults who have been
:
:Let us limit ourselves to cases of adult immersion in a foreign
:language. Even trained phoneticians can't seem to rid themselves of an
:accent. A case in point is Dr. Lehiste, whom you mentioned in a
:previous note. She has an extremely sharp mind, a good memory, and more
:knoweldge of phonetic detail than 99.999(etc.)% of the human population.
:She has lived in the US for several decades. She may be a busy woman,
:but her business is language. Why does she still speak with an accent?
So. Her main point of study was the phonetic detail (according to
you), not how to speak english without an accent, or not even to
learn english.
: The fact remains that no
:amount of practice or immersian can make an adult the equal of a child.
What fact? It seems to me what you are saying is that the observed
evidence shows that no adult can lose their native accent. Which is
like saying "That man listens only to jazz because that is all I hear
him listening to".
Just because there is no observed evidence does not prove your theory.
I say because any child can learn his native language any adult can
learn the childs native language also. My proof is that any adult can
do what ever any child can do.
On the serious side, what if we taught the adult to use the phonetic
system of the language (perhaps by using nonsense syllables in that
language) before letting him see, hear real words. And then insure
that he picks up the correct accent also. You would also have to
teach only in the target language, you don't want him to used his
native language on anything. Saying anything in his native language
would result in forty whips with a wet noddle. This would take
considerably longer then the present methods, but the results, I
predict, will raise some doubt whether or not an adult can learn the
language as a child does.
A big problem with this is proverbs or metaphors. When I am speaking
japanese, english grammar and semantics kind of creep in and take
over when my japanese is not adequate. Often this brings laughter
or consternation. How do you prevent this interference between the
two languages? I wonder how a child does it? If he does he must have
some kind of marker marking the grammar and semantics, japanese or english.
I'm tired of the arguments, it can't be done because it hasn't been
done in the past. If that were true than there would be a lot less
Steven Jobs and Bill Gates in the world.
mark
--
edwards@vms.macc.wisc.edu
{allegra, ihnp4, seismo}!uwvax!uwmacc!edwards
UW-Madison, 1210 West Dayton St., Madison WI 53706
------------------------------
Date: Fri, 4 Dec 87 22:24 EST
From: Richard A. O'Keefe <quintus!ok@unix.sri.com>
Subject: Re: Language Learning (anecdotes)
Maybe it's my ignorance showing, but everyone seems to be asking the
question "children can learn their native language perfectly (as in:
without an accent), can adults learn languages perfectly too, and if
not why not?". It seems to me that the first part of this question
is false: children do NOT learn to speak the same language as their
parents. This is particularly clear in the case of slang and metaphor
(I *still* don't know what "twenty-three skidoo" meant), but I quite
certainly haven't got the same accent as my parents, and in my own
country "age dialects" seemed more obvious to me than "regional" dialects.
Speakers of a language where "silly" once meant "happy" shouldn't rush
to assume that children have some magical skill lost to adults.
------------------------------
Date: Sat, 5 Dec 87 21:09 EST
From: das@locus.ucla.edu
Subject: Re: Language Learning (anecdotes)
In article <2048@uwmacc.UUCP> edwards@unix.macc.wisc.edu (mark edwards) writes:
> I say because any child can learn his native language any adult can
> learn the childs native language also. My proof is that any adult can
> do what ever any child can do.
I wouldn't be so sure about that last sentence. Look at "hemis" (people who
have had one hemisphere of their brain removed). If this removal occurs before
a certain age (around 3 to 5, I think), it appears that the person seems to be
more "normal" than someone who's lost a hemisphere at a later age -- that is,
those brain functions that are normally performed predominantly by one
hemisphere can be done almost as well by the other, if the need arises before a
certain age, and not as well if it arises after that age. This isn't precise,
but it does illustrate that what the brain can do changes with age, and not
always in the direction of increased power.
This says nothing about your contention -- just your proof.
-- David Smallberg, das@cs.ucla.edu, {sdcrdcf,ucbvax}!ucla-cs!das
------------------------------
Date: Sat, 5 Dec 87 22:10 EST
From: Murray Watt <clyde!watmath!utgpu!jarvis.csri.toronto.edu!utai!murrayw@rutgers.edu>
Subject: Re: Language Learning (anecdotes)
In article <435@cresswell.quintus.UUCP> ok@quintus.UUCP (Richard A. O'Keefe) writes:
>Maybe it's my ignorance showing, but everyone seems to be asking the
^^^^^^^^^ Possibly, your correct.
>is false: children do NOT learn to speak the same language as their
^^^^ Not 100% the same
>(I *still* don't know what "twenty-three skidoo" meant), but I quite
>certainly haven't got the same accent as my parents, and in my own
Why then can different regional accents in North America be traced to
different groups of immigrants??? If your parents are new immigrants
then you might not have thier accent. Instead you have the accent of
your teachers, your freinds' parents, and your favorite television
character.
>rush to assume that children have some magical skill lost to adults.
This was not a "rushed" conclusion and no one thinks it is a "magical"
skill.
Murray
I wouldn't believe in magical skills either.
------------------------------
Date: Sat, 5 Dec 87 14:56 EST
From: wales@CS.UCLA.EDU
Subject: Multiple regional dialects and acquiring "native" accent
222 1159 <9693@shemp.UCLA.EDU> Article retrieved; body follows.
In article <2363@tut.cis.ohio-state.edu> paul@tut.cis.ohio-state.edu
(Paul W. Placeway) writes:
>In article <2913@bcsaic.UUCP> rwojcik@bcsaic.UUCP (Rick Wojcik) writes:
>> Here is a case in point. My wife learned French by study+immersion
>> in young adulthood. She has achieved fluency to the point where she
>> can fool other French speakers into thinking she is French--but not a
>> speaker of the listener's dialect. This deception is almost never
>> maintained in extended conversations. An accent is always detected.
>Actually, the "story" I was thinking of is similar, but with a big dif-
>ference: I am told that Dr. Lehiste (who's native language is Esto-
>nian), when traveling in Germany, regularly fools native speakers into
>thinking that she is German, but from some other region. From what I
>have been told, this effect is true, even for extended conversations.
It would be interesting to know whether native German speakers think
Dr. Lehiste is from some *specific* other region of Germany, or simply
from "somewhere else, I'm really not sure where".
In areas such as Germany and England, where a large number of regional
dialects abound even today, it should be easier for a good mimic to get
away with being mistaken for a native speaker from "some" region other
than that of the people he is conversing with. This would be partic-
ularly true when large numbers of people in a country already learn a
"standard" dialect (Hochdeutsch, Received Pronunciation, etc.) in school
anyway: people would then be more used to hearing said "standard" spo-
ken with various different accents, and a foreigner's acceptable margin
for error (within which his residual accent would not stand out as for-
eign) would thus tend to be larger.
This phenomenon is probably true for North America as well to some
degree (Canada probably somewhat more than the US). I've listened very
carefully to Canadian speakers from several parts of Canada (Toronto,
Ottawa, Edmonton, Calgary, southern Alberta, and Vancouver) for many
years. Even among speakers from the same area, there is still a fairly
wide mix of pronunciation features -- primarily either a varying amount
of "CBC English" (roughly the dialect of southern Ontario) or a varying
amount of "General American" (particularly in Alberta south of Calgary,
but present to some extent just about everywhere, probably because of
US TV, which most Canadian households get via cable).
As an "experiment" of sorts when I visited Vancouver last August, I
mimicked the average dialect of the region as closely as I could. Doing
this involved aspects of pronunciation, word usage, and cultural
knowledge, and although I've spent several years studying the issues
involved, I still wouldn't say it was *easy*. However, I am confident
that what was coming out of my mouth was definitely *not* a crude cari-
cature of general Canadian speech, and my experience (see below) seems
to have confirmed this.
I know that I "fooled" at least two British Columbia natives, on separ-
ate occasions, into thinking I was Canadian. Even after fairly extended
conversation (15-30 minutes), each was quite surprised when I told them
I was from Los Angeles. (One middle-aged woman from Vancouver finally
picked up on a statement I made to the effect that I was visiting the
area, and her comment was, "Oh, you're not from Vancouver?")
One person did say I sounded obviously American to him. However, he
already knew me, knew I was from California, and in his presence I was
probably a bit more self-conscious than otherwise about the way I was
speaking -- so I think I can say he "doesn't count".
Most people didn't appear, as far as I could tell, to notice one way or
the other (though I can't say for sure, since I can't read minds :-},
and I didn't go around asking people where they thought I was from based
on my speech!). Most often (especially in the stores), what gave me
away in the end was my MasterCard or Visa -- since there are only a
handful of major banks in Canada, and my cards from Wells Fargo Bank (a
major California bank) looked noticeably different from anything most of
the store clerks had apparently ever seen.
No one took the initiative of asking what part of the US I was from, or
whether I was enjoying my visit to Canada, or anything similar, except
when I had already mentioned where I was from (or pulled out one of my
funny-looking American credit cards :-}). I made all my minor purchases
with Canadian cash, by the way, so I never gave myself away by pulling
out US money and asking about the store's current exchange rate.
No one (except for the above-mentioned acquaintance who already knew me)
claimed to identify me as American on the basis of my accent. Further,
even when people did know or find out I was not a native, not one person
ever asked me about my accent; nor did anyone say anything like "Yeah, I
kind of thought you were from the States by the way you talked." This
could conceivably be because they were "being polite" -- or because
enough bits and pieces of my native California dialect slipped through
(or were noticed once my true origin was revealed) that the people in
question did in fact think I had an "American accent". However, as I
said, I do know that I lulled at least two people -- and probably more
-- into assuming I was "one of them", so I suspect that whatever "Yankee
drawl" (a common Canadian characterization of General American speech)
was still to be found in my conversation was likely no more than what
has managed to find its way into the average Vancouver mouth via the US
mass media.
I assume one big reason I was able to get away with doing this "accent
switch" was that there is already enough dialectal variation in Canada
that I didn't need to do a 100% perfect job of imitating a given set of
features precisely in order to avoid detection. It still wasn't easy,
as I already mentioned. There are numerous detailed differences between
the "average" dialects of Vancouver and L.A. which have taken me several
years to ferret out -- things which people normally never notice (and
which aren't even discussed in most of the books or linguistic journal
articles which describe Canadian speech), but which still do contribute
to the overall sound of a native speaker and will make one's speech
sound at the very least "somehow strange" if they aren't there. But
there still is some "margin for error".
Then again, lots of people have "tin ears" for accents. Last year, when
I visited Calgary, I ran into one young lady who said she thought the
people on the US prime-time sitcoms spoke pretty much the same way she
did (she was quite mistaken). And, a old roommate of mine from Edmonton
told me that he couldn't really tell Canadians and Americans apart by
their accents -- unless, for example, the person did something like pro-
nounce "been" to rhyme with "bean" (something which is virtually never
heard in the US, but which many -- though not all -- Canadians do).
I won't try to go into more detail here and now (this article is long
enough already as it is!), but I'll be happy to elaborate on any of the
above issues via e-mail for anyone who is interested.
-- Rich Wales // UCLA Computer Science Department // +1 (213) 825-5683
3531 Boelter Hall // Los Angeles, California 90024-1596 // USA
wales@CS.UCLA.EDU ...!(ucbvax,rutgers)!ucla-cs!wales
"Sir, there is a multilegged creature crawling on your shoulder."
------------------------------
End of NL-KR Digest
*******************