Register  Login  Active Topics  Maps  

Stephen Krashen, an interview

 Language Learning Forum : Polyglots Post Reply
52 messages over 7 pages: 1 2 35 6 7  Next >>
s_allard
Triglot
Senior Member
Canada
Joined 3690 days ago

2704 posts - 5424 votes 
Speaks: French*, English, Spanish
Studies: Polish

 
 Message 25 of 52
01 March 2013 at 12:27am | IP Logged 
schoenewaelder wrote:
Majka wrote:
   Children get a ton of corrective input adults don't get - everybody and their dog feels free to correct children.


If you live in France, you may continue to benefit from this, well into your adult life. I often (ok, maybe half a dozen times) have seen non-native interviewees on TV being corrected as to the gender of a word they have just said.

(I think it may be due to the fact that if someone does actually speak the language to a high level, it is quite "startling" when they make minor mistakes, and you instinctively correct them. I suspect at a lower standard, they tolerate the mistakes)

This is a very good observation. I think there are two reasons this sort of correction happens in French. First of all, gender is so important for the rest of the syntax that it is extremely annoying to have to listen to bad syntax continuously.

Secondly, when done discreetly, this sort of correction is well appreciated because non-native speakers, especially at high levels of proficiency, want to be corrected. Nobody like to keep on making the same mistakes.
3 persons have voted this message useful



slucido
Bilingual Diglot
Senior Member
Spain
https://goo.gl/126Yv
Joined 4935 days ago

1296 posts - 1781 votes 
4 sounds
Speaks: Spanish*, Catalan*
Studies: English

 
 Message 26 of 52
01 March 2013 at 5:19pm | IP Logged 
Aaron Myers interviews Dr. James J. Asher, creator of the Total Physical Response technique (TPR):

http://youtu.be/B7K0KctKJtc
2 persons have voted this message useful



patrickwilken
Senior Member
Germany
radiant-flux.net
Joined 2793 days ago

1546 posts - 3200 votes 
Studies: German

 
 Message 27 of 52
11 April 2013 at 7:39pm | IP Logged 
emk wrote:

As for Krashen, I have a great fondness for his work. He's one of those smart and entertaining people who devote their careers to explaining a single idea. I always appreciate that sort of researcher, even when I suspect that they're only 80% correct.


All the people I know who have learnt English to +C2 level (e.g., my wife, my mother and my last boss) have all said their success was due to reading (my boss says watching four hours of TV soaps at night also helped). None of them have done anything else formally past A2 for learning. When I tell them I do Anki cards everyday etc they just smile at me in a sort of tolerant way. When I ask them about grammar rules they look at me strangely and just say that say that they don't do grammar and know any formal rules. All of them speak at a level that is indistinguishable from a native.

So I am confident that Krashen is on to something, even if it's not the full story.

At the same time it's probably important to understand that his ideas fit within a certain cognitive neuroscientific frame, which is certainly not proved at this point.

Some ideas:

1. The idea that the mind/brain is composed for a set of specialized computational processing modules has been around for a long time (at the moment fMRI suggests that might be something like 30 separate processing modules for vision in the human brain - no one knows precisely how many). Fodor's ideas about Modularity of Mind are helpful here.

2. Chomsky introduced the idea of the brain as a Language Acquisition Device and you could think of the LAD as one giant language module in the brain.

3. Another concept that floating around is "cognitive impenetrability". Which is basically the idea that modules can't receive input from higher cognitive modules. The classic example would be for visual illusions, in the Müller-Lyer Illusion you can know the the lines are the same length cognitively, but can't affect the processing in the module of the brain that makes one line seem longer than the other.

So basically Krashen hypothesis is that the area of the brain that learns language (words, grammar) is cognitively impenetrable to areas that can cognitively learn grammar rules and the like. Presumably because the language area is simply unable to understand/accept/process the sort of input that comes from explicit grammar rules. It can only understand natural language sentences and the like.

As a cognitive scientist this seems like a completely reasonable hypothesis, but I guarantee that no one knows whether it true or not.

There is a huge and active discussion going on in cognitive science at the moment about the proper relationship between explicit knowledge (e.g., explicit grammar rules) and this knowledge's ability to affect other areas of the brain. There is no firm agreement about how correct the concepts of modularity or cognitive impenetrability are. They are probably correct for certain things and not for others.

I find it fascinating that such a theoretical discussion can have such practical implications for language learning.

2 persons have voted this message useful



Volte
Tetraglot
Senior Member
Switzerland
Joined 4699 days ago

4474 posts - 6725 votes 
Speaks: English*, Esperanto, German, Italian
Studies: French, Finnish, Mandarin, Japanese

 
 Message 28 of 52
11 April 2013 at 10:02pm | IP Logged 
1) See also Norman Doidge's "The brain that changes itself"; neuroplasticity is also a major factor.

2) You could, but it's probably more useful to think of the brain as consisting of parts that strengthen, weaken, and remap based on use, and language as a set of statistical processes that make sets of neurons fire together.

3) The lines look the same length to me, though they didn't the first time I saw the illusion, many years ago.

Krashen advocates massive amounts of i+1 input, which is compatible with a number of ways of analyzing the brain, including 2).

1 person has voted this message useful





emk
Diglot
Moderator
United States
Joined 3792 days ago

2615 posts - 8805 votes 
Speaks: English*, FrenchB2
Studies: Spanish, Ancient Egyptian
Personal Language Map

 
 Message 29 of 52
11 April 2013 at 10:08pm | IP Logged 
patrickwilken wrote:
All the people I know who have learnt English to +C2 level (e.g., my wife, my mother and my last boss) have all said their success was due to reading (my boss says watching four hours of TV soaps at night also helped). None of them have done anything else formally past A2 for learning. When I tell them I do Anki cards everyday etc they just smile at me in a sort of tolerant way. When I ask them about grammar rules they look at me strangely and just say that say that they don't do grammar and know any formal rules. All of them speak at a level that is indistinguishable from a native.

So I am confident that Krashen is on to something, even if it's not the full story.


I think Krashen is right about the necessity of massive amounts of input. Nobody ever gets taught how to negate the mandative subjunctive in English (except a few people majoring in linguistics and a tiny fraction of advanced ESL students), but everybody seems to learn it anyway.

But I think Krashen is wrong about the sufficiency of massive comprehensible input, and this can be shown trivially. I know a young woman who's lost virtually all her active skills in her L1, to the extent that she doubts she can put a grammatical sentence together. But at the same time, her receptive skills are excellent. She can watch police dramas with lots of slang on television and understand almost everything (I'm guessing she has somewhere from C1 to native comprehension).

She has the massive comprehensible input. But she lacks any reason to actually speak her L1. And so she understands without being able to speak. To me, a single counter-example like this seems fatal to any theory that doesn't discuss the role of two-way communication in language acquisition.

So that's one reason why I argue that Krashen absolutely has the main idea right, but that he's still missing some important details at the edges.

patrickwilken wrote:
So basically Krashen hypothesis is that the area of the brain that learns language (words, grammar) is cognitively impenetrable to areas that can cognitively learn grammar rules and the like. Presumably because the language area is simply unable to understand/accept/process the sort of input that comes from explicit grammar rules. It can only understand natural language sentences and the like.

As a cognitive scientist this seems like a completely reasonable hypothesis, but I guarantee that no one knows whether it true or not.


<geek hat on>

Intuitively, I strongly doubt that people can read grammar rules, understand them intellectually, and then transfer that knowledge directly into their automatic speech. If people could really learn like that, they'd surely be better at subjects like math and computer science. But the reality is that people are laughably bad at dealing with the kind of typed, recursive tree structures that appear in grammar.

That said, I also doubt that language acquisition is completely cognitively impenetrable. This is a really interesting topic, and I don't have time to go into in detail. But just because people can't transfer explicit grammar rules directly into speech doesn't mean that there are no other ways to achieve the same goal. Some possibilities:

1) Can the brain accept explicit examples of acceptable and/or unacceptable sentences and learn rules from that? What about rehearsing acceptable sentences? No need to teach grammar, just tag certain data points for special interest.

2) Can you create comprehensible input the hard way, using either translations or explicit knowledge of grammar to understand something? What if you re-expose yourself to the now-comprehensible input on a regular basis?

3) Can approximately correct intellectual knowledge be used to bootstrap the acquisition of detailed subconscious knowledge?
3 persons have voted this message useful



Volte
Tetraglot
Senior Member
Switzerland
Joined 4699 days ago

4474 posts - 6725 votes 
Speaks: English*, Esperanto, German, Italian
Studies: French, Finnish, Mandarin, Japanese

 
 Message 30 of 52
11 April 2013 at 10:36pm | IP Logged 
1) To some extent.
2) Yes.
3) To some extent.

These would be my answers from personal experience. 1) can clearly go wrong (witness school systems where people need to memorize lots of words and sentences, but students rarely gain basic communicative ability in their L2), but I've found it useful. 2) is usually an incredibly slow and painful way to do things; the results sometimes stick extremely well, and sometimes don't stick at all. 3) can be used for bootstrapping sometimes, and works as well as a chocolate teapot other times - I've picked up a couple of rules of Mandarin tone sandhi intellectually over the years, but couldn't get even individual tones right when I started seriously trying 3 days ago, and doubt I'd recognize tone sandhi if I heard it, much less produce it correctly; trying to explicitly think about tones still trips me up entirely when I try to produce them.

2 persons have voted this message useful



patrickwilken
Senior Member
Germany
radiant-flux.net
Joined 2793 days ago

1546 posts - 3200 votes 
Studies: German

 
 Message 31 of 52
11 April 2013 at 11:55pm | IP Logged 
emk wrote:

But I think Krashen is wrong about the sufficiency of massive comprehensible input, and this can be shown trivially. I know a young woman who's lost virtually all her active skills in her L1, to the extent that she doubts she can put a grammatical sentence together. But at the same time, her receptive skills are excellent. She can watch police dramas with lots of slang on television and understand almost everything (I'm guessing she has somewhere from C1 to native comprehension).

She has the massive comprehensible input. But she lacks any reason to actually speak her L1. And so she understands without being able to speak. To me, a single counter-example like this seems fatal to any theory that doesn't discuss the role of two-way communication in language acquisition.


My mother knew a woman from when they were both Lithuanian refugees after the war. They both emigrated to Australia. A few years ago the woman had a stroke, and she woke up much to the dismay of her English speaking husband only being able to speak Russian, which she presumably learnt sometime before 1942, and had reportedly not spoken 65 years since. Thankfully after some time her other languages re-emerged.

As far as I have been able to find out her Russian language ability was dormant before the stroke (Russian wasn't even her L1 - it was a second language she learnt as a child for a few years) and was only activated after the stroke suppressed her other stronger language areas for Lithuanian and English.

So I guess I would interpret your friends lack of ability to speak as perhaps due to having a weakened/inhibited language representation through non-use, which allows her to retain her passive listening ability, but not active talking ability. I feel at B1 in sort of this position myself. I can watch movies quite comfortably now, but I can't speak very easily (or at least if I do speak I catch myself making grammatical errors that inhibit me).

But even if that is wrong, it doesn't mean that Krashen's main point, which I take to be that knowledge of explicit grammar rules aren't able to be used to train up the language generator in our heads, is.

emk wrote:

Intuitively, I strongly doubt that people can read grammar rules, understand them intellectually, and then transfer that knowledge directly into their automatic speech. If people could really learn like that, they'd surely be better at subjects like math and computer science. But the reality is that people are laughably bad at dealing with the kind of typed, recursive tree structures that appear in grammar.


I agree completely. I think just that the codes are different. Whatever generates grammatical speech in our brains, doesn't code grammar in the sort of explicit rules we learn from grammar books. So in that sense the language generator area of our brains is impenetrable to the geeky grammar rule expert part of our brains.

But I was thinking that maybe one thing we could say to the language area is that it's made a mistake (i.e., send an error signal back), which might allow the language area to adjust its code appropriately. The error signal could originate from someone telling you are making an error, or by you consciously recognizing you've made an error. So in that sense knowing grammar might be quite helpful for not only filtering the output as Krashen suggests (which slows speech) but also rewiring the processing itself.

emk wrote:

That said, I also doubt that language acquisition is completely cognitively impenetrable. This is a really interesting topic...


Yes. Because it tells us what, if anything, in addition to Krashen's comprehensible input we can do to learn a language.

emk wrote:


1) Can the brain accept explicit examples of acceptable and/or unacceptable sentences and learn rules from that? What about rehearsing acceptable sentences? No need to teach grammar, just tag certain data points for special interest.


Interesting ideas.


emk wrote:

2) Can you create comprehensible input the hard way, using either translations or explicit knowledge of grammar to understand something? What if you re-expose yourself to the now-comprehensible input on a regular basis?


This seems to happen all the time when I am reading, when a sentence suddenly clicks and it's meaning suddenly emerges. It's hard to imagine that I am not learning the sentence after I understand it.

emk wrote:

3) Can approximately correct intellectual knowledge be used to bootstrap the acquisition of detailed subconscious knowledge?


I don't think that's what you mean exactly, but when I started learning German I read a simple grammar which was very useful in allowing me parse sentences as I started reading so I had comprehensible input. So in that sense knowing grammar rules is helpful, but of course that's a much simpler set of grammar that you need for correct output.

Edited by patrickwilken on 12 April 2013 at 12:00am

1 person has voted this message useful



patrickwilken
Senior Member
Germany
radiant-flux.net
Joined 2793 days ago

1546 posts - 3200 votes 
Studies: German

 
 Message 32 of 52
12 April 2013 at 12:11am | IP Logged 
Volte wrote:
1) See also Norman Doidge's "The brain that changes itself"; neuroplasticity is also a major factor.

2) You could, but it's probably more useful to think of the brain as consisting of parts that strengthen, weaken, and remap based on use, and language as a set of statistical processes that make sets of neurons fire together.

3) The lines look the same length to me, though they didn't the first time I saw the illusion, many years ago.

Krashen advocates massive amounts of i+1 input, which is compatible with a number of ways of analyzing the brain, including 2).


1. Neuroplasticity is amazing. The cognitive scientist Stanislas Dehaene made this really interesting point. If you look in the brain there are areas specialized for the neurocomputation of reading. what is amazing is that this ability is completely a cultural phenomenon. In our DNA there is presumably some coding for language processing, but there can't be any DNA coding for reading. That's too recent evolutionarily. So every time we teach a child to read we are demonstrating how plastic our brains really are.

2. I agree, but I don't think that changes Krashen's point. The area doing the language statistics doesn't necessarily understand explicit grammar rules, and so what's the point of learning them?

3. If you are jaded by the Müller-Lyer, try out my favorite:

http://web.mit.edu/persci/people/adelson/images/checkershado w/checkershadow_illusion4med.jpg

Does knowing that A and B are the same shade of grey help you see them that way?



Edited by patrickwilken on 12 April 2013 at 12:14am



1 person has voted this message useful



This discussion contains 52 messages over 7 pages: << Prev 1 2 35 6 7  Next >>


Post ReplyPost New Topic Printable version Printable version

You cannot post new topics in this forum - You cannot reply to topics in this forum - You cannot delete your posts in this forum
You cannot edit your posts in this forum - You cannot create polls in this forum - You cannot vote in polls in this forum


This page was generated in 3.9063 seconds.


DHTML Menu By Milonic JavaScript
Copyright 2020 FX Micheloud - All rights reserved
No part of this website may be copied by any means without my written authorization.