Noam Chomsky (33:26) and Stephen Krashen discuss modern linguistics.

General discussion about learning languages
Online
Kraut
Black Belt - 2nd Dan
Posts: 2601
Joined: Mon Aug 07, 2017 10:37 pm
Languages: German (N)
French (C)
English (C)
Spanish (A2)
Lithuanian
x 3204

Noam Chomsky (33:26) and Stephen Krashen discuss modern linguistics.

Postby Kraut » Mon Jun 14, 2021 9:15 pm

https://www.youtube.com/watch?v=148KDh4tfrI

Noam Chomsky (33:26) and Stephen Krashen discuss modern linguistics.
5 x

User avatar
Querneus
Blue Belt
Posts: 836
Joined: Thu Dec 01, 2016 5:28 am
Location: Vancouver, Canada
Languages: Speaks: Spanish (N), English
Studying: Latin, French, Mandarin
x 2269

Re: Noam Chomsky (33:26) and Stephen Krashen discuss modern linguistics.

Postby Querneus » Wed Jun 16, 2021 1:18 pm

Kraut wrote:https://www.youtube.com/watch?v=148KDh4tfrI

Noam Chomsky (33:26) and Stephen Krashen discuss modern linguistics.

This bit around minute 11 was amusing, on teaching grammar:
Stephen Krashen wrote:I've been misquoted as saying "don't ever teach grammar in a language class", "teach grammar; go to jail", "it's a sin", "don't ever use it". Not quite my position. My position is, it's really hard to do! You're not going it make much progress with it. [...] If you want to use grammar, there are daunting conditions that have to be met that make it very tough.

We think consciously learning grammar is only available as an editor or a monitor. [...] The conditions [...] 1) you have to know the rule, consciously, [...] Linguists always point out they don't know all the rules [...] You've got to know the rule and that is really impossible to do. 2) You've got to be thinking about correctness [...] 3) You have to have time. Normal conversation does not give us enough time. [...]

In the real world, we have found that the common view of language teaching will not work. The common view of language teaching?, you teach people a rule consciously, so they can master it consciously, and you expect them to master it right away, and then you practise. With output. [...] Like you learn the 3rd person singular, and then they give you sentences, where you have to say it again and again and again, until you, as they say, "automatize" it, and then you'll be able to use it.

There has been a real mistake, a real confusion of cause and effect. What we think now is going on is: if you acquire the language, if the conditions are right for language acquisition, the effect will be acquisition of language. The cause is language acquisition, the effect is knowing the rules of grammar, and knowing the rules of vocabulary. And this happens only in one way, and here is the major hypothesis: we acquire language in only one way, and that's when we understand it.

(He uses the English 3rd person singular -s as an example of something that's acquired really late, and that many advanced L2 speakers don't use correctly in spontaneous output.)

At minute 33, Chomsky compares learning a language to learning to walk:
Noam Chomsky wrote:At one point you talked about the change from language acquisition to language development. And I think that's a very significant idea. Language isn't really learned; it just grows in the mind. [...]

It's almost like learning to walk. [...] At a certain point a child stands up, starts moving, figures it [...] Nobody's taught you this, but, a lot of calculation and computation goes into simply walking down the street. You don't know the rules, you couldn't know the rules, maybe some biologist could figure out what they are. But that's not the way you pick up walking.


At minute 49, Krashen amusingly elaborates on his belief that accent acquisition is perhaps not affected by the Critical Period as much as by plain group membership (apparently politics is probably what prevents us from sounding like natives), which for him also means there's not much of a point in worrying about it:
Krashen: Critical Periods, um... I have a hypothesis about what's going on. I'd like to believe, I'd love to believe that the language acquisition device never shuts off, that it's always there.

And I started thinking about this in the 1960s, when I was teaching at a school in Ethiopia, and I met a guy there called Gerald Mosbak[spelling?], from the UK, and we had an interesting conversation. He said that when he was in secondary school they all did French, and he hated it. It was poor methodology, boring, he hated the teachers, he hated the French language, he hated the French culture. Now to make things worse, he had to do an oral examination in front of all your teachers at the end, which is the worst torture you can do for an impressionable teenage guy speaking another language. So he decided he would show his disdain for the language, the culture, and the school. He came in by humiliating everybody. He came in dressed French with a beret, with a glass of some liquid pretending it was wine, and he did what he thought was foolishly exaggerate a French accent. He said, Ah ! Bonjour monsieurs, mesdames, je suis très content d'être ici avec vous. Ah, comment ça va aujourd'hui, hein ? And his teachers said, "Well! Why weren't you doing that during the year? That was fantastic. Your accent is great."

He could do that trying to make fun of his teachers, but he couldn't do that ordinarily. I found later that Peter Ustinov, famous wonderful actor, could do movies in French, where his French was totally convincing. He said, "I can't do that outside, when I have a conversation."

Conclusion: language signals what group you're a member of.

That's its, y'know... Accent is a powerful way of doing that. And if you're not a member of the group that uses the language, you have a hard time using that accent. When I meet someone from the UK, I don't suddenly start speaking with a British accent. I find it extremely difficult to do. Versions I've heard all my life I cannot imitate, or I'm very uncomfortable doing so.

It's club membership. Maybe accent does not obey the Critical Period. Maybe we have what we could call an "output filter" instead.


Chomsky: it's true.


Belmekki: Interesting. Thank you very much [...] So what's the relevance of the Critical Period for English-speaking proficiency among immigrants? Don't you think the same way kids acquire-...?


Krashen: Yeah. I think the implication is to not worry about it at all. They're going to get the accent of their friends, people they feel comfortable with. They're not going to get the accent of the teacher, that's been shown over and over again. And there's not much we can or should do about it.

This will make them comfortable when the need comes out or the feeling comes out that they should talk in a slightly different way. Okay, my colleague [insert name] told me that when she was a young girl, she became a tomboy. She's not anymore. But that was the role. There was a special language they used, a special way of speaking that she wouldn't use now.

I'm telling you this, because you know all about it, it happens to all of us. Don't worry about it. Teaching it won't help: I've looked at the research and teaching accent is the same as teaching grammar. You've got to be focused on certain forms, it's limited to simple rules, etc. and when your mind is off it, you can't do it. So I would say, let's not worry about it.
8 x

Cainntear
Black Belt - 3rd Dan
Posts: 3469
Joined: Thu Jul 30, 2015 11:04 am
Location: Scotland
Languages: English(N)
Advanced: French,Spanish, Scottish Gaelic
Intermediate: Italian, Catalan, Corsican
Basic: Welsh
Dabbling: Polish, Russian etc
x 8667
Contact:

Re: Noam Chomsky (33:26) and Stephen Krashen discuss modern linguistics.

Postby Cainntear » Sat Jun 19, 2021 1:30 pm

Stephen Krashen wrote:I've been misquoted as saying "don't ever teach grammar in a language class", "teach grammar; go to jail", "it's a sin", "don't ever use it". Not quite my position. My position is, it's really hard to do! You're not going it make much progress with it. [...] If you want to use grammar, there are daunting conditions that have to be met that make it very tough.

This is supremely weaselly.

I mean, he says he didn't say not to do, he says he only says it's hard, then he sets out impossible conditions, finally saying it's impossible, but somehow saying he says not to do it is "misquoting" him...?
If he says it's impossible, he at least very heavily implies not to do it.

Krashen wrote:The conditions [...] 1) you have to know the rule, consciously, [...] Linguists always point out they don't know all the rules [...] You've got to know the rule and that is really impossible to do.

That's a huge leap of logic. Not knowing all the rules doesn't imply you don't know any.
For example, a rule that is often left unsaid is that indirection can be used to express politeness.
eg "Shut the door!" is more direct and rude than "can you shut the door?" "Could you shut the door?" is in turn less direct that "can you...?" and comes across as more polite.
It's not precise, it's not complete, but it's true, and it's helpful.

And yet... do we talk about that in language classes? Many beginners' classes now avoid imperatives and teach "would you...?" type requests in order to make their students appear more polite when talking to natives, but they're only teaching them to be polite by default, not to understand how politeness works.

How many low-level learners out there simply "acquire" conditional language as though it was just the imperative?

In the real world, we have found that the common view of language teaching will not work.

This is a perfect distillation of everything about his rhetoric that winds me up. Who's this "we"? He's co-opting unnamed others into his argument. Clearly he's talking about some kind of minority if the "common" view is wrong. And if the common view is so wrong, why is that not self-evident?

The thing is, he's straw-manning. Yes, in practice we often don't give people enough time and practice to fully internalise the rules, but that's not because of a genuine belief that once you tell someone how to do something once they can now do it perfectly every time. Teaching fails because of a combination of factors, and rather than attempting to fix the holes in teaching, he just proposes that everyone throws away everything and starts again doing something totally different.

Querneus wrote:(He uses the English 3rd person singular -s as an example of something that's acquired really late, and that many advanced L2 speakers don't use correctly in spontaneous output.)

That's another thing that annoys me about him -- constantly talking about examples that show why old methods don't work, without actually presenting any evidence that his immersive "acquisition" idea makes things any better, and instead only offering hand-wavey excuses to say why people who fail to acquire these features in immersive settings are not proof he's wrong.

Noam Chomsky wrote:At one point you talked about the change from language acquisition to language development. And I think that's a very significant idea. Language isn't really learned; it just grows in the mind. [...]

It's almost like learning to walk. [...] At a certain point a child stands up, starts moving, figures it [...] Nobody's taught you this, but, a lot of calculation and computation goes into simply walking down the street. You don't know the rules, you couldn't know the rules, maybe some biologist could figure out what they are. But that's not the way you pick up walking.

Yeah, but when I wanted to learn to Salsa, I got someone to tell me the steps. Maybe I could have just watched dancers and tried to imitate them, but it would have taken me a fair bit longer, I'm sure.

Querneus wrote:At minute 49, Krashen amusingly elaborates on his belief that accent acquisition is perhaps not affected by the Critical Period as much as by plain group membership (apparently politics is probably what prevents us from sounding like natives), which for him also means there's not much of a point in worrying about it:

Yeah, that kind of sounds like a doctor refusing to treat a patient because they're dying. Well yes, doc, which is why we want you to intervene.

My personal view on accent is that yes, people see it as an important marker of identity, but there's no reason they should. But when I'm speaking in my own language, my accent identifies me as part of a sub-group of natives and therefore an insider. If I don't modify my accent at all in a new language, the accent identifies me as an outsider.

Is that the self-identity I want to express to others?
Do I want to be the "other" -- one of "them" rather than one of "us"?

This is an idea a teacher can get across if they discuss stuff with students in their own language rather than pointing at their eyes and ears.
Belmekki: Interesting. Thank you very much [...] So what's the relevance of the Critical Period for English-speaking proficiency among immigrants? Don't you think the same way kids acquire-...?

Krashen: Yeah. I think the implication is to not worry about it at all. They're going to get the accent of their friends, people they feel comfortable with. They're not going to get the accent of the teacher, that's been shown over and over again. And there's not much we can or should do about it.

Did he really cut him off before he'd finished asking the question? Seems like the presenter was trying to ask something about adults, and he answered with regards to kids...

Pair of dinosaurs...
5 x

User avatar
lysi
Yellow Belt
Posts: 98
Joined: Sat Mar 27, 2021 1:34 pm
Location: Ontario, Canada
Languages: English (N), French, Mandarin (Beginner)
x 300

Re: Noam Chomsky (33:26) and Stephen Krashen discuss modern linguistics.

Postby lysi » Sat Jun 19, 2021 2:31 pm

Cainntear wrote:
Krashen wrote:The conditions [...] 1) you have to know the rule, consciously, [...] Linguists always point out they don't know all the rules [...] You've got to know the rule and that is really impossible to do.

That's a huge leap of logic. Not knowing all the rules doesn't imply you don't know any.
For example, a rule that is often left unsaid is that indirection can be used to express politeness.
eg "Shut the door!" is more direct and rude than "can you shut the door?" "Could you shut the door?" is in turn less direct that "can you...?" and comes across as more polite.
It's not precise, it's not complete, but it's true, and it's helpful.

This is one of the things that Krashen seems to like to do a lot, that is to say, attack the strong interface position (that explicitly learned knowledge becomes implicit knowledge naturally by practice).There is however a weak interface position where explaining a rule can help acquire a particular structure, with students who are explained a rule before shown input consistently acquiring the rule better than with just input and no explanation. This does involve the problem of operationalization of implicit and explicit knowledge. How exactly are you supposed to test if knowledge is implicit or explicit? This was something Krashen was criticized for originally but it actually is possible to test, particularly with time based tests.

Cainntear wrote:
In the real world, we have found that the common view of language teaching will not work.

This is a perfect distillation of everything about his rhetoric that winds me up. Who's this "we"? He's co-opting unnamed others into his argument. Clearly he's talking about some kind of minority if the "common" view is wrong. And if the common view is so wrong, why is that not self-evident?

The thing is, he's straw-manning. Yes, in practice we often don't give people enough time and practice to fully internalise the rules, but that's not because of a genuine belief that once you tell someone how to do something once they can now do it perfectly every time. Teaching fails because of a combination of factors, and rather than attempting to fix the holes in teaching, he just proposes that everyone throws away everything and starts again doing something totally different.


Language learning has changed quite a bit since the 1980's when Krashen first created his theories, but he continues to repeat the same idea of "we've tried everything and it just hasn't worked". I think this is part of why he's so popular online. People hear him repeat this (or his supporters) and assume it's true and conflate their (usually) bad language learning experience in school with the grammar heavy and no input methods he attacks (like the grammar translation method, which he is actually right to attack, since it's a method without a theory), but really, most language courses don't teach like this, and haven't taught like this for ages. It's also a fairly attractive idea that language learning in the past was just difficult because they didn't do it properly, but NOW we have a scientific method and you get to be part of this new revolution that makes everything easier. I don't mean to say by this that language learning hasn't gotten easier, it certainly has consistently gotten easier over time but that's a different topic.

I think the title of the video is fairly funny. Krashen's theories haven't been "modern" for a while now.
Cainntear wrote:
Querneus wrote:(He uses the English 3rd person singular -s as an example of something that's acquired really late, and that many advanced L2 speakers don't use correctly in spontaneous output.)

That's another thing that annoys me about him -- constantly talking about examples that show why old methods don't work, without actually presenting any evidence that his immersive "acquisition" idea makes things any better, and instead only offering hand-wavey excuses to say why people who fail to acquire these features in immersive settings are not proof he's wrong.


He does actually have a real justification for why people fail to acquire features with only input, though, it's the affective filter. Their affective filter was just too high. They didn't want to acquire it enough I guess.
1 x

Cainntear
Black Belt - 3rd Dan
Posts: 3469
Joined: Thu Jul 30, 2015 11:04 am
Location: Scotland
Languages: English(N)
Advanced: French,Spanish, Scottish Gaelic
Intermediate: Italian, Catalan, Corsican
Basic: Welsh
Dabbling: Polish, Russian etc
x 8667
Contact:

Re: Noam Chomsky (33:26) and Stephen Krashen discuss modern linguistics.

Postby Cainntear » Sat Jun 19, 2021 5:41 pm

lysi wrote:
Cainntear wrote:and instead only offering hand-wavey excuses to say why people who fail to acquire these features in immersive settings are not proof he's wrong.


He does actually have a real justification for why people fail to acquire features with only input, though, it's the affective filter. Their affective filter was just too high. They didn't want to acquire it enough I guess.

I class that as hand-waving. It's unmeasurable, unprovable and unfalsifiable.

Basically it boils down to saying to other teachers "Your successful students learn despite you, your unsuccessful ones fail because of you. My successful students learn because of me, my unsuccessful ones fail despite me."
3 x

User avatar
Querneus
Blue Belt
Posts: 836
Joined: Thu Dec 01, 2016 5:28 am
Location: Vancouver, Canada
Languages: Speaks: Spanish (N), English
Studying: Latin, French, Mandarin
x 2269

Re: Noam Chomsky (33:26) and Stephen Krashen discuss modern linguistics.

Postby Querneus » Sat Jun 19, 2021 5:52 pm

Cainntear wrote:Did he really cut him off before he'd finished asking the question? Seems like the presenter was trying to ask something about adults, and he answered with regards to kids...

Pair of dinosaurs...

No, he didn't cut him, rather, Belmekki's connection got cut out from the call for some seconds. It's a problem Belmekki kept having during the videocall.

I guess I don't care enough about Krashen to either defend him or do a chorus attacking him, but was amused by the anecdotes at least.
2 x

User avatar
Saim
Blue Belt
Posts: 676
Joined: Sun Nov 15, 2015 12:14 pm
Location: Rheinland
Languages: Native: English
Others: Catalan, Serbian, Spanish, Polish, Hungarian, Urdu, French etc.
Main focus: German
x 2314

Re: Noam Chomsky (33:26) and Stephen Krashen discuss modern linguistics.

Postby Saim » Mon Jun 21, 2021 2:07 am

Cainntear wrote:Yeah, that kind of sounds like a doctor refusing to treat a patient because they're dying. Well yes, doc, which is why we want you to intervene.

My personal view on accent is that yes, people see it as an important marker of identity, but there's no reason they should. But when I'm speaking in my own language, my accent identifies me as part of a sub-group of natives and therefore an insider. If I don't modify my accent at all in a new language, the accent identifies me as an outsider.


Absolutely. It should of course be noted that the fact that identity being an important factor in accent acquisition doesn't mean that there is no other factor, or that nothing can be done besides trying to identify with the target language community or just give up. It seems pretty clear to me that phonetic consciousness (having some kind of conscious understanding of the ways the two phonetic systems differ) can also help you avoid L1 interference in your L2 accent.

The three major aspects of accent acquisition seem to be articulation ("put your tongue here"), perception (learning to reliably "hear" L2 phonemes and major allophones and identifying the differences between the L1 and L2 systems), and psychology (issues of feeling authentic, like you're mocking the L2 community, identity, etc.). While it may be that when pronunciation is discussed in language classes and textbooks there is a sort of hyperfocus on articulation (if there is any discussion of pronunciation at all), that doesn't mean we should ignore the other two aspects. And in fact my intuitive sense is that perception is just as important as identity in this, although that is just based on my own anecdotal experience.
4 x
log

شجرِ ممنوع 152

Cainntear
Black Belt - 3rd Dan
Posts: 3469
Joined: Thu Jul 30, 2015 11:04 am
Location: Scotland
Languages: English(N)
Advanced: French,Spanish, Scottish Gaelic
Intermediate: Italian, Catalan, Corsican
Basic: Welsh
Dabbling: Polish, Russian etc
x 8667
Contact:

Re: Noam Chomsky (33:26) and Stephen Krashen discuss modern linguistics.

Postby Cainntear » Mon Jun 21, 2021 5:47 pm

Saim wrote:
Cainntear wrote:Yeah, that kind of sounds like a doctor refusing to treat a patient because they're dying. Well yes, doc, which is why we want you to intervene.

My personal view on accent is that yes, people see it as an important marker of identity, but there's no reason they should. But when I'm speaking in my own language, my accent identifies me as part of a sub-group of natives and therefore an insider. If I don't modify my accent at all in a new language, the accent identifies me as an outsider.


Absolutely. It should of course be noted that the fact that identity being an important factor in accent acquisition doesn't mean that there is no other factor, or that nothing can be done besides trying to identify with the target language community or just give up. It seems pretty clear to me that phonetic consciousness (having some kind of conscious understanding of the ways the two phonetic systems differ) can also help you avoid L1 interference in your L2 accent.

The three major aspects of accent acquisition seem to be articulation ("put your tongue here"), perception (learning to reliably "hear" L2 phonemes and major allophones and identifying the differences between the L1 and L2 systems), and psychology (issues of feeling authentic, like you're mocking the L2 community, identity, etc.). While it may be that when pronunciation is discussed in language classes and textbooks there is a sort of hyperfocus on articulation (if there is any discussion of pronunciation at all), that doesn't mean we should ignore the other two aspects. And in fact my intuitive sense is that perception is just as important as identity in this, although that is just based on my own anecdotal experience.

Conversely, when pronunciation is not discussed at all, that's usually indicative of a philosophical belief that exposure teaches perception, and articulation follows from perception.

However, I personally think that articulatory phonetics is so little discussed that "hyperfocus" is almost always an overstatement. In self-teaching books, articulation is often thrown in as an introductory chapter before lesson 1, throwing everything at the learner in one go and not attempt to programme in progression and development. Or to put it another way -- articulation is "told" to the learner, not "taught".

I personally believe that good perception is a result of good articulation -- in learning to pronounce properly, you teach your brain what the proper sound is; you develop a strong map of the phonemes which you can then map your received input to.

You'll never learn to perceive the two Ds in Madrid the same if you process them with the map for the English D and voiced TH sounds, but if you pronounce them both the Spanish way, the articulation of the two is so similar that they just feel similar in your mouth.

[Edit: error in italics tags corrected]
Last edited by Cainntear on Sun Jul 11, 2021 9:32 am, edited 1 time in total.
2 x

User avatar
Saim
Blue Belt
Posts: 676
Joined: Sun Nov 15, 2015 12:14 pm
Location: Rheinland
Languages: Native: English
Others: Catalan, Serbian, Spanish, Polish, Hungarian, Urdu, French etc.
Main focus: German
x 2314

Re: Noam Chomsky (33:26) and Stephen Krashen discuss modern linguistics.

Postby Saim » Sun Jul 11, 2021 12:31 am

Cainntear wrote:Conversely, when pronunciation is not discussed at all, that's usually indicative of a philosophical belief that exposure teaches perception, and articulation follows from perception.


I'm not sure that that's (always) the assumption. I think a lot of people think that pronunciation is essentially unlearnable, or not worth bothering with beyond some vague surface details (and usually approached from the orthography - "x letter produces y sound that is the same or similar to z sound in your native language").

I'd also avoid conflating the idea that we need to do explicit perception training with the idea that perception solves itself through exposure alone.

I personally think that articulatory phonetics is so little discussed that "hyperfocus" is almost always an overstatement. In self-teaching books, articulation is often thrown in as an introductory chapter before lesson 1, throwing everything at the learner in one go and not attempt to programme in progression and development. Or to put it another way -- articulation is "told" to the learner, not "taught".


That's true, thanks for pointing that out. It seems that pronunciation is hardly ever integrated into a holistic approach, just discussed as an aside at the beginning of learning.

I personally believe that good perception is a result of good articulation -- in learning to pronounce properly, you teach your brain what the proper sound is; you develop a strong map of the phonemes which you can then map your received input to.

You'll never learn to perceive the two Ds in Madrid[i] the same if you process them with the map for the English D and voiced TH sounds, but if you pronounce them both the Spanish way, the articulation of the two is so similar that they just [i]feel similar in your mouth.


I think that's probably part of it, especially when it comes to consonants. However, vowels and tones are much harder to pin down and seem to operate on a range of values in opposition with each other than the more concrete values of consonants. Personally I've found that even fairly simple things like the ɛ~e and ɔ~o distinction in many Romance languages need a lot of exposure and explicit perception training for me to incorporate into my own speech, and that focusing on articulation just doesn't cut it. Tones/pitch accent I find even harder to acquire than these vowel distinctions. Then again this is in principle an empirically solvable question, and I imagine you have a much deeper knowledge of what the current state of research is on this than I do.

Now that I think of it, I actually do do a combination of articulation and perception training and it seems to give pretty good results. I guess given my academic background in linguistics I take the articulatory part for granted, which would could lead to mistakes in teaching approach if I ever end up back in a pedagogical role.
1 x
log

شجرِ ممنوع 152

Cainntear
Black Belt - 3rd Dan
Posts: 3469
Joined: Thu Jul 30, 2015 11:04 am
Location: Scotland
Languages: English(N)
Advanced: French,Spanish, Scottish Gaelic
Intermediate: Italian, Catalan, Corsican
Basic: Welsh
Dabbling: Polish, Russian etc
x 8667
Contact:

Re: Noam Chomsky (33:26) and Stephen Krashen discuss modern linguistics.

Postby Cainntear » Sun Jul 11, 2021 10:24 am

Saim wrote:
Cainntear wrote:Conversely, when pronunciation is not discussed at all, that's usually indicative of a philosophical belief that exposure teaches perception, and articulation follows from perception.


I'm not sure that that's (always) the assumption. I think a lot of people think that pronunciation is essentially unlearnable, or not worth bothering with beyond some vague surface details (and usually approached from the orthography - "x letter produces y sound that is the same or similar to z sound in your native language").

Not always, no; hence "usually". But a very large proportion of the learn-by-exposure does go that way -- Krashen, the Silent Way etc.

Krashen has suggested that we all acquire accent really quickly, we just don't perform accent because "affective filter". It's a remarkably odd thing to state, because if that were true, it would mean that listening was pretty much trivially easy rather than a very difficult task.

It seems to me self-evident that learners struggle to perceive phonological features (even before you look at any of the literature that demonstrates how real these struggles are).

I'd also avoid conflating the idea that we need to do explicit perception training with the idea that perception solves itself through exposure alone.

What do you mean by explicit perception training? There's a lot of evidence that minimal pairs training doesn't readily generalise into a language skill. I do close listening with some of my students, but I wouldn't describe it as "explicit training", merely "focused practice".

That's true, thanks for pointing that out. It seems that pronunciation is hardly ever integrated into a holistic approach, just discussed as an aside at the beginning of learning.

I remember reading a piece by a teacher saying the worst thing you can do is teach a new learner to count, because numbers have evolved in most languages to be clearly distinct, so you end up going through most of the phonemes of the language in one sitting if you count as far as ten.

He suggested that a course should always be structured to introduced the phonemes systematically, and sticking with the numbers, he pointed out that in Spanish, you could start with "diez" and "cien" are quite different, but each only has one phoneme the other doesn't, making them a better pair to start off with than "uno" and "dos", which only share one phoneme, and if you move on to "tres", that means introducing the Spanish R in your first lesson which many learners struggle with. This seems a bit excessive to me, because surely you'll be introducing other phonemes in different situations anyway, and you can use some of them in your numbers, but it does illustrate the principle. If the first two uses of UA are "cuatro" and "cuando", why potentially have them several weeks apart? If you can structure the course so that the phoneme gets used in two different words in the first lesson it appears in, surely that's better?

By the same token, I wouldn't teach the alphabet in one go. If you introduce CDs, that's two letters learned in a meaningful context (assuming the target language uses CD, that is!) and then you can continue to DVD. If you're teaching English you can add in TV, and now you're up to 4 letters without ever stepping through ABCs, and if the student is from a language background that uses the Latin alphabet, they're already primed for the pronunciation rule for B, G and P before they've encountered them.

I personally believe that good perception is a result of good articulation -- in learning to pronounce properly, you teach your brain what the proper sound is; you develop a strong map of the phonemes which you can then map your received input to.

You'll never learn to perceive the two Ds in Madrid the same if you process them with the map for the English D and voiced TH sounds, but if you pronounce them both the Spanish way, the articulation of the two is so similar that they just feel similar in your mouth.


I think that's probably part of it, especially when it comes to consonants. However, vowels and tones are much harder to pin down and seem to operate on a range of values in opposition with each other than the more concrete values of consonants. Personally I've found that even fairly simple things like the ɛ~e and ɔ~o distinction in many Romance languages need a lot of exposure and explicit perception training for me to incorporate into my own speech, and that focusing on articulation just doesn't cut it. Tones/pitch accent I find even harder to acquire than these vowel distinctions. Then again this is in principle an empirically solvable question, and I imagine you have a much deeper knowledge of what the current state of research is on this than I do.

It may be that I'm misremembering my own experiences but as I recall it, the way of thinking about it that helped me is the whole concept of the phoneme map as areas with fuzzy boundaries. As I recall it, there's a strong tendency in languages to avoid making any sounds in the boundary areas (except when ill or drugged!) but it's OK for the phoneme to exist absolutely anywhere in the clear area. I always made a habit of trying to find the typical "central" characteristic of the phoneme first and not bother much about accent, so not only would I be understood, but when I heard the different allophones, I'd recognise that they were just variations on a theme.

I've always felt, both intuitively and intellectually, that without reference to the core physical characteristic or making a sound, it's near impossible to intuit the patterns from sound alone. I'm a firm believer in the theory that perception is based on a mirror neuron response, and that perception of phonemes is based on mental reconstruction of physical movements, and that the acoustic qualities of a sound are not in and of themselves sufficient to perceive and acquire the phoneme. I can't see how perception training can be explicit without referring to the movements responsible for producing the sound in the first place -- i.e. there is no perception training without articulatory phonetics.

Now that I think of it, I actually do do a combination of articulation and perception training and it seems to give pretty good results. I guess given my academic background in linguistics I take the articulatory part for granted, which would could lead to mistakes in teaching approach if I ever end up back in a pedagogical role.

Yup. The hardest part of teaching is always putting yourself in the learner's shoes. I'm always either underestimating or overestimating my students!
1 x


Return to “General Language Discussion”

Who is online

Users browsing this forum: emk, Kraut and 2 guests