Correction... how useful...?

[Cainntear]

I've not followed every post, but I too question the value of MT. Whilst I have to always admit that people learn differently.

I argue that people's apparent differences are superficial. What "macro" techniques the teachers give us may change, but I do feel that what we as good language learners actually do -- the thinking strategies we employ to get through the techniques -- are very similar.

It does seem the method is being repeatedly touted as something greater than what it is.

Oh, absolutely it is. But that doesn't mean it isn't a very big step in the right direction.

Like Irena I went to have a listen to the French course after reading about experiences with it here. I found it limited and that Michel Thomas himself isn't the best imitation model for the language.

Listening to something teaching a language you already know doesn't help you analyse the teaching method... unless you listen to the whole thing, catalogue the language taught and learned and compare it to what another course could teach in a similar span of time.

How's your Spanish? I can recommend using MT's Spanish course if you haven't got very far with it yet, and it will take you a very long way -- in fact, in your case, you've probably got enough of a variety of vocabulary in other languages that you'll find the course very useful.

I'm also flummoxed in a way because MT does a lot of correction, whereas Cainntear seemed to think this doesn't work (am I wrong about that?).

I started off this thread by saying that delayed correction doesn't always work, but that immediate correction is far more successful. MT doesn't make a note of student errors in the course of 40 minutes then deliver the corrections of all those errors in a 10 minute session.

I would assume the similar issues with MT as with Pimsleur, both positive and negative. That they pursue communicative speech. I'm limited in my comments because I have done neither of these for French, but I do know that the issue of the gulf between spoken and written French would in this case (and in any language really) put the learner at a bit of a disadvantage.

In the case of MT, the goal is command of grammar and it is taught by using using translation tasks that are focused deliberately and vary enough to not be substitution drills, but don't vary as much as Pimsleur tasks do. Pimsleur is effectively teaching phrases for memorisation and therefore doesn't have the range.

Nowadays practically all learners know they need audio input so the those inclined to book learning make that valuable effort; whereas those who feel they can make most progress lying in the sun listening to audio are still a bit recalcitrant with regard to the value of checking grammar and using written material to propel themselves.

Yes, people who use audio materials and don't pay attention are not going to learn a lot, but you can use an MT course that way. Pimsleur you can go into auto-pilot and echo the language from memory.

But you can't use MT that way, because there's no real repetition so you can't just memorise the right answers.

Pimsleur had at its heart various ideas that were in vogue when it started, and it owes a lot to the audio-lingual methods pioneered by the US army and took it further into the realms of memory schedules.

MT was not like that.

Reading information is surely the fastest way of getting it. I know I can read an article faster than I can listen to one.

Define "getting it". You might be able to read fast, but the physical act of reading is different from "getting it".

[Cainntear]

For one thing, there's a bit of an obsession with meaningless exchanges which are what I term "expository language". For example, consider a language with three locations -- I'll use "here" to mean "where I am", "there" to mean "where you are" and "yonder" to mean "where neither of us are". Similarly "this" (one near me), "that" (one near you) and "yonder" (one near neither of us).

I used to get Pimsleur short courses out of the library for fun, but they used to have all sort of dialogues that were like "is the bank here (where I am)?" "No it's there (where you are)," or "Do you want this one (near me)?" "No, I want that one (near you)."

The whole thing basically teaches the words incorrectly.

I've done two Pimsleur courses and I've never seen any dialogue of that type. Are these 'short courses' something else? Is there really any confusion as to what 'here/there', 'that/this' means when the context is clear, as it very much is in the limited and clearly defined dialogues of Pimsleur?

The short courses I'm talking about were a mixture of the first boxed set of a larger course in popular languages and the World Traveler series of languages that were too niche to merit a full release.

Thinking about it I'm pretty sure the thing in Pimsleur was usually about where the bank, station or hotel was (probably also a common thing in the World Traveler series, even if less common in the regular courses) and the constant "Is the bank there [near you]?" "No, the bank is here [near me]," really wound me up.

I think the "Do you want this [near me]?" "No, I want that [near you]," may actually have been in various other courses... possibly the "MT method" courses by other teachers who clearly didn't get what MT was doing. I'm also pretty sure it was in a Welsh audio course that I used for a while.

[Le Baron]

Listening to something teaching a language you already know doesn't help you analyse the teaching method... unless you listen to the whole thing, catalogue the language taught and learned and compare it to what another course could teach in a similar span of time.

Yes, this is likely true. Though it does also offer another perspective beyond the learner perspective. Whether it's a valuable perspective is likely debatable.

How's your Spanish? I can recommend using MT's Spanish course if you haven't got very far with it yet, and it will take you a very long way -- in fact, in your case, you've probably got enough of a variety of vocabulary in other languages that you'll find the course very useful.

I did actually give it a go, but it wasn't for me. I think I'm past that stage now. Though sometimes I've used the last few stages of a course to extract some value. I imagine with MT one has to follow the entire thing to develop a feel for the approach.

I started off this thread by saying that delayed correction doesn't always work, but that immediate correction is far more successful. MT doesn't make a note of student errors in the course of 40 minutes then deliver the corrections of all those errors in a 10 minute session.

Ah yes. I agree.

[Querneus]

All I know is that for seven or nine years I tried to correct some of my father's basic pronunciation errors, typically cases where he pronounced English words by reading them out as in Spanish, and it just never worked. So I've simply given up on correcting his pronunciation now.

For example, I tried to get him to pronounce "mountain" and "account" as [ˈmaʊn.tn̩] (MOUNT-n) and [əˈkʰaʊnt] (uh-COUNT) instead of [ˈmountain aˈkount] (MOHN-tyne, uh-CONE-t), and it just never worked. Which I find strange because I find I am myself very amenable to corrections of this sort, but I guess people vary. Not necessarily an age thing: my mother is much better in comparison.

[Irena]

All I know is that for seven or nine years I tried to correct some of my father's basic pronunciation errors, typically cases where he pronounced English words by reading them out as in Spanish, and it just never worked. So I've simply given up on correcting his pronunciation now.

For example, I tried to get him to pronounce "mountain" and "account" as [ˈmaʊn.tn̩] (MOUNT-n) and [əˈkʰaʊnt] (uh-COUNT) instead of [ˈmountain aˈkount] (MOHN-tyne, uh-CONE-t), and it just never worked. Which I find strange because I find I am myself very amenable to corrections of this sort, but I guess people vary. Not necessarily an age thing: my mother is much better in comparison.

It may be that your father doesn't think this is important. "If you understand me, then why should I fix it?" Or, it may be that he just has trouble noticing the difference. You say two things that sound obviously different to you, but the person you're speaking to thinks you said the exact same thing twice.

[Cainntear]

All I know is that for seven or nine years I tried to correct some of my father's basic pronunciation errors, typically cases where he pronounced English words by reading them out as in Spanish, and it just never worked. So I've simply given up on correcting his pronunciation now.

It may be that your father doesn't think this is important. "If you understand me, then why should I fix it?" Or, it may be that he just has trouble noticing the difference. You say two things that sound obviously different to you, but the person you're speaking to thinks you said the exact same thing twice.

Which do we take as most likely, then?

If the first is true, it's unprovable and doesn't move us onwards. If the second is true (which I believe to be the case) then the problem is that he is using the same mental processing of phonemes for two languages with fundamentally different phonemic systems. Phonemes are part of the phenomenon of "filter of perception", and we generally understand pretty well what people with different accents from ourselves say to us in our native language, but it's a rare talent to be able to mimic someone using their own phonology -- like people with non-rhotic accents are perfectly capable of understanding rhotic accents, but they're usually rubbish at mimicking them (hell, it's very rare for an English actor to be able to recreate a Scottish R consistently).

Querneus's dad's behaviour is not universal (obviously), but I believe that the way he could have been taught that would have made a difference was to build a phoneme map. Those of us who don't "need" to be taught in a way that builds a phoneme map are those of us who instinctively build a phoneme map. Being taught in such a way that we can't help but build a correct phoneme map would have been good for Querneus's dad, but I say it would also be good for Querneus's mum, because it would have been quicker and more effective.

It's also why I'm hesitant on recommending MT Spanish to people deliberately aiming at Spanish-as-spoken-in-Spain. MT Spanish teaches seseo (ce/z pronounced the same as S) whereas everyone I knew spoke with distincíon (ce/z pronounced differently from S).

I could understand them by treated [s] and [θ] as allophones of the same phoneme (as I'd been taught it), but attempting to do so left weird things up to chance. Like for some reason I'd always pronounce "especial" as though it was "ezpesial". Thankfully I caught this early enough that I didn't have a lot of words that feel together and I could build a better phoneme map with few words to move. I'd read about the problems of phoneme separation early on -- the OU English course talked about Cockney English (an accent that evolved out of French medieval colonists so has no /h/ phoneme) and the weird arbitrarity of correction of the Cockney accent in schools. Because they literally can't tell the difference between no sound and /h/, the kids in school would seemingly randomly add H before vowels or drop H. I say "seemingly" randomly, because there was an observable pattern: kids asked to read out "heart" would say "art", and vice versa. This was statistically relevant so not just pure chance, but at the time I was studying, no-one knew what the actual mechanism that caused it to happen was. I am not aware of whether anyone's gone deeper into it since.

To me, though, the fact that I said "ezpesial" so consistently was linked to this somehow.

[Cainntear]

It may be that your father doesn't think this is important. "If you understand me, then why should I fix it?" Or, it may be that he just has trouble noticing the difference. You say two things that sound obviously different to you, but the person you're speaking to thinks you said the exact same thing twice.

Which do we take as most likely, then?

If the first is true, it's unprovable and doesn't move us onwards. If the second is true (which I believe to be the case) then the problem is that he is using the same mental processing of phonemes for two languages with fundamentally different phonemic systems.

And crucially, the second is a provable hypothesis, which means it can be tested. The first is not provable, and is a call to inaction -- "don't improve your courses cos everyone who doesn't learn doesn't want to".

[Irena]

And crucially, the second is a provable hypothesis, which means it can be tested. The first is not provable, and is a call to inaction -- "don't improve your courses cos everyone who doesn't learn doesn't want to".

Of course it's provable. Just ask the father. If he says "yes, I understand that I'm not saying it quite right, but I don't care," then it means he doesn't care. And if he says "what you said sounds exactly like what I was saying," then it means he can't hear the difference. What's so complicated about that?

[Cainntear]

And crucially, the second is a provable hypothesis, which means it can be tested. The first is not provable, and is a call to inaction -- "don't improve your courses cos everyone who doesn't learn doesn't want to".

Of course it's provable. Just ask the father. If he says "yes, I understand that I'm not saying it quite right, but I don't care," then it means he doesn't care.

Have a look back at this thread and look at how we've almost swapped places here.

I repeated what someone else said and you basically told me you can't believe what people say.

You said that I couldn't trust the tears because (in summary) the people in question might be deluding themselves as a self-defence strategy.

I can bounce that one back at you now, because "I don't care" can more easily be a self-defence strategy than tears. "I don't care" can be hiding a fear of low ability. It can be hiding a history of abuse, which causes a failure to identify with that abuser's own language. Even a simple whip with a cane in school for making a mistake qualifies as abuse.

But I am not taking the criers' words for it. I'm taking a long view based on having been in the classroom with them repeatedly -- often for months -- and having a conversation with them where they describe in detail what they actually do... which is to follow the instructions more closely than I do.

So you said I shouldn't believe what they told me. I followed that up by trying to point out all the above. You said again that I shouldn't believe them.

And now you're saying to believe a simple answer. If you were to do what you're asking me to do, you would believe that the people crying were people that were genuinely trying. Can you believe that?

And if he says "what you said sounds exactly like what I was saying," then it means he can't hear the difference. What's so complicated about that?

The problem is that if you ask someone to listen for the difference between two things, you're inviting them to start on conscious analysis.

Recent studies have questioned the wisdom of minimal pair practice -- i.e. listening to an audio and deciding if it's "ship" or "sheep", or "ship" or "chip". Findings (from a Japanese university unless I'm mistaken) have suggested that students who did regular minimal pair exercises did significantly better at a minimal pairs test at the end of the study, but that they're comprehension and production outside of a dedicated minimal pairs exercise was no better than people who hadn't spent months doing minimal pairs.

Basically, when someone is listening consciously, they override the filter of perception and actually pay attention to the specific sounds.

Like if I asked you to repeat something I said, you would not normally mimic my accent. If I asked you to copy my accent, you would struggle less if I asked you to do that before saying the sentence I wanted you to copy, and you would struggle more if I only asked you to copy my accent after you'd heard the sentence, because you wouldn't have been listening for that information first time round.

[Irena]

@Cainntear

Oh, the father could certainly lie. (I'm not talking about Querneus' father in particular, since I know nothing about that particular person. I'm talking about people in Querneus' father's position in general.) He might say "yes, I understand that I'm not saying it quite right, but I don't care," when he does in fact care but is ashamed of his poor pronunciation. But even then, it probably means "please don't bother me with this again." If you choose to bother him with it again, then tread lightly, and understand that you're running the risk of damaging your relationship with him by refusing to drop the matter.

And as for those crying students: if they were being graded on their performance, it's perfectly possible that they were crying because they were about to get a bad grade. That could happen regardless of effort. If they were not being graded but were still frustrated with their lack of progress, then they probably were putting in a lot of effort. In that case, there are various possible explanations for their lack of progress. Teaching methodology is one possible explanation. Lack of brain power is another. Seriously, why do you think states give citizenship to immigrants who only managed to reach the measly B1 level in the language of their adopted country? Because it would appear that for a great many otherwise competent human adults, it is extremely difficult to reach a higher level, even after a decade of residing in a country in which their target language is spoken. Some of this adult incompetence can be overcome by choosing a sufficiently easy language first, and then possibly taking on a more difficult one later, as a more experienced language learner. For example, if you are an adult native English speaker, it might not be the best idea to go for Scottish Gaelic first. Try Italian first, and then Scottish Gaelic. Still too hard? Then try Esperanto first. Still too hard? Then it's probably hopeless.