Not Learning From Mistakes

I've been talking a fair bit recently.  First I pointed out that the emotional power of correction of mistakes is often overstated on the basis of a few exceptional cases, and then I pointed out that in the classroom it's underemployed for fear of embarrassment.

Not being corrected

But here's another perspective on the whole thing: some mistakes just don't get corrected.  There's several reasons for this:

1. The other person doesn't want to be rude, so continues to nod politely rather than cause potential embarrassment by commenting.
2. The other person doesn't understand you and says so, but can't correct you because he doesn't know what you're trying to say.
3. The other person understands you, so doesn't see the need to mention the error.
4. The mistake is noticed and mentioned but not corrected, because the other person suffers what is known as "language blindness".

Each of these is troublesome in different ways.

The first couple you can do nothing about, but it's really frustrating if your conversation grinds to a halt after five minutes when you discover that neither of you has a clue what the other is talking about.

From the third, it is tempting to conclude that if an error doesn't stop the other person understanding, it's an acceptable error (a widely held belief among communicative approach teachers).  But in general this isn't true, because it ignores the fact that what is unambiguous in one context may be very ambiguous in another.  For example, many learners of English have problems pronouncing the past suffix -ed (except in -ted, -ded) and drop it.  In the sentence I walked there but it was closed, losing the "ed" doesn't make the tense ambiguous (I walk there but it was close), because "was" clearly marks the tense, but if you only say I walked there and it comes out as I walk there, suddenly it's very ambiguous indeed.  The problem gets worse as phrases, clauses and sentences get more and more complicated as you proceed through the language, and as error builds on top of error, language gets less and less accurate.  One of the main ideas in the communicative approach is that you can "get by" with flawed language and that accuracy will take care of itself later on, but by saving up mistakes for later, they militate against improvement, which is a shame for the students....

And finally, language blindness.  This is when you know what is being said or what you want to say, but you can't find the words or the structure.  In translation it happens when the structure you're translating from blocks you from seeing the appropriate target structure, and it's something that you can't think your way out of, because the material you're working from seems only to get stronger when you think about it.  In the situation of conversational corrections, you've understood what the other people are trying to say, but when you try to correct it all you can hear is what they said, and in the end you cannot give them any hints as to what they said wrong.
Incorrect "correction"

And even if you do get corrected, how do you know the correction is correct?
Take a sentence like *I am going walking yesterday.  There are two likely intended sentences: I was going walking yesterday and I am going walking tomorrow.  Now with this sentence, most correctors would offer both options.  However, in general, listening is a subconscious act, so when listening to someone speak we hear only one thing.  If the corrector misunderstands the error, he will obviously give an incorrect form in response.

Also, you have to remember that very few people genuinely know how they speak - most of us only know how we were told we should speak, and I have heard native speakers "correcting" foreigners for using a (descriptively, statistically) correct grammatical form by providing an outdated (prescriptive) school-book form that they themselves don't use.  Or when pressed for a translation, they give something that is almost correct, but it subtley inappropriate.


All in all, correction during conversation is more than a little hit-and-miss.

While I feel that the idea of fossilised errors is a gross exaggeration, it's still better to get started on the right form as soon as possible to build up good habits.

If you rely on conversational corrections to teach you correct grammar, conversations will become a drag.  Maybe not for you (if enjoy the process of puzzling through), but most people don't have the patience to put up with it for long and you'll find yourself going through conversational partners very quickly indeed.  This is fine if you live in an area with lots of speakers of your target language, because you can always just hang about in a bar until another speaker comes along, but if you're trying to maintain a friendship (or even a romantic relationship), the language will soon start to be a barrier to communication.

Overall, it's just much quicker and more efficient to learn in a structured way where one thing leads to (and supports) another, rather than having a scattergun approach of learning whatever comes up even if it is in no way linked to what you already know.

And it is best to get your teaching from an informed speaker of the language - that is to say someone who not only speaks the language, but has studied it and is consciously aware of the subtleties of grammar and usage, of connotations and register differences.  (And yes, that means that sometimes a non-native speaker can be a better teacher than a native.)

Teaching from mistakes

I find it curious that despite the claims that we learn best from our mistakes, many teachers are reluctant to take advantage of this in teaching.

Well, not reluctant, I suppose, but it's more a matter of wanting things both ways.

In your standard EFL class, we don't typically correct spontaneous errors on the spot (and there are certainly many circumstances where correction would break the flow of the conversation), instead giving "delayed feedback" -- a short period at the end of the class where several of the "big" mistakes of the day are put up on the whiteboard and discussed and corrected as a class.

One of the reasons for this is to avoid drawing attention to the invidual or causing any embarrassment.  However, this means it's too late.  Ten or fifteen minutes later, the student doesn't have the same emotional tie to the sentence and any correction is purely academic.  What makes this into a particularly fruitless endeavour is that most of the time they know the correct answer in theory, but they just fail to apply the correct rule in practice, so the error isn't addressed properly.
But if we look at the various confusions I mentioned last time -- married vs tired, embarrassed vs pregnant, having a cold vs having constipation -- we see that correction is at its most effective when it connects on a very vivid, immediate, emotional level.

I'm not saying we should go out of our way to embarrass and humiliate learners, but that we shouldn't be afraid to take advantage of the humour or absurdity of an error in making the correction more memorable.

For example, I know one fellow Gaelic learner who hasn't really grasped the correct use of possessives, and is wont to say things like "tha mi cat agam a' dol dhan bheat", which means pretty much "I am one of my cats on my way to the vet".  Now, you can correct her with "tha an cat agam" and she might repeat it, but 10 seconds later, she'll be saying it wrong again.  So why not just tell her she's called herself a cat? 

Ok, so you have to build up a certain rapport with the other party to make sure you're doing it in a good natured way, and they feel you're laughing with each other rather than laughing at the other person.

But delayed feedback is almost always too late: a mistake can only lose its power to embarrass when it no longer feels like that particular student's mistake, and if it's not that student's mistake then it's just another restatement of the rule. 

Hang on a minute... let's get this straight.  Delayed feedback isn't correction at all?!?  Not on a personal level.  On a technicality, yes, it is correction, but on an emotional, personal level, it's not.  Or at least not always.

So either correct in time to make a difference, or else find a way to teach that avoids errors in the first place.  (And that's not necessarily as hard as it sounds.)

Learning from mistakes...?

I've heard it said many times that the best way to learn a language is to make mistakes and be corrected, because this correction somehow "personalises" the learning.  But to me, this is a logical absurdity.

Here's an example from my personal experience.  In Spanish, there is a special word hay that is equivalent to the English there is.  You therefore do not translate there is verbatim (which would give you *allí está).  In Spanish, they also have a word "demás" which is used for other(s), meaning the rest of a group (not other as in different).

Now I often mistranslate the others (the rest of the guys) as *los otros, but I never mistranslate there is as *allí está.   This means I can say hay correctly despite never being corrected and I keep getting los demás wrong despite fairly frequent correction, and conceptually one is no more difficult than the other.

This is only one example, but in general when someone corrects my Spanish, it's for one of a closed set of mistakes that I make all the time.  Being corrected seems to have absolutely no direct effect on my errors.

The only effect is when I subsequently choose to work consciously to eradicate that mistake.  Or perhaps more accurately, when I consciously work to learn the correct form, because more often than not, an error isn't the result of learning something wrong, but actually an indication that you haven't learnt it at all.  I get hay right because I learned it early on, I get los demás wrong because I never really learned it, and even now I'm only "aware of existence" -- I still don't feel I've learned it.

This isn't to say that errors and correction are valueless, not at all.  Corrections have no special ability to make language stick, but they at least indicate a gap in your knowledge and they often give you a starting point for filling that gap, so you certainly should listen to and take note of any corrections you're given.  What you shouldn't do is ascribe magic powers to corrections and believe that they are a substitute for other ways of learning -- this will only slow down your progress while making conversations in your target language far less enjoyable for both parties.

Why is this idea so persistent?

There is a small set of mistakes that people only make once.  For example, when a Spanish learner tries to say she's tired (cansada) but instead says he's married (casada), it is an instantly memorable situation and unlikely to happen again.  Of course it's a situation that's quite embarrassing, which might be compounded if she now mistakenly says she's pregnant (embarazada).  And she'll never do that again.  Similarly a Spanish speaker suffering a cold is unlikely to forget that being constipated is not the same as having a bunged-up nose (constipado in Spanish).

So these mistakes certainly do lead to better recall of the correct form, but this is not because of how correction works as a generally applicable learning strategy, but rather a consequence of the vocabulary in question.  Saying translación (an archaic variation of translación - movement/transfer)instead of traducción (translation) doesn't have the same comedic value (and in fact isn't likely to obscure the meaning in context) so doesn't have the same potential to stick.

Most often it is these extreme cases, these outliers, that are used to convince us of the efficacy of the technique, but don't be fooled: learning the correct form from the word go is far more effective than making it up as you go along and picking up a catalogue of corrections.

Mechanics' Meaningful Music

It is often claimed that an adult cannot learn the sound system of a new language.  This claim is followed by the caveat that some adults do, but these adults are dismissed as exceptional, and non-typical.  Certainly, they are exceptions, because most people don't, but there's a big difference between "don't" and "can't".

A sound system is composed of phonemes, which are often defined as minimal units of meaning in sound.  Every human is capable of producing a whole range of sounds, regardless of their language, and to process every single detail of the sound produced would simply be too much for the brain, so we bundle the sounds up together, and even though the sounds of the T in "try" and "butter" may be slightly different (or completely different, depending on your accent), we still recognise them as being the same thing.

As a general strategy, this works.  The adult human meets lots of people with slightly different accents, but the phonemes are all roughly the same, so the detail of the differences is irrelevant.

As a language learner, though, this starts to pose problems.  Our brains believe that only certain sounds are meaningful, and therefore discard any information they believe to be irrelevant.  If you have a language with two phonemes equivalent to one in your language, you will not believe the distinction is meaningful.  Just take a look at Japanese, where they have one phoneme equivalent to the English L and R.  Many Japanese learners of English cannot hear the difference between "law" and "raw" or "appear" and "appeal" without active concentration.

Most people get stuck in this rut their whole lives, and this is used as evidence that you can never learn the sound system.  But if we step outside the world of language, we might just find reason to be more optimistic.

It's amazing what an experienced mechanic can determine about a car or other machine just by listening to it.  Sounds that to you or me would just be squeaks and squeals are to him a full description of the workings and faults of the engine.

Mechanics develops this skill over time through a mixture of direct instruction and experiential learning.  The engines they work on give constant feedback that develops into a meaningful structure -- if a given whine co-occurs with a drop in revs, the two become associated and the sound takes on meaning.

But by this reasoning, surely language itself should give a meaningful framework to sounds?

Unfortunately, it would appear not, and it isn't actually that hard to see why.

Language has evolved to have a certain amount of redundancy, a certain level of "fault tolerance".  It is very difficult indeed to find any complete sentences that function as minimal pairs (ie that differ by one phoneme only), particularly within the restricted language set that most beginners are faced with.

Going back to my earlier examples, "law" is a noun, "raw" is an adjective.  There will always be enough information in the context to tell the two apart.  "Appeal" and "appear" are both verbs, but the usages are distinct.

Essentially, I believe that the average learner is never really forced to build a meaningful framework for these differences.  The end result is that they get deeper and deeper into the language, building more and more coping mechanisms that a native speaker would never rely on.  The model of the language they build is wrong, and while they can understand most things they hear, the person they are speaking to often cannot understand them because, as I said, the native speaker doesn't employ the same strategies as the learner.

What is needed is for the teacher to force a meaningful framework, and the only way I can see that happening is through early teaching of pronunciation.  If a learner has to pronounce the difference between ż and ź in Polish, and is corrected when using the wrong one, his brain will know there's a meaningful difference.

It may not be fashionable, but some negative feedback is definitely necessary....