29 May 2011

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....

27 May 2011

Learning is fun

In every sphere of teaching, there is a tendency to try to "make learning fun".  This is done through games and the selection of "relevant" or "interesting" material.

I have always been of the opinion that most of the games presented in classes are a distraction from learning rather than an aid to learning, and that most of the attempts to make subjects relevant or interesting tend to obscure the point being taught.  In maths and physics at high school, I used to get given all sorts of contrived scenarios that boiled down to a pretty simple calculation, and many of my classmates would be puzzled by what the question was asking.  The "relevance" that was supposed to be helping motivate us became a hindrance.  And yet the teachers saw the difficulty as a good thing because in the real world, we wouldn't just be given a sum.  Well in the real world, we'd have a lot more variables to deal with.  Personally, I couldn't see any benefit.

I have always been of the opinion that one of the most mentally enjoyable things on the planet is learning.  "You would say that," people tell me, "because you're good at it.  Other people are different."

But to answer that is to miss my point, because learning is pure mental stimulation, and mental stimulation is (to simplify horribly) the basis of enjoyment.  I argue that there is no-one who doesn't enjoy learning.  "But what about the people who don't do well in school?  They don't enjoy it!"  But if learning is a universal pleasure, this is looking at things the wrong way round: underachievers don't fail to learn because they're not enjoying themselves, they fail to enjoy themselves because they're not learning!

On a recent trip to a charity shop, I was lucky enough to stumble across the book Towards a Theory of Instruction by Jerome S Bruner.  In the first chapter of the book, published in 1966 but based on earler papers and lectures, he says:
We discovered one point of especial value for my own future inquiry.  There is a sharp distinction that must be made between behavior that copes with the requirements of a problem and behavior that is designed to defend against entry into the problem.
 He says that the nature of the poor performance of the children he was studying...
was not so much a distortion as it was the result of their working on a different set of problems from those the school had set for them.
He then goes on to point out that
Once our blocked children were able to bear the problems as set -- when we were able to give them a chance for conflict-free coping -- their performance was quite like that of other children
Essentially, he tells us that kids' learning problems seem to be pretty much absolute and digital -- you've either learned or you haven't, which would suggest that changing the way the problem is stated isn't going to make any positive difference to the underperforming students, because they still won't use the appropriate strategy.

But Bruner's legacy is the idea of "discovery learning": that people don't need to be taught things, and that they learn better by discovering things for themselves.  It's an idea that has been horribly distorted and misrepresented over the years.  Certainly every description of this that I have ever read takes a much harder line than Bruner himself, and even Bruner's own studies on this were within a very confined field.

In the book, Bruner deals mostly with elementary maths, ostensibly because of the clarity of expression and the fact that all readers will be familiar with the basic arithmetic under study.  Using cubes to work out areas and volumes is fairly standard, but often the full potential is ignored -- Bruner's studies went on to explore recombinations, and then eventually on to elementary calculus, all at an early primary level.  Bruner was surprised and impressed by how little explicit instruction the children needed, and the additional concepts they explored without being asked.

But the physical constraints here were leading, and they also meant feedback was immediate.  When high schools try to employ discovery learning we get into dubious practices such as "discovering the boiling point of water".  You stick a thermometer in a beaker of water and put it over a bunsen burner.  When it boils, you read it off.

But wait... at what point is water officially "boiling"?  I'm in my 30s, and even though I'm a competent cook, I still couldn't identify with confidence the actual point at which a pan is "on the boil".  So as a teenager I was staring at this beaker, trying to decide when it was right, and scribbling down several numbers.  At the end of the "experiment", we all gave our numbers, and we were all wrong.  What exactly did we discover?  The teacher had to tell us that it was 100.  Yes, that's right, we were using Celsius, the temperature scale defined by the boiling and freezing point of water.  This exercise was pretty ridiculous....

So far, so irrelevant to the language learner.  Well, that's all background.

The idea of discovery learning has infected the language profession too.  But language is pure abstraction -- there is no physical reality to count or measure or explore.  The notion of "correct" language is so vague that it is exceptionally difficult to stumble upon by accident.  And whereas most physical experiments can be reattempted without prejudicing the results, every reformulation in language gives the listener a partial understanding.  Three or four attempts to speak make not result in a single correct sentence, but the other person may well know what you mean by the end of it.  You never need to discover the correct answer.

But as I said, most modern advocates of discovery learning are far more hard-line than Bruner.  In Toward a Theory of Instruction, Bruner didn't use the term very much at all.  Instead, Bruner focused on learning as a process of increasing abstraction, starting at enactive (physically carrying it out), moving through iconic (typified by diagrams) and finally becoming symbolic (including linguistic descriptions of the problem).  To Bruner, the point of physical learning seems to have been the idea that it is required to form the understanding of the concept.  On the simplest level, you can't really learn the word "biscuit" if you've no concept of baked goods.

This ties neatly to the work of one of Bruner's contemporaries: David Ausubel.  Ausubel proposed something called an "advance organiser".  According to Ausubel, the main thing was to prime the student to receive new information, and much of that was about showing why something wasn't new at all.

My dad taught chemistry, and he was big on advance organisers.

But an advance organiser doesn't have to be physical, like discovery learners think -- it can be conceptual.  He taught the wave equation (velocity = frequency * wavelength) by analogy to a factory conveyor belt.  Each item on the belt was a wavefront, and the gap was a wavelength.  Most kids want to have frequency increase when wavelength increases, but the analogy makes it clear why this can't happen.  Needless to say, he didn't have physical access to a baked bean cannery to carry this out in, so he did it on the blackboard.  Under Bruner's structure, it is therefore iconic, and he's skipped the enactive phase entirely.

So what does this mean for the language learner?

Well, a suitable advance organiser can bypass the "enactive" discovery stage if we already have a suitable analogy at another level of abstraction.  The physical reality of a conveyor belt is so easily understood that my father only needed to evoke the idea -- the "advance organiser" for the wave equation.  With language, we can go one step further -- we already have an advance organiser in the symbolic domain: ie our native equivalent.

Let's look at an example.  In EFL we tend to talk about 1st, 2nd and 3rd conditionals and teach them by example.
1st: I will do it if you tell me to.
2nd: I would do it if you told me to.
3rd: I would have done it if you had told me to.

Many teachers avoid translation, as they see at as a dangerous thing.  But what if we stop saying "translation" and start saying "by analogy to your native language"...?

Michel Thomas teaches the conditionals in his Spanish course, and he does it entirely by analogy to English.  Even the 3rd conditional, often considered hopelessly difficult and very advanced, becomes simplicity itself, because the structure in both languages is almost 100% equivalent.

Another device he uses is when teaching "to wait".  In the Romance languages, this doesn't take the "for" of the English "waiting for" someone.  So he takes the word "await" and uses it as an advance organiser, saying that in French, Spanish and Italian, you "await" someone.  But he still says "wait for" too, because he is evoking both the meaning and the form.

One of the most striking things about Thomas's courses is how little of the material in them is specific to any situation or context.  Thomas taught only the most general and reusable language, and by playing with the structures, he gave his students an incredible level of control over the language.  When he demonstrated his techniques in an English high school for a TV documentary (The Language Master), one of the regular teaching staff had this to say:
The revelation is that it's the learning process itself that motivates these kids, the mastery of the stucture, the mastery of part of the language is the thing that keeps them going, keeps them enthusiastic.  And we lose sight of that in the way we teach. ... We think we capture their interest by finding them interesting materials that are supposedly related to their interests outside in the world generally, and maybe we miss the point.  And I think he's probably onto something very important here.
Which leads us back to where we started: learning is fun.

What triggered this post was actually getting a link to an article on computer games (of all things!) in the Guardian several weeks ago.  To quote:
our growing love of video games may actually have important things to tell us about our intrinsic desires and motivations.
Central to it all is a simple theory – that games are fun because they teach us interesting things and they do it in a way that our brains prefer – through systems and puzzles. Five years ago, Raph Koster, the designer of seminal multiplayer fantasy games such as Ultima Online and Star Wars Galaxies wrote a fascinating book called A Theory of Fun for Game Design, in which he put forward the irresistibly catchy tenet that "with games, learning is the drug".
Games can sell themselves on superficial features like graphics, soundtracks and clever media campaigns, but in the long run, the fun in any game derives from the fact that learning stimulates the brain.

So while the experts in fun are telling us that it's the learning that matters, the experts in learning are trying to look elsewhere for fun....

20 May 2011

"Say what sounds right."

Bad advice has an annoying habit of sounding like good advice, and this little phrase really is something of a wolf in sheep's clothing.  It's definitely appealling when someone points out that that's what we do in our own language.

I wouldn't argue that my end-goal isn't to be able to simply say what sounds right, but I just can't see how that end-goal affects my learning path: nothing will never "sound right" until I've learnt it, so how can I learn by what "sounds right"?

The consequences of this are not insubstantial, because if I've not learned it yet, what's going to sound right to me?  What sounds right is something that I have learned, but this will be out of context.

Take for example the verb "start" in English.
You can "start something".
You can "start doing something".
You can "start with something".
You can "start by doing something".

If you learn only two of these, then only those two will "sound right".

Saying "what sounds right" traps you into what you know and stops you expanding your language.  What you need to do is stop and think, and use the appropriate form, even if it isn't familiar enough to "sound right" yet.

Here's another example.

French has a feature called "liaison" -- certain final consonants are silent but reappear when followed by a vowel, but only if the two words are tied together syntactically.

In the word "vous", the S is normal silent, but in the phrase "vous allez" (you go), it has a /z/ sound.
Now, the past participle of to go is "allé", which is pronounced identically to "allez", so when you ask "êtes-vous allé" (have you gone), if you go by what "sounds right", you might pronounce that /z/.  But in this question "vous" and "allé" are not syntactically bound together and liaison should not occur.

The idea of "what sounds right" reaches a very messy conclusion in Scottish Gaelic. A single syllable consisting of a schwa before a noun can be one of three things: "the", "his" or "her".  "his" always causes initial lenition (soft mutation of the first consonant) of the following noun, "her" never does.  As "the", this form can occur before masculine and feminine nouns in certain cases, and causes initial lenition, and only with certain letters.

Many teachers suggest that you learn noun gender by "what sounds right", by agreement with the article, but your ear will be exposed to the various case-inflected forms and possessives, so what sounds right might not be "the boy" at all, but "her boy" or "his boy".

13 May 2011

The problem with podcasts

I'll get to the point quickly.  The problem with (language learning) podcasts is quite simply that they are modelled on radio.

Podcasting started with a bunch of early technology adopters who wanted to play at being a DJ.  They produced programs (semi-)regularly playing music, talking about minority interests etc.  They built a successful little community, and everyone suddenly wanted to hang out with the cool kids.

The language profession has always looked for ways to employ new media to produce new ways of delivering classes, and suddenly they had something new.  Or so they thought.

I'll indulge myself with a bit of repetition: the problem with (language learning) podcasts is quite simply that they are modelled on radio.  Radio -- that's nothing new.  Language learning hasn't been popular on the radio since people called it "the wireless".  They found it didn't work very well, but moved onto TV.  However, there haven't been many major TV language projects since the 80s (except in the field of English teaching).  Again, it didn't really work.


Because radio and TV are effectively the same thing, under the hood.

They are casual media.  You tune in, you tune out, as and when you like.  Programs have to be clever to avoid this: soaps are popular with producers because they retain their audiences through extended story arcs.  A more profound effect is that professional musicians are losing favour in TV prime-time to amateurs -- because a talent show has winners and losers, and with a knock-out format you come back to root for your favourite, even if you don't like the other guys.

Dedication to a single program, to tune in week after week, isn't easy, which is why most TV shows fail.  Which shouldn't be a problem for a language course, because people doing it are going to dedicate the time to it, right?

But holding your viewers isn't worth anything if you can't pick up new ones.  About 10 years ago there was a sci-fi series called Farscape.  It was very well written, and as a book it would have been an outstanding success.  But the problem was that it kept building on itself, so it was difficult for new viewers to understand, or for people to come back to viewing it after a break.  Though the series had a strong following, the few viewers it did lose were never replaced, and it got cancelled before it reached the final series.

And this is the problem for the language program on TV or radio: by its very nature it cannot build an audience: you have to watch from the start.

So why is this a problem for podcasts?  After all, we can download old podcasts -- we're not just restricted to this week's....

True, but once the program has been written as though it were a weekly radio program, there's a certain cognitive dissonance with playing catchup.

Every podcast has its jingles, it's little chatty banter, and all too often they end with "see you next week" or something of that ilk.  I for one find it unpleasant to listen to that when I'm doing an episode every day.  In fact, I find a lot of the banter irritating if I'm only listening to one episode a week.  The banter is part of the radio style, it's not alway there for pedagogical reasons.

So I have to ask myself: why write a language course modelled on radio programs, when instead you could write a language course modelled on language courses?  It's not like there's a dearth of materials out there to model yourself on: Linguaphone, Pimsleur, Michel Thomas and several other successful products exist on the market, and among them, some have survived a notably long time, unlike the short-lived fad for radio programmes.

Many of these don't have jingles (unlike the irritating audio accompanying many school textbooks), none of them are interspersed with chatty banter and very few have inane encouragement (phrases like "keep it up, you're really doing well" in a patronising voice really wind me up).  Lessons usually start with a lesson number and a title, and nothing else.  You may be given advice on the recommended frequency of use, but it's not rammed down your throat by someone gushing "see you next week! Cheeeeriooooo!"

So please, people; write language lessons, not radio programs

One of the most popular figures in language methodology is a certain Stephen Krashen.  People love him.  People quote him.  People even refer to what he does as "research", but Krashen himself has been involved in very little research, and most of the papers he writes cite papers which refer to other papers.  And many of the papers he cites were written by him.

Krashen's view of teaching is massively oversimplistic, and unfortunately that appeals to people.

I've had a pop at Krashen in the past (in the latter part of my post Expository vs Naturalistic language), and I'm not the only one.

The journalist Jill Stewart got stuck into him over a dozen years ago about bilingual education in her Los Angeles Times article Krashen Burn.  In it, she attacked his views on the education of Spanish speakers in the USA as being not only contrary to the evidence from teaching practice, but also diametrically opposed to the principles he professes for adult second language acquisition, even though he suggests adults should learn "like children".

In fact, Krashen's theories are so all-pervasive that Timothy Mason, when working as an English teacher trainer in the Université de Versailles St Quentin, dedicated most of a semester-long degree-level course to deconstructing Krashen's claims and rebutting them with references to real research and other academic opinions.  And he's now put the lectures on-line for all to enjoy. [Edit 2015-07-30: the pages appear to have disappeared from the site, but are available via archive.org's wayback machine.]

As I said, Krashen's theories are popular because their apparent simplicity appeals to both the teacher and the learner.  But more insidiously, Krashen claims there is really no such thing as "learning" a language, saying instead that you "acquire" it.  Now logically, if there's no such thing as "learning", then there can be no such thing as "teaching".  This is particularly appealing for the teacher, because if there's no such thing as "teaching", there is no such thing as "bad teaching"; instead, we have the idea that the teacher gives the student the opportunity to acquire the language, and can't really be blamed if the student doesn't take it.

(Actually, can we still call the language learner a "student"...?  Or even a "language learner"...?  If there is no learning, there is no study, so surely the appropriate word is "acquirer"?  This may seem like a simple game of semantics, but my understanding of the word "student" is someone who actually works at learning.  By continuing to use the term "student", Krashen's followers risk unconsciously passing all the blame for failed learning to the students.)

I understand all this, yet I am still baffled as to how the gaping flaws in Krashen's reason are still so hard to point out to people.

Krashen says we don't learn by production, we learn by listening and understanding.  Yet it is self-evident that this simply isn't enough in the real world.  Think of any immigrants you know.  All over the UK we have had wave after wave of immigration, particularly since the second world war.  There are loads of people who have lived here since the 70s who still haven't "acquired" English to a native-like level, with continued native-language interference.

As Timothy Mason points out, choosing not to correct mistakes "may be seen as a perfectly rational judgement on the part of the learner, who decides that any further investment in perfecting his grasp of the L2 will not pay sufficient dividends in added communicative and social power."

But it's more than this.  Certain errors cost more to fix at a later date than others.  I addressed one of these a few weeks ago: the falling together of phonemes.

Using the same example as in my previous post, we can predict that a French beginner of English will have difficulty distinguishing T from unvoiced TH, and D from voiced TH.  We can intervene and make them see the distinction, but only through production.  There is simply too much redundancy in language to be able to force the student to need to discriminate.  Word pairs such as "this" and "diss" differ so much in usage that discriminating the phonemes is very rarely going to be required to comprehend the sentence.

And as I said in the earlier post, my French friend can pronounce all the phonemes of English (with a bit of an accent) and he can even hear them when he listens for them, but they are missing from his model of the language, and in order to learn to pronounce every word correctly, he would have to relearn all his vocabulary.  He can function perfectly well in English, so learning correctly would certainly not pay "sufficient dividends" compared to the time he would have to invest, so it is "perfectly rational" that he doesn't.

The same goes for many people's grammar.  Little errors like missing the word "to" from "need to" rarely confuse anyone, and so there is no impetus for correction.

Current thinking is that it is sufficient to focus on survival language and "getting the message across" without any concern for accuracy.  Accuracy, they tell us, is an advanced skill for advanced students, and isn't worth the effort for most people, who just want to have a nice holiday.

But accuracy must be an early focus or it will never be achieved.  Grammatical and phonological accuracy is cheap and easy to start with, and gets more difficult and expensive the later it's left. 

So we have to teach, and we have to teach accuracy, or the student will never achieve it.

10 May 2011

I've always been a bit concerned about learning other people's mistakes, but this never fully explained why I always feel a bit funny about talking to other learners.

Well, I think I've worked it out.

Last night, I was listening to a German woman talking Italian, and because I didn't have the Italian accent as a cue (she has a strong German accent), my brain didn't switch into Italian mode properly, and it felt like I was listening to Spanish-with-errors instead.


06 May 2011

Repeating to learn or learning to repeat?

Repetitio est mater studiorum, it has been said, or repetition is the mother of learning to put it in a way even I would understand.

I was on a train last week and dozed off.   As I woke up, I had a simple phrase stuck in my head: "repeating to learn or learning to repeat?"  It's a crystallisation of something that I've been thinking about for a while.

When I look at wordlists, I see the danger that you associate the words in them with the list rather than with the intended item.  (When I was in high school, I couldn't remember the names of any berries in French without recalling "strawberry = fraise" first and then continuing down the list from there -- everything above strawberry in the list I was given was a full-sized "fruit", so I must have mentally subdivided the list into two...)

When I look at recorded drills, I see the danger that repeated drilling of the same order leads to the same sort of "learning the list" as with wordlists.

But on top of this, there's something slightly different and slightly more insidious.

All too often, the tasks we set as teachers or that we are set as learners can be carried out without engaging in meaning or true language, which I term as working mechanically.  The result is that we can produce a correct utterance without actually meaning it.  In the immediate term, this gives an illusion of learning.  What I'd never really thought about was how learning tasks build on this mechanical ability.

In terms of language learning, these mechanical skills are a dead end because they rarely lead to fluency, so I didn't think you could build on them.  But then I started noticing how there is a certain skill to carrying out the mechanical tasks, so the tasks can become slightly more complicated, and the student learns to perform such tasks quicker and quicker.  In the long term, then, the illusion of progress is maintained, even as the student's linguistic competence improves very little.

So there's two things here: learning of scripts instead of learning language, and learning mechanical skills vs learning linguistic skills.  While it is important to remain conscious of the distinction, you can sum up both the erroneous strategies as "learning to repeat".

So teachers, when you set a task that focuses on repetition, ask yourself this: are your students repeating to learn, or merely learning to repeat?