The persistence of origin myths

It's quite hard to challenge the orthodoxy in most fields, and language is no exception. This only seems fair - you need a lot of evidence to disprove an existing theory... or do you? I suppose that depends on what evidence there is for the existing orthodoxy.

The orthodoxy of language origin theories is often distinctly lacking in evidence, as in many cases it is tied into notions of ethnic origin, most of which are being proven wrong even as we speak.

A great many national ethnic origin myths are based on the idea of conquest as an annihilation of the existing population. For example, take the Anglo-Saxon invasion of England and south-east Scotland. Received wisdom is that the Norsemen started invading the territory roughly corresponding to modern Denmark, and started killing and displacing the locals, who fled across the North Sea and started killing and displacing the local Celtic Britons on the east coast of Great Britain.

That gives us the creation myth of England, that was built on hearsay.

I mean... did the Romans kill and displace the locals? No, they ruled over them. We were asked to believe they were the exception.
But did the Greeks kill and displace the locals? No, they ruled over them. Another exception, perhaps? The Moguls in the Indian subcontinent? The Mongols?
And more recently the Ottoman Empire? Or indeed the French or British Empires?

The fact of the matter is that the overwhelming majority of well-documented "invasions" in history have simply involved the installation of a new elite over the existing population. The genocide perpetrated in the Americas and Australia is actually pretty unique in history.

So why are we expected to assume that all the poorly-documented invasions are so different from the documented ones? Why shouldn't the Anglo-Saxon invasion be more like the Romans?

Well, thanks to DNA, that old orthodoxy has been turned on its head. Genetic testing suggests that there's a heck of a lot of Celtic ancestry where the old kill-them-all theory of historical invasion would have left us with only Anglo-Saxon blood.  On top of this, in the old Anglo-Saxon homeland around Denmark, they still have predominantly Anglo-Saxon DNA, where we were supposed to believe that the Norsemen had killed or exiled all the Anglo-Saxons.

But we shouldn't have needed DNA evidence. We know that the Anglo-Saxon kings of England claimed descent from King Arthur, a figure not from Anglo-Saxon mythology, but from British Celtic mythology. This alone should have proven conclusively that a huge portion of the English population was drawn from Celtic stock. Not to mention that the Danes have never looked all that similar to the Norwegians and Swedish. The Norse Vikings that invaded the west of Scotland were towering red-heads, while the Danish Vikings that invaded Northumberland were normal height and had dark hair.

This naturally has direct consequences in the origin myths of languages.

The best documented case of language birth brings us once again back to the Romans.

We know a heck of a lot about Classical Latin, but we know very little about the colloquial speech of the common man in ancient Rome. We know it was different, but we don't know how much. We also know that when they went out and conquered other nations, the locals tried to learn Latin, but did so imperfectly. We know that these imperfect Latins eventually developed into the Romance language family we know today (French, Italian, Spanish etc). Crucially, while we know very little about the early stages of the Romance languages, this is because Latin remained the language of the elite, so all literature was in Latin.

This means that the single best-attested example of language evolution revolves around an elite language hiding changes in the common language.

We can compare this to creole languages. Even today, many creole-speaking countries are run by elites that prefer to speak in French or English, and resist efforts to raise the profile of the local creole language.

All our most reliable data on language birth comes from languages where a large population have imperfectly learned the language of the elite, and where the elite language and the common language have existed in parallel for a significant period of time. It stands to reason that this should be our default assumption for all languages.

And yet our standard model of language development is still based on time periods: Old followed by Middle followed by Modern.

Given all that, it came as no surprise to me that someone recently published a paper proposing that a lot of so-called "Middle English" was actually contemporaneous with "Old English", and that the loss of grammatical complexity wasn't due to the Norman invasion (the Normans were a tiny minority elite ruling over a huge population) but rather the result of the Britons imperfectly learning the Anglo-Saxon of their rulers. The reason he gave for the apparent abrupt change was quite simple: the Anglo-Saxon elite were disenfranchised and Norman French became the language of literature. Middle English literature was not written by the descendants of the old Anglo-Saxon elite, but by the descendants of the peasantry, or people of Norman descent who learned it from the peasantry.

Shouldn't that have been our default assumption all along...?

Language learning professionals vs Professional language learners

The internet, they say, "democratises" human activity. We no longer need to go to the ivory towers of academia to learn; we no longer need experts as intermediaries: we can collaborate with one another.

This is true, certainly, but with it comes a certain set of dangers.

First, the "wisdom of the crowds" is generally fallacious, and we either get a mass of people who constantly contradict each other flat-out or we get little cliques that share and reinforce each other's views, to the exclusion of all new information.

More importantly, though, people want to defer to experts. This means that it's actually quite easy for someone with the right patter to set themself up as a "lay expert". Once they do so, they gather a little clique of the "wisdom of the crowds" type who will support and propound the self-appointed expert's proclamations.

There are many such "experts" in language learning. Typically they say they teach not on "dry, academic grounds" or the "received wisdom of the establishment", but "from experience". Their argument is simple and appealing: I have learned a language, therefore I know how to learn a language. But wait... haven't millions upon millions of people learned a language too? Why you and not them?

I call these people "professional language learners". Wouldn't we all like to learn languages as a job? Wouldn't that be great? I know I'd love it... except that's only of benefit to me, so really there's no reason for anyone else to pay me to do it. I find it difficult to stomach that there are people out there who make their entire living by writing and making videos about their language learning, and kidding themselves and their audiences on that they're giving some immensely valuable and unique insight into the learning process.

But they're not.

Their advice is at best vague, and very often even inconsistent and self-contradictory (eg Sid Efromovich's video that I discussed recently). Vague advice can be followed to the letter, and still have you doing something almost entirely the opposite of what was intended. As advice it's at best useless, at worst detrimental. Why am I failing? What am I doing wrong? Frustration sets in. Maybe I'm just no good at languages.

But why is the advice vague? Is it a fault in the author's use of English? His writing composition? Maybe, but mostly it seems to me that these people don't actually fully understand their own process. There is much to be learned from these people, but only if they're willing to discuss it, so that we can help each other tease out the details.

This is why we need to defer to language learning professionals, people who have trained, and studied, and (hopefully) taught. But most importantly, they are in a position to experiment. They can try something on one class, identify the weaknesses, then try it on another class in an adjusted form. Did it work better? Then it's better. A professional language learner only has a sample of one, and therefore cannot identify the changes that make things better.

That is not to say that all language learning professionals are always correct; sadly, language teaching standards are pretty poor at present. Many teachers and academics continue to parrot outdated and/or unproven theories as gospel, but if they can express their views more clearly, then at least you'll be better able to follow them if you choose to.

Of course, a lot of language teachers aren't really experts anyway. Most professional teachers of English as a foreign language have a four week certificate that is essentially a walk-through of typical classroom techniques, and no real in-depth analysis of what works, when it works, why it works or how. Simply being a teacher does not make you an expert, and very few teachers would ever try to claim otherwise. A real expert is someone who has dedicated multiple years of their life to both academic study of the topic and real life application.

It takes a lot of time and a lot of money to become an expert, and for those of us who are going through all the slog of trying to become genuine experts, it's kind of galling to see these guys walking the easy route and getting pretty handsomely rewarded for it. 

Video commentary: 5 techniques to speak any language

Sometimes it's nice to run with a theme, and although I'm mixed it up with a few unrelated posts in between, I started discussing what we can learn from learners a couple of weeks ago, and followed up with a discussion on the limits of my own self-awareness. I figure it makes sense to build on this theme and have a critical look at what others say.

I came across the following video last month, where Sid Efromovich presents his "5 techniques to speak any language" at TEDx UpperEastSide. You can watch the video (~15 min) now to make up your own mind, or skip straight down to my analysis.



Now, the first thing to note is that the video blurb says that Sid grew up in Brazil, which tells us that either he is an exceptionally good learner (I never noticed any flaws in his English) or that he was a childhood bilingual -- unfortunately not even his personal website tells us which. Not knowing his starting point makes it quite difficult to evaluate how suitable he is as a model for any individual (if he's natively bilingual in Portuguese and English, that wouldn't be particularly unusual, but for those of us who were brought up monolingually, it doesn't match our world).

Now lets move on to the actual talk, and Sid's rules.

1: Make mistakes

My first reaction when he gave the title was "here we go again," as I expected the usual line about how we learn from our mistakes. Yes, we are more than capable from learning from our mistakes, from which the hardliners conclude that there is no learning without mistakes... and yet the things I've learned best were learned right first time.

Thankfully, that's not what Eric goes on to discuss. Instead, he talks about how we all have a "database" of sounds and structures that our brains identify as correct, and that everything outside that database is flagged as "wrong" by the brain.

This throws us into a little paradox where our fear of making mistakes causes us to make mistakes.

His example is the Spanish letter R, which he gives with the word "Puerta". The Spanish R is markedly different from the English R, but close enough that a beginner's brain will try to replace the wrong-sounding Spanish R with a correct-sounding English one, which Eric refers to as the "closest relative sound".


In this rule we have something of real value to the learner. I've discussed the difficulty of dealing with foreign phoneme maps many times, but I never made that last logical step and talked about perceived wrongness and correctness, so a big thank you to Eric -- this idea alone made the video worth watching for me.

Caveat emptor

However, in that same example of puerta, Eric starts to show the limitations of his language awareness, as he appears to be claiming that it is only the Spanish R that isn't in the English "database", which is simply untrue.

The P, though close, is subtly different, as Spanish is a language more typically distinguished by aspiration, and English by voicing (in the pairs P/B, T/D, C/G). It's a very subtle difference, but a difference nonetheless.

The Spanish UE doesn't exist in English, and any attempt at approximating it is going to be wrong -- one typical English pronunciation of Puerto Rico is /ˌpwɛərtə ˈriːkoʊ/, whereas the Spanish is /pʷeɾto ˈriko/.

Finally, there's that last syllable ta. This has a clear vowel, because all Spanish syllables do... but not all English ones. It is exceptionally difficult for an English speaker to pronounce a clear vowel in the syllable directly preceding or following the stressed syllable, because in English these are almost always reduced to schwa -- that feeble little "uh" sound.

So there's the first alarm bell... his pronunciation is good enough to show that he internally knows the difference, but he's not consciously aware of everything he does, which (as I keep saying) is the limitation of anyone who claims to speak from experience.

 

2 Scrap the foreign alphabet

Rule number 2 is a doozy. Eric reckons that the foreign alphabet "will give you wrong signals," at least for languages in the same script as your own, and he is correct. It is difficult for an English speaker to see the letter "I" as representing it's so called "cardinal" value (the Latin I) because that's not the sound it has in English.
So far, so logical. What is the alternative?

Some (not Eric) propose learning aurally first, ignoring writing completely. (I disagree, because there are differences that a beginner might not be able to hear in the spoken language, but they'll be seen in the written language.)

But Eric is in favour of writing down. Does he suggest learning the IPA? No, instead he proposes using your own pseudo-phonetics, based on the sounds of your own language.

His example for this whole section is the Brazilian currency: the Real. It's a great example, because it is immediately misleading, having as it does the same spelling as an English word. His suggested phoneticisation, though, is very troubling: hey ouch . In writing it this way, he's using the same "closest relative sounds" he warned against in rule 1, which is a total contradiction. A Brazilian Portuguese R is not like an English H: it's a guttural sound, more similar to the German/Scottish CH (Bach/loch)... but not even quite the same as that. The L isn't quite the w-glide of ow/ouch, and even if it was, you're going to have to find a different notation for it in different situations (in can occur after vowels that an English W wouldn't appear after). And there are other problems too, but that'll do for now.

The main thing is that having presented only two rules, we already have a contradiction and fundamental incompatibility, throwing into doubt Sid's credibility as an instructor.

But setting that aside for a moment and looking at this rule in isolation:

Avoiding the native spelling is by no means a necessary step in learning a language.

Myself, I have at times learned to pronounce a language from its written form before making any serious attempt to learn the language, and to me that's far more useful, because I'm able to look at dictionaries and phrasebooks to learn new language, whereas if I'm stuck with my own idiosyncratic phonetics, I won't have access to any external sources whatsoever. I say learn the alphabet, but learn it right. Even if idiosyncratic phonetics did work (and I don't think they do), the cost in terms of isolation from materials is simply too high.

3 Find a stickler

He reckons you need someone who'll correct you, and you probably do, but he doesn't address the serious limitations of this, or how to avoid them.

First of all, correction is ad hoc. You say what you want, you get a correction. Fine, but language and a system, and therefore effective correction has to be systematic. A lot of the corrections you get from a native speaker are going to be extremely different from what your mistake was, and therefore don't really correct the source of the error, just the superficial form.

Secondly, if you make an error, you may be misunderstood. If someone says "I will do it yesterday," do they mean "I will do it tomorrow" or "I did it yesterday"? Where an error is ambiguous, the correction given has as much chance of being totally wrong as being right.

But worst of all, sticklers often teach things that are outdated. For example, the English you and I as subject isn't current in most parts of the English speaking world, but a stickler will "correct" you and me even if says it himself. Similarly for there's three things. Or he might "correct" can I...? to may I...?

How do you avoid these things? How do you choose your stickler? Well, for the first one, you're not looking for a language buddy, you're looking for a teacher. For the second and third, you're looking for a good teacher. You need someone who is an expert not only in the language, but in showing others how to become experts.

4 Shower conversations

Eric then suggests talking to yourself, a technique lots of people use. For many of us, that language practice is in the form of a silent internal monologue. The acknowledged limitation with this is that it's not practising pronunciation.

Eric's idea of having your conversations in the shower gives you a safe area to practice pronunciation, but the bigger point he raises is that by having a conversation as opposed to a monologue, you're more likely to identify gaps in your knowledge.

I'm not sure that this is actually true, and as a result he glides by what should have been his most important point, and worthy of being a rule in and of itself: mind your gaps.

When you're speaking, it will be your natural tendency to avoid and skirt round gaps in your target language knowledge, and doing so is part of being a successful user of the language. However, it is all too easy to become so proficient at avoiding gaps that you stop looking for them, and stop learning new things. The really successful learners continually seek out gaps in their knowledge and plug them with new information. Obviously it's less embarrassing to identify those gaps when you're on your own than when you're in a conversation, so it's a good starting point, but on the other hand, the reality is that you cannot say anything to yourself that you don't expect, so you can't find as many gaps speaking to yourself as you can when speaking to others. If you can be bothered, you can carry a notebook to jot down any gaps you come across, but while I had such a notebook for years (for Spanish), I only ever noted down 2 or 3 things in it...

5 Buddy Formula

Sid's final rule is a "formula" for finding the best "language buddy" for practise. Now I object to the trite, twee abuse of the term "formula" here, because it's just a rule that he chooses to write with an equals sign:

Target language = best language in common

He raises a very good point in his justification for this: that clear communication is the motivation for learning a language, and that if you know you can get your message across more easily in another language, you're going to switch to that other language.

You will get no argument from me on that -- I often find myself banging into that wall, and I often caution against so-called "immersion" classes where the class has a common language anyway (eg. immersive Gaelic for English speakers) because it risks conditioning people to view the language as a barrier to communication instead of a means of communication.

But more than that, Sid offers no advice to dealing with the big problem that all learner-learner conversations carry: learner errors. If one English learner says pod-ae-to, and the other says pod-ah-to, they're both wrong, and there's no stimulus for correction. Or perhaps one of them gets it right, then the other gives a well-meaning "correction" that teaches them the wrong thing.
...unfortunately, no feedback on errors.

Overall

So what can we take away from Eric's advice? Even if his rules were all unarguably correct, they alone are not going to teach you a language -- in fact, they're pretty peripheral to the main learning process.

But his rules aren't correct. Rule 2 is a dogmatic assertion that glosses over a very complex issue.

Neither rule 3 or rule 5 is wrong per se, but neither is essential, and Sid's description is inadequate as advice for the learner, because it doesn't give any concrete advice on how to circumvent any of the problems a learner will face.

Rule 4 fails on similar, but slightly more interesting, grounds. Again, he has failed to really give enough advice on how to avoid potential pitfalls, but in rules 3 and 5, those were pitfalls I don't believe he'd really thought about himself. With rule 4, he does talk briefly about the pitfalls. He identified a problem, and identified a suitable solution. He has used that solution and it should be useful to many others. But there is nothing unique in the shower conversation that forces you to identify your language gaps. If his rule had been "Mind the gap", his shower conversation could have been given as one possible technique to address it, and then he would have been forced into providing a description of how to structure a shower conversation such that it addresses the goals of a "mind the gap" rule.

Is language like science...?

Quite often, when I talk about the rules of language, I find I get hit with the response "Language isn't like science!" When I talk about teaching language systematically, people say "Language isn't like science." When I talk about language in schools, I'm told it's destined to fail because "Language isn't like science."

Well, I'm in the middle of trying to sort through a lot of old stuff that had been stored in the loft, and I came across a piece of paper on which I had hastily scrawled the following:

Language isn't like science.

Why?

It's about choosing the rules, not knowing the rules.

Now this is not a statement of my belief; rather it's my attempt to understand the logic behind the statement, so for anyone other than myself to get my full meaning requires a bit more explanation.

The reason people say "language isn't like science" is because of their misconception of the nature of science. To them, science has been presented as a series of rules to be memorised. They have been conditioned to think that the end goal of science is to be able to regurgitate the rules on demand, because that's all that was required of them in school.

That is not science.

Science is the art of investigating natural phenomena and finding explanations and models for them. These explanations and models are mostly a combination and application of existing scientific rules, and sometimes of identifying and creating new rules.

Or to put it another way: science is not about the knowledge of rules, it's about the application of rules.

But would the same statement not hold for language too? Language is not about the knowledge of rules (we can all agree on that) but the application of rules, surely?

Science is very often taught badly, in that there is such a focus on the rules themselves that students never get the chance to integrate those rules into a working body of scientific knowledge. This leaves the student able to recall the rule or law by name, but not recall the rule when addressed with a problem that requires that rule in order to reach a solution. You cannot solve a useful scientific problem this way -- the only type of problem that can tell you explicitly which rules are required to solve it is a problem that has already been solved, and science is about creating new knowledge, not repeating the known ad nauseum.

A good course in science will instead train the student in identifying the characteristics of a problem domain and noticing patterns that relate to particular laws or rules: they will teach them how to select the appropriate rule for the given situation.

That, I contest, is the very same process we go through when we try to formulate an utterance. We have a bank of words and grammatical rules at our disposal, and we have to select the appropriate items from it to express the message that we want.

So language is a lot like science, and the objections typically raised against grammar teaching are systemic problems that also affect science teaching. It's a problem that the late, great Richard Feynman recounted in his memoir Surely You're Joking, Mr Feynman?, when he talks of his experience on sabbatical placement in Brazil. It's a problem that affects all education systems to a greater or lesser extent.

But the problem comes when reformers attempt to throw the baby out with the bathwater: "rules teaching has failed," they tell us, "so we need to do away with rules."

That, to me, is a ridiculous philosophy. How can you choose which rule to apply if you don't know what rules exist? How can you search for it if you don't know what it is?

Let's be clear, I do not have to be able to recite the present tense endings of regular -ARE verbs in Latin in order to usefully "know" the rule, but that doesn't mean I shouldn't be taught them. I initially learned Spanish, for example, by the explicit teaching of the endings, and the explicit teaching of rules like 2s = 3s+"s" and 3p=3s+"n" (NB: this is my notation, not the way I was taught the rule!), and not by memorising the list of conjugations or a table. But that was still explicit teaching. I did not learn by osmosis, I did not learn by exposure, I did not learn by magic. I was told what my range of choices was, then given sufficient opportunities to make those decisions that I eventually could make the decision subconsciously.

Learners and learning: "I wish I'd known then what I know now..."

These are the favourite words of many internet polyglots, and continuing on the theme of what we can learn from learners, they are one of the most important reasons why you should take any advice from someone who has "been there" with a pinch of salt.

The problem with this philosophy is simple: there's no guarantee that you would have been ready to learn that then. I'll use my own learning techniques as an example.

What I wish I'd known

I took French and Italian as high school subjects, and didn't learn all that much in either, considering the time cost. When I started learning Spanish and Gaelic, I was simultaneously studying English language at degree level with the Open University. When I moved on to degree-level Spanish, I was annoyed that none of the linguistic concepts I'd been taught in the (mandatory!) English module were used to speed up the learning process. Too many things were left undescribed and confusing to my fellow students, when a ready explanation was available.

The English course didn't just occupy itself with the traditional grammatical concepts of parts-of-speech and syntax, but addressed language in terms of multipleframeworks: Halliday's systemic functional linguistics and metafunctions, the idiom principle vs the open-choice principle (language as a series of set phrases vs a series of free grammatical choices), lexicogrammar, corpus linguistics and collocation, directness and indirectness, agency and affectedness... the list goes on.

Knowing each of these concepts allowed me to disambiguate or disentangle my own confusion about new word or phrase forms, and I know that I would not be as successful a language learner today without having undergone that course of study. (This of course is in direct contradiction to a significant number of language learners who flatly refuse to accept that conscious study has any value whatsoever.) Now I have often wished I had learnt all that earlier... but would I have been ready? If I had not already learnt (a little of) two foreign languages, would I have been as open to the instruction I was given? Or if I had not been actively engaged in learning two languages simultaneous to that study?

I cannot know, therefore I cannot say for sure, but I believe that it would have helped. I temper that with the knowledge that the course I took would be overkill for most learners, but that there are certain concepts that help immensely. (eg indirectness as a form of politeness; compare; "shut the door","can you shut the door" and "could you shut the door" -- the imperative is direct and rude, can is somewhat indirect, and could is very indirect, making it polite. This same rule about indirectness holds for most European languages, but not e.g. Japanese. It is very easy to draw a learner's attention to the presence or absence of this pattern, but many courses fail to even attempt this.)

So I would advise learners to learn this stuff (if I only knew of a good book!) but I would also caution them about trusting my advice blindly.

What I might have wished I'd done

When I bought myself a DVD player, it was because of language. I had taken a French level test with the OU before starting my study, and while my reading and writing were at a pretty good level, my listening level was disproportionately low (incidentally, this worked in my favour in the end, as otherwise I might have started with French rather than English) so I needed to improve my ear.

Fast forward 9 years, and I now find myself watching Gaelic TV or French films and noticing unusual turns-of-phrase on the fly, by comparing the soundtrack and the subtitles. This, as it turns out, is a highly effective way of learning the sort of stuff that doesn't make it into the textbooks.

So would I recommend it? Not really. 9 years is a long time, and it has taken me a lot of practice to get to the point where I can do this -- when I started, the subtitles would soak up my attention, and I wouldn't just "not understand" the audio track, I wouldn't hear it -- my mind blocked it out. It took me several years to stop blocking it out, and years after that to be able to consciously follow both languages, and even now I'll generally slip into either reading or listening -- it takes active concentration to do otherwise.

Now I could have forgotten all the time and effort and told myself that I obviously hadn't worked hard enough soon enough, and that if I had, I would have learned even better and quicker than I did, but I simply don't believe that's true. I don't think this technique, which is now one of the most useful in my arsenal, is actually worth the effort of learning to most people.

This serves as a useful reminder to me that everything I do now is a refinement and improvement of things I've been doing for years. These techniques can't just be "done", they have to be learned; which means that I can't just "tell" people to do them, I have to teach them... or keep my mouth shut.


So there we have another problem in learning from successful learners: all too often they will simply tell you what they do now, with no real conscious understanding of how hard it was for them to reach the point where they could.

A plug or a polemic?

Whenever a book on language is released, it is traditional for the author to try to get a column in a newspaper raising some pertinent language issue, then launching into what is little more than a sales pitch for his book, describing how it addresses the problem.

It was refreshing then, to see an article in the Guardian where an author decided that the best way to use his 15-minutes of fame was not to plug his wares, but instead to pull out a polemic on the state of linguistics in general.

Harry Ritchie started off with a pertinent language issue, as is tradition. His issue was the often overlooked problem that our culture of "talking properly" in schools is actively disengaging children from their learning. (Most language experts agree on this fact, by the way; it's just the political and school systems that reject this.) His book, as far as I can gather, is simply an attempt to explain real language patterns in a clear and engaging way (I have not read it, so cannot comment on the success or otherwise of this) and he could have continued in the typical manner by throwing up multiple examples drawn straight from his book, slowly pushing the reader into wanting to know more.

But he didn't. Instead he went tangential to the content of the book, delivering a finely-tuned polemic about the state of the linguistics world, and pinning the blame on the door of Noam Chomsky.

And he is right to do so.

I was twice introduced to the world of linguistics: once through the Introduction to Machine Language unit in the University of Edinburgh's artificial intelligence department, and once through the Open University module The English Language: Past, Present and Future.

Both courses were very good, which should be no surprise giving the academic stature of Edinburgh's AI department and the OU's English department. However, both started by looking at structural grammar through the view of Noam Chomsky's generative grammars.

Given the research and refinement in the half-century since Chomsky, it should be no surprise to find that his model is primitive and of little use, and yet it is still taught as the sina qua non of modern linguistics. I've written my own little piece about this before, so I'll not repeat myself too much, but I'll say this:

Chomsky wrote a model of grammar that was based almost entirely on reading language one-dimensionally and sequentially, a model that so badly fits real language that it allows the generation of meaninglessness such as colourless green ideas sleep furiously, and then extrapolated from that nonsense that meaning and grammaticality were entirely separate things.

He wrote a demonstrably broken model, then used the brokenness to try to draw conclusions about the real world. And people treat him like some kind of genius for  it...!

What can learners tell us about learning?

Last month, David Mansaray appeared on the How-to-learn-any-language forums discussing a new series of podcasts about language learning. His previous interviews focused exclusively on polyglots, but now he wants to expand that scope:

"I plan to interview polyglots, linguists, teachers, expats, successful students, interpreters, lexicographers, etc.."

It's definitely a good idea. There is a lot to be learned from polyglots, but sadly, there seems to be very little that can be learned from the internet polyglot community.

The problem is simple: most are happy to tell you what they do, but a great many are not happy to let you find out what they do. You can listen to what they say, but you can rarely probe or challenge it, which is a nuisance, because very few of us are ever fully aware of what we do.

Stepping away from the world of languages, consider this story I was told in my university days. There was a cheese factory near Edinburgh, and one of their employees specialised in determining when the cheese was ripe enough for packaging and selling. He was nearing retirement, and the company hoped to make a machine to do his job rather than training a replacement, so they called in experts on AI and machine learning to try to create a robot for the purpose. They asked the man how he determined when the cheese was ready, and he told them he prodded it with his thumb, and he knew if it was ready based on how springy it was. They set up a machine that bounced a little thumb-sized probe on the cheeses and measured their springiness. The problem was, there was absolutely no correlation between the measurable springiness of the cheese and the expert's judgement on whether it was ready or not. Eventually they discovered that in prodding the cheese, the man had been breaking the surface, which released a scent that he subconsciously detected. As computers can't smell (yet), the project was abandoned, and an apprentice hired.

Going back to language, for years I derided those who talked about "shadowing" other people (ie repeating audio books or films verbatim), thinking it was a passive process that wasn't truly "linguistic". I didn't shadow, anyway. But then I remembered that I used to shadow DVDs in Spanish. Not religiously, not obsessively, but from time to time. I did it, and I can't say it didn't help me -- it might have, it might not have. I can say I believe that it is of minor utility at best, but I have to be careful not to say what it "is" or "isn't".

If I can enter calmly into a discussion with other people, I can often be made to realise that I have fixated on one thing and completely ignored something else, and I can in turn make them realise that they have done the same thing. The end goal of such a discussion shouldn't be to walk away with a better understanding of each other's techniques but for each to walk away with a better understanding of their own. When we enter such a conversation with the view that we are correct and the goal of convincing everyone else that we are, we add no value.

Interviews can be a starting point, but only if the interviewee is engaged in discussion rather than merely lecturing those he sees as less informed.