Saturday, 10 November 2012

It's not easy and it takes time.

Of all the EFL quackery out there, perhaps the most commonly employed by advertisers is the notion of learning a language in a extremely short space of time using some fantastic new method.  This closely parallels weight loss and fitness advertising, and both of these play on the human tendency to want something but not want to have to work for it. People want the gains but they don't want the pain. Unfortunately the value of being able to do something difficult is that it is difficult to do. 

Advertisers get round this because "learn a language" or "become fluent" mean different things to different people and there isn't really a way to objectively measure this. For people who don't speak another language this might seem odd as the notion of "fluent" might be, to them, connected with the idea of a native speaker, but it's not that simple. Non-native speakers rarely reach levels comparable with native speakers and it's questionable if that is even a desirable target for many. Like most things in language learning, when you scratch the surface you find things are a bit more complicated.
The Guardian recently had an article with the headline "How I learned a language in 22 hours" which uses this same fuzziness in order to get away with a very misleading headline. The author later adds several caveats:
It goes without saying that memorising the 1,000 most common words in Lingala, French or Chinese is not going to make anyone a fluent speaker. That would have been an unrealistic goal. But it turns out to be just enough vocabulary to let you hit the ground running once you're authentically immersed in a language. And, more importantly, that basic vocabulary gives you a scaffolding to which you can attach other words as you hear them

This though, is quite different from the claims made earlier in the article:
When I asked Ed if he thought it would be possible to learn an entire language in such a minuscule amount of time using Memrise, his response was matter-of-fact: "It'll be a cinch."

The article also claims that the learning happened in just 22 hours with the headline "How I learned a language in 22 hours". What we're talking about here though is the time Foer spent studying the words. But as he notes in the article.  
Cognitive scientists have known for more than a century that the best way to secure memories for the long term is to impart them in repeated sessions, distributed across time, with other material interleaved in between.
So Foer tells us how important the time distribution is (learning goes on when we're not studying, for example, when we're sleeping) but yet implies that this process only took, in total, one day. I suppose the headline "I learnt some basic words in a foreign language over a period of three months", wouldn't have made such a good headline.
The piece is ostensibly a huge advert for an app called memrise and to some extent Foer's books.  I download the app and was pleasantly surprised to find that memrise was free and had no ads. It's quite a fun app too, but it's quite limited in what it can do. For instance you can't check your pronunciation listen to any of the language, nor can you practise writing characters or making sentences. It's basically just a app for memorising things.
Memory techniques like using mnemonics or the method of loci can help us to store information in our brains. Language learning however isn't just about learning vocabulary items and switching them between languages. Take this phrase in Japanese:
yoroshiku onegaishimasu
It is used, in some contexts, daily and it has no simple English translation. If you tried learning Japanese from English, you wouldn't learn this phrases because it doesn't exist in English. The reason I mention this is because there is an interesting part later in the article when Foer writes:
I told him, "Omona, nayoka Lingala malamu mingi te. Nasengeli kozala na mosalisi koloba Anglais" – "Look, I don't understand Lingala very well. I need to have a helper who speaks English."

Now I don't know this language but as it's not related to English, it would be very surprising if they also used "look" in this way. Any Lingala speakers out there please feel free to comment.
In order to speak a language successfully you need to be able to process what is being said to you almost instantaneously and be able to formulate an appropriate response in almost the same amount of time. You need to understandable and you need to understand the grammar and pragmatics of the language. It's not easy and it takes time certainly more than 22 hours. In a memorable article on "principles of instructed language learning" in which Rod Wllis lays out what we know for sure about language learning, the claim made with the most certainty is this:

If the only input students receive is in the context of a limited number of weekly lessons based on some course book, they are unlikely to achieve high levels of L2 proficiency

If you want to get in shape then exercise and eat less. If you want to learn a language then study it and practise a lot. There's no magic solution.

well the 22 hours article/advert continues to buzz round twitter. The makers must be very happy and memrise has whizzed to 3rd place in the educational app chart.  

Amazingly this isn't the first Guardian piece on Memrise, in fact they had one back in March another in January, another mention here and another plug here -all part of their memory series. Amazingly in an article just titled "what you like" a reader apparently felt the urge to write in and say how much he liked memrise in December 2011. Just fancy!

Machine Translation
Another article that caught my eye and also had some rather suspicious claims was this one on, forwarded by @ScottThornbury no less.  I won't dwell on this one too long suffice to say its headline "machine translator speaks in your own voice" is very misleading. firstly the speech recognition software, as the guy says "makes a lot of mistakes" -you can see a number of them on the screen as he's going along, and secondly and more importantly, it doesn't speak "in your voice" at all. I'm sitting here listening to this Chinese translation, and it sounds like a generic Chinese computer voice to me.  


  1. Love this website. I’ve really enjoyed the postings so far.
    I think you might be preaching to the choir on this one though. It'd be hard to persuade anyone (let alone language teachers) that you can learn a language so fast.
    As you say, the guy said he’d memorised 1000 words in 22 hours – so just words - and the 22 hours was divided into lots of very short learning periods. Setting out to learn a language by memorizing a dictionary sounds like a silly way to approach to the task (Where’s the context? What about collocation and syntax?) It does sound like he didn’t have any other options though. I think the fact that he was able to learn so many words is testament to the power of employing mnemonics and good recycling – oh and working smart rather than long, of course.
    Do you know how memrise orders the words? Something that bothers me about flash card apps is many seem to group things by class. So the learner might see a picture of an apple then an orange then a pear and a banana etc. Something we do know about learning a language is not to try to learn a bunch of similar vocabulary items at the same time. It may appear to be more organized, but it confuses. But probably memrise presents things randomly or maybe he started at A and worked through to Z if he was memorising a dictionary!
    I liked the article and think the thing I’ll take away from it is remember to raise the subject of mnemonics in class – and stress the importance of regular recycling. Maybe something I should also be taking away is the value of a large vocabulary.

    1. huge thanks for the comment. I find it hard enough to believe that anyone reads this blog let alone might "love" it.

      I've been playing with memrise and it seems the lists are made by users (not 100% sure though) and the order, as far as I can tell, is just random...or the order they were inputted in.

  2. hi i agree with vicky hollet that this is a great website to read and help counter some unsound thinking in elt.

    as a clarification of "flashcard apps" those with spaced repetition such as the open-sourced anki are quite good tools to memorise vocab. the memrise app may be another good tool by leveraging social media aspects and gaming techniques. though i don't know how successful using other people's generated mneumonics can be?

    recently i have been trying out the open-source learning with text, as one way to build up my vocab cards in anki.

    maybe u have already read the language log post on that tech demo by microsoft? -


    1. thank you Mura,

      Your point about other people's mneumonic is interesting but I think you can make you own, can't you? I'll check out those web sites.

  3. Great piece on this- shame it isn't in the Guardian alongside the article!
    I had a good look at the Memrise website and I felt that it was a good way of memorising things; the game element definitely helps. AS yo say, the techniques are nothing new, but setting it up as a Farmville type thing is, and it's a good idea.
    However, the Learn Chinese seemed to have much better 'mems' than many of the other sets of vocab, and I agree with Mura that other people's mems may not work for you.
    Then there are, of course, all the limitations you mention- it isn't really learning a language, though I agree it could be a good starting place. AS Vicki mentions, a large vocabulary definitely IS useful- though, again, you'd need to be selective. The first word in one of the Spanish sets is 'greeting'. When would a beginner ever need to say that?!

    1. thanks for the comment,

      I've been playing with it now. It does have some good features as you said, though in farmville the plants actually grow when you water them and you can see the image. But is the human mind always quite as regular? I'll reserve judgement as I'm still playing with it.

  4. It's all too easy to get taken in by magic claims but I'm airy sure that every teachers would agree that the real magic element in learning a language is time. If you spend enough time engaging with a language then you will learn it, of course the better the methodology and language learning techniques you use the more productive that time will be, but it still takes time.

    When learning Russian one of my techniques was to use a frequency dictionary and research each word I came across. I'd mark ones I'd come across and tick ones that I knew how to use. Over time I had to untick some words as they could be used in different ways. One of the nice features was that I picked up some very natural ways of speaking with features of hesitation and ways to stress language points.
    Of course introductions didn't appear near the start so (related to your last post on frequency and time spent on a word) perhaps the language learners frequently use is different to day to day usage, (beginners meet lots of new people and when they travel they have to introduce themselves more than your average person has to do.)

  5. Thanks Chris,

    I agree that time is important, but input volume matters too. Also distance. I learned Japanese in 4 years which I thought was impressve until I met a Chinese girl who did it in 1 year 5 months!

  6. By distance do you mean how similar the languages are? Good point about input volume as well as time.

    1. yeah. So Chinese wouldn't find Japanese too taxing and we shouldn't (cough cough) find French too difficult.

  7. mallingual, I think your article is very well put together and necessary for consumers to make their decisions regarding the app. I agree that its a tad limited, for example while learning how to spell Brad Pitt's name may help in recognizing cyrillic letters it will eventually be useless in the long haul. Though I think overall if the concept were applied slightly more academically the app would be an even greater tool in language learning.