Xmmm wrote:Imagine two identical twins who decide to learn the same language but won't be able to practice together because they're kept in solitary confinement and subjected to random electrical shocks for the duration of the experiment. Both finish the duolingo course in 90 days, then they go in separate directions:
1. Twin #1 does anki 30 minutes a day. He creates his own cards.
2. Twin #2 reads simple news articles, childrens books, graded readers and watches dubbed shows including cartoons. In addition he follows bands and tries to decipher song lyrics when he finds stuff he likes.
I think that both twins HATE language learning in any form after 90 days with Duolingo (and random electrical shocks). But let's assume they have survived and still want to continue the experiment. Then the question is what the twins have learnt during that period. If they can read news broadcasts and understand newspaper articles then their passive skills are at a level where they both are ready to do extensive activities, and both would profit from it. If not, then they most likely reason would be that they know too few words and the logical solution would be to do some Anki or wordlists or any other kind of focused vocabulary training you might come up with.
There is one point I would like to stress, namely that you can't just throw all words into one pot. I would not recommend SRS or wordlists for learning grammar words. The most common words are often irregular, and they are used in complicated constructions. Here the best way would be a combination of some kind of grammar study alongside with intensive text studies. I know that some learners hate grammar and pin their faith on discovering the rules along the way - like small children allegedly do. But hey, do small children do that? I guess their parents teach them the basics by repeating sentences endlessly and weeding out errors - and that's also a kind of grammar course. Actually I can't see the point in inventing the rules from scratch when you can cheat and look in a key. To my best knowledge there has NEVER been published any proof that pure discovery from scratch functions better than a combination of cheating and critical evalution.
Then there is a grey zone, where the biggest problem is that the words in the zone are used in unpredictable idioms. You can study them in a structured way, but in the long run I see extensive reading and listening as the key to learning to use the idioms idiomatically - and that includes avoiding those that hardly ever are used. Let me guess... may a couple of thousand words are used in idioms. But personally I only remember idioms if I know what they mean literally, and that means that I still must know the words they contain - so even at this level both Anki and wordlists are definitely relevant.
And the last group, which contains many thousand words... well, I like doing wordlists from a dictionary because I can run through the whole alphabet and decide which words I want to learn. And I do want also to learn supposedly rare words - I have some slightly nerdish hobbies, and for me it is not irrelevant to know words for scientific or cultural oddities. But I also know from experience that it is possible to get through more words in a given time if you don't have to wait for each of them to pop up in something you read or hear. With my weakest and newest languages I mostly do wordlists with words from my reading, but the balance is later changed towards words from dictionaries, simply because I get more from working that way.
And what then about twin two? Actually the real problem with both Anki and wordlists is that they boost your passive vocabulary, and those who try to avoid them may do so because they are more interested in getting a limited, but active vocabulary, and that's definitely a valid reason. But then twin 2 also has to do more than trying to decipher simple news articles, childrens books and graded readers and watch dubbed shows including cartoons.