Anybody would probably agree that it’s required to have listened to the language to be able to speak it. But for how long? Instead of other activities? Is it detrimental to read at the same time? Will the world explode if we at some point choose to read along and aloud?
As I have said, I would never listen to 2,000 hours of TV. Even after living in Japan for 5 years, and speaking it all the time for business, I found TV drama difficult to understand.
As for how long the silent period, in my case, very long. Two years at least for Russian, and still it is mostly silent, although I do talk a couple of times a week with Russian tutors at LingQ if I am not too busy. Korean, I do not intend to speak for many months, although I tried out our only Korean tutor once, and will try out anyone other tutor who shows up for Korean.
The issue is clearly not about what activities are harmful or cause the world to explode. The issue is where we choose to spend our time, since the time we have available for language learning is limited. I would not spend my chorusing, or repeating phrases, or reading out loud. It is not that these activities are harmful, I just think they don’t do much for me, or not as much as more reading and listening and vocabulary acquisition. I am not so concerned about prosody, since if I listen to enough of the language, when I do start speaking I am quite confident that my pronunciation will be adequate and continue to improve. I prefer to spend my time on interesting content and word and phrase acquisition. I feel I get a better grasp on more of the language that way.
Talking of spending thousands of hours doing activities, here’s a quick rant I want to get off my chest:
I had a quick look at All Japanese All The Time, and the author recommends you listen to 10,000 hours of your target language. My first thought was “!Aack! Surely that’s a typo!” Then I thought, well, after spending 10,000 hours of my life doing ANYTHING I would expect to have become a Supreme Grand Master at it. Isn’t any method that tells you to spend that long doing something fairly fatuous? Like saying “If you spend 10 000 hours cleaning your house, you’ll have a REALLY clean house?”
He also recommends learning to write 10,000 authentic sentences in your target language. This, by my calculations, is equivalent to learning to to write out two entire novels. Doubtless you would learn much from the exercise, but would it be an effective use of your time?
Or maybe I should go to bed now…
I am amazed that his method involves so much work, yet he does have a loyal following. I prefer focusing on content of interest and learning words and phrases as I go. An hour a day, occasionally more and occasionally less is all the time I have. I admit that I do not have the discipline to follow his method, I also have other interests, believe it or not.
As for the 10,000 hours of listening at AJATT, based on what I recall from reading through the site, he recommends listening basically all the time. Have dramas playing in the background while you’re doing other things. Listen to music in the target language. Have material playing while you sleep. If you did all that, you could listen for 100 hours a week and then be finished the 10,000 hours in 2 years. IIRC, his basic point was just to encourage people to listen more or less all the time - it is All Japanese All The Time after all:-)
I think the 10,000 sentences was related to SRS. He was saying you should keep adding sentences to your SRS, and then you’ll need to get up to 10,000. He had this “sentence-mining” approach where you scour material for sentences to add to your SRS. If you did that, you could probably add dozens of sentences a day and hit the 10,000 before too long.
Even if you added dozens of sentences a day, you would need some time to review them. You may even need to review them more than once in order to be able to use them. I think this would be quite time consuming, although probably quite effective.
As I said, it is not something that I would have the discipline, time or inclination to do, but it probably works.
Yes, I used to try to add sentences to my SRS, but I just found it too time-consuming. Especially because I had to type them in. If you used online resources and were able to copy and paste sentences, it might not be too bad. After a while I just started to wonder if I was really making the best use of my time - especially as I found I was often trying to keep up with my SRS rather than reading and listening.
The other good point the AJATT site made about flashcards was that they should all be in the target language. I think he collected sentences from wherever, but the answer portion was the definition/example as copied from an online Japanese dictionary. I’m becoming less of a fan of the “Japanese sentence / equivalent English sentence” cards I’ve been using. But I think that, like you, I don’t have the time, patience, or inclination to build up my own deck in that fashion.
Well, I’ve thought about AJATT’s 10,000 sentences meme and decided to create LingQs of a handful of sentences per lesson, with their phonetic rendering and their meaning in the hint box.
I’m not listening to Japanese while I’m asleep through. My ears need down time.
@skyblueteapot,
That’s a really good idea 
There’s no reason why one couldn’t do the 10,000 sentences method within the context of LingQ. I sometimes link entire sentences in French, but only the ones I don’t really understand. AJATT was also about including sentences that you were thought were interesting, or that you’d like to be able to say, not just ones you couldn’t understand.
And when we link words here, there’s no reason we can’t include target language definitions of the word - that would be very close to AJATT style vocabulary cards because we’d have the word, the surrounding text (although not a complete sentence), and a target language definition.
I don’t know which SRS program Khatzumoto is using, but if you’re going to add thousands of sentences, and want to review them according to the spaced repetition idea, the flashcard system here at LingQ just isn’t as good as (for instance) Anki.
I haven’t yet got the import function to work here, while Anki allows me to import as many entries I want in one go, from a tab separated txt.file (other formats are also supported). You can have manually added entries (for Chinese at least) translated automatically. How about that!
So, if AJATT/10 000 sentences is something you want to try out, the tools provided here are probably not the best.
I’ve got the import function to work on LingQ! Mostly. Copy and paste into a word document, then convert to a CSV file. YOu have to watch that you get the right encoding and commas in the input sentences can cause it to go pear-shaped.
I might try ANKI, but I don’t really like flashcards. I review vocabulary a page at a time in LingQ. Sentences work just fine there.
Khatzumoto was using Mnemosyne (IIRC) and then switched to the SRS that he developed. I think he provides some basic sentences to get people started, so they see what cards should look like.
I’ve come to the conclusion that I’m not interested in building my own deck, however I’m open to pre-made decks that illustrate the usage of vocabulary and grammar. Hopefully, in the future, decks that like will be professionally prepared and entirely in the target language (at various levels of difficulty). I like the Core 6,000 deck and the 8555 (I think) Sentences decks available for Anki although I wish they were entirely in the target language. I also wish I’d discovered them a long time ago. Standardised decks are much more useful in the early stages - I spend a lot of time just deleting things I already know.
But I’m also thinking of just abandoning my SRS and following Steve with his Random Repetition System.