@Tobberoth

Quote:Sure, people make mistakes, but there's no way someone is going to keep conjugating every single verb incorrectly all the time.

Twas simply to illustrate a point. The mistake is not corrected by simply being acknowledged. And if it is, it takes far longer than giving correct examples (input).

The mistakes people make vary depending on the person. "Say me" vs "Said to me" is clearly a conjugation error. I've heard this and many others like it be repeated by those who are learning in their teens and those who only started well into adulthood.

Quote:Yeah, it takes time and dedication and you will keep making the mistake when speaking quickly,but it's not going to happen that you hear you're doing it wrong and forget it because you're making mistakes in the next sentence and the next one again.

I'm not sure what you are saying here. Something about the "because." Could you restate that, please?

Quote:You're not talking about early output, you're talking about "learning words from a dictionary then attempting to say something".

I disagree, but say that is the case, how's this different from any other premature output? A person has very little grasp on what he's trying to use, and thus prone to making more errors. Without sufficient input the errors will continue.

Please excuse me if I'm being horribly naive, but I thought the gist of the input theory isn't that you should avoid output entirely, but that you should, as much as possible, only say things that you know how to say. Like, once you've been exposed to and understand some thing in some context, feel free to use it, but don't go applying it willy nilly where it doesn't apply. And if you do, make extra special effort to focus on the new input you receive when somebody that actually speaks the language corrects your fool self.

Of course, that's just my impression of what I've read.

My current thinking is that the best time to start controlled practice of output is at the intermediate stage. 'Intermediate' being an arbitrary distinction--once you've accumulated enough of the basic building blocks of the language that you don't feel like you're trying to operate with too little fuel, and are comfortable enough with self-study or 'cooperative' study that you can practice applying what you've learned with control (namely the ability to analyze and correct what you said before 'grading' your progress and moving forward).

Mcjon01 Wrote:Please excuse me if I'm being horribly naive, but I thought the gist of the input theory isn't that you should avoid output entirely, but that you should, as much as possible, only say things that you know how to say. Like, once you've been exposed to and understand some thing in some context, feel free to use it, but don't go applying it willy nilly where it doesn't apply. And if you do, make extra special effort to focus on the new input you receive when somebody that actually speaks the language corrects your fool self.

Of course, that's just my impression of what I've read.

You've read wrong! How could you gain such a logical, sensible idea from such a straitforward and dogmatic approach?

Dare ye output, dare ye fail --

Anti-Moon 3:16

Nestor Wrote:My current thinking is that the best time to start controlled practice of output is at the intermediate stage.

More sensibility? Yee haft anger'did Anti-Moon-Weh.

phauna Wrote:Listening to Japanese while you are mindfully doing something else is pointless, and definitely a distraction.

I humbly disagree. Having the television run all day did lots for my Japanese.

Evil_Dragon Wrote:

phauna Wrote:Listening to Japanese while you are mindfully doing something else is pointless, and definitely a distraction.

I humbly disagree. Having the television run all day did lots for my Japanese.

Me too. I think that--huh? what? Did someone... ? 日本語speakmasuka?

nonpoint Wrote:igordesu is one of the cultists you guys speak of. He seems nuts, just read his comments! Freaking CULTIST!!

Boo. Hoo. I just got called an extremist by some random person on some random internet forum. I guess that means it's all over for *me* now ~rolls eyes~....

By mindful I mean, if you are listening to Japanese while doing some computer programming, or maths problems, or something difficult, like Khatzu seems to do then that won't help at all. You can listen while you make your breakfast, though, or have a shower etc.

As for college classes, I don't live in America, this is the internet, if you can't find good education in your country then I don't see what that has to do with classes intrinsically being bad. Don't assume I'm talking about America, this is the internet.

As for polyglots, no they don't seem to take classes, perhaps they are a special case. However every single person that learned English and attained English fluency that I know has been to classes for a long time, and rates them as worthwhile. I don't know any English learners who haven't been in classes.

Many specific language mistakes that people do are based on first language interference. Koreans and Japanese are going to have a hard time with a/an, the and plurals no matter what, because there is really no equivalent in their languages. Not all language learners are made equal or start from the same point. Also laziness to correct mistakes does come into play once learners get to a certain level of communicative ability.

As for only outputting exact Japanese sentences that you saw in an SRS, I think that's impossible. You might as well just buy a phrase book and memorise that. Also, a key learning necessity is to actively use the language, ie make up your own sentences with effort. I don't think anyone can do that without mistakes. If you aren't striving to make those sentences that are a little above your level then your study time is not active.

Tobby,

ぐらい is technically not correct for points in time, but seems to be one of those "native mistakes" that I hear from my friends a lot. You too?

ゴローーーーー

my 2 cents

phauna Wrote:

ghinzdra Wrote:Anyway I m always extremely interested in critics and very open minded . So I m all ears.

Well in that case, no amount of input will allow you to easily and faultlessly output.

In some senses this is true, you need to practice the action of speaking to become good at it, but without a word outputting engine that is built up using input you will never be able to speak like a native. We can see this principle by looking at the differences between good and bad native speakers. Everyone can speak perfectly in terms of intonation and pron, however, those that have read a lot tend to be able to speak much more eloquently.

phauna Wrote:Listening to Japanese while you sleep will not help you at all, it may even hinder your sleep and therefore your retention.

Like Nukemarine says, this is mostly the time during which you go to sleep and when you wake up. In particular I think the time when you wake up is effective, as you are not chopping the sentences with your brain and analysing them, just letting them flow in. Also, whilst it is certainly arguable whether Japanese audio while you sleep has no effect, it is hard to say this with certainty. You can be woken up by noise, in particular if someone calls your name, so it seems logical to assume that what you hear is being interpreted on some level. Maybe it's worthless, but maybe not.

phauna Wrote:Taking classes in Japanese is a very good thing to do, as long as they are good classes.

What are good classes? I went to an often praised, sometimes talked about as the best Japanese school in Japan (YAMASA). It was a crap (teacher talks 90% of the time, students talk one each). If a school like this isn't good, where are the good classes? Certainly the ones I took in Australia were much worse which is saying something.

Even English classes, which generally are much better than those of other languages in that they are fun and communicative, do not take into account the input hypothesis and are highly grammar focused. On a number of occasions I saw choices made at the school I taught at to make more money at the expense of the students progress.

There may be some good Japanese classes somewhere in the world, but they are a needles in a barn of haystacks.

phauna Wrote:Listening to Japanese while you are mindfully doing something else is pointless, and definitely a distraction.

You are definitely analysing what you hear on some level, since if you hear your name you will notice. In particular I think the intonation will go in, much the way you can bop along to a song while doing something else. More importantly, there will be moments where your mind wanders and you pick up sentences here and there.

However, I think listening to Japanese while reading or writing English is not helpful. You can get your languages tangled up and in my case it gives me a headache.

phauna Wrote:Over-kanjifying your srs sentences will make your Japanese seem unnatural, even if it may help retention a little.

Does Katzu recommend this? I agree with you here. I think you should just use whatever the source you got it from had. You need to learn when to use kanji and when not and if you just kanjify everything then you lose this information.

phauna Wrote:Eighteen months is an unrealistic time frame for fluency, even for a European language.

Fluency is a vague concept, but I disagree. Just looking at how far I have come when I have had the time to do "AJATT" completely I think this is possible.

phauna Wrote:Making mistakes while outputting will not cause permanent damage to your Japanese.

It is hard to say definitely what will damage your L2, but it is undoubtable that some people who should be basically a native speaker are far from it. Why this is so is questionable.

For example I have read the emails of a 60 year old Japanese man who had been in America for 40 years and had forgotten Japanese. They were in English and there were a number of mistakes and unnatural sentences. It was immediately obvious that it wasn't native.

I personally think if you are using the input method, much like children do the shear weight of input you get will fix up any errors you are making. In this case mistakes are not a problem. However, if you try to learn a language using grammar and consistently make the same mistake over and over, even if you then get a whole lot of input you will keep making the same mistake without a lot of concerted effort.

phauna Wrote:By mindful I mean, if you are listening to Japanese while doing some computer programming, or maths problems, or something difficult, like Khatzu seems to do then that won't help at all. You can listen while you make your breakfast, though, or have a shower etc.

I find trying to listen to Japanese on headphones while working to be quite detrimental to my work efficiency. I imagine this all depends on your occupation - I have the same one as Khatzu so I don't know how he can do it. If you do something like turning wood or whatever I can't see it being detrimental - just be careful not to lose a finger.

I also for example find that my sentence SRS review time increases noticeably if I have a TV show running at the same time. That's a direct measurable decrease in active learning performance.

As for listening when going to bed, I have a hard enough time switching off and falling asleep without such a thing.

thermal Wrote:

phauna Wrote:Listening to Japanese while you sleep will not help you at all, it may even hinder your sleep and therefore your retention.

Like Nukemarine says, this is mostly the time during which you go to sleep and when you wake up. In particular I think the time when you wake up is effective, as you are not chopping the sentences with your brain and analysing them, just letting them flow in. Also, whilst it is certainly arguable whether Japanese audio while you sleep has no effect, it is hard to say this with certainty. You can be woken up by noise, in particular if someone calls your name, so it seems logical to assume that what you hear is being interpreted on some level. Maybe it's worthless, but maybe not.

Why not listen to audio while falling asleep, but not throughout the whole night? As long as the audio's not so interesting that you don't fall asleep til it's over(that's happened to me :p)

There's also this: http://forum.koohii.com/showthread.php?p...6#pid40446

vosmiura Wrote:As for listening when going to bed, I have a hard enough time switching off and falling asleep without such a thing.

Not falling asleep works me, I just need my mind to shut up and listen. That way I get at least 20 minutes of active listening before passing out.

nest0r Wrote:There's also this: http://forum.koohii.com/showthread.php?p...6#pid40446

This links to an article about a study done by a PhD grad in New Zealand. It was touted by Katz and celebrated by AJATT followers as evidence that listening to massive amounts of Japanese, even though incomprehensible, is the key to fluency because it builds the necessary neural tissues.

I looked into it at the time. I want to point out here that this is incorrect information. The study says nothing about listening to massive amounts of input, investigates nothing about neural tissue development and does not claim that aural input is the main ingredient in language proficiency. There is no new knowledge in this study.

The narrow point of the study is that we are often unable to hear sounds in a new language that are not found in our native language. Exposure to the new language makes us able to hear those new sounds, therefore it becomes easier to learn that language. The study says nothing about the effects of listening to the language once the ability to hear the new sounds is achieved. [If you reread the article with this interpretation in mind, I think it works - the headlines of articles about this study were often quite misleading]

The experiment involved non-Russian speakers recognizing a list of Russian vocabulary words. The basic idea is: one group was exposed to Russian before before being tested, but not the other. The group with exposure did better on those words which contained sounds not found in their native language. The study did not involve any brain imaging equipment, etc. I guess it's speculated the brain forms new neural connections when it learn to hear new sounds.

In fairness to his readers, Katzumoto might want to include a note of correction in his blog relating to this article. I kind of breaks my heart to think there are some teenagers out there dutifully attached to their headphones and shutting out their families and friends for months while they 'become Japanese'. This study shouldn't be used by them as evidence that this is necessary or even helpful beyond a certain point. (Immersion is helpful, but it involves more than passive listening.)

[In hindsight, I should have posted something there earlier. But, you know, sometimes it's easier to just stay out of it...=] Thanks for reminding me Nest0r]

Indeed. Well, my response was 'I'll wait and see', though I felt doubtful about the claims about neural tissues and feel there's some truth to it in regards to our ability to train our ears to sounds in a 'reduced listening' fashion. Until recently my main recommendations for listening/immersion would have been mostly for the initial stages (that is, outside the audio incorporated into Anki that I used to adhere to zealously before moderating it, re: dictation and repetition), but now w/ Nukemarine's thoughts I feel inspired to try moving my previous momentum-based planned redundancy via frequency lists in the SRS to a place outside Anki via audio splices. But I ramble.

I also thought the same thing when I first came across the article, Thora. But I also thought developing listening ability to recognize each phoneme and allophone could take much longer than the article suggests. I thought it might take an adult person a year or even more of constant exposure to acquire a decent listening skill for basic sounds depending on similarity between the target and his or her first language.

But come to think of it, Japanese learners may not need to do passive listening that much if they had already done a ton of listening when they started studying the exotic language. I always wondered why English speakers who are learning Japanese are usually better at listening than English learners in Japan. I even suspected the phonetics of the Japanese language could be a little simpler than that of English. But if most of the learners had already listened to Japanese for some time before ever starting language learning, then it's not surprising at all. Most of the English learners in Japan were just forced to learn the language and have listened to little English spoken by native speakers in real life.

It seems that passive/active listening improves your listening even if you already have a basic phoneme recognition skill because a spoken language has a lot more than just phonemes. And I do think passive listening is helpful to some extent, though it's never as effective as active listening. But it may not be as helpful as Katzumoto claims it to be.

Nestor, could I ask you to record all your posts and send me an audio file? I'll then splice and dice them into my SRS and listen to them all day and all night. Exactly how many thousand hours do you think it will take before it becomes comprehensible input?

Thora Wrote:Nestor, could I ask you to record all your posts and send me an audio file? I'll then splice and dice them into my SRS and listen to them all day and all night. Exactly how many thousand hours do you think it will take before it becomes comprehensible input?

It's terrible, I still can't be clear. Well, I do have a 50,000 word text file of deleted, unsent, or recent comments. I can throw it into Textaloud! ;p

Thanks for the critical examination Thora. Far too often people make causal statements without enough evidence to back it up, then people see that it references some study somewhere and instantly believe the causal statement without analyzing the study. In truth it's always hard to make a causal statement from one study due to confounding variables.

thermal Wrote:Also, whilst it is certainly arguable whether Japanese audio while you sleep has no effect, it is hard to say this with certainty. You can be woken up by noise, in particular if someone calls your name, so it seems logical to assume that what you hear is being interpreted on some level. Maybe it's worthless, but maybe not.

The science on that matter is very clear actually. While being promoted by many self-help books and websites, the notion that you can learn new information by listening to it while you sleep lacks any basis in scientific evidence. It's easy to design studies to test whether any retention of material exposed to during sleep takes place. Such studies have first been conducted by Charles W. Simon and William H. Emmons in 1956 and they showed no such effect and the notion has been pretty much dead since then. Later studies failed to show any measurable effect either and from all that is known about the brain that result isn't surprising.

It's as simple as that, while you sleep you can't "listen". Processing new information requires a high level of brain activity which just isn't there when you sleep. Waking up from loud noises makes sense (you wouldn't want to be eaten by other animals), but the leap from "noises can wake you up" to "the brain decodes foreign languages while you sleep and without you noticing any of it" just doesn't work.

Of course it would be awesome if we could learn something by... doing nothing. It's tempting. It's part of human nature to look for a ways to improve without having to do much or anything at all for it, sadly that's a fact that's happily exploited by all sorts of quacks. :/

That being said, it's also clear from the literature that memory is being consolidated during sleep and that enough sleep helps with the retention of newly learned information.

Audio at night keeps me awake. But I found that if I put on something fairly soothing for 20 minutes or so, I'll listen to it, and then go to sleep quite promptly when it's finished. This seems like an improvement, since I haven't usually been able to fall asleep in 25 minutes otherwise ;)

Maybe it would also be useful to play something more energetic in the morning, automatically from 20 minutes before I'm supposed to get up. I just need to figure out how to do that.

I don't need your science to tell me it working, I trust my gut, and my gut trusts Khutzumoto (says I'm hungry to).

Feel free to believe whatever you want oregum, I'm just putting the information out there for people who are interested in the issue and may not have as much faith in their gut as you do.

It's pseudo-study. Full stop. It honestly doesn't need to be done. I believe passive listening while awake helps. I say this because when I'm doing passive listening as well as other things, I find that from time to time I'll comprehend the background audio and go "oh snap! I understood that!". I think this is important because when you can comprehend something without actively listening to it, it means you're understanding of it is at a decent level.