AJATT vs Organic: Audio Immersion. Can you really learn like a child?

Both AJATT and the organic approach favor intense audio immersion. Which one is best for you? ▼See More ▼

Visit my Patreon community:

Red Kokeshi Angel Patrons who help this work to be possible:

Tayruh
Full Name
Mayka M
Valerie
Jean-Marc Giffin
Tasty Treeko
John Buhler
Edward Nicholes Jr.
Sergio O Parreiras
JuanjoPP
Tascha Keettel
Bibo
Steven Davis
Daniel Schulz
Liane Degville
Lola
Joe Wright
Rembot
Colin Jervis
Arzar
Kathy Worley
Mirnes Selimovic
Nico.Nico

Sincere thanks to all my patrons, supporters, students and fans

▼Please visit us at KawaJapa

[elmhurstenglish]

I feel the main difference between AJATT/MIA and Organic Japanese is the difference in emphasis on intuition vs logic. Largely, the AJATT/MIA relies on one's ability to intuit rules.

If I were to show someone a series of numbers with some underlying pattern enough times, they'd intuitively know what number came next even if they couldn't consciously express the rule itself - in this instance, they have to: 1. derive a subconscious understanding of the rule as well as 2. confirm the validity of it through repeated exposures.

However, someone could be shown the rule from the outset and then be shown the same series of numbers, and in this instance they'd only have to: 1. confirm the validity of it through repeated exposures.

The latter is far less taxing, but neither of the methods let you get away with not exposing yourself repeatedly to the source material itself (learning requires feedback, and the source material is the feedback). I'm surprised there's any contention with the latter being inherently superior (as it tackles the language from bottom-up and top-down simultaneously). It's no different than learning to drive a car: sure you could just get in the car and over time get used to how everything works and then be able to drive, but why give yourself the headache when you can just have instruction (top-down; conscious application; logic; model) and then *also* get used to it through driving as well (bottom-up; feedback; intuition; experience).

If there is an issue using top-down and bottom-up simultaneously, it's usually because the conscious model used in top-down is horribly inconsistent with the bottom-up experience. Despite that, this doesn't mean the *idea* of using both is harmful, it just means that the model you used needs to be changed. I find your work on here, Dolly, to provide the most accurate model of the experience thus far. Interestingly, the intuition approach is very Yin and its converse is very Yang - as is said, both are required in a perpetual and harmonic dance.

[akkimylo]

As someone who can to a certain extent learn with this method, I have to say that your video highlighted some key points on why certain people can or cannot do so very accurately! I've been watching some Japanese YouTubers playing a game I enjoy and my comprehension is laughably low, around 20 or 30 percent and I've noticed that I'm slowly learning words or even getting used to grammar points in the manner you described. It has been pretty enlightening to understand a little about how that may happen described in terms of linguistic intuition and tolerance of ambiguity, as I am lucky enough to have both to an extent that helps me learn. While this video isn't directly teaching something in japanese, I believe that it will help many people pick out the best study method for them in the chaos of learning resources and suggestions that are coming from everywhere. I wasn't expecting to watch this entire video but you had me hooked!

[GarrisonMorton]

3:35 "That isn't a ball" I giggled a bit there.

[martimf.]

動画を作ってくれてありがとうございました！
ここからもよろしくお願いします!

[yesudesu]

Doing both seems like the right choice

[tamusic]

Interesting video. It got me thinking about a few things.
1) My SO has low ambiguity tolerance, and I have a relatively high ambiguity tolerance. This definitely matters in how the two of us learn languages (and it also affects how well we tolerate not being able to fully understand others. I'm *much* less troubled by that than my SO).

2) As for learning as a child, in addition to what (correctly, I believe) was pointed out in the video, a child has someone there to tell what everything around her is called, nearly all of the time. Parents, other children.. will an adult delving into AJATT have that? I think not. And then there's the very real issue of the enormous amount of time invested, as Time&Spaces commented earlier.

3) The video mentioned "attaching new words to [a] pre-conceptualized world" - well, yes. But not necessarily to the full extent. When I learned English I learned new associations for everything (except during the crude and limited English lessons we had in early school). Conceptualized I guess, but my associations for words in English basically never used the same association as my native language - it sometimes took years for me to suddenly notice that words in the two languages were cognates - they were similar, they meant the same, but I had entirely different (but still correct) associations in my mind.

4) I guess one could say that I to some extent learned English in kind of an "AJATT" way. No studying at all, in fact. I learned all the grammar, structure, metaphors and nuances intuitively. But I can't do that with Japanese. I'm so much older now than when I learned English, and my memory and maybe other faculties aren't what they used to be, but maybe it's more because there's less to grasp on to when learning Japanese compared to learning another European language.
Which leads to:

5) I'm following the recommended (paraphrased:) "learn through anime (etc) slowly, until you understand, and keep listening to the audio as and when you get it" method. With the hope that, at least for listening, my comprehension will get lifted enough that I can move to more of what I did with English back then. We'll see. There's the reading issue which isn't easily solved though, so this may take a while. But darn, my SO is Japanese and it's shameful that I still can't speak more than simple phrases.

[MedK]

I've been trying the AJATT/MIA method for a couple months now, and I don't think I'm making any progress at all.

You've made me realize that that was because I have a low language intuition ability and I can't tolerate ambiguity at all (Your example with the whole French thing was _really_ relatable).

So my questions are: now that I know that, what exactly do I do? I've been watching your grammar videos, but I'm pretty sure that on their own, they're not enough to teach me Japanese; I need vocabulary too, right? I don't know any words at all. How do I go about learning vocabulary?

[tjstarr]

I think a balance needs to be struck between immersion in the language, listening and perhaps repeating what you hear even if you don't understand it, and learning grammar on the other hand. Unlike children, we are better equipped to understand the grammar of our own language vs. that of other languages. Of course, sometimes we notice the differences so much we can't see the similarities. But, when I say learning grammar, I don't just mean learning terminology, either Western grammatical terms or Japanese traditional grammar. You can learn that Wa is the Topic marker, and Ga is the Subject marker, but still not know how to use them. You can learn that the verb stem that comes before the -nai suffix is the Mizenkei, but that won't necessarily help you unless you are studying the history of the language. By learning grammar, you need to intuitively understand why you are saying what you are saying, so you can break apart long, complicated phrases like -nakereba ikemasen. I think this channel does a great job covering what you need to know in terms of Japanese sentence structure and other things.

[Kinorashi]

I have been thinking about the concept of language conceptualization ability from hearing all the people telling me that Rosetta Stone is a terrible way to learn Japanese. I couldn't understand why people were coming to this conclusion. I am wondering if the level of success with Rosetta Stone Japanese is a sign of someone's language conceptualization ability. The Defense Language Institute Foreign Language Center (DLIFLC) has a test whose purpose is to rate someone's language conceptualization ability. It test your ability to understand a made up language. The languages you can study are divided into four groups and Japanese is in the hardest group for English speakers. Only people that score high enough on the test can take a language in that category.

[CHOCOLATIONZ]

Thank you for saying that AJATT is not a method for everyone. I have seen one youtuber trying to claim it like the only best method on earth and other methods suck.

[Ryodakun]

I know AJATT can work. In fact I pretty much did AJATT with the english language back in the day. And I knew way less than 30% of the language. Maybe 1-2% at best. But despite it working out for me, looking back I can't help but wonder, why would I do it like that again? It took me a really really long time to get to the point where I felt comfortable in english. I feel like there's no real reason to do it that way because it's generally quicker to do immersion and study at the same time. Sure I can listen to 10 hours or more of Japanese podcasts or tv shows a day and eventually, after months I will possibly finally pick up why the particle "に" used in the way it is, due to recognizing the pattern.
Or I can simply study it and understand it in 10 minutes, for example by watching your videos and then do my 9 hours and 50 minutes of immersion.
Kanji and Vocab are already 2 things that take a long long time to get used to, why would I add grammar on top of that? Sure, you need practice to get used to grammar as well, but unlike kanji and vocab, it has the option to be learned through logical analysis which makes the whole process that much quicker.


If I take 2 people of about the same intuition for languages and I put one on AJATT and one on your approach, I guarantee after a year, while both will probably be decent, the AJATT learner will be behind.
I think on your approach the learner will be at a point where they can comfortably use the language everyday in about a 12 to 18 months.
With AJATT I think it will take between 24 to 30 months at least. I am just guessing here, but with english, a language much closer to my native language and no added baggage like kanji, it took me 2 years to get comfortable with.
If I added some grammar study on top of that, I could have avoided many pitfalls which I had to work out through trial and error, much more quickly.


Not bashing AJATT here. I think AJATT would work just fine for me. I have a reasonably good intuition for languages and therefore could probably do it like that. But again... Why would I do that?

[tiocsti]

I suspect the struggles most people have with ajatt/mia come down to tolerance of ambiguity (especially as you just start out where your comprehension is limited to some foreign loan words and a few words here or there which have reached general understanding in your native language). People with those problems will often resort to enabling subtitles (in the ideal case after you've immersed with it in your target once) which cuts your immersion per hour in half. Since we have no objective means to assess linguistic intuition, it's not really possible to evaluate that contention. ajatt is not based on learning how children learn (other than being based on a lot of input), so that point seems a bit suspect. The fact that there's no science supporting low linguistic intuition making massive input based approaches less practical (these problems should be universal problems to mass input approaches, and not limited to mia/ajatt) makes me somewhat skeptical.

[dycedargselderbrother]

I view this realm of knowledge as being actively developed, so much so that terminology isn't even stable, e.g. MIA being a new generation of AJATT despite AJATT itself not even being that old. More has probably changed in the development from AJATT to MIA than has in the academic method for decades or longer.

In the end I don't think organic and AJATT are incompatible. I view AJATT as recognizing that constant input is the way to learn language but that the "formulae" and various ratios and approaches are still in flux--obviously since Matt just put out a video where he revised his opinion of a 50/50 mix of text/active audio to 70/30 (plus potentially unlimited passive audio). I've read comments saying that Matt says to ignore grammar and structure, but I'm pretty sure I read on the MIA site or saw in a video that it's fine but not to spend something like more than 10% of your time on, which, again, I don't think is fundamentally incompatible with what's presented here. That said, there is a definite difference on emphasis.

The organic method's effectiveness can even be explained in AJATT/MIA terms. For example, by parsing an anime line-by-line and sticking with said anime for consistent vocabulary, you're creating a lot of so-called +1 comprehension opportunities. Also the approach isn't really all that different in theory from sentence mining. The main difference is how much time is spent in Anki. With the organic approach, instead of reviewing last episode's sentences in Anki, you're "mining" new ones from the next episode.

I think there will end up being multiple potential approaches and that there will eventually be appeals to students to figure out how they learn and use that to guide them down the appropriate path.

[a.m.]

Carefully go through anime with Japanese subtitles, and try to understand as much as you can. Thus, when you listen to the audio, you will understand what's going on, and will be drawn into it

[saaah]

Hi,
Some sites and channels (not necessarily just for Japanese) recommend avoiding the use of dictionaries as a rule. They insist that we should always choose material that is easy to us, to maximize the ability to learn through context without constant dictionary lookup interruptions.
What are your thoughts on dictionary usage? And what are your thoughts on choosing easy, rather than more challenging content?

[kaylaautry]

Hello Dolly sensei.
I've been trying to find the proper video to ask this question on, but none of them seem like the right place, so I apologise if this is off topic.

But, firstly, in relation to this video (kind of), and not actually what i initially came to ask about, I actually stopped immersing after 1 month. It wasn't because i couldn't understand (I was immersing with audio all the time, even when sleeping), but because i was picking up on speech patterns, the "spaces" between words and particles and such, even 相槌 patterns (to the point where I was unknowingly mimicking 相槌 outside of immersion in English conversations). Basically, hearing Japanese became like hearing English to me. I could have something playing In the background, and suddenly All my focus would go to parsing and analyzing what i was hearing, whether i wanted to or not, whether I could understand or not. I don't know if this Is good or bad, but it has caused me for the past month not to deliberately immerse. Somehow, hearing Japanese in Japanese has made things hard for me.

However, I am trying to break out of this, and that brings me to what i actually want to ask. Today i decided to play a Japanese game /in Japanese/. It's a game i was interested in before so i figured, "why not". I don't think which game it is is necessary. But, basically, of course, I hardly understand anything, and what I do understand doesn't contribute to understanding what's going on in the given dialogue. I've decided to go about playing the game ( by playing i mean understanding) by taking screenshots of a word if it appears commonly, then putting the sentence into Anki. Of course for this particular game i pretty much have to wing it until the dialogue is over. I'm kind of worried I'll go through without even paying attention to the Japanese, if I'm honest.

But, anyway, for Japanese media In general, do you think this Is the way to go? Or do you believe i could be doing something better?

Thank you in advance for responding.

Also, was the red ball thing a Matt vs Japan reference?

[zamyrabyrd]

My intuitive grasp of language is quite low because my brain cannot help but analyze. In the end, though, what I learn sticks better since it has been examined inside out and upside down. I find that when listening to children's stories with Japanese subtitles, for instance, I have to go back several times after I got the general gist of the sentence.
There are frequently more than one concept or kanji I don't know. I have to stop to pick them apart and even look them up in a dictionary. It takes time to integrate everything but there is usually some point I missed even though I did understand the sentence. Repetition is a way of examining an object from different angles.

[shaheemgoldberg]

Hello, Dolly! I first off want to thank you for making this video. Even if I don't agree with a lot of what you say, I can appreciate someone thinking critically and not just blindly accepting what others say. You have my respect in that regard. Okay, the first point I would like to address is being able to learn like a child. I agree that we are not exactly like children. Heck, I'd say in a lot of ways we have an advantage when it comes to language. Yes, it is true that kids are "conceptualizing" the world while we are simply attaching new labels to what we already know. However, when it comes to actually acquiring the language, what kids and adults do is pretty much the same. I believe you are underestimating the power of our brains. All humans (including androids *wink*) are masters of pattern recognition. The basis of intelligence IS pattern recognition. When you expose yourself to the language a lot, your brain is going to pick up on patterns and form models unconsciously. This happens as a child learning your first language and as an adult learning a second. This also ties into the difference between learning and acquisition. Acquisition deals with unconscious processes in our brain. When one exposes him/herself to copious amounts of a language, the act of recognizing patterns and forming models of the language is acquisition, not learning. Learning is more deliberate. With learning we are CONSCIOUSLY forming models taking in information. That is why I don't like to say I "learned" English, but rather I "acquired" it. So, to wrap up this point, yes, there are differences between us and children. That goes without saying and I think most would agree with that statement. However, when it comes to acquisition and pattern recognition, it is quite frankly the same with children. Okay, now onto your reasons for why AJATT isn't for everyone. First reason you have is "High level of linguistic intuition". Admittedly, your French example had me a bit perplexed. You claim that you were able to read books and that your French was "objectively" better than hers, but you couldn't understand "practical" French being spoken to you. Your friend was able to pick up this practical French because she had better linguistic intuition. This seems really odd. To be able to read books in any language, you would have to be at a fairly high level to do so; Books contain some of the most difficult language in *any* language. I am not trying to bring you down by any means, Dolly, so please don't take it that way. However I think there may have been a misjudgement in your ability and your friend's ability in French. Intuition is something that is developed through experience. In order to develop linguistic intuition, you need to "get used" to the language. You need to put in a lot of time into the language and give your brain a change to acquire it. Once you acquired, you develop a "sense" for the language. In other words, intuition. If I were to guess, your friend may have had the upper hand in this case with practical French because she was exposed to more of "practical" French than you. More exposure leads to better linguistic intuition, this is not predetermined. Anyone can develop linguistic intuition overtime. Example, assuming you are a native/fluent English speaker. Allow me to ask you a question. Which sentence sounds better or more correct to you? "Honey, I made dinner!" or "Honey, I created dinner!". I am confident that 99-100% of people would chose the first option. Why, because you have heard that variation of the phrase millions of times. Your *intuition* led you to choose that...Now let's move on to your next point. Your next reason for why AJATT isn't for everyone is "high tolerance for ambiguity". I personally have a pretty low tolerance for ambiguity; at least I had a very low tolerance of ambiguity in the beginning of learning English". In the beginning, I understood jack squat, and having to learn watch a bunch of English while understanding 1% of my immersion sucked. However, over time as I immersed more and more and mined more and more sentences, I understood more and more and it became less of a painful experience. I think with AJATT you have to push through this in order to get it to work. And THIS may be why AJATT may be hard for many. However, I still think AJATT can work for just about anyone willing to do it. With enough effort, I think anyone can overcome this "ambiguity" problem. Now to add a little on my personal experience with AJATT I learned English using AJATT. I understood pretty much nothing when I started, and I wasn't a hikikomori like some AJATTers. Most days I only spent 2 (sometimes 3) hours actively immersing everyday. My life was not dedicated to English. I didn't throw my life away for English. Even though I didn't have as much time as a NEET would, I still manged to reach a fairly high level of English in about 3.5-4 years. I can speak English without too much effort, read novels no problem, and just enjoy my ability overall. I am looking forward to your response, Dolly. I am not learning Japanese (nor do I ever plan to), but I like your channel. Cheers.

[futurez]

Just one other thing... The 'concept' learning was interesting. I wonder though, we already have those concepts in our brains as adults, so it's just a case of attaching a new sound to them. I guess that might take more time, but I think I'm right in saying we rehearse things in our mind as we read/hear them, so I wonder if this rehearsal, when he see/hear the new word for 'ball' comes close to the actual physical experience a child has with first encountering the concept? I'm guessing it could.

Maybe that's where an adult's talent lies, maybe the best 2nd language learners are those who are better able to rehearse such things in their minds? I know elite altheletes do the same kind of thing - they imagine carrying out movements in the mind, and I believe science has shown that this has almost as big of an effect as actually doing the movement.

REM sleep, and deep sleep, is probably doing the same thing. So you may learn a new sound for an already established concept, but perhaps during sleep there's a process going on which is acting out the concept using the new sound? I don't know, I'm just speculating now, but I don't believe it's quite as simple as saying adults don't learn the concept with the sound, whereas children do. It's hard to know what goes on in the subconscious brain.

Anyway, sorry about the long comments. Your videos are really making me think. 🙂

[JeanOfmArc]

This is a very interesting and philosophical video. I really liked the ball analogy. Would you mind crediting/pointing a link to the video you mentioned? I'd be interested to hear more from their perspective as well.

As far as learning goes, I'm probably middle-of-the-line when it comes to the scales you were talking about. If you'll indulge me, I'll explain my situation a bit. I went to live in Japan 10 years ago now, and lived there for 2 years. After I left Japan, I hadn't really used/studied it at all. But suddenly at the start of this year my interest in learning more sky-rocketed (for various reasons) and I'm back into it.

10 YEARS AGO:

When I started learning Japanese, I did the old phrasebook/textbook method. And although it DID, in fact, help me to say some meaningful sentences quickly, it ultimately wasn't that useful as a learning medium because I had no understanding of the language. It was more like "if you make these sounds, people will know THIS EXACT thing", which isn't very useful when you're trying to have a conversation. You can only say "私はAが好きです。私はBが好きです。私はCが好きです。" so many times before they'll politely find an excuse to go somewhere else.

I also did hundreds of Japanese--to-English word flashcards. And although it did help me to memorize the words, it wasn't fun (there's not much that's engaging about seeing "Expression A" and then "Meaning of A" over and over) and I'm not sure the rote memorization was a very efficient way to pick up the vocabulary either. I passed the JLTP3 (the old 3, not N3) and after that I stopped any kind of formal "study".

It wasn't until I decided to stop doing language books and flashcards and just started having conversations with Japanese people that it really started to be fun, engaging and efficient.

NOW:

I've been more intentional about learning language theories and efficient ways to learn. AJATT is one that I've looked into a lot, Comprehensible Input is another, and of course your Organic method. However, because I fall in the middle of the scales you talked about in your video, I don't think I'm 100% on any 1 method.

Ambiguity: When I went to Japan, I had conversations with people in which I only understood about 70% of what they were saying, and could only easily express about 60% of what I wanted to say. Even now I'll watch anime from start to finish and only understand 30-40% of the words. However, I can infer a lot of information based on context, visuals, etc. and end up understanding about 70-80% of what's going on as a result (via linguistic intuition), and I am fine with that (ambiguity tolerance). I think this stems from the facts that: A) I am already fluent in French from having done all my schooling in the language, and B) I am a musician, and music is inherently both ambiguous in meaning, yet universally understandable as a medium.

Certainty: However, I am also a computer programmer. I love logic, because it's how I understand how something "works". And so your lessons, and other teachers who take more logical approaches, are my favourite to watch and learn from, because they feed my desire to understand EXACTLY why each word was chosen and why they are in the order they are in. Or in the case of Japanese, why they are omitted. :)

I feel like the combination of both approaches is wonderful, as I feel like I am honing in on understanding from 2 different directions, using means that I did not use the first time I was learning. I'm not sure if there's a theory that combines approaches like this, but so far, I'm loving it WAY more than when I first started learning Japanese.

It remains to be seen if my Japanese will substantially improve as a result of these combined methods, but at least I'm having more fun doing it. In fact, I'm having so much fun that I'd rather be learning Japanese than playing games these days, which is amazing and surprising to me.

Anyhow, thanks, as always, for sharing your thoughts. You're incredibly insightful and express your ideas very well.