MIA or WaniKani

oh noes I missed all the debate

sad life


I had the same dilemma as you @_josh, but I ultimately chose WaniKani for the above. I think a lot of the really hardcore MIA people are younger than me, or spartan disciplinarians with Zen like powers of concentration and focus. They spend lots of time optimising their learning approach, messing about with optimising Anki, often while simultaneously documenting their progress and writing manifestos about how to do things the right way.

I have a full time job, friends and a great relationship. I also have other hobbies. So while I used RTK and WK together for some time, I ultimately loved that with WK somebody had done the grunt work for me. It gives me an accessible, carefully planned route of progress through kanji and a chunk of vocabulary that syncs up perfectly across all my devices.

I also saw that video by Matt and got spooked, especially since a lot of what he said makes sense. I have realised however that some of what he perceives as downsides of WK are not deal breakers for me. One example of this is the dreaded :japanese_ogre: MEMORY INTERFERENCE :japanese_ogre:. I think for me personally, I realised that to some extent, this is just part of the Japanese language. There are tons of homonyms, similar looking radicals and kanji. WaniKani so far as been pretty gentle with this, but when ‘interference’ does arise, I feel like I am tackling it right away rather than being afraid of encountering it in the wild somewhere down the road. I am also tackling readings right away and can very quickly link up what I am learning with vocabulary I picked up in my listening and immersion efforts. That means that WaniKani doesn’t operate in a vacuum for me; it does exactly what Matt advocates and clicks onto existing bricks of knowledge I came across elsewhere, just like little lego blocks.

This next part might not sound like it’s related to WaniKani but hear me out: I’m a Linux user. I feel like though they are two very different domains, there’s MIA/AJATT types in the Linux world too. It’s a world full of endless customisation, tinkering, optimising, obsessing over tiny details and there is really NO limit to the amount of time you can pour into Linux as a hobby (looking at you, Arch Linux, Gentoo, Linux from Scratch). I certainly take full advantage of that and customise all my systems exactly like I like…but then I just get to work and use my hardware for whatever I bought it for.

If default settings work for me, I don’t change them. WK feels something like that to me - a system with good defaults. If I don’t like some aspect of it, I can easily change things up with great plugins that users on the WK forums painstakingly and generously developed. And if that’s not enough, there’s a lot of good third party tools being made that integrate into WK (though to be honest I can’t afford to use them all…those monthly fees add up).

If I have trouble or need extra help with some other aspect of Japanese learning, I can get involved in WK’s weird friendly and engaged community. That includes @mamimumason taking time to prompt and respond to my horrendously improvised beginner sentences (with the patience of saint) and when I am comfortable enough I can also dip into book clubs aimed at various levels. In the grand scheme of things, WK and its forums are a platform that is open and accessible to all. Sure there’s some serious power users and tinkerers here too but I don’t feel like I’ve accidentally barged into some ivory tower or retreat for extremely opinionated hermits that will spit on me if I’m not into the same ascetic and elitist approach to education. It’s ok to have fun here.

Building on the above, to me the MIA world seems like it’s the the GIT GOOD school of Japanese language learning. If you’re very competitive, youthful and boisterous I suppose that might work for you. But I am a more collaborative and congenial type of learner. The only person I have anything to prove to is myself…and really I just want to enjoy the process of learning rather than beat myself up over it.


My god, I wish I could spend hours upon hours making and customizing flashcard decks.
But I can’t. I’m an adult now, and that means I’m a slave to the clock :frowning:


I guess R(reduced?)RTK is what it was known as “lazy kanji”, if so, It did existed and was frowned upon with people doing it getting the usual comments in the tone of: “why would you go for a half assed way of learning japanese when Matt is saying to go the other way… duh!!” :poop:

I had already attempted RTK before WK. Managed to get to chapter 15 before giving up on that. At some point I was doing both but felt It wasn’t helping much and decided to spend time doing something else, albeit reading, which in my case were Graded Readers for Japanese learners, that turned to be easy enough to follow something like 3 months from starting with WK. This way with a tool that allowed me to learn kanji/vocab and a way to put it into practice I didn’t care much about those “right vs. wrong way of learning videos”.

I did a lot of the active / passive listening meanwhile, as I found ways to watch content without understanding a large portion and still feel engaged with it. But then I just couldn’t find a suitable way to read native content until I reached at least a critical mass of kanji / vocab.

Whatever goal you have with the language, be it reading books, watching shows, engage into significant dialogues with someone from the other culture or just passing an exam like the JLPT (which I’ve learned can be a goal in itself), whatever method you use, if you see yourself progressively closer to reaching it each day, then it’s a valid method I feel these days.

Now if you don’t have any specific goals it’s way too easy to get lost in pointless discussions, but tbh I think that’s beyond learning languages and more related to how you achieve anything in life, which often includes time and dedication. :wink:


I did not read this whole thread so it may have been said before, but I think with AJATT and MIA you have to consider 3 things:

  1. Studying Japanese 4-5 hours a day and cutting yourself off from the world is just dumb, you can study Japanese, make progress quickly and have a life at the same time.
  2. I’ve seen people upload videos to YouTube of them speaking Japanese after these methods and they sound like they are repeating phrases they heard and not like they are thinking about what they want to say and expressing their thoughts. Also they grossly overuse and misunderstand particle endings, I head ね and よね after basically every sentence and あの and ま constantly it sounds weird. For example something like - 今日は今日うん今日ですねまあの自己紹介ですよね6月ぐらい日本語を勉強していますよねまそうですよねぼくはXですよねまね It sounds really weird.
  3. The two biggest supporters of these methods on YouTube, Matt VS Japan and Brit VS Japan both failed to use this method and apply it to other languages (Chinese and German).

I just want to say that I think I’ve gained a pretty decent kanji reading intuition from WaniKani, especially with the help of this script to point out less obvious phonetic components.


Well, I am the same, usually. I prefer intuitive learning since it suits me most.
I have tried that with Japanese for about year though and found that not being able to read was a major obstacle. Especially if you are not in contact with the language everyday in other ways, then reading usually is a very good way to intuitively learn language patterns, phrases, vocabulary and, well, basically to think in that language in a way.
So I decided to do both, try to continue learning intuitively (watching anime, playing games with Japanese voices but English text etc) AND at the same time using learning tools like WaniKani and BunPro.
So I draw on various sources for studying Japanese, not just one.

For me it feels a little doing music, or some kind of art like drawing Manga or so.
You can be intuitively good at it if you have a knack for it. But only putting in some extra “techniqe-training” makes you really good at it and lets you advance better.

But that’s just my point of view.


because i don’t know how to read books and wanikani is fast if you’re committed



I say both.
I follow MIA now as well. After many years of WK I can absolutely say it works.
If you learn the stuff from WK and pair it with in context native material, it’ll stick.


Even assuming AJATT/MIA works the way its proponents say it should (and I doubt there’s any scientific evidence for that, just theories and anecdotes), I’d say WK has these advantages:

  • Your knowledge will at least be vaguely useful for real Japanese right away. You’ll be able to look at least a few common Japanese words and know how to pronounce them even at level 1. MIA’s focus on spending several months doing nothing but indexing kanji as out-of-context symbols means you’ll have make a long-term time commitment with very little short-term reward.
  • WK’s content is organized in a reasonable way for you, with no work required from you. Its mnemonics are of a bit more questionable value, but at least they’re made with no work required from you. MIA encourages you to spend a lot of time fiddling with Anki even to index the kanji, even before you get into the massive Anki timesink that is their “sentence mining” process. Again, it’s a major time investment, without much short-term reward.

It’s important to note that MIA is meant to be a complete roadmap for going from nothing to fluency, which WK isn’t designed to be at all. There’s really no overall roadmap that WK fits into, other than maybe the big Tofugu post about learning. So WK versus MIA as a whole really isn’t a terribly valid comparison; it’s more like WK versus Anki, or WK versus Remembering the Kanji.

If MIA works at all, I’d guess it’s probably just because it requires you make a massive time investment by design. Learning a language is a massive time investment any way you go about it, but there’s really no way to do MIA without that investment. Whereas it’s at least possible to be lazy with WK and get poor results out of it.


Cool, thanks! It’s helpful to have people who have done it say it works.

I think the idea of MIA is so you learn the language how you would learn your native language. It has you do things they think are most efficient from the start so you can eventually master the language, which is what I’m going for. Like I said I’m going to try it out (because I have a ton of time to invest) and I’ll report back whenever and say how things went. I love WK, but this way of learning seems to fit more of my personality and how I learn.


Best of luck! :grin:


So coming to this late, but since I have a slightly unique perspective, I’ll share.

I did RTK years ago (like ~10). I liked it at the time, got ~1500 kanji to mature in Anki. That, together with kana and grammar from Tae Kim got me to a state where I could “read” (understand, or at least decode) quite a bit of Japanese text. But I certainly couldn’t speak it, or understand spoken Japanese.

Readings/comprehension never came, though. Some basics were easy, but I never developed that “intuition” for it, despite frequent input.

WK, in the 3-4 months I’ve been here, has built me more of that intuition than several years of trying post-RTK. Why? Because by learning the readings in the context of vocabulary words here, I’ve found a lot of other vocabulary, either not in WK or at a higher level than I’ve achieved so far, that I already know how to read and can often infer meaning of due to knowing the kanji meaning and reading together from WK.

Even better, I’ve found a lot of these encountered words, I read in Japanese first. And more and more often, that’s all I have to do. I have the concept of what it means without translating to English terms. This is very new, very rare, but it’s pretty profound to me. I am understanding words directly in Japanese.

RTK was great, don’t get me wrong. I can progress in WK much faster because of my RTK pre-knowledge. I basically ignore the WK mnemonics for meaning, since most of the kanji I know already or pick up/relearn quickly. But it will never, can never, get you to the point where you will be understanding Japanese in Japanese.

If I was doing it over, I would have done WK without the separate step. It would have taken longer than WK after RTK, for sure, but it probably wouldn’t take as long as RTK followed by WK as I’m doing. And I wouldn’t have delayed getting to this point where Japanese starts to make sense as long as I have.

TL;DR Despite doing RTK, I didn’t develop reading intuition until I learned a lot of readings. I learned those by doing WK. And that jump started my learning to an immense degree.


Immersion is certainly valuable, half of my Japanese studies have simply been reading or listening to Japanese even if I couldn’t fully understand it and I’d say it’s greatly helped my grammar understanding.

However there’s a limit to what you can learn with immersion. Even Japanese natives have to rope learn Kanji at school, there’s no way I could learn kanji solely through immersion so I think Wanikani is a very valuable tool to help you learn the shear amount of kanji there are.

You can always learn the Kanji with Wanikani and then immerse yourself in reading to solidify your understanding later


Sort of… these two goals are kind of at odds, though, and MIA definitely attempts to prioritize efficiency over native-like learning. (I think one of Matt’s videos points out that it takes Japanese children six whole years to get to the point where they start learning kanji, and who has that kind of time?, and adult learners should be able to do better, or something.) As others have pointed out, Japanese children don’t memorize 2000 kanji before they learn how to speak a word of Japanese, and they certainly don’t spend hours drilling isolated sentences in Anki.

I do think that most of what’s good about MIA and AJATT comes from the parts that do try to emulate native-like learning, though. Understanding words and ideas monolingually is invaluable, in my opinion, and definitely worth prioritizing as early as you can. Immersion in native material instead of fakey textbook Japanese is also critical as early as possible.

But, I’d argue that if you want to really learn like a native, you should be acquiring ideas in something like the way actual children do. Don’t try to decode words from whatever complicated definitions you get from the old CD-ROM dictionaries for grownups the AJATT people happened to find lying around; learn them from dictionaries meant for children and written for children’s level of understanding. Or, even better, learn them from books literally written for babies, who learn most of their basic vocabulary before they can even lift a dictionary.

I’d also hypothesize that spending hours reading and re-reading every complete level-appropriate book you can get your hands on is probably better than reading the same sentences in Anki over and over for hours. That’s certainly how I learned most of my vocabulary in my native language, at least.


Love what you’ve written. This also reminds me that I’d still like to work my way through reading all of the ehonnavi corpus :scream:. New Years goal ftw!!


MIA is what I call myself when I haven’t done WK reviews in weeks: Missing In Action


If you want to learn like a native you should hire a full-time tutor who tells you to say “いただきます” all day.


There should totally be an Anki plugin that does this.