This blog post was brought to you by the generosity of AJATT's patrons!

If you would like to support the continuing production of AJATT content, please consider making a monthly donation through Patreon.

Right there ↑ . Go on. Click on it. Patrons get goodies like early access to content (days, weeks, months and even YEARS before everyone else), mutlimedia stuff and other goodies!

What If RTK Isn’t Working For Me? Early Cut-Off Kanji

So, usually, I give abstract advice, because:

  1. It lasts longer, and
  2. You need it more

But it’s certainly not all you need.

I can cook, spice and feed you sentence packs. I can give you the raw frozen fish of how to learn a language, or I can give you the fishing technique of how to become the kind of person who can make learning decisions for themselves again — the way you used to before you were schooled.

There’s a place for all three approaches, and today I’d like to talk about one: early cut-off kanji (early switching to MCDs).

Newsflash: The RTK/Heisig method doesn’t necessarily work for everyone out of the box. That doesn’t mean it’s bad and it doesn’t mean you’re bad; some clothes just don’t fit some people. Truth be told, it didn’t even work for me (in its raw form), and I promote it unreservedly, like gospel; I personally used a heavily modified and personalized variant on the Heisig method for my initial kanji acquisition. So there.

What I’m about to say it going to sound like heresy. Especially to fellow Heisig grads. But it shouldn’t and I’ll tell you why: the point has never been to follow a specific or pretty or elegant or popular or widely accepted or even logical-sounding method; Heisig caught a lot of hell in the 1970s not simply to give us his method but to set us free from being forced into anyone’s method or tradition — ever again.

So here it is:
If you’re plodding through RTK (Remembering the Kanji) and hating it, then maybe it’s not for you. And maybe you need to cut off early and go straight into MCDs (formerly the “sentence” phase). Sure, you’ll have fewer kanji under your belt, but so what? You can pick them up later, along the way. If your personality is such that the feeling of having to go through 2000~3000 kanji first is just sending you into paroxysms of emoness, then…yeah, don’t…Cut off early, maybe at, say, 1000 kanji, start MCDs, picking up other kanji along the way, either deliberately or in context.

Your first 1000 kanji are a strong foundation, at 1000, you finally “get” in your bones what kanji are truly about and how they work and are structured. It’s not like you don’t know any kanji just because you don’t know 2000 yet; you just know less than some people do at a certain stage; it’s sort of like being born prematurely — it’s just…that was your time to come out.

Is this method-breaking-method “ugly”? Yes. It is ugly and messy. It lacks the conceptual beauty of getting all one’s ducks in a row and doing all sorts of things perfectly first. The theoretical mathematician inside you is balking at the messiness, but now is the time to let the get-‘er-dun’ engineer inside you take over.

Is knowing more kanji earlier better? Sure. But that’s not the issue here. If you’re hating the kanji phase so much that it’s slowing you down, you’re already on a path of not knowing Japanese period, let alone kanji. There are bigger things at work than you polishing apples for non-existent teachers. There are no gold stars for coloring in the lines any more; we can get computers to do that; MSPaint can do that. Only you can make your learning process fit you perfectly, because only you are you.

Is “getting through” all the Heisig kanji first prettier? Is it neater, is it tidier, is it more “disciplined”? Sure it is. But human beings do not grow in straight lines. Human beings are crooked and lumpy and unique. There are many luxuries you can afford, but you cannot afford the luxury of mental decorative towelery — all your methods have to earn their keep, getting themselves wet and dirty and your hands dry; you can’t afford to be stuck to methods just because they’re neat and tidy and pretty and that’s how your “supposed” to do it and that’s how Kenyans on the Internet tell you how to do it. They have to work for you. Don’t commit to a method. A method is just a means. It’s like a paper towel — ya use it, and then you throw it away when you’re done with it, and everybody’s happy and nobody cares.

It’s like getting a massage — at some point, you stop getting the standard massages you’re “supposed” to get, and you get the kind that fit your body perfectly; over time, your masseur gets to know you better and all that. It’s not like you didn’t try to learn kanji the Heisig way; it’s not like you dismissed it out of hand simply on hearing how it works, without making an attempt. No, we’re past that kind of juvenility. You tried. You’re trying. You’re one of the good people. And it isn’t working for you, and that, as the sappy therapist types stereotypically say…is okay. It’s mmm kay.

Learning methods were made to serve you, not you to serve the method. If it — your method, your process — sucks, then bend, break and change the rules until it doesn’t suck any more, and one of the quickest ways to do that might be to fast-track yourself into MCDs.

Breaking the rules is not the end of the world. Giving up, is.

  10 comments for “What If RTK Isn’t Working For Me? Early Cut-Off Kanji

  1. tony
    January 16, 2014 at 22:10

    I have got to over 1200 kanji through RTK. I usually get about 80-90% with Anki repetitions. I also use the Memrise version of the course ( to help me remember the key words too. Sometimes, like this afternoon, I feel like I am not learning in the right way and going through it too quickly. Some days I do rush myself to meet a certain target, other times I enjoy making stories and take my time. I bet this is fairly common anyway.

    I can’t wait to get onto MCDs. I have had a cheeky look at one or two MCD decks that people have shared on anki and read Khatzumoto’s free samples to get me geed up. I am only wondering whether I will later on make time to do RTK3. I hope so.

    Great website, Khatzumoto. Thanks so much. Your methods remind me of James Hamilton’s Hamiltonian system of language learning from Victorian times that I have been researching. He also felt the common way of language learning was awful and relied on grammar, which he did not. He instead provided word for word translations of texts that the teacher recited (one word L1 one word L2) until the student could repeat the words. The idea was that eventually through repetition the L2 word would be learned more naturally. He also used keywords, so each L2 word always had the same L1 translation and maintained that with few exceptions words have only one meaning. He found this convenient.

    Anyway, once I get off RTK I will combine your 2 methods. I will get interlinear translations from Mangajin and Hiragana Times and work with them while sentence fishing generally.

    Cheers again.

  2. January 16, 2014 at 22:20

    I’ve been slowly coming to this realization myself (i think) and being the weak creature that i am it feels as great as always to have you speak out loud the thoughts I’ve been having. I’m currently at 1500 or so 漢字 after a year of messing around with Heisig but it doesn’t make me feel as bad as it used to! I’ve come to the realization that i just don’t like to srs that much and that I rather read a 漫画 instead. I’m still waiting for that first sentence that I just have to put in surusu. But until that time comes I’ll just continue to read books and 漫画^_^

    It’s so true what you said about the getting a backbone in your kanjis though. Like, eventually, if you just continue to fool around with kanjis you’ll become so familiar with breaking them down in a way that’s comfortable to you that, you basically can add them in any order you want. I mean lets be honest. The 2000 常用漢字-list doesn’t really have a meaning anyway. There are a lot more 漢字 after that but they just become more of a pleasure to learn as well 😉

    One thing I just wanna share with you all is a thought on the common practice to ask your 日本友人 to correct your mistakes. It’s been mentioned here and there in ajatt and I just wanna make an alternative statement based on my own experiments.
    [b/]It’s not a big deal, maybe [b/]
    You almost never learn a word by hearing it once anyways, and when i ask what a ginger root is called, or if i’m corrected on how to say some word, since my brain is used to the wrong pattern I’m gonna make the same mistake again anyways and one minute later I’ll ask how to say 生姜(薑(しょうが)) again.
    So my point is that, if it bothers you to ask questions and be corrected all the time, don’t sweat it! You’re gonna learn it all eventually so just do the fun stuff and eventually you’ll know the right words anyways. The same with pronunciation really. Eventually you’ll just hear how strange you sound compared to all the other 日本人s and correct your speech on your own 😉 Shadowing is the shit and it’ll help you make funnier jokes 😀

  3. Linval
    January 17, 2014 at 02:48

    Good article. I went through RTK in its entirety about a 2 months ago, and I still suffer from time to time from the “can write from keyword, but can’t always quite recognize the kanji in context” syndrom. However doing good ol’ Heisig was a fantastic way to demystify the big bad kanjis, who went from being a messy tangle of strokes to actually making sense. And I think that Heisig is worth doing just for that reason – it is an easy, soft way to getting used to what people consider to be the biggest roadblock to learning japanese. “Demystifying the Kanji” would be a good sub-title for the book methinks…

    Heisig’s biggest strength (and weakness) is that it is almost irrelevant if the keyword makes sense or not – some kanjis in the book sport weird keywords that only make sense when the kanji is used in compounds, or really obscure sub-meanings, so they are never totally disconnected, but they might seem weird at time. However, I think this is a complete non-issue as the true aim of the book is not to learn vocabulary, but to make sense of the kanjis in a more… meta way, if that makes any sense.

    So I WOULD recommend people to go through Heisig, or at least through enough so that things like element placement and stroke order starts making sense (which it will, eventually and effortlessly). However it is far from mandatory, as you’ll encounter kanjis in the sentences anyways. This is where you actually learn vocabulary. RTK definitely makes the sentence phase smoother, but is not a hard requirement. And if you can’t be bothered to create a story for every kanjis in the book, there are many ready-made shared anki decks that’ll do the work for you.

    Also it is worth remembering that Heisig devised this method in a time and age when SRS didn’t exist, so his method with the heavy emphasis on imaginative memory can be twisted beyond recognition and still be efficient, thanks to the incredible power of the SRS.

    Phew, that came out longer than I expected… sorry for that.

  4. Caren
    January 20, 2014 at 11:46

    I hated RTK within the first few chapters and moved on to Slime Forest (a game with a similar concept as RTK). With that, I learned about 1600 and then got tired and moved on to sentences (I don’t like MCDs much) – as well as a few other alterations but I won’t go into details, but it basically included learning/reviewing kanji as I saw them show up.

    My japanese still improved tremedously. I actually started to enjoy playing real japanese games. It’s more important to have a method that keeps you going, than to have the “right” method and get bored!

  5. beneficii
    February 10, 2014 at 02:50

    I’m thinking of where the kanji will just be visible and I will just write it and move onto the next kanji. I can read a lot of kanji through long absorption, but I can’t write nearly as much as I want to from memory.

  6. Bill
    February 17, 2014 at 17:37

    I have been completed RTK for a fairy long time and now just do SRS for it at maintenace: which is still 50 to 150 cards reviewed daily. Does anyone have an opinion on whether it is unnecessary to maintain at this point if it does feel like somewhat of a roadblock to my daily further study or would I be squandering that knowledge I’ve already assimilated?

    • tony
      February 18, 2014 at 11:39

      I also finished RTK last week and had the same thought as you just now. I have spent my entire morning doing reps, and not sentences. My new tactic is to make sure I score 100% on anki by restricting myself to clicking either hard or good. If I don`t quite get it and click hard I also write the kanji out 3 times and read my story aloud. This should eventually keeps reps low but it is not ideal. Of course I could break up my study and just do kanji during occasional breaks rather than in one long session. If I had a smart phone I would perhaps do that, but it is just me and the computer for the now.

  7. April 10, 2014 at 07:25

    It’s perfectly true that not one size fits all, as often as the masses want to lead us to believe. And that idea reigns supreme with language learning.

    Although I had completed the awkward “Kanji” stage years ago, I understand that a lot of learner just want a push-button system. But if you have passion for the language (and experiment with the methods that you learn about), then the target language(s) will be a breeze. Thanks for the post, Khatzumoto!

  8. Chasmandingo
    November 4, 2014 at 14:49

    I am almost there myself. I am at 1800 and am losing the will to Heisig. I am fairly certain that I am in need of starting on MCD’s and coming back to the last 200 a few at a time as I learn sentences. I feel that it will jump start my interest and love of Japanese once again. Plus I can continue to review my kanji and just go with the flow!

  9. Bill
    June 13, 2018 at 07:01

    My problems with the AJATT RTK method are:
    1. It appears to be trying to teach/practice how to write Kanji. Writing is output. We should be doing input before output with reading/writing, just as we are with listening/speaking.
    2. When I see a character, I prefer to have a sound in my head (even if it is only one of the sounds) of the character. I don’t want to be thinking of some weird keyword “nightbreak? really?”

    So I’m using Heisig’s kanji order but learning the two most common words with each character. So instead of 3000 keywords/stories/mnemonics, I’m learning 6000 real words that I see in real Japanese content. This works better for me.

Leave a Reply

Your email address will not be published. Required fields are marked *