I was just thinking today about how “dirty” and fluid and lazy Lazy Kanji is. And I was thinking about the obnoxious certainty that anti-RTK people have about how RTK couldn’t possibly work.
And it hit me that it may have been this article (about the differences and similarities between AI and cybernetics) by a guy called Paul Pangaro that got me thinking in the direction of lazy kanji. For curiosity’s sake, here are a couple of interesting quotes:
“information (or intelligence for that matter) is an attribute of an interaction rather than a commodity stored in a computer”
This quote in particular (from Humberto Maturana’s Biology of Cognition and Autopoiesis and Cognition: The Realization of the Living) really got me thinking [emphasis added]:
“Learning is not a process of accumulation of representations of the environment; it is a continuous process of transformation of behavior through continuous change in the capacity of the nervous system to synthesize it. Recall does not depend on the indefinite retention of a structural invariant that represents an entity (an idea, image or symbol), but on the functional ability of the system to create, when certain recurrent demands are given, a behavior that satisfies the recurrent demands or that the observer would class as a reenacting of a previous one.”
What does it mean to “know” a language? What does it mean to “know” a kanji? It’s a question of worldview, I guess. To put it a little simplistically, it’s a question of whether one sees things statically or dynamically. And it’s important that each of us gets somewhat clear on an answer to these questions, because that answer will decide whether life is tingly and fun, or toadly sucky.
If one defaults to one of the certification-seller-approved answers and decides that: “knowing” a language means being able to take boring, arbitrary, authority-presuming tests on it, administered by people you neither know nor like, then that’s fine, too — it’ll just…suck.
If one defaults to one of the school-approved answers and decides that: “knowing” a language means doing insomnia-curing textbook assignments on it and avoiding FUNBUN (for-native by-native media) like the plague (or rationing out FUNBUN as a “reward” for “real learning”, as if FUNBUN were some kind of fattening dessert instead of the nutritious vegetable dish that it is), then that, too, is fine…it just kind of sucks.
Translated specifically into language-acquisition terms, Maturana seems to me to be saying: you don’t get good at a language, you get used to it — it’s a habit; it’s a relationship. It’s as if the language were a person. You don’t learn to “read”, you get used to text; you develop text reflexes. Not only is the language itself in flux, but we are in flux with respect to the language.
In the late 1970s, James Heisig freed us from stifling and ineffective ideas of what it means to know a kanji. To the extent that unquestioned, unfettered attachment to the RTK method threatens to itself become a new orthodoxy, the Lazy Kanji/SRS combination could be a new “mini-shift” in kanji perception: you don’t “know” kanji, you get used to them — just like your phone number. (So used to them, in fact, that you can read and write them in any situation you want or need to).
Having said that, Lazy Kanji is actually built on the core RTK insights, namely that:
- the component logic of kanji should be exploited to learn each character bottom-up, and each “family” of characters in order of increasing structural complexity.
- each kanji represents a concept that transcends any one language, and this concept can be represented using a word or words in your native language or L1.
So, to me, it just represents a natural progression rather than a “revolution” or anything drastic like that. Heisig’s always going to be top dog in my book. He brought us so far. He broke the mold.
In any case, I think the ultimate lesson is this: screw doing things “the” right way. Your way doesn’t have to be “right” — it just has to work. Maybe we’ll add that to the “Three Laws of Language Acquisition” 😀
whoa whoa whoa, toadly is my word, you can’t take it, bro
I have to say, that quote from Maturana really blew my mind. It’s such a great description of recall that it’s sorta changed the way I think about approaching recall. Recall’s not about “knowing” things, it’s about being able to respond to certain events in a way that is measured as “correct”. Like a reflex. That’s awesome.
A++ would read again
ha. You are really going deep on the language acquisition talk now.
I think the the ‘keep it simple’ stuff is more suited for my mind. I’ve tried to put together a few more games for my Japanese learning/goal achieving in general. (click my name if you feel like taking a peek.)
By the way, you should let Momoko know that her fun articles are still wanted.
お元気で！
Sorry, the link in my name above is to my latest posts. I didn’t think about the games post getting moved back over time. My bad to those of you who got sent astray.
Here’s the link to the games I was referring to:
hatchjapan.wordpress.com/2010/09/30/mini-games-for-a-game-for-life/
5 stars. Would buy from this seller again.
Nice post as usual^^
Question: Do you think going through RTK1 without storys, but with the lazy format, is advisable ? I have a premade deck with multiple meanings in Heisig order which links to kanji.koohi.com. I find that I spend almost 50 % of my time on that site reading stories while I add new cards. But recalling a meaning without a story feels hard… Especially with so much similiar Kanji introduced in a row.
Hey Khatz,
I think you might appreciate this:
www.iq.harvard.edu/blog/sss/archives/2005/12/bucld_statistic_1.shtml
A lot of what I’ve read in statistical learning seems to match what you talk about with AJATT.
Dear Ole,
I managed to do the kanji very quickly (under a month) using no explicit mnemonics, whilst posing the kanji as the question and the keyword as the answer.
I recommend, however, that you name the radicals. I also recommend that you put the names of the radicals in under the keyword on the answer side of your card.
I find that, doing this, I invented “natural”, my-brain tailored, obscure and almost meaningless mnemonics for many kanji, which I can reel off easily at any time. I think people spend too much time trying to think of acceptable- “cool” or “appropriate or whatever- mnemonics. In truth, the best mnemonic is the one which you mind most easily links with the kanji and its meaning.
For example, I named the components of the “snake” kanji- 蛇-“insect”, “crown” and “spoon”. For me, as soon as I see the “spoon king of insects”, I imagine a snake wearing a crown of spoons, lording it up over some flies (get it? Lord of the Flies?).
Is that sensible? No. Does it make me look clever? It makes me look ridiculous. Does it work? Oh yes. Oh yes it works.
And you don’t waste time trying to invent mnemonics you think you’re going to find useful. Handy.
Sincerely,
Someone replying to you.
Dear Ole,
I neglected to mention something addressing your other concern- will it be hard to remember kanji without stories, especially when so many similar kanji appear at once?
I set mine to appear in a random order, having automatically generated my deck already. This solved the problem of too many similar kanji appearing at once- rarely were the kanji appearing related at all. I also dumped all 3007 kanji available on koohii into my deck, instead of doing it in “books”, which I found to be an almost arbitrary distinction in the digital world. I think Heisig was misguided to notice to problem of interference and then just lump a load of really similar stuff together at once.
You don’t initially commit new things to memory as fast as you do when you explicitly go through a list of kanji- you have to see them more often first. After that, after “getting in your head”, it’s all the same. None-the-less, I developed my method in response to feeling that mnemonics took far too long, and find that it’s easier to just accept having to see stuff over and over (about five times?) than to stress over inventing stories.
I also, to try to make the best of the ability of pre-scanning material as one does when one applies mnemonics to kanji, just went through the kanji in order, checking their meanings, and writing them out, saying it out loud. I did this for about 20-40 kanji a day, writing them out about 5 times each.
At some point, I assumed, this approach would converge with the other- new kanji appearing would, though I had not remembered the keyword exactly, be “familiar”. I feel that this idea was a success.
The final thing to note is that you are not doing tests, now. The fact that you are on this website indicates that you have an interest in autodidactism. You are not testing yourself, you are teaching yourself. Don’t worry about failure, and don’t think that, because you have to fail more often using the above method, that this means it’s bad. I found it was actually very quick, because though there were many failures, they went by at great speed. I was able, therefor, to come to successes quickly.
Sincerely,
Someone replying to you AGAIN.
I have to admit that I (eventually) tried lazy kanji after a while, since the reviewing using English keywords eventually became kinda pointless (having learnt how the characters were actually used and what they meant). Personally, I like the format, and it was worth the experiment.
Although, Ole, I’d still learn to produce characters from the keyword initially were I starting afresh. Full active recall >>> passive recall, especially when initially learning words.
i am about to finish up rtk1 (less than 200 left! should be done by the end of the weekend) the conventional way (except i didn’t make up my own stories 99.9% of the time — i took them from kanji.koohii.com)
i plan on starting sentences, but i also want to do rtk3 fairly soon (i was thinking when all the rtk1 cards become “mature” in anki, although i might start sooner because why not)
i am wondering, will lazy kanji significantly affect my writing ability? seems like it wouldn’t — once i had hit a certain point in rtk1, i didn’t have a problem writing kanji i hadn’t seen yet within the book, but encountered a lot elsewhere (like 桜 or 円). i am considering lazy kanji (kanji -> keyword flash cards without memorizing someone else’s story, and instead making my own if needed like the guy who responded to ole mentioned) for rtk3 because it seems like it would be quicker now that i have a good basis (and a good intuition for stroke order). also, it seems like there wouldn’t be too many new primitives in rtk3 since many kanji have already been covered in rtk1.
thoughts/ideas/gripes?
as i am thinking about this i like the idea of it more and more. also, if i were to find a nice list of kanji that didn’t appear in rtk and their english meanings (does anyone know of such a resource? it seems kind of far fetched, but it would be nice to have, and i wouldn’t have ever imagined the following rtk has, so maybe it exists?), it seems like i could add those in the deck and continue learning without much trouble/effort, when it would have been more annoying before.
p.s. sorry if this is kind of ramble-y — i was kind of thinking in text as i was writing this i guess haha
Yo Khatz! This is very, very much in line with both my thinking on and my experience of using Lazy Kanji + the Kendo Mod. As you know, I had gotten around a quarter of the way into RTK1 when the fates conspired to make it very difficult to continue that way. Fortunately, you broke out the original Lazy Kanji post right around that time, I did my experiments with it, found a way to hack it so that it would keep it from becoming a visual memory game, and the rest is history. While reccuperating from an illness I put together a deck of all the RTK1 Kanji with my Modification, made it a shared deck, and began working through it. I found I could learn double the number of new kanji in a day and it was completely effortless. No, at first, as I was learning this way, I had a “fuzzy” recall, not perfect and immediate clarity where a story pops in my mind the second I see the kanji and I can give the meaning or if given the meaning immediately write the kanji. That took a few reviews. Further, I think if my only interaction with the kanji was during my Anki reps, I’d need to see it more often than the algorithm dictates or my recall would drop. Because the very parts of the technique which make it easy and truly effortless also prevent you from having to recall as much information on each review, you don’t form some crystalized representation of kanji=keyword=mnemonic story. Instead you have an organic collection of fuzzy concepts that that gain their solidity through USAGE and study. We don’t learn kanji in isolation from the rest of the Japanese language, even if we learn it first. And we aren’t learning kanji so that when given the English keyword on the test we can remember a stupid story and write the kanji out. We learn the kanji so that we can read and write real, flipping, JAPANESE. 日ー本ー語！！！ I’m not here to learn Heisigarian (though my props to the man, he paved the way, for sure, his method is totally valid and workable, thousands of people have used it, yada yada), But anyway, I’m not here to learn Heisigarian, I’m here to learn Japanese. And the only reason I am learning Japanese is so I can bloody well use it, not take some test. It may be that some day I will be required to take a test in or on Japanese, and I’ll do fine just like I did on tests in school. In fact, I’ll do better than the person who memorized the answers, because I will have consumed the language and made it a part of me.
When I was in school I NEVER studied. Not a bit. I read a ton. I wrote papers. I read the assigned reading, then on top of that I read a bunch of other books, articles, papers, etc about whatever from the assigned reading grabbed my interest and made me want to dig deeper. I asked lots of questions and paid attention during lecture and discussions. But not once, in my four and a half years at university, did I stay up looking at flash cards or pouring over a textbook looking at the passages I had highlighted. I didn’t have time for that. I had two kids, both under four, and a full time job working about 45-50 hours a week. But, I could do tons of reading, because I could find audio versions of damn near most of what I was studying (English and Philosophy major here), load them on my ipod and listen like crazy in the car, while I was working, at night while I wrote papers, took care of the kids, etc. Even most of the science courses I was required to tkae to meet my gen ed. requirements had much of the material available as audio. I graduated with a 3.87.
Similarly, I “learned” the kanji effortlessly, almost like osmosis, in a similar way by just working through my Lazy Kanji + Mod Deck which is publically shared on anki and you can read about here. When I was “done” was I fuzzier on kanji than someone in the same position who worked through RTK the “hard” way. Yes. But, it didn’t matter because as soon as I started doing sentences, working through vocab on smart.fm, and doing some really, really serious reading (I participated in, and placed highly, in the Read More or Die contest), as soon as I did those things, all that fuzzy fluid knowledge began to congeal in all the right places, and continues to do so as I learn more each day. I hope it never becomes some static, dead, inert “object of knowledge,” because as soon as it does I am no longer learning, or using the language, I’m just learning disconnected, meaningless and rather boring “facts.” No thanks, I say, I’d rather go read some manga.
to follow up on my last comment and asking if there is a list of kanji beyond rtk
while i didn’t find that specifically, i found an xml file (kanjidic2.xml — the same one that jim breen’s dictionary uses) with a with over 13000 kanji, their english meanings (when applicable), readings, jlpt levels (when applicable), etc. etc.
i got most of the way through writing a script that will parse it into an anki deck (text format — it might be usable with other things, like surusu; i don’t know) when i decided i don’t care to know 13000 kanji and that 3000 will probably be enough (and if it isn’t i can easily learn the special cases when they come up without the rest getting in the way)
that being said, i am close enough to finishing the script that i am willing to finish it this weekend sometime and upload it if anyone else is interested.
so, would anyone like me to finish this and upload the deck?
@kalek
< 挙手>
Anyone has has been learning x language for long enou– scratch that, anyone who has been learning X subject for long enough changes up their X knowledge acquisition methods for the better. Your learning evolves with you. But you gotta start with the original tool and lay the ground work before you can evolve it. (I can hear people saying stuff about vacuum cleaners and brooms and awls and power drills, yeah, I get it, shut up and follow the metaphor for a bit).
Once you’ve gone through Heisig for awhile, the oldskool way (paper and pen if you must), figured out the radicals, made some unforgettable stories and SRS’d them enough time to have your next rep of 日 be 3.4 years from now, then you can start doing lazy kanji-type ideas. Until you get to that point, and 隣 is still only 3 months away, you may want to hold off on going lazy, but you can still evolve and tack on the reading of となり and りんしつ on that card, just to help you along 3 months from now so neighbor doesn’t trip you up with I dunno, 最寄り or something silly.
If you go with lazy kanji from the start, oh fark, it might work, but I doubt it. (same with mono-lingual dics from the get-go and all sorts of other *crazy* ideas 😉 ). But chances are slim.
In any case, uh… (すまん、二つ目の金麦を）basics, don’t avoid the basics to jump to shortcuts which you can’t fully justify or understand yet (unless you have loads of time to waste and it is FUN to waste while doing!, since you don’t, keep reading). The Heisig system is obvious and clearly explained why it is advantageous and (despite school systems) been working fantastically for 30 years+. It also sets a rock solid foundation for learning the Kanji’s writing and meaning. Lazy Kanji is a branch of that, made by Khatz for his own personal evil schemes and may just work amazingly for him (同様に俺は日本のAVだけを見てる) but he had to reach a certain level before making that funky juice work for him. Just like how Cloud can’t Omnislash in FF7 until he’s used all the other limit breaks and gone on some damn quest to find an item to teach him how to use it first. Or something like that.
This applies to everything. Including baked goods.
Especially baked goods （日本には竃はちょっと珍しいでしょう？だから炊飯器でケーキとかを作ったり）
以上
Jaybot7, I personally know several people who, starting from zero, d/led my lazy kanji deck and have since learned all 2000 kanji in that way, without ever even cracking open Heisig’s book.
笑 Did you guys even read the article?
Who cares if it’s “right”, just as long as it works for you.
It doesn’t have to make sense.
Khatzu, I’m going to agree with you completely on almost every point, and disagree on one minor point that doesn’t matter at all. すみません! 🙂
I agree with your Heisig point #2, but I think #1 — breaking down the kanji and learning them bottom-up — is pretty obvious. If I knew that I needed to know all the kanji, that’s exactly how I’d go about it.
I think the genius concepts from Heisig were recognizing that (a) to be good at Japanese, you need to learn all of the joyo kanji, and (b) if you find a good way to go about it, it’s entirely *possible* for anyone to learn them in a reasonable period of time. I never would have guessed that (b) was true, which means I never would have admitted (a) to myself.
Even today, it seems that almost everybody (even Japanese people and Japanese teachers!) think (a) you don’t really need kanji, or you don’t need *all* the kanji, and (b) if you do want to learn the kanji, it’s a long and painful process that will take you 12 years.
When I took Japanese classes for a couple months once, our teacher (a Japanese woman) wrote hiragana on the blackboard but mentioned that it was awkward to read, and that no Japanese person would write like that, except maybe in books for very young children. I never could quite reconcile to myself having to learn to write something that even Japanese people had trouble reading… Good thing I found AJATT!
@kendo sweet! The cycle may be complete then.
Do the best with the tools you have. (I knew I was coming back to the power tools at one point). If The power drill works better than an awl or a broom works better than a vacuum cleaner, *use it*.
The only reason I slightly doubt them (stop listening to me) is the same way I slightly doubt pre-made sentences. Pre-made sentences (that you don’t mine yourself) don’t seem to have the same personal connection and seem harder to remember. 何だか… ね。
I’ll humbly go away now.
Great post as usual~
I can tell Im forming a ‘relationship’ with Japanese because when I spend too long listening to english, I start to miss Japanese. A lot. Which makes it easy to immidiately enter my immersion when I get the chance. Also, according to my computer, Anki is now my most commonly opened program 😀
One last thought on Lazy Kanji (I’m totally missing the point of this post, yada yada), while I’m actually positive it will work great for reading the Kanji for input and reading, it doesn’t have the automatic cloze deletion that normal heisig has. Which won’t ever pose a problem until… at some point… you reach an output stage and you have to write やくそく もくてき or any てき or any かなう for that matter in context and if you *may* have trouble bringing forth the kanji from your head, without any particles or other kanji but a blank sheet of paper in front of you.
It may sound rare now, but it will happen someday, and on that day, you’ll curse out loud and blame me, I’m sure of it 🙂
It’s ok Jay, I am too. lol. TBH I just enjoy a good debate from time to time, and like sticking up for ideas that I’ve adopted, especially when they are unpopular because they are new. I think that’s all well and good as long as the atmosphere remains positive. It’s when debate becomes arguement, when the negativity in the conversation threatens my freedom to experiment, that I simply walk away. Thankfully, that hasn’t been an issue in this conversation because we all get along in perfect harmony here in the AJATT cult. Must be something in the kool-aid. ;P
Anyway, two quick points:
1. You’d be absolutely right, IF, we were learning kanji in a vacuum. But, like I pointed out in my initial reply to Khatz, the Lazy Kanji deck is only the 1st place I encounter and use these kanji, not the only place. If one is doing sentences as Khatz reccomend, then one is regularly writing kanji outside of your regular Lazy Kanji reviews. Further, the SRS is certainly not the only place where one see’s, reads, and writes kanji. If one is truly doing All Japanese ALL the Time (and I recognize the irony of that claim in this post, but bear with me) then one is living in an ocean of kanji. And that’s what Khatz is talking about in this post. Learning kanji isn’t something you do once, get a crystalized piece of information that then sits in your brain to be reproduced on demand. No, knowing kanji EXISTS in the ACT of reproduction. Like I said, I’m learning Japanese, Not Heisigese. Lazy Kanji, or RTK, or whatever, is only a starting point, to become familiar with the kanji and comfortable with them, so that when you go to learn words and sentences that use them it isn’t overwhelming and your mind has a little sticky spot or hook on which to connect with the words and phrases and sentences as you learn them. Anything you need to be able to write will eventually be there. If you have to go look it up this time, you put it and some examples and stuff in the SRS to get some extra practice with it and so you make sure that next time you need it, that behavior module is ready to run and you can whip that kanji, that word, that phrase out exactly how you need it.
Anyways, that’s a pretty long “1,” so I’ll leave “2” and “3” for another time (I didn’t even have 2 and 3 in mind when I wrote “1,” so your not missing out on anything. I was just gonna make some more reasons up. lol. yeah, I just lol’ed at myself.)…
See this? This is me not engaging in debate (特に英語で、日本語だったら多分):
I agree with everything you said 🙂
ちょっと注目だけをしたかった。といっても、元々カツ正しい。プレイをしなくちゃいけないよ。一杯。
So old i know buuut..thought i’d add a 2012 viewpoint on things.
1) It amazes me how fiercely ppl will guard certain ideas, especially if those ideas aren’t their own. Heisig is great..but Heisig is also so 1970’s (no offense to him). Ppl learned kanji before Heisig, and thousands of ppl learn it without him. Point being, its not “Heisig or nothing,” as some comments are trying to dictate. Its not SRS or nothing either; to be frank, all you need to do is expose yourself to a language with the *intent* of understanding it. I’ve known ppl who went to Japan, no Heisig or SRS’s at all, and learned the language within a year. Yes thats possible. No its not magic. Its common sense. A guy could be wringing his hands and pointing a blazing house behind him, and I wouldnt know a single word he was sa ying; yet, I bet I could *understand* what he was trying to communicate to me (omfg my/that house is on fire HELP!) As you hear/see a language enough, as in get enough exposure, you’ll be able to figure out what all the individual sounds/words mean. But that comes with exposure, thats all. Just simple exposure. You dont need fancy tools or clever intellectual devices for that…you just need to DO it.
2) Problem with most ppl is that they don’t want to fail. Because of this, they spend most of their intellectual time inventing “fail-proof” mechanisms. Of course nothing is fail-proof, so when it fails, they inevitably toss it aside and call it “ineffective,” or “not useful,” or “xyz is better.” Here’s a hint..you will fail. Thats just the bottom line. Get used to failing, because everyone fails during the learning process (especially with something as nuance heavy as language). Rather than trying as hard as possible to not fail, just accept failing. Who cares if you can’t recall that kanji for the 5th time…i bet it looks way more familiar than it did before you started. You know what else? I bet you’ll remember that kanji better than any other kanji you’ll ever study. Fact is we learn from our failures, and our failures tend to stick out in our mind better than our successes. So use whatever method you find easiest to do. Thats it. Failure rate is entirely irrelevant, because failure is inevitable. Just because you failed recall with Hesig 1 time before learning, and the other guy failed it 5 times before learning on a another method, doesn’t make your method better. After all, 3 months from now, whos to say you’ll remember it better than the guy that failed it 5 times? Just do whatever is easiest (it doesn’t have to be hard, seriously) and enjoy learning.
This is horrifically long so i’ll hush now. All of this was to say that there is no debate. I mean really, you can’t argue something doesn’t work when it, in fact, does work for people. That is just silly. Any method works. I mean it. Anything works as long as two things are true:
1. You are actively learning (trying to understand and comprehend)
2. You are doing it
Its really that simple. How did you learn your native language? SRS’s? Heisig method? Cramming? Note cards? None of that. You just exposed yourself to the language (not like you had a choice), and your childlike curiosity allowed you to learn. Learning a 2nd language is no different; it only seems harder because we’re adults now, so things have to be 10000x more complicated because we are *supposedly* more intelligent. I find that debatable. Just do something, anything, and try to learn while doing it. Eventually, you will learn. Period. Full stop. End long dissertation.
It sounds like you think your opinion would not be well received, but to be honest, I agree with you. I think you’re right that Heisig is not the only way to learn kanji, and an SRS is not essential to being able to learn vocabulary. In fact, I would take your first requirement for learning – You are actively learning (trying to understand and comprehend) – and actually say that I think for some of the time, even this is optional.
However, I would at the very least like to defend the idea that Heisig and SRS aren’t useless. It’s not that Heisig is the only way through the kanji, it’s just that it’s a way that implements a logical breakdown of the kanji mountain. No, it won’t work for everyone and millions of people have successfully learned the kanji without it; it’s just one way that some find useful.
Secondly, I would like to say that I completely agree with you on the fact that people are afraid of failure, but in order to grow and develop, failure is necessary. I would also like to note that SRS actually hinges on people’s ability to fail. If you want to use an SRS, you’ve got to be capable of failing hundreds, or even thousands of times (I’m good at this 😉 ).
Anyway, Heisig’s RTK and SRS are just tools in a person’s learning toolbox; useful, but not entirely necessary or the only way… at least that’s my two yen.