RECENT TOPICS » View all
Hi everyone,
I am looking to start RtK soon, and would like some advice.
On the one hand, I think it's worth investing a bit of time to make sure I am setting off in the right direction, but at the same time I've found it discouraging to see so many conflicting opinions about the "best" way.
I've been reading a lot on these forums, and I didn't want to post a new topic about this, but I have probably spent at least 30 hours just reading people's opinions about how to study Japanese and Kanji, and very little time actually studying Japanese. The trouble is, a lot of the advice seems to be aimed more at people who have either finished RtK or are at least a good way through it. For example, some posts have some great information, but honestly it's a bit hard to understand for someone who doesn't know what all the terms mean yet.
In any case, here is some of the advice that I've picked up so far, based on what people have said on these forums and other places:
- Use an SRS system. Done. I've started to use Anki for other things I'm studying too, and I like the theory behind it. Time will tell if it works as well as I hope it will.
- Just do RTK1 for now -- don't try to supplement it with Kanji readings or other things that will slow you down. I can see the logic in this.
- Stick with the original keywords. At first I was interested in some of the lists of alternative (improved) keywords, but I think that's because I had some notion of the keywords giving me deep insight into the meaning of each Kanji. However, after reading more I got the impression that the keyword is just something abstract to hang further learning on and you're never going to find a single English word which perfectly corresponds to every Kanji anyway.
- Go at a reasonable pace -- not too quickly. As I haven't started yet, I'm not sure what pace is going to be best for me, but hopefully I'll settle into some kind of routine.
- Don't try to learn a tonne of vocab at the same time -- you're better off learning it once you know the Kanji to link it to. Again, I can see the logic in this.
Feel free to tell me that some of this advice is silly and I shouldn't listen to it.
A couple of the things I'm still a bit unsure about are:
- I found some cool spreadsheets. Should I pay much attention to them at this stage? As I said above, some of the advice I read was to stick with the original keywords. Should I worry about errata, or changing the kanji order so that "helpful kanji to learn earlier" come before the kanji that feature them as components? Or is it just easier to not worry about these things as I go through for the first time?
- Which Anki deck is best? Some have lots of fields with helpful extra information, but obviously if I'm just trying to do "keyword -> kanji" then it's superfluous, at least for now. Is there any point to choosing an advanced deck now, modifying the card layout to only show the keyword/kanji fields for now and then if I want to later I can add other fields in?
Any help would be appreciated. Thanks!

ファブリス wrote:
LOL
Haha, yes I agree, just do it, although it's really not as scary as the photo suggests. Really.
It looks fun, though--weeeeee!
If you have RTK1 and anki, then you're all set. You need pen and paper, too. Also, I would suggest learning the kana, and then start learning grammar. Build up your vocabulary, too, if you want. RTK1 does not really need to be done on its own. Maybe just make sure that the other activities don't cut too much into your RTK1 time.
As everyone has said, you just need to do it. The way you do it isn't nearly as important as the fact that you do. If you've spent hours reading about how to do it, those are hours that would have been better spent actually doing it. Thats why people have given such short answers; it may seem unhelpful, but the time your spending reading these posts (including mine (Seriously, stop reading
)) is time that would be better spent actually doing RTK. The reason there is lots of conflicting advice is because people find different methods easier, so you should experiment and find what works for you.
In answer to your questions, I think most of those spreadsheets are best ignored for now. The only ones that look useful are the ones that are labeled "While going through RTK" , but you can ignore them too if you want. I wouldn't reccomend breaking from the RTK order (though you don't necassarily have to do the whole thing) because it makes the kanji alot easier to learn. If you want you could do RTK Lite, though.
I didn't use anki for RTK, so I can't help you there.
P.S. Lol at that picture XD
Last edited by Splatted (2011 April 20, 10:30 am)
Even if you don't do any new cards at least do your reviews. Stay Consistant until you are comfortable ramping it up. Being Consistant will get you farther faster.
What I can tell you is that you should never put yourself under stress DOING the cards. Just do it. No matter how much. See, if you do 10 kanji everyday -- hey, calculate that per year!
I disagree with the sticking-to-the-Heisig keyword idea, since you can easily customize keywords on this site or @ home with Anki.
Use Halpern's keywords instead, they are more accurate. Dr. Heisig - despite his genius - is no lexicographer.
And do away with one keyword per kanji: you want to deal with polysemy as early as possible and you need multiple keywords for some kanjis. It helps disambiguate glyphs with homosemic primary meanings.
It also solves the problem of the near-synonyms that Dr. Heisig chose to adopt to keep up with his rule of 1 different keyword per kanji. This was clearly wrong, as it often introduced nuances in English meanings that were totally spurious in the kanjis.
One last thing: you know that RTK does not teach you Japanese, don't you? It is just astonishing to me to hear of people that expect to be able to read books in Japanese after RTK. RTK is just the first step on a long, long road.
louischa wrote:
And do away with one keyword per kanji: you want to deal with polysemy as early as possible and you need multiple keywords for some kanjis. It helps disambiguate glyphs with homosemic primary meanings.
I think that's a rather confusing advice. If I had had to rate my knowledge of several keywords everytime I drew a kanji, I would probably still be doing the thing. See, you're bringing up context when the idea is to learn the basest form/morpheme possible. That's against the spirit of the method (counterproductive).
Which is probably why Heisig based his keywords on etymology and not lexicography: it can be argued to facilitate compounds learning afterwards (no overlap on the previous signified: real world wins, keyword fades).
Last edited by EratiK (2011 April 20, 7:44 pm)
EratiK wrote:
I think that's a rather confusing advice. If I had had to rate my knowledge of several keywords everytime I drew a kanji, I would probably still be doing the thing. See, you're bringing up context when the idea is to learn the basest form/morpheme possible. That's against the spirit of the method (counterproductive).
Not at all, and you are totally ignoring my point about synonymy. Multiple meanings makes story-smithing much easier, and if you look closely, Dr. Heisig himself encourages polysemic references for several glyphs: ri <-> computer, and many others. It is therefore part of the Heisig method, and is certainly not counterproductive. Sooner or later, you will have to learn those extra meanings, so better to familiarize yourself with them at an early stage. The availability of spreadsheets with Halpern meanings makes the idea very easy to implement.
louischa wrote:
Not at all, and you are totally ignoring my point about synonymy. Multiple meanings makes story-smithing much easier, and if you look closely, Dr. Heisig himself encourages polysemic references for several glyphs: ri <-> computer, and many others. It is therefore part of the Heisig method, and is certainly not counterproductive. Sooner or later, you will have to learn those extra meanings, so better to familiarize yourself with them at an early stage. The availability of spreadsheets with Halpern meanings makes the idea very easy to implement.
Isn't the point of RTK to remember the Kanji rather than 'know' them? I could imagine coming across the same Kanji with a different meaning later on in my studies much better than learning all of them at once. Plus I'd believe it's a lot more effort to create elaborate or over-complicated stories.But in the end it's up to the individual, if you find it easier one way then go for it!
Last edited by raz789 (2011 April 20, 8:38 pm)
Okay, my last post for this thread. This is getting tiring. It is impossible to offer advice on this forum without being immediately misinterpreted and contradicted by Peter, Paul and Mary.
Look at the stories for 曲. You see that many story writers are aware this has a dual meaning: "musical composition" and "curve". If you look at these stories, they easily incorporate both meanings and are neither overly complicated nor long. You can do it in one sentence only.
All I am saying is to type "MUSICAL COMPOSITION" and "CURVE" in the bloody "CUSTOMIZE KEYWORD" box. This will not cause you to finish RTK1 in 10000 years, nor will it violate the spirit of the method, nor will it cause your brain to melt of exhaustion, because this is what Dr. Heisig himself is doing when he suggests to link 与 ("BESTOW") to an extraneous meaning of "SLINGSHOT".
Then you'll thank yourself for two reasons: A) you'll have two meanings instead of one to play with if you encounter the same glyph in a later kanji (which will facilitate story writing - a fact I experienced first hand), and B) when you'll study real vocabulary, you'll know how to disambiguate the meaning from the context. Much easier to do than to learn that what you thought was "BEND" could actually mean "MUSIC"! Whaaaaat???
Is that more clear? Signing off now.
louischa wrote:
Okay, my last post for this thread. This is getting tiring. It is impossible to offer advice on this forum without being immediately misinterpreted and contradicted by Peter, Paul and Mary.
I think they just have a different opinion (though that does not mean that they have not misinterpreted or contradicted you). The posts seem ok to me.
Thank you to those encouraging me to just get cracking. I'm certainly looking forward to it.
I think some debate about these things is good. People such as myself want to know the opinions more experienced people have before devoting hundreds of hours to something.
I saw other people point out two issues with using Halpern's keywords elsewhere on the forum -- they are not unique, and they are not always as memorable. louischa, if I've understood you correctly, you're suggesting that you should use multiple keywords instead of trying to avoid duplicate keywords for kanji with similar meanings?
Some people seem to think it is worth a few minutes of extra effort (searching Katsuo's great spreadsheet) to just patch up the most glaring inconsistencies. I'm still somewhat interested in this, but I don't want to end up in the situation where I can't just "study another 20 kanji" because doing so would require me to sit down and mess around with searching a list of keywords to choose better ones etc.
Also, if I did use different keywords, is it possible that I'll run into problems later on e.g. if I am using a shared Anki list?
I suppose I should ask the following questions:
- Has anyone here successfully completed RtK using alternative keywords from the start, and can say that it has helped them in the long run?
- For those who have successfully completed RtK with the original keywords, did you have much trouble adjusting to alternative meanings later on?
And louischa, yes, I realise I won't be able to just read Japanese after completing RtK.
codebeard wrote:
- For those who have successfully completed RtK with the original keywords, did you have much trouble adjusting to alternative meanings later on?
(I did it with the original keywords)
Do you mean like learning the original keyword for 足(Leg) but out in the real world see it having the meaning of both leg (あし) and the meaning of being sufficient in 足る? If that's the question, I don't have much problem with it. When I see new vocab with familiar kanji, I consider it a bonus that the meaning is similar/identical to the connotation of my keyword, not an expectation.
codebeard wrote:
- For those who have successfully completed RtK with the original keywords, did you have much trouble adjusting to alternative meanings later on?
Short answer, no.
Just keep in your head that kanji aren't words and heisig keywords aren't meanings (although they're pretty good most of the time). Smack through RTK as quickly as you can (without burning yourself out) and move on.
zigmonty wrote:
Just keep in your head that kanji aren't words and heisig keywords aren't meanings (although they're pretty good most of the time). Smack through RTK as quickly as you can (without burning yourself out) and move on.
Ditto.
And in my case, yes, I think my brain will indeed melt if I use multiple keywords for each kanji. But that's just me.
louischa wrote:
Okay, my last post for this thread. This is getting tiring. It is impossible to offer advice on this forum without being immediately misinterpreted and contradicted by Peter, Paul and Mary.
Hey dude, I was just pointing out my own opinion, I'm not here to convert anybody. It's not my fault if your ideas are opened to criticism; that's what a forum is all about.
All I am saying is to type "MUSICAL COMPOSITION" and "CURVE" in the bloody "CUSTOMIZE KEYWORD" box. This will not cause you to finish RTK1 in 10000 years, nor will it violate the spirit of the method, nor will it cause your brain to melt of exhaustion, because this is what Dr. Heisig himself is doing when he suggests to link 与 ("BESTOW") to an extraneous meaning of "SLINGSHOT".
Well that's debatable, and in my perception of RTK, the more simple the better. But that just my style. Plus your Heisig example is poorly chosen, since bestow and slingshot are two different symbols (additional stroke). Polysemy only applies to identical signifiers.
It's something a beginner would have to figure out on its own (I was pretty much set after kanji 750)(original keywords). Because over 2000 kanji, you'll see for yourself what works and what doesn't, and you can always adapt. For example, I did most of the (few) changes on the go. Some keywords (especially the primitive ones) won't seem fit to you. It takes less than 2 minutes to find a better one (even less if you consult an etymology or lexicography list), and another less than 2 minutes to see if it doesn't pop out later. Some say it breaks the flow, but others say they welcome the break. You can only find that out through practice. Good luck! ![]()
Last edited by EratiK (2011 April 21, 4:40 am)
I went through RTK a few years ago, and the spreadsheets I made in the "While going through RTK" section are things I would personally have found useful. Of course that doesn't mean everybody will.
The Kanji Name Comparison spreadsheet may be useful if you normally have a computer handy while studying. It can be put in RTK order, and a quick glance will tell you how keywords from other sources compare to RTK. If the latter appears unusual, you may wish to investigate further before choosing/creating a story. Halpern gives the best overall meaning(s) for each kanji, but I would usually stick with the RTK keyword while bearing Halpern's in mind in choice of story/ nuance, etc.
In case it's inconvenient to use the spreadsheet above, I posted a short list of the main keywords I'd consider changing (which you already linked to above).
E.g. one of the strongest candidates for change is 校 number 1278 "exam". Much of the common vocabulary for this character relates to "school", and little or none to "exam". "School" isn't used elsewhere in RTK, but "examination" (easily confused with "exam") is used as a keyword. Conclusion: change the keyword to "school".
Changing RTK keywords brings some consequences, e.g. you have to check the new one isn't used later in the book. Also you can't use the stories on this site (though a few people have submitted stories for alternative keywords).
Regarding the other links in my "While going through RTK" post, the first one can be useful for simplifying characters, and also gives you some extra kanji "for free". However, they are mostly less common characters.
The second link gives details of unnamed combinations, e.g. after 品 "goods" & 木 "wood" appeared together for the third or fourth time (215:燥;671:操;1369:繰;2035:藻) I wished I had known and combined them to begin with. Not all unnamed combinations are as common or useful as that one though; I listed every one I could find for the sake of completeness.
Incidentally, I often added the information above in the relevent story section on this site, but the links above collect it all together.
codebeard wrote:
- For those who have successfully completed RtK with the original keywords, did you have much trouble adjusting to alternative meanings later on?
No, not at all, the keywords are meant to fall away anyway. I did change many of the "primitive meanings" names, though. Just be careful if you plan on doing that. If you change a primitive elements name, you will have to use it in later stories as well, if a kanji contains the same primitive element. For instance 使う the left part is called "person" by Heisig. You can change the name to Mr. T if you like. Or you can name it after one of your parents, or your girl / boy / best friend. Just remember to use the "person" in all the stories that contain that "primitive element."
And as the others have told you already, just do it, and don't worry. It is easy as 1, 2, 3. Start with 一 This is the roman number 1, lying on the side, which is written from left to right. And work your way up to the magic number 2042. Or, what i would suggest, do the supplement as well - containing the new list of official use kanji.
There are not many mistakes you can make, you can easily make them without even noticing it. So if you are uncertain about something, ask, ask, ask! Even if you think it is a stupid questions, ask, ask, ask! And have fun, because it is a fun process, once you get the hang of it. And it is also one of the easier parts of learning Japanese, making yourself familiar with the writing of kanji. ![]()
Last edited by Nagareboshi (2011 April 21, 6:18 am)
I went through the books with the original keywords, and they were okay. I don't think the Halpern keywords are any huge improvement, if an improvement at all, from the ones I've seen, but they might help a bit in the short term.
Using compound keywords might help with all the various similar keywords. I get a bit frustrated from time to time with string, cord, thread, etc, whichever of those are true keywords.
I don't keep track a list of the keywords, so please spare me the comments on which of those are and aren't keywords.
Over the long haul, any of those approaches should be okay if you get through the book. When I see 使, I think つかう, for example, not the keyword, which I've forgotten at the moment. I think that'll be true of any approach, and the various usages of the kanji will come over time.
IMO, the only keywords you'd have much use of changing are noted by the community in the story section.

