Not sure wether this is the right location for this question so sorry if im on the wrong place.

Anyways , My kanji study so far :
I study Kanji using stories that include the Meaning and the ON yomi to save time assuming that this would teach me how to read kanji much faster.

But now that I reached kanji 200 and being totally exhausted to a point were I limited my study to 10 kanji a day , I started to wonder is this method really faster? I tend to remember the meaning of the kanji but the on sometimes goes beyond my head and just simply forget it, but after allot of repeating I eventually get the on and the meaning in my mind.

BUT that takes more time than it sounds.

What do you Master Kanji studying jedis think is faster :
Learn the meaning first and later on study the On?
OR learn the meaning and the ON at the same time ?

The great thing about RTK is that it can be done quickly (within say 3 months), so you can get on with the real learning. So if it's slowing you down too much ditch the ON yomi. If it takes much longer to complete RTK, you may as well just be studying kanji the traditional way in parallel with other study. I learned ON yomi with the kanji and didn't find it so hard but I wouldn't consider it essential. You could always do it on a second pass if you wanted. Alternately you could just learn it for those for which it's easy to include in the story. Get RTK done quickly, and if you can remember an ON yomi for the majority (or even just some) of the kanji then that's an added bonus. If you can't it's no big deal. One thing that helped me was using a consistent mnemonic for each ON reading. I just made them up as I needed them or borrowed them from a list I found online (google kanji damage).

10 a day is a little slow if RTK is all you're doing. The faster you can do RTK the less important it becomes in the long run because you'll start remembering the kanji more deeply through actual use. I stopped reviewing RTK after I had about 95% of the cards in the fourth or fifth box on this site's SRS. I've forgotten a lot of the stories. They really don't matter once you've learned enough readings and vocab, so don't take forever on RTK. Also I don't recommend RTK3. I did it and it's largely a waste of time. RTK1 is enough to demystify kanji and then you really need to move onto expanding your vocabulary.

At the beginning, I used to concentrate on both ON and remembering the meaning. The pro is that it's easier to remember the character if you remember the reading because often it's in the same 'grouping' and parts of it may be contained in another character. The con that it can be slower. I did both up until about 1400. Now, I put more emphasis on remembering the character and don't put so much emphasis on the ON. The ON is just in my deck so that if it stuck easily/effortlessly, I will continue to remember it.

I used RTK1 and left the readings for later, and that seems to have worked well. I think that trying to do too much at once is a bad idea.

As you learn words in Japanese you'll find that you'll pick up the on-yomi readings for Kanji as you go along. You'll run into more and more compounds with the same kanji and the reading will sink in. The reason why doing it like this is better is because if you look up the readings in a dictionary you'll find every possible reading however some of those readings may be so rare that you only use them in like 3-4 instances. For instance, 性, the most common reading for this is as the first kanji in a compound is せい, even as the last it is pretty common. However there are a handful of instances where it is しょう. So you'll spend the time learning both when really you could probably get away with just learning せい. There is an even better example of this that I found where saw this one kanji as the same reading again and again, and finally I ran into this one uncommon word that it is in and learned that for this SINGLE word, it has a different reading, so I'm always failing it in my deck.

Sorry for the huge bump but it's a good / often asked topic, I think. I'm in a similar boat as the OP, where I questioned "Why do RtK and simply learn the meanings when I can, at the same time, learn the readings as well?" As I started to learn the readings along with the kanji and its meanings, though, I found that a lot of the time I would see examples where neither readings were used, or it was a little daunting trying to remember 5 different Kun readings, so I'll try a continuation of RtK I think.

For those who have completed 1 and moved on to reading or vocabulary, do you have a set method for learning the readings, or do you just come across a word with the kanji you know, learn the reading for that specific word / compound, and move on to new words?

When through RTK1, a lot of people seem to advocate just learning the readings that occur in words as you come across them, and are very happy with it. For myself, I find that this works beautifully for Kun readings, but for the On readings I am finding a systematic approach means less forgetting. Search for all the old threads on 'movie method', 'memory palace', 'kanji chains' if you are interested.
I have started with a very loose version of the movie method. I work with some pretty sloppy stories, in the hope of charging forward and not wasting too much time on my method. One disadvantage of this has been that I often get some interference between my RTK stories and the On story. I may find I have ended up using a different nuance or even different meaning of the keyword in the 2 stories, and it can be slightly confusing. Even if not, I have to remember the first story for the meaning and then jump to a different story for the reading - at least until it becomes more automatic and the stories themselves start to fade into the background.
In short – movie method/kanji chains/whatever = OK, but I am beginning to feel that something like what is described below by nadiatims might be simpler and cleaner.

nadiatims Wrote:I and others have had some success with this and I think it's worth trying even if you abandon it later. Most common kanji have more than 1 reading (sometimes as many as 10 or so). I would advise only learning 1 common on-yomi reading for each character. The on yomi reading is the chinese derived reading of a character and is used mostly in compound words. These readings are mostly monosyllabic so it's easy to make mnemonics for them. So as you learn new kanji, look up the on-yomi (usually written in Uppercase or Katakana in dictionaries) and make up sound primitives for each new on-yomi reading and incorporate them into your stories. An example would be 缶 which has the on
-yomi KAN. I used the sound primitive Ghengis KHAN (mongol warlord) to represent the sound KAN and used him in my stories for any character read KAN.

http://forum.koohii.com/showthread.php?tid=6294


Is this what you have been doing? If so you might consider continuing, but, take her advice (in this thread), and simplify – Kun readings and more than one (or occasionally two?) On readings are going to bog you down and don't seem necessary. On the other hand, the advice about racing through it, I'm not sure about. If you are learning Japanese as well as the kanji, I say - relax. Even if you were doing 6/day you would still be through in about a year. Take longer if you want. It's not a race. At that point, having mastered over 2000 kanji for recognition and writing +/- some important readings would not be a bad effort . You will also have picked up a lot of Kun words in that time, I expect.



I think the important point is, quoting Katsuo in the thread above, "The main disadvantage of using those methods is that you are more on your own and won't be able to use most of the stories on this site ". I'm sure I would not have had a hope of getting through RTK1, without relying on other people's RTK1 stories. Probably, that's why not many people don't try such an approach the first time through - because the idea isn't very obvious in the early stages and the group support does not exist, and because it may be more work in the short-term (especially if you try to include too much), not to mention the fact that there is no clear pattern to follow when choosing images to represent your On reading. It is a pity though. Perhaps if someone were to post useful stories and a few more took them up and added to the effort it could produce something worthwhile.

PS: Another quote from Katsuo "Somebody (not sure if they are a member here) once made a list of sound mnemonics that can be used to associate kanji with their on-yomi." http://www.ravenbrook.com/project/etp24/...mnemonics/
And there is the kanji damage site for ideas.

Plus: I am wondering why it is considered bad form to revive an old discussion thread. Would it be better to start something new? Surely not ?? Discussion about this sort of stuff already seems too scattered. Don't understand...

Great reply, rachels. The advantage of rtk now is more apparent to me after doing some self study. Like I said, once I came across the fact that kun or on readings were sometimes numerous, then the idea of learning readings per word became more appealing. I think if you go into rtk assuming that there are just a couple of set in stone readings for each kanji then you are more inclined to try to learn the readings.

And the stories for kanji that people post here are indeed helpful. If only they could be exported to the rtk iOS app easily!
Edit: I just discovered the extremely useful csv function so ignore that last part.
Anyway thanks again.

pompom Wrote:Sorry for the huge bump but it's a good / often asked topic, I think. I'm in a similar boat as the OP, where I questioned "Why do RtK and simply learn the meanings when I can, at the same time, learn the readings as well?" As I started to learn the readings along with the kanji and its meanings, though, I found that a lot of the time I would see examples where neither readings were used, or it was a little daunting trying to remember 5 different Kun readings, so I'll try a continuation of RtK I think.

For those who have completed 1 and moved on to reading or vocabulary, do you have a set method for learning the readings, or do you just come across a word with the kanji you know, learn the reading for that specific word / compound, and move on to new words?

From what I understand RTK is designed for remembering Kanji. RTK doesn't include the ON and KUN readings. I have learned that learning the KUN reading on them own can be more tedious than learning the Kanji. I suggest to study the KUN and ON reading independent of the RTK. It could be use as practice for remembering to write the Kanji correctly. There is a book called 日本語能力試験　漢字ハンドブック that can be used. The new version publish 2011/04/11 will have all the new Kanji, and is available from Amazon Japan.

My method is backwards to that used in RTK.

RTK2 covers the on-yomi for the RTK1 characters. I thought the sections on the sound primitives was useful, but other people prefer learning pronunciations in context. I have no idea which works best, but learning words in context at some point is extremely useful, IME.

Check out Textfugu's approach to learning Kanji (there's free lessons).
Maybe you can adapt your own method a bit.

I found that raw RtK wasn't "it" for me either.

I have the on-yomi on the question side, and don't usually pay much attention to it; but sometimes I work it into my RTK story, after I've failed a card by writing a different kanji than the one wanted. Even though it's on the question side, new on-yomi words containing those kanji do seem to be sticking more easily than other vocab. It's hard to predict whether this would be worth the time on a larger scale.

On further thought, using that method from the beginning would possibly make RTK easier. So if it also helps with learning vocab, does that mean you could get the bonus for free or better-than-free...? (Is it likely to do the writing memory any harm? In some cases it will give a hint as how to write the kanji.)

----

For example,

Question: argument ロン
Answer: 論
Old mental image: Socrates having an argument at the botany conference.
New mental image: Socrates having an argument with Ron Weasley at the botany conference.

Of course, you can do whatever you like and if learning both at the same time feels better to you then go for it, you'll get there eventually either way.

But, for pure efficiency, I feel that it is much better to learn the meaning alone first. 'meaning alone' is a deceptive phrase actually - you are learning simultaneously the elements that make up a kanji, the order that they are drawn in, the final appearance (and unless you are taking great pains with your font settings you are probably getting a mix of 'drawn' and 'print' styles as you move between sources and/or draw from memory).

That's quite enough to memorize already without adding a totally unrelated sound into the mix.

Now, you might think that you're doubling your efficiency but many people find that learning ON readings is pointless anyway and instead learn vocabulary words as a whole ('these kanji spell this word, this word has this pronunciation', instead of 'these kanji are pronounced thus and thus, and pronounced together make this word') ... if you find yourself learning that way later, your memorized ON readings will be mostly wasted.

Even if you -do- use your memorized ON readings, I think it would be more efficient with study time to memorize one and then the other - if you drop a 98% recall rate to a 90% recall rate, then you have 5 times more failures - 5 times more reviewing to do!

That's much worse off than the 'twice the work' that you're trying to avoid. Now, I don't really know that those numbers are accurate for anyone (I don't have a study available) but my own experience is that my recall rate working through RtK is over 99% and my recall rate with traditional flashcard memorizing ON/kun/meaning was more like 80%.

Trying to say what that means for anyone else is impossible, of course, but the point I'm making is that you should think about how much more reviewing you are adding. For me, even a 1% increase in my failure rate would be doubling my review efforts and completely wipe out any imagined advantage.

From experience, it's smart to studying things separately. You focus on one thing at a time. I did RTK then I went onto the sentence phase and I later added a vocab deck and a production deck. I believe once you reach a certain level, then you can probably handle a bit more Jp learning but that's only when you've reached a level of comfortable with the language(I.e. advanced reading/listening/kanji and can withstand long-hours of listening/reading just like in your native-language)

Old thread I know, but I would like to share my experience. I finished RTK1 2 1/2 years ago and left it so I started again and I just finished RTK1 + supplement with 1 onyomi each. My strategy was simple, follow heisig/community primitives and give each onyomi a location as I came across it.

Before I started
1)I made an alphabetical list of the onyomis with the number of kanji that use them so I'd know how vivid and distinct a place should be.
2)I also wrote my own flashcard application in vb (awesome programming for babies) using the kanji lists online from the super fantastic Katsuo (really thankyou) and kanjidict. That way I could review 3 sided cards: Question - symbol - onyomi, and choose my own intervals 1-2-5-10-20-40 days, and a study section with kanjidic entries and one-click links to this site and jisho.org common word search. And the question could be anything I wanted so I very often just used the kunyomi, or keyword and onyomi for similar keywords, or a japanese example sentence to distinguish hot weather and hot items and if at all possible no English. I'm also working on a basic android counterpart but I am lazy totally new to java (and god damned case sensitivity) so I'm going slowly.

For each kanji I would:
1) jisho.org common word search to find the best onyomi
2) same search to check heisigs keyword suitability and change it if i wanted.
3) check RevTK for stories or inspiration for my own and set it in my onyomi location.

After studying a load of new kanji I would review them about 3-4 times in one session, almost until the story wasn't needed, before letting them climb up the SRS staircase to long-term memory.

From my previous run I knew that making things at all similar was a disaster, a clear visual minds eye image of a physical object for each primitive was the most memorable for me, nothing abstract like "stretching" instead I used those extending zig zag mechanism cherry picker since it looks like the primitive, and "illness" instead use sick caveman like the community, footprint/stop I used a bright orange traffic cone instead. Using my own primitives did make many stories unusable for me, but then I would have a clear graphic picture in my head and not a lot of the community stories provided something that satisfied me anyway.

The advantages are that now I can type the onyomi into my computer and the kanji I want pops up, no keyword or radical searches like before, also I can read most words and guess the meaning IF there is no okurigana, I am no longer scared of my denshi jisho and finally learning new words is really really simple if you know the 2 kanji and their readings. The disadvantages are that the kanji by themselves are still rather useless, there is no wonderful rainbow at the end. so my next step is to learn at least one real word for every symbol, so like 800-1000 words of 2-4 symbols each.

I was getting like 99% on almost all my not-new reviews so I started marking cards wrong if I'm slow of the reading if slightly off, so recently I've been getting 90%-95%. I honestly do not remember the symbols from last time but the kanji did feel very friendly and easier to work with this time round, I think my brain's kanji mnemonic lobe was very well practiced and I had good experience picking good stories for me.

If anyone would like the app they can have it and maybe the source code if I have the strength to give it away, I'll just got to make it cleaner and put in loads of comments and big big credits to Heisig, Fabrice, Katsuo, and Jim Breen. It's no RevTK or Anki but I enjoyed making it and it works.

EDIT: As for speed, I found it took about the same time, some primitive/location combinations are inconvenient, like a boat nowhere near water or a mountain in a train station meaning stories take more time but for 90% the onyomi is only a backdrop, there is no extra primitive so it adds very little extra difficulty. And for determining where an onyomi should be, just use anything, I used sen as a train station's platform, eki as the main lobby where you buy tickets, shou has 69 kanji so I used my dads big back garden, kan had 40ish i think so I used my big local shopping centre, I don't want stories to overlap too much. Kou, 60 sothing, was my japanese house's living room which was a bit of a mistake actually, too small, so I started making all my primitives 6 inches tall.

How much time do you have per day to study? That'd be a big factor in making this decision, IMO.

The first time I was unemployed and it took me 6 weeks with breaks, pretty much 60-70kanji in 5-6 hours a day before I'd burn out and random days off. Second time I did it between living a life, so reviews everyday and mostly 6-7 hr days, 50-100 kanji, (my max was 130 in a day) and mostly 0 others, 12 weeks in all and that was up to 2300 kanji, 2199 rtk1+supplement+high freq extras.

I would have done exactly the same if I were doing it in 7 hour days or 1 hour evenings. I just thought I've got the magic heisig imaginitive memory skills now to fit in onyomi with very little extra effort so why not. Time per story was probably a bit less due to experience, time per kanji was possibly a bit more due to word searches and keyword adjustments.

If you're doing rtk at home along with a textbook Japanese course then the pros and cons have that have been discussed, heisigs order is not for frequency or joyo grade, it is for the easiest memorization of all 2200 in one go.