I don't know if kanji is the 'hardest' aspect of Japanese, but not out of any conviction to the contrary; I just don't like to look at any of this stuff in terms of difficulty and rather as just something that has to get done. That said, Kanji are definitely a huge obstacle standing in the way of learning an already difficult language (at least with Chinese, the grammar is similar, and the characters typically only have one reading). I think kanji also serve some rather obvious benefits, but that's neither here nor there. Regardless, I'm constantly bombarded with reports on absolutely raging illiteracy from Japanese-speaking non-natives, including those actually living in Japan, to an extent that just doesn't exist for any other language. Naturally almost all of this IS second-hand, but trying to speak with Japanese people and Japanese learners online in the past has found overwhelming use of romaji by both parties, even in private conversations when I clearly have kanji in my profile, which has reinforced the idea for me of the zero-expectation of actual literacy in the language. As a member of the 'Japanese-learning' and hopefully a future member of the 'Japanese-speaking' groups, I find this frustrating and bordering on humiliating, and hope to strive for hyper-literacy just as a counter.
So, whether Kanji are a particularly difficult component of the language or not, they are apparently difficult enough.

It's also worth keeping in mind, when considering the opinions of those who have learned the Kanji but still struggle with mastery over the language, that everything seems easy after you've learned it. I can't think of anything I've already learned in Japanese (including, now that I'm nearly finished with RTK1, the kanji) as being anything other than absolutely basic and natural. That didn't stop it, however, when I stop to remember, from being mind-shattering the first time around...

By the by, if you're learning kanji by sight as you discover them in words, how are you learning to write them? Do you look it up when you encounter it and memorize the stroke pattern?

@Nightsky

I have only just started recently using an SRS like Anki so I can't say much about it in terms of the method you described.  However, I did attempt to learn compounds through readings (although more time was spent learning admittedly out of context) as I explained, so it wasn't like I was completely ignoring context.

Under your method, how are you able to remember the positions of radicals in the Kanji?  Without Heisig, I frquently forgot which should be on the right/left or top/down.  And don't the radicals themselves appear to be arbitrary strokes in your mind?  I used to omit the occasional stroke or confuse certain radicals without thinking of them as having some symbolic meaning.  How do you keep them lodged in your mind?  I'm sure that you have learned to recognize a great many vocab (as I did), but can you honestly produce them perfectly every time without consulting a dictionary?

I'm not exactly sure how the Japanese themselves learn Kanji but I assume it is a combination of rote memorization and compound reading exposure.  Most Japanese that I have met frequently have problems remembering how to write the more difficult kanji, and they must consult a dictionary in order to do so.  The nice thing about the Heisig method is that this type of forgetting is much less common when you have good stories planted in your head.  Compared to the Japanese, we might have these keywords, assuming they don't fade, rattling around our heads.  But maybe that's a small price to pay for near perfect production accuracy.

If you have had continued success under your method, then more power to you.  And, of course, I wish you the best of luck in your future studies.

Last edited by Dragg (2008 June 07, 10:30 pm)

Hey NightSky, why don't you do some field study and see what all the fuss is about? Download the free pdf of the first couple hundred kanji of RTK1 and use this website for a few weeks. I bet you'd have fun! It couldn't hurt, right? ^^

NightSky, what do you find to be the most difficult aspect of Japanese?

I think I said once before, we have as many opinions as there are people because what works best varies from person to person. For example, I have tried SRSing sentences containing kanji, and also SRSing isolated kanji to try and memorize the readings. The first was definitely more successful. But if others have had a different experience, I can't argue with the facts.

What does become a problem is if people are stubborn or hot-headed and refuse to try another method to see if it works. You can't say RTK doesn't work for you unless you've tried it. You definitely can't say it doesn't work for me, because it stands against the plain fact that it does work for me.


As for why people might consider kanji as the hardest part of learning language:

(a) Without RTK or something as effective, it's impossible to read and write the language. You can learn all the grammar and vocab you want, but you still won't be able to pick up a magazine and read.

(b) Depending on your individual strengths and weaknesses, you might find any of the grammar, the vocab, the idioms or the kanji the most difficult for you. However, because for languages with sane writing systems you still have to learn all those other things, Japanese and Chinese have the added monumental task of learning a ridiculously large alphabet. So whether it is the hardest aspect for you or not, kanji is the biggest stumbling block because it is one additional burden.

Personally for me I've always loved kanji. Once I got to about 400 I was unable to learn any more at an appreciable rate unless I a workmate suggested RTK. Now I've quadrupled the number of kanji I can at least recognize, and doubled the number I can use in reading and writing.


As for the value of learning the keywords (well actually I try to use Japanese keywords where I can), what you are learning is an "alphabet". It's just like you learn a bunch of maths symbols which have keywords like "plus", "minus", "degrees",  "integral", "infinity" and so on, and then you can use them to read and write maths. It's the same with RTK keywords. It's just something to help you identify a symbol until you get to know how to use it.

Anyway, in the vast majority of cases, you're learning the character's meanings, so how can that be a bad thing? Just so long you are aware that many are only one of a number of meanings, while a few need to be taken with a grain of salt, and a small number are only ever used in compounds and only have a very vague meaning of their own.

nac_est wrote:

I find the Japanese "language", as opposed to kanji, to be actually quite easy. The grammar is easy and the pronunciation is straightforward. Yes, it's all very different from our western languages, but that doesn't make it more "complex" than kanji.

I couldn't agree more. 

Kanji is complex because of the shear amount of them, but the actual language itself is pretty awesome.  Discovering that verb endings contained constants opened some doors wide open (from "る?　endings to "て?、　?たい?、　?た・だ?、?ました?　etc etc etc, depending on tense, politeness and intention)

@MeNoSavvy

I used to think I could get by in Japanese without ever learning how to write the kanji.  However, kanji that look almost exactly the same as one another can have vastly different meanings.  Through learning via reading recognition alone, its harder to pick up on these subtleties.  For example, a few tries of writing with a pen was all I needed to cement the difference in my mind between "samurai" and "dirt".   Before writing it out, it was much harder for me to remember that the only difference was due to a slight elongation of the cross stroke.  Because Kanji is much more visual-spatially complex than Western script, writing becomes more important as a way to wrap your head around it, even if just for the purposes of good recognition.

If Kanji were always obviously unique-looking and distinct from one another to the casual observer, I would definitely be inclined to doubt the necessity of hand-writing practice in the computer age. But as it is, that's just not the case with many kanji.

Last edited by Dragg (2008 June 10, 1:05 pm)

yukamina wrote:

NightSky, what do you find to be the most difficult aspect of Japanese?

Already answered previously...

raichu wrote:

(a) Without RTK or something as effective, it's impossible to read and write the language. You can learn all the grammar and vocab you want, but you still won't be able to pick up a magazine and read.

I STRONGLY disagree....... impossible is far to pessimistic, hard maybe, impossible? no

Anyways we're getting off topic, I made this thread to find out who that special someone was that said he memorized 100 kanji a day without Heisigs method.... if anyone does memorize 50+ kanji a day, please do elaborate... I want to know how long you spend per day and how much attention you spend to each kanji (personally, I write out 1 line of kanji on my paper and move on).

So please, anyone who does 50 or more a day , give us some tips (assuming you don't spend all day on kanji )

lazar wrote:

yukamina wrote:

NightSky, what do you find to be the most difficult aspect of Japanese?

Already answered previously...

raichu wrote:

(a) Without RTK or something as effective, it's impossible to read and write the language. You can learn all the grammar and vocab you want, but you still won't be able to pick up a magazine and read.

I STRONGLY disagree....... impossible is far to pessimistic, hard maybe, impossible? no

Anyways we're getting off topic, I made this thread to find out who that special someone was that said he memorized 100 kanji a day without Heisigs method.... if anyone does memorize 50+ kanji a day, please do elaborate... I want to know how long you spend per day and how much attention you spend to each kanji (personally, I write out 1 line of kanji on my paper and move on).

So please, anyone who does 50 or more a day , give us some tips (assuming you don't spend all day on kanji )

I will tell you this: 'normal' people DON'T learn 50+ kanji a day by rote-memorization. (Though I assume some people might have another method than either Heisig or rote-memorization, I just can't think of what such a method would exist of.) People can try to do so, but it requires an immense amount of rehearsing and reviewing in the weeks that follow that first pile of 50+ kanji up. Otherwise you would simply forget the ones you have 'learned' within a very small amount of time. You WOULD have to spend all day on kanji or you WOULD forget a lot of these kanji as soon as you didn't review them enough again.

However, ofcourse it IS very much possible to learn all kanji you would want to learn through rote-memorization. (Can't believe Raichu begged to differ, ten-thousands upon ten-thousands of people have proved you wrong already.) It just IS a tad slower.

Personally, I would take 40 kanji a week (including on- and kunyomi readings and all of that jazz) as a normal pace for anyone learning Japanese by rote-memorization. I'm basing this number off the amount of kanji professional students of the Japanese language in the Netherlands have to learn. For they attend the world-renowned Eastern Languages department of Leiden University and thus I assume they know what they are doing, trying to get a bachelors / masters degree. This, together with all of the other things they have to learn (culture, history, grammar, words, listening, speaking, writing academical papers, etcetera.) is already experienced as a HEAVY workload. And no, that is NOT because we Dutch are principally lazy as a lot of people would like to believe .

I guess that's all I have to say about rote-memorization. To sum it up: Not as fast as Heisig, but not impossible to use when learning Japanese.

(P.S. The reason why I started writing this in the first place is because I thought you aimed at rote-memorization when you asked your question about 'Who remembers the person that said he doesn't use heisigs method but still manages to learn a bazillion kanji a day?. Only later I thought of the fact that you might not be aiming at rote-memorization at all, so I'm afraid my post is a tad off-topic or incomplete then. But at least you have my objective take on rote-memorization .)

Last edited by roderik (2008 June 11, 1:59 am)

@MeNeSaavy
I understand the calculations for what your time is worth, having been a freelancer for a number of years. (though never at those rates of pay ) However, my kanji writing is not in addition to other studies, it's worked into the studies I already have, so there's really no additional cost. This gets us back to the problem of rote memorization. If I was to learn the kanji by rote, there would be a time cost, as you suggest. I am making it a natural outgrowth of the learning I am already doing, ergo a free benefit.

Hi Let me clarify what I was saying above. I think Heisig is a good system, however I find his claims as to it's effectiveness a little over the top. I haven't met anyone who has learnt all the kanji in 4 weeks by following his book. Nevertheless I certainly think his ideas are incredibly valuable. I strongly advocate people use Heisig or similar but alternative systems based on verbal and/or visual memory aids. All I'm suggesting is that for some students it might be best not to spend too much time trying to exactly being able to reproduce the kanji in writing. Just learn to recognize it. Even heisig himself in the introduction to his book says that acquiring fluency in writing the characters requires a lot of practice (I'm paraphrasing here), but that remembering the characters themselves does not require that they be written.

So what I'm really saying is that unless you want to learn to write the characters well, you can dispense with writing them over and over again.

Also I strongly advocate using an SRS system, and incorporating your mnemonics into the system, so each time the kanji appears so does the memory prompt. This constant reinforcement will assist in memorizing the kanji. Eventually the prompts can be removed from your SRS.

Systems which may be an alternative to Heisig include Henshall (which I think is too academic), Bodnaryk (which I think his pretty good, and he includes stories for all the kanji). Also Marc Bernabe has produced some books called Kanji in mangaland which is based on visual rather than verbal memory. It looks good but I haven't looked at it in detail. Also I'm not sure if the intended 3 volumes are finished yet. One thing about Bernabe is his method introduces the kanji basically in the same order as is taught in Japanese schools (and only teaches the Kyouiku Kanji).

As far as the effectiveness of Heisig, 4 weeks is definitely possible if you are doing kanji full-time.  There was a period of about a week when I learned about 75 to 100 kanji per day working 6-10 hours per day.  This was during Christmas break from classes last December, and I also had no work at the time.  I wasn't able to keep up with this pace because I found it psychologically strenuous, but I'm sure some people could do it so long as they are willing to tough it out.   

I'm not familiar with Kanji in Mangaland but it sounds very similar to kanji pict-o-graphix.  How does this system based on visuals distinguish between similar looking kanji?  How about when the kanji meaning is abstract?

For example, the imaginitive memory of Heisig's system can very easily make transitions from the similar kanji of "water" to "eternity" to "swim" by thinking about little kids loving to swim in water for an eternity.  But how does a purely visual system coherently do this without inefficiently and confusingly changing the visual images completely?  Another example which is probably even more difficult would be the transition between "stop" and "correct."

As far as Henshall's book, I'm not sure why this is even being mentioned.  Its the same as Heisig in the sense of having stories and keywords.  The only meaningful difference as far as presentation is that he lists the phonetic readings and a few vocab compounds for each kanji, but they are completely out of context. 

No matter what system you use, I completely agree with the necessity to at least learn to write out each kanji at least once, or at the very least learn how to write every primitive/radical that compose them.  Otherwise, I think even the most visual learners will have problems reading the commonly-seen stylized fonts of Japan and other people's handwriting, even if you yourself don't handwrite.  If you are constantly remembering only in terms of concrete visuals instead of with strokes as well, these situations will be headaches for you.

  I agree with MeNoSavvy in the sense that, if you only care about recognition, any practice beyond just writing it out a few times is probably not ultimately necessary, but if you have the time, extra practice can certainly only help you.

Last edited by Dragg (2008 June 12, 1:22 pm)

One of the guys at AJATT (Wan) - was able to learn all 2042 kanji in just over 20 days, averaging 100/day - that's the fastest I've heard of. I don't have the dedication to do that much!