I recently bought the Kodasha Kanji Learner's Dictionary. I find it really clear and interesting. I like its emphasis on one key meaning, to which most compounds are in some way related. Makes memorization easy and Kanji very logical. I did wonder, since it is a similar emphasis to Heisig, whether there were discrepancies, and if they would cause problems with my reading later on.

I found this review of Kanji Methods:

http://nihongonotto.blogspot.com/2007/11...udied.html

"Hester's analysis of Heisig keywords against the core meanings in the Kodansha's Kanji Learner's Dictionary suggests that as many as 20% of Heisig's keywords do not match the core meanings of the characters"

My questions are:

a) Is this claim accurate?

b) Is there a comprehensive list of the discrepancies somewhere?

I found a partial list here:

http://groups.google.com/group/sci.lang....a82879c4e8

c) Should I worry?

d) Are any of the other methods worth trying ( Henshall, De Roo, Kanji ABC, Everyday Kanji), and if so, should they be used as alternatives to Heisig, alongside it, or after it?

My main concern is having at some future point to unlearn meanings. I'm something of a slow unlearner.

J

a) Maybe. Lots of them are really funky. 20% seems incredibly high, but yeah, some of the keywords only have the slightest connection to the kanji. As Heisig said, they're chosen as mnemonic cues.

b) I don't know.

c) No. If you go into it knowing that most kanji have multiple meanings and the one you're using as a cue might not be the most prominent one, then you'll be alright.

d) Do Heisig, then dive into Japanese head first. Pick up readings as you go along. Many of the Kanji are impossible to classify in a word and only exposure to their various compounds will you get some sense of them. And often it's not something you can put into words very well - at least in my experience.


Edit: I'd also note that there is an RtK deck available for Anki that includes most/all of the JLPT vocabulary for each kanji. I always had Anki display this vocabulary on the answer side of my card, which I would glance through sometimes. Not too active, just if there was some confusion about how a certain kanji might be used. So I used the keyword, but also had immediate exposure to its many uses. This helped me really pick out those few keywords that didn't make any sense (and even change a few).

[Katsuo]

- With Heisig's method each kanji needs a unique keyword. Sometimes the most suitable keyword for a particular kanji has been already been alloted elsewhere.
- Some of the less suitable keywords have been revised in later editions.

Bearing the above two things in mind, I think the proportion of keywords that could be improved is quite small.

Jeromin Wrote:b) Is there a comprehensive list of the discrepancies somewhere?

I made such a list. (Edit: See this Google docs spreadsheet.)

Quote:My main concern is having at some future point to unlearn meanings. I'm something of a slow unlearner.

In most cases you will be adding extra meanings rather than unlearning anything.

c) NO. The goal of heisig is not to teach you everything about the kanji. It's to get a good handle on them before you start other studies.

Slow unlearner, ha, ha, ha! Me too. Heck, I don't call myself ゆっくり亀　for nothing.

It is important to remember that *a kanji is not a word*, it is a symbol, which is a condensed expression of a related set of concepts. And equivalent concepts may or may not exist in English.

In many cases, Heisig's keywords are tangential to the central meaning of the kanji, or in some cases just bizarre and useless. "decameron" is an example of particularly useless. I get rather frustrated with him, despite the brilliance of his method.

Sometimes I try to account for such discrepancies in my story. Here is my story for Frame 607: Apply 応:

Heisig's keyword doesn't APPLY this kanji directly. 応用する （おうようする）means "to apply" as in 彼らは新しい科学技術をその産業に応用した。"They applied new technology to the industry."　 The most common translation of 応 (on it's own without 用) might be "RESPOND". As in responding to a question or responding to a situation. Story: Men are CAVEmen at HEART. When confronted by emergency situations they RESPOND by APPLYING their primitive power.

I have not had a problem making mental corrections after learning a less-than-ideal keyword. Though when learning new kanji I sometimes look up the kanji or ask my wife as I am writing my stories.

yukkuri_kame Wrote:In many cases, Heisig's keywords are tangential to the central meaning of the kanji, or in some cases just bizarre and useless. "decameron" is an example of particularly useless. I get rather frustrated with him, despite the brilliance of his method.

I'm puzzled as to why you would cite that particular example. 'Decameron' is a fair keyword for the character.

onafarm Wrote:

yukkuri_kame Wrote:In many cases, Heisig's keywords are tangential to the central meaning of the kanji, or in some cases just bizarre and useless. "decameron" is an example of particularly useless. I get rather frustrated with him, despite the brilliance of his method.

I'm puzzled as to why you would cite that particular example. 'Decameron' is a fair keyword for the character.

Ever hear the words 下旬 and 上旬？

Decameron is a very useful keyword. Captures the meaning very well.

i'll agree, anytime i see one that makes me think "really..." i check. such ones that are giving me a bit of a time figuring out are 炎 and 淡.
As far as Decameron goes, as a keyword, i would rather have "10 days" or "10 day span"
he does not define "decameron" in the text, so i had to look it up.
Though, much worse are the keywords seduce and spot.

Can this be the "complaining about keywords you don't like" thread?

向 = yonder
However, it seems to mean "oppose." I don't know if there already is another kanji for oppose.

Sample words
向こう
立ち向かう
向こう側

They keywords --don't matter--

Sure, it can be helpful if they match up, and if they don't, then go ahead and change it to something you see more fitting. It's not going to change the method or the experience. If you find words that work better for you, by all means, use them!

The keywords are just --keywords-- and aren't meant to tell you what the kanji "mean." They're just there as signals to bring the kanji to mind.

Just to say Heisig explains somewhere (in an interview?) he didn't base his keywords on core meaning but on etymology (because he wanted to capture the original iconic image, like the imaginative method does), that's where most of the discrepancies come from (if you look up every etymology, you'll see Heisig is quite accurate). He didn't use core meanings probably to avoid confusing the drawing of the kanji with actual single kanji words.
But as Asriel said, keywords are just keywords. In the end, they don't really matter.

I agree that the keywords don't much matter.

"Yonder" seem like a fine keyword for ”向こう”。

GinRyuu Wrote:As far as Decameron goes, as a keyword, i would rather have "10 days" or "10 day span"
he does not define "decameron" in the text, so i had to look it up.

I'm puzzled.

Why is it a problem to look up an English word with which you're not familiar?

The Hesig keywords become really, really annoying when you start learning real live Japanese words ™ and you end up in situations like this:

APPLY -> Well, that's something like 適する or maybe 使用する or 適用--crap, is it 適, 使, or 用?
応　<<-- Actual Hesig character.

Oh, like 応募. How the heck was I supposed to figure that out? I was totally barking up the wrong tree.

When this happens, you have two options:

- Aggressively revise keywords with ones that better fit your sense of what the character means.
応 is totally 応える反応 (spell it in kana so as not to spoil it) or REACT
APPLY isn't a bad keyword for 用

Like stories, this will end up being very subjective. I don't have any problem with YONDER -> 向, even if "yonder" = あそこ = 彼処.

I think this is the better choice, but it's not the one I ended up taking.

- Stop reviewing. It worked for Hesig. It's not all that bad.* But, you could end up in the situation I'm in now: I need to return to handwritten kanji review, but now I have around a thousand to re-learn and hundreds of those will require re-keying. Better to have done that as it was needed.

(* Yes, despite how a year or so ago I gave others a hard time about wanting to quit reviews. I was wrong. It's bad. Not fatal. I still think its better to catch up reviews and reset failed cards (leech threshold to 1).)

If you're just starting out, I wouldn't worry too much about this. Hesig's keywords got me really, really far. It's true, they don't matter--you can replace them whenever you need to.

Well, In chapter 31, in the introduction there is a paragraph that says '.... fears about the inadequacy of the key words should be greatly allayed. For in much the same way that the character slowly finds its way into the fabric of memory and muscle memory. the keyword will gradually give way to a key concept distinct from the particular English keyword used to express it. Hence the substitution of a japanese word or even a number of words will prove no stumbling block. quite the contrary it will help to avoid confusion between keywords with family resemblances.'

So basically, it shouldn't be a problem. I"m sure confusion will set in later on, but I think that illustrates that the point of this method of keywords is just a starting point to learn how to *write* the character which i think it does a great job at. It will be replaced anyway with a japanese meaning (some concepts and meanings in japanese just don't have an equivalent english word anyway, just as some english words do not have a japanese equivalent, but that doesn't matter in the end. it's just a starting point). So just do it and don't worry.

You're eventually going to forget all the keywords anyway. I certainly don't know any of them anymore.

That is, unless you actually keep up with your reviews. After a while I just stopped. It's hurt my writing a bit, but not my reading really.

Katsuo Wrote:- With Heisig's method each kanji needs a unique keyword. Sometimes the most suitable keyword for a particular kanji has been already been alloted elsewhere.
- Some of the less suitable keywords have been revised in later editions.

Bearing the above two things in mind, I think the proportion of keywords that could be improved is quite small.

Jeromin Wrote:b) Is there a comprehensive list of the discrepancies somewhere?

I made such a list. (Edit: See this Google docs spreadsheet.)

Quote:My main concern is having at some future point to unlearn meanings. I'm something of a slow unlearner.

In most cases you will be adding extra meanings rather than unlearning anything.

A really excellent and complete list.
I've revised the first 500 and most of my corrected keywords were a perfect match with the Kenkyusha/Helpern column. I even found some new ones I now know I've missed.
Thanks for the hard work and for sharing, you certainly made my life easier.

to the contrary, if you look up definitions for decameron, the closest thing you will find is a collection of novels.

If you look up 'decameron' where? How about a dictionary?

In any case Heisig defines its use in the text.

From The Oxford Dictionary of English:

Decameron dɪˈkæmərən

a work by Boccaccio, written between 1348 and 1358, containing a hundred tales supposedly told in ten days by a party of ten young people who had fled from the Black Death in Florence. The work was influential on later writers such as Chaucer and Shakespeare.

And how do i get 10 days from a collection of tales between 1348 and 1358?
----straying from the topic

From the next part of the definition where it says "supposedly told in ten days by a party of ten young people"

GinRyuu Wrote:As far as Decameron goes, as a keyword, i would rather have "10 days" or "10 day span" he does not define "decameron" in the text, so i had to look it up.

This is false. He tells you what it means, where it came from, and how to use it. I don't think it would be good to use the number 10 in the keyword when 十 is not in the character, plus it wouldn't convey the sense of ‘bound’ his explanation conveys.

See also this thread on the forum, titled ‘Decameron’: http://forum.koohii.com/showthread.php?tid=6809

Edited out the actual story since that'd be a bad habit to get into on the forum.

The Decameron (subtitle: Prencipe Galeotto) is a collection of 100 novellas by Italian author Giovanni Boccaccio, probably begun in 1350 and finished in 1353.
also just checked the story, he does not say "10 days" anywhere in it. Maybe this is because i am using an older version?
But this is completely off the point. This was supposed to be about keywords, not a long discussion of what is and what is not a decameron.
Heisig's keywords can be vague, there is no doubt about this, right?

GinRyuu Wrote:The Decameron (subtitle: Prencipe Galeotto) is a collection of 100 novellas by Italian author Giovanni Boccaccio, probably begun in 1350 and finished in 1353.
also just checked the story, he does not say "10 days" anywhere in it. Maybe this is because i am using an older version?
But this is completely off the point. This was supposed to be about keywords, not a long discussion of what is and what is not a decameron.
Heisig's keywords can be vague, there is no doubt about this, right?

As I pointed out in the link, right at the beginning of that wiki entry it says: “The title is a portmanteau of two Greek words meaning "10" (δέκα déka) and "day" (ἡμέρα hēméra).”

It's in the Heisig book, twice. I think you're misreading it? Because you've so far missed the “10 days” in three different definitions of Decameron even though they're right in front of you? Is English your native language? That could be the problem. Some NNES prefer to use custom keywords from their native languages.