I was watching some horrible kung-fu movie last night, and there's a shot of a plaque that is subbed "black dragon"- but I recognized the kanji for "black" before the prompt appears. Later, there's a shot of a woman's grave, and I recognize the kanji for "woman" and "grave" in a compound.

That's what RTK does. I haven't learned a single reading of a character past the very simple keyword, and already I can randomly pick out characters, and sometimes even put together the general sense of something. Heck, sometimes I even piece together bits on a Chinese menu from the similarities.

The more RTK you learn, the less daunting Japanese seems as a whole. You go from "holy cow how will I ever remember how to write that" to "oh, hey, imperial edict (詔 #342), I remember that guy".

[ファブリス]

Yeah, it is very encouraging, especially if you go to Japan. You're just another tourist.. but somehow you don't feel so lost. You can recognize train station names between signposts and a map, etc. You can have a decent guess at many simple words written on machines, electronics, food, ... Fun!

For me the big test will be next August when I will spend 3 weeks traveling around Japan with a friend who basically only knows how to say 'konnichiwa', so finding everything will basically be up to me. I'm starting RTK2 today and I've done a few Pimsleur units, and I will soon have a lot more time off work so I can study a bit more. So by the time August rolls around I'll hopefully be ok.

With RTK, now I can understand the sings at Chinese restaurants... xDD The one near my house is called "Huasan Restaurant" (華山酒家). But now I know it means "Chinese Mountain"... ain't that great?..xDD... Others are called "New China", "Golden Dragon" and similar two-character compounds...

thecite Wrote:

Tobberoth Wrote:

undead_saif Wrote:Why do you guys keep saying that the keywords aren't important? Usually, the keyword is one of the meanings for the kanji, sometimes it differ, so when you finish RTK1 you'll be able to write and recognize 2042 kanji, and you'll know a meaning for most of them.

Nah, the keyword is extremely vague in most cases. It's helpful, but not important. I would never rely on it for meaning, but it is definitely useful when going for understanding of compounds.

The keyword is only helpful if you know the Japanese word it relates to. When I want to use a Kanji I don't know, let's say, むずかしい、I know 難 is 'difficult', and can confidently use it thinking it's むずかしい. That being said, the keywords should really only be used to help you learn from RTK2.

Okay, so...wait. RTK doesn't include pronunciations, that much I understand. It only provides a single word (one of many) that is often one of the common meanings associated with the character, but sometimes it isn't? Sometimes the word has nothing to do with the usage?

So...is RtK just about form? Does it offer the meanings of the radicals? Is it best combined with a good Kanji dictionary so that I can try to assimilate some possible meanings? Should I just be focusing on the shapes?

I'm sorry if this is something a lot of people ask, but I feel kind of lost right now.

@A_Broken_Pencil

The biggest benefit of RTK, at least in my view, is learning how to easily learn the proper writing of new kanji that you encounter. Heisig's method of creating stories for each kanji makes learning how to write new kanji incredibly easy. The second biggest benefit of RTK is learning how to actually write over 2000 or 3000 kanji (depending on whether you complete just RTK1 or both RTK1 and 3).

The difference between those who've learned kanji through RTK and those who haven't is that an RTK learner is more likely to immediate identify the radicals (common stroke patterns shared across several kanji) that a kanji is made of, and therefore have a much easier time learning how to write the kanji. Yes, there are other methods that emphasis learning kanji with the help of radicals, but the Heisig approach differs in its use of stories. Most other methods that focus on radicals lack a memory aid to help learners remember the combination of radicals and where those radicals are positioned relative to one another. The story aspect of Heisig's method makes it much easier to remember the combination of radicals and how they're positioned. In turn, this makes remembering how to write a certain kanji a far simpler task.

@A_Broken_Pencil
Why don't you check out the free RTK sample linked on the homepage (and @Khatzumoto/AJATT and elsewhere)(or is it on the RTK off site?RTK wiki page?)? You could make your on idea first hand.
I think it includes the introduction where he explains the method.

The main benefit I get from RTK is that Kanji don't look like a mess of impossible lines anymore. I can distinguish one from the other.

I like crossword puzzles. One thing you have to keep in mind that is that the clues are just that: clues, not definitions; otherwise, the "answers" don't always make sense.

I think RTK needs to be approached the same way. RTK's only purpose is to learn the kanji, not the Japanese language, so I take all the "keywords" with a grain of salt. I treat them as clues to how to write the characters, not as vocabulary.

[Katsuo]

A_Broken_Pencil Wrote:Okay, so...wait. RTK doesn't include pronunciations, that much I understand. It only provides a single word (one of many) that is often one of the common meanings associated with the character, but sometimes it isn't? Sometimes the word has nothing to do with the usage?

The keyword is always related to usage. Well, more than 99% of the time; there are only two or three cases out of the two thousand kanji where I can't quite see where the keyword came from.

Bear in mind that one part of the method is that keywords must all be different so as to keep each kanji distinct in your head. Sometimes the most suitable keyword has already been used for another kanji, and in those cases either a synonym or minor meaning is chosen.

' Wrote:So...is RtK just about form? Does it offer the meanings of the radicals? Is it best combined with a good Kanji dictionary so that I can try to assimilate some possible meanings?

RTK takes the tradtional 214 radical set and expands on it (because some components and combinations aren't covered the traditional set). Heisig often chooses the same standard radical meaning, but also frequently changes them to ease memorization. I made a table comparing the two.

I'd suggest just sticking to Heisig's names as checking each one in a kanji dictionary will slow you down a lot. However, if you want to do a quick check, I made a spreadsheet comparing the names that other sources give to kanji. If the RTK keyword is completely different to those from the other sources then it might be worth investigating further.

Well, colour me convinced.

I picked up Volumes 1 and 2 about a week or so, but have been reticent getting around to them, since I still feel odd knowing how to write a kanji, but not how to pronounce it. Last night I finally gave it a go, and...wow. I've never been able to memorise so many characters so quickly before, and with almost no need for reviewing.

Trying it is loving it.
Yeah, the first 500/1000 really all feel this way (imo).

Katsuo Wrote:I'd suggest just sticking to Heisig's names as checking each one in a kanji dictionary will slow you down a lot. .

Seconding this. I was runingn out of time, and stopped trying to look things up, and learned the last 1200 in a month, when it took me most of three months to learn 800. In retrospect, I wish I had not bothered to look up any of them in the many Kanji dictionaries I have.

On the other hand now that I have learned them, I am far too lazy to go back and fill in the blanks on really learn some vocabulary for those Kanji I just used the Heisig word for.

In this respect, I imagine that if learning Japanese (not just learning Kanji) is the goal, looking up each character in the dictionary would help, maybe, possibly, because once you know a kanji you are not likely to go back and study it. But that's probably better not to do anyway.

On the other hand, RTK has some tremendous gaps in it that make "getting through with it and done with it" make sense. "Cram your head with the Kanji and then learn the language" makes sense when you see those holes. And it also helps you hate the silly keywords less.

When doing RTK it helps to banish the words "meaning" and "definition" and think of keywords as concepts or ideas. Using a dictionary will inevitably get you mired in meanings and definitions, and impede your progress in learning to recognize the kanji, which is the point of RTK.

However, the benefit of knowing the "meaning" is you don't pick the wrong meaning based on a keyword with a vague meaning itself. This in turn ensures your story is tailored to the correct flavor of the kanji.

There are spreadsheets and apps that include the various meaning of kanji in RTK order, so I don't think one needs to go to a dictionary for the most part.

Actually, I should have said "a regular dictionary". For keyword clarification I use Halpern's New Japanese English Character Dictionary, which is totally concept-oriented; in fact, its headwords are keywords and are referred to as such. It's been very helpful, especially since I put the Japanese app on my iTouch and can quickly access the appropriate NJECD entry number.

Anna B Wrote:Actually, I should have said "a regular dictionary". For keyword clarification I use Halpern's New Japanese English Character Dictionary, which is totally concept-oriented; in fact, its headwords are keywords and are referred to as such. It's been very helpful, especially since I put the Japanese app on my iTouch and can quickly access the appropriate NJECD entry number.

You have found the big version of that?

That is an amazing dictionary, but it is out of print in the US.

The mini version is handy for lugging around, but not nearly as nice as the big one.

If it is ever available as an app on on iTunes I would love it.

[Katsuo]

The KLD (compact version of NJECD) was available on the App Store for a while, but has mysteriously disappeared. "Enfour", the company that made it, has many other apps still available, so I don't know what's going on.

A similar dictionary that is available is Jishop. Like Halpern's it gives keywords to the kanji and breaks examples into themed groups. Overall there are less examples per kanji, but on the other hand it covers far more characters. When Jishop was first released it got low ratings on the App Store, but the interface and look-up have improved greatly with various updates. There's also a free version that covers 1,006 characters.

Anna B Wrote:When doing RTK it helps to banish the words "meaning" and "definition" and think of keywords as concepts or ideas. Using a dictionary will inevitably get you mired in meanings and definitions, and impede your progress in learning to recognize the kanji, which is the point of RTK.

I feel like this is something I can't quite wrap my head around, despite the fact I'm loving Heisig so far.

I mean, what's the point of reaching the intermediate stage of 'I know this usually signifies abundance, and I can draw it to look pretty shweet' without just putting your foot down a little harder on the pedal and changing it to 'This Kanji means 'very, very many', except when it's in this word, in which case it means 'forever'. It's usually read as　ばぁっ九トン　or ろっつ but a few names read it like くらすたばんちゅ。 Also, I can draw it very well.'

I know there's probably some logical thread I'm failing to follow, but right now it seems like completing RtK without adding the info in RtK2 is stopping halfway, like learning Japanese vocabulary in romaji. Maybe it's because my overall goal is to one day be native-level fluent, but right now I don't mind the idea that it might take a few extra months (maybe even an extra year) just so I can 'know' each Kanji as well as the average Japanese person would.

Plus, I'm still a little concerned that due to the primacy effect and my own knowledge of how my brain tends to run, that Heisig's keywords (no matter how useful) will override the later-learned genuine connotations. I know Kanji are imbued with suggestions rather than concrete meanings, but I feel like learning an inaccurate suggestion would be more confusing than having two characters with the same keyword.

Like, Heisig says that the radical　十　sometimes has the connotations of a needle, and bases many of his stories on this, but when I look it up, both online and in a physical dictionary, no-one else seems to mention this. 針 itself does contain the general shape of a cross, but...if I run into a new kanji and it contains the 十 shape, I feel like I'm going to assume it might possibly have something to do with needles or pointy things in general, when it apparently never will.

So I guess what I'm saying is, can I take a peek at the aforementioned spreadsheets just to check that it wouldn't work better for me to fiddle with one or two of the keywords before I commit them to memory?

Sure, but keep in mind Heisig went there, and he choose an optimum route. Of course, it seems a little dated sometimes, but around 90% (95%?) of his work is fine (IMO).

What you've got to understand is the relationship the primitives have in order to create meaning in a kanji is the same 2 kanji have to create meaning in a word. So you have to get the hang of this kind of abstract thinking in the kanji in order to remember words with utmost efficiency later on (still IMO).

The whole point of Heisig is because very few words are made of only one kanji (am I wrong?), so you cannot say in most of the cases "this kanji means X word with Y reading" because it doesn't work that way. Chinese writing is about context. The only fixed thing you can rely on is the sphere of meaning. In English it's easy, because the base of words are letters, pronounced all the same way, and that don't have meaning. In Japanese, the base of words are kanji, and they are all homophones (since there isn't accents like in Chinese), so the only discriminating factor is meaning, that's what you can rely on.

So to go around the problem, think of keywords as idea/concept, and don't bother with readings yet (because it will just burden your brain and make you learn less efficiently (readings can change in a kanji, but meanings are about the same, whatever the readings)). The quicker you learn the kanji, the quicker you'll start learning vocab.


EDIT: and my goal is native-like fluency too.

You could add in the most commonly used onyomi as an added primitive in your stories if it really bothers you that much.

Each onyomi can act like a primitive, and you just incorporate that into the story somehow. Or you could even add in kunyomi if you waned to. Look up the movie method.

A_Broken_Pencil Wrote:I mean, what's the point of reaching the intermediate stage of 'I know this usually signifies abundance, and I can draw it to look pretty shweet' without just putting your foot down a little harder on the pedal and changing it to 'This Kanji means 'very, very many', except when it's in this word, in which case it means 'forever'. It's usually read as　ばぁっ九トン　or ろっつ but a few names read it like くらすたばんちゅ。 Also, I can draw it very well.'

The best way to put your foot down a little harder on the pedal would be to go through RTK1 faster. It's designed to be a quick process which gives you a head start on the rest of the field. Adding in various readings and attempting to memorize various exceptions will just slow you down and make the process less efficient. And trust me when I say it gives you a massive head start: I'd say at least 90% of compounds have a clear X + Y = Z vibe in terms of meaning, which makes them so much easier to learn. You'll pick up all the readings automatically as you learn words.

harhol Wrote:The best way to put your foot down a little harder on the pedal would be to go through RTK1 faster. It's designed to be a quick process which gives you a head start on the rest of the field. Adding in various readings and attempting to memorize various exceptions will just slow you down and make the process less efficient. And trust me when I say it gives you a massive head start: I'd say at least 90% of compounds have a clear X + Y = Z vibe in terms of meaning, which makes them so much easier to learn. You'll pick up all the readings automatically as you learn words.

I agree with this. 100%.

(I mean, If you already know a bunch of spoken Japanese to the degree that you can get by without English in Japan (very very rare case) then you can get a lot out of messing with dictionaries with RTK. It will slow you way, way, way, down, but you still can benefit from it. But if you don't know enough Japanese to do only Japanese , just use use the keywords and learn it like an alphabet.)

kapalama Wrote:

Anna B Wrote:For keyword clarification I use Halpern's New Japanese English Character Dictionary, which is totally concept-oriented; in fact, its headwords are keywords and are referred to as such.

You have found the big version of that?

That is an amazing dictionary, but it is out of print in the US.

The mini version is handy for lugging around, but not nearly as nice as the big one.

If it is ever available as an app on on iTunes I would love it.

It's available from Amazon resellers for about $60 new, and under $30 used.

@A_Broken_Pencil: I sympathize, because I started out wanting to do the same, thinking it would be the most logical and efficient approach. And I tried it from various angles, but in the end went back to straight RTK, for several reasons, both structural and pedagogic. (Um...is it pedagogy if you're teaching yourself?)

EratiK Wrote:The whole point of Heisig is because very few words are made of only one kanji (am I wrong?), so you cannot say in most of the cases "this kanji means X word with Y reading" because it doesn't work that way.

Recently I learned that "Lawer" (べんごし) is 弁護士 - "Valve Safeguard Gentleman"