I apologise if this has been asked before, but just how many Kanji do Heisig's books actualy cover? I'm looking through several different reviews/summaries; some of them seem to imply that the series covers only the Joyo Kanji, some state it contains "about 3,000" (so I'm guessing Jinmeiyo as well, right?), and one or two seem to say that Heisig's work covers all existing characters.

So I guess what I want to ask is:

-How many Kanji does Heisig cover in the series (all three books)?

-Are the characters 'promoted' to Joyo earlier this year included in any of his books?

-If not, are there any books I should plan to move onto after I've finished Heisig? (Which may well be a few years from now)

Thanks in advance for any help you guys can give. =)

Hey.

Book 1, 2 and 3 take care of 3007 kanji. You must add the 126 recent add.

By then, any new kanji you might encounter, you will know how to learn by yourself. If you like systematic approach, you can download a frequency list of the 4000 most frequent kanji, and learn the ones you do not know from there.

I think all the old joyos are taken care of in the books (I'm a noob).

( and existing kanji are more than 50 000, but even KANJI scholars don't usually go farther than 10 000. Durmous says you need to master 4000 kanjis to master technical japanese, but the most frequent 1000 are used 90% of the time, that should be more than enough to get you started... )

Yep, RTK1 = 2042, and RTK3 gets you up to 3007.
I'm not sure about the new joyo...

--Waiting for Katsuo to come in with all his wonderful graphs and charts--

[Katsuo]

Heisig recently released a supplement to cover the new joyo kanji that aren't in RTK1.

The old joyo set (used from 1981 to this year, 2010) had 1,945 kanji of which all but one were in RTK1, the other being in RTK3.

The new joyo set (introduced this year, 2010) has 2,136 kanji, adding 196 and removing 5 from the previous set. The 196 new characters can be broken down as follows:
RTK1: 39
RTK3: 134
Not in RTK: 23

The supplement to RTK1 mentioned above can be downloaded for free. It includes all the new joyo kanji that are not in RTK1, i.e. a total of 157 characters (134 + 23).

EratiK Wrote:Hey.

Book 1, 2 and 3 take care of 3007 kanji. You must add the 126 recent add.

By then, any new kanji you might encounter, you will know how to learn by yourself. If you like systematic approach, you can download a frequency list of the 4000 most frequent kanji, and learn the ones you do not know from there.

I think all the old joyos are taken care of in the books (I'm a noob).

( and existing kanji are more than 50 000, but even scholars don't usually go farther than 10 000. Durmous says you need to master 4000 kanjis to master technical japanese, but the most frequent 1000 are used 90% of the time, that should be more than enough to get you started... )

4000 kanji eh? Interesting, I've always been the type of person who wants to learn a lot of kanji, I initially was just going for 2000, but now i'm at that mark. I want 3000. I have a feeling even after 3000 I will want to go farther. Although 10,000 is a lot, maybe if I go there slowly over the next decade or so(10 years). Although I do plan to head into mandarin after japanese, so the possibilities of kanji are endless for me to learn. Just don't know when to "stop". But I guess that's learning for you, you never stop but you can get fluent in all skills though.

I don't know any Japanese professor/ doctor who knows anymore than 4000. Maybe they aren't telling the truth but I seriously doubt anyone does that. The highest level of the 漢字検定　（kanji kentei: Kanji proficientcy test）covers 6000 characters and has a pass rate of 15%!!! The first book covers the old 常用漢字（じょうようかんじ） list and the third book is comprised of roughly 1000 kanji for advance studies and names.

I meant KANJI scholars (sorry for the unclearness everyone).

EratiK Wrote:I meant KANJI scholars (sorry for the unclearness everyone).

yea i know. 4000 is basically my goal for kanji. I'd say ultimate as you don't really need to know that much to become fluent. But I guess it's something to aim for. The kanji kentei stuff, which tests for 6000. A lot of people said it's really not a practical thing, the things you need to know to pass that test are just way too much...
i.e. knowing modern vs old kanji

EratiK Wrote:I meant KANJI scholars (sorry for the unclearness everyone).

If you want to just learn characters, stop studying Japanese and study Chinese where they write in only characters.

If you want to study Japanese, then stop with RTK 1 and learn how to use them in written Japanese, and then learn others as you are exposed to them. Japanese people cannot write even 2000 characters. They know how to read most of what they know but cannot write them. And they cannot even read properly many place names. They know them if they live there, but there is a very good reason that Japanese is no longer using Kanji for many newly incorporated towns: Japanese people simply cannot read a good percentage of place name Kanji. If they live there, they can read it. If they don't they can't. Japanese people from small towns have to get Furigana imprinted on their business cards for their town names as well as their own names (there is no enforced regularity in name readings, whether they be Place of People names.)

Japanese does not have 10,000 kanji, because once you start getting anywhere near 10,000 you are just calling Chinese characters "Kanji" which is just misunderstanding what Kanji is. Kanji is the parts of the *Japanese* languages written non-phonetically. There are tens of thousands of Chinese characters which are not Kanji because they are only used in Chinese. (Though they may appear in Japanese written material when proper names are written using the Chinese native characters). Those are not Kanji, those are Chinese characters.

It's like calling Greek philosophical terms 'English': Some have entered the English language, most have not. And just because they appear in English texts does not make them 'English', just because printed in the midst of English words.

kapalama Wrote:Japanese does not have 10,000 kanji, because once you start getting anywhere near 10,000 you are just calling Chinese characters "Kanji" which is just misunderstanding what Kanji is. Kanji is the parts of the *Japanese* languages written non-phonetically. There are tens of thousands of Chinese characters which are not Kanji because they are only used in Chinese. (Though they may appear in Japanese written material when proper names are written using the Chinese native characters). Those are not Kanji, those are Chinese characters.

It's like calling Greek philosophical terms 'English': Some have entered the English language, most have not. And just because they appear in English texts does not make them 'English', just because printed in the midst of English words.

I have never seen this definition of "kanji" anywhere else, and it's certainly not how the word is used in Japanese. What do you mean by "the parts of the Japanese language written non-phonetically"?

Almost any Chinese character can theoretically have a Japanese reading, and many of them do when you read older kanbun works, so how a Chinese character is used in the modern written language shouldn't have anything to do with whether you call it a "kanji" or not.

kapalama Wrote:Though they may appear in Japanese written material when proper names are written using the Chinese native characters. Those are not Kanji, those are Chinese characters.

As a matter of fact, if Chinese characters appear in any way in a text in Japanese (proper Japanese names, or Japanese location names), yes, they are kanji (operative definition). But yes, I agree, the corpus of ("pure") Japanese kanji probably doesn't exceed 10 000, but it would take an other scholar to confirm.

Which makes me wonder, how do the Japanese call the characters used in Chinese (language). Don't they call that "kanji" too? (it's not a rhetorical question)

Quote:It's like calling Greek philosophical terms 'English': Some have entered the English language, most have not. And just because they appear in English texts does not make them 'English', just because printed in the midst of English words.

Sure αρχε or τυχε are not English. You're talking before the borrowings are borrowed. But once it's borrowed (used), it's no longer Greek (I mean it is ALSO Greek), it's become English (too). I mean the Greek signifier as seeded an English signifier which in return seeds an English signified. Are the English signified and the Greek one identical? Probably not, but for an operative definition as sign (same signifier and same denotated signified), yes they are.
It's all I'm saying.

EratiK Wrote:Which makes me wonder, how do the Japanese call the characters used in Chinese (language). Don't they call that "kanji" too? (it's not a rhetorical question)

Yes, they don't have a separate word for hanzi.

Quote:But yes, I agree, the corpus of ("pure") Japanese kanji probably doesn't exceed 10 000, but it would take an other scholar to confirm.

I'm curious what you mean by "pure" Japanese kanji. If you're referring to 国字, I doubt they number much more than about 200.

JimmySeal Wrote:

EratiK Wrote:Which makes me wonder, how do the Japanese call the characters used in Chinese (language). Don't they call that "kanji" too? (it's not a rhetorical question)

Yes, they don't have a separate word for hanzi.

Yeah, if you look at wikipedia the main article on 漢字 is about Chinese characters (and maps to the English article "Chinese character"), and there's a separate article 日本における漢字 which maps to the English article "kanji".

As for the second aspect of the post a few posts back, you seem to be confusing technical vocabulary, loan words, and borrowed written symbols. What constitutes a "loan word" as opposed to a foreign word is not always clear, but my usual test is that if a word can be found in a native language dictionary, it is a loan word that belongs in that language, regardless of the origin. Of course even that's not clear cut because some dictionaries have words that others don't.

JimmySeal Wrote:I'm curious what you mean by "pure" Japanese kanji. If you're referring to 国字, I doubt they number much more than about 200.

I meant Japanese kanji in Kapalama's sense: the 50 000 and such are found in Chinese sources. Therefore, considering only Japanese sources (in Japanese)(which is even shorter in time and in quantity), you shouldn't get more than 10 000 of the previous (possibly even less (the keitei 6500?), but I'm just supposing).

yudantaiteki Wrote:As for the second aspect of the post a few posts back, you seem to be confusing technical vocabulary, loan words, and borrowed written symbols. What constitutes a "loan word" as opposed to a foreign word is not always clear, but my usual test is that if a word can be found in a native language dictionary, it is a loan word that belongs in that language, regardless of the origin. Of course even that's not clear cut because some dictionaries have words that others don't.

Apologies for my English.
If we're talking philology, you can use the dictionaries of the time to say if a word is loaned or if it's just the symbols that were, but it might not be the best method, since there were so few of them (dictionaries). I admit the line is blurry here. You'll have to consider a certain time lapse (10 years?) from the first attestation.

My point was, as Japanese language changed under the influence of adopting a foreign writing system (the on readings), it was therefore lawful to consider them to be borrowed terms when they were stopped being used as phonetic transcriptions (and also included meaning); and lawful too to call them kanji.

But this is JUST and ONLY my own opinion. I'm no scholar and I don't intend to imprint my views on anyone.

Katsuo Wrote:Heisig recently released a supplement to cover the new joyo kanji that aren't in RTK1.

The old joyo set (used from 1981 to this year, 2010) had 1,945 kanji of which all but one were in RTK1, the other being in RTK3.

The new joyo set (introduced this year, 2010) has 2,136 kanji, adding 196 and removing 5 from the previous set. The 196 new characters can be broken down as follows:
RTK1: 39
RTK3: 134
Not in RTK: 23

The supplement to RTK1 mentioned above can be downloaded for free. It includes all the new joyo kanji that are not in RTK1, i.e. a total of 157 characters (134 + 23).

Here's what I want to know: if those 23 kanji that are not in Heisig are so important as to be included in the new joyo set, why weren't they in Heisig in the first place? In other words, why did Heisig include so many kanji in his third book that won't become joyo kanji but neglected those 23? I'm just curious.

I honestly cannot see the point in learning any more than about 2500-3000 kanji. You're far better using all that energy to learn vocab and hone in on other skills. I know about 1800 kanji and often read very complex academic papers in Japanese. Very rarely do I come across a kanji that I cannot read, and usually it's just someone's name or something extremely abstract. Maybe if you have a fetish for the names of tiny country towns or historical figures knowing any more than 2500 will come in handy.

kusterdu Wrote:Here's what I want to know: if those 23 kanji that are not in Heisig are so important as to be included in the new joyo set, why weren't they in Heisig in the first place? In other words, why did Heisig include so many kanji in his third book that won't become joyo kanji but neglected those 23? I'm just curious.

My vote is for "because they're not actually all that important". I suspect the frequency dropoff means that there are quite a lot of kanji with a frequency that puts them out of the top 2000 but not by much, so if you're extending the set then there's a wide choice of fairly-rare-but-not-incredibly-rare kanji to pick, and it's not surprising that the set Heisig picked doesn't overlap completely with the new-Jouyou. Some of the extensions to Jouyou were just for the benefit of placenames, too, so if that wasn't in Heisig's criteria for picking RTK3 kanji they won't show up. In particular, I think RTK3 got all the "doh, how did they miss that one?" kanji like 頃,誰,俺 and 丼, which are the ones that you might actually care about.

I think that for second learners of Japanese once you're above 1500 or so the chances are very likely that your kanji knowledge is well ahead of your vocabulary anyway, so it's not really all that important either way.

Thora Wrote:eratik, there's a list on Katsuo's website of the 23 new Joyo kanji not in RTK1 with a word for each. You might get a sense of the need for those kanji from that list. I don't think it's pop songs. Yes, the supplement Katsuo linked will be incorporated into new editions.

Yup, I know, thanks. Virtual bs already flushed.

If you're interested in learning more kanji, that's fine, IMO. You might enjoy Shirakawa Shizuka's (白川静) book 常用字解. It traces the history of the characters, which is fun, and it's fairly easy to read, so it might be good practice. I agree that in practical terms, 3000-4000 characters is already well into the college level. Depends on your goals, and what you like doing

bucko Wrote:I honestly cannot see the point in learning any more than about 2500-3000 kanji. You're far better using all that energy to learn vocab and hone in on other skills. I know about 1800 kanji and often read very complex academic papers in Japanese. Very rarely do I come across a kanji that I cannot read, and usually it's just someone's name or something extremely abstract. Maybe if you have a fetish for the names of tiny country towns or historical figures knowing any more than 2500 will come in handy.

Agreed here -- I sometimes wonder if I even know 1800 but it depends on your definition of "know", I guess. I do come across kanji I can't read in papers sometimes but I can often tell the word from context. I don't like it when academic writers purposely use obscure kanji for no good reason, though.

The additions to the Jouyou list are very arbitrary, from what I can tell. Just copying some things I wrote in older threads:

-----

Quote:Incidentally the 196 list includes 9 kanji not on the 191 list (柿哺楷睦釜錮賂毀勾), and drops 4 which were on it (哨聘諜憚). It would make sense because, to me at least, those nine are a lot more familiar than the other four.

It all just seems so arbitrary to me -- 哨 was used on the staff room chalkboard in the school I worked at in Japan, and 憚る and 諜報 are both words I know. On the other hand, 錮, 勾, and 毀 are totally unknown to me. I'm sure they show up in some things, but apparently not in anything I read.
-----

-----
All in all, the additions seem just about as arbitrary as the original list -- most of them are comparatively common among rare kanji, but there are a couple that I scratched my head over (Why does 瑠璃 need to be Jouyou? 冥 was added a couple of years too late, now that Pluto's not a planet anymore. 辣 seems to have been added only for 辣油, but that's already written in katakana about half the time and there are so many food related terms that use non-Jouyou kanji already I don't see why this one in particular was chosen. Maybe the compilers like ramen. I guess they added 丼 as well, and　酎 for 焼酎 and 酎ハイ. Perhaps the committee met in 居酒屋...)
-----

And IMO the most important:
-----
The truth is that you will never be able to study and memorize every single kanji that you will ever encounter for your life. The more kanji you study, the less useful each additional kanji becomes, and once you reach a certain point, you are much better off just learning kanji from things you read rather than trying to learn them from a list. The idea that you're ever going to learn 8,500 characters is pretty silly, and there's a strange belief that you need ridiculous amounts of kanji to read "scholarly" or "classical" things, which really isn't true. Depending on what you're reading there's probably a set of specific kanji you need to know, but even then it's not going to be efficient to just get some list from a random place that has no connection to what you're studying.
-----

From my experience, the kanji from the 23 non-heisig new Joyo kanji that are worth knowing are:

拉 璧 籠 喩 嗅 傲 羞 貪 楷 哺

I think I see most of these characters rather often, and several of them usually without furigana.

籠 is basically freebie if you already know 篭 and that 竜 is a simplification of 龍. I think 籠 is actually far more common in actual usage.