yudantaiteki Wrote:I really just want to break the mental block so many people seem to have where they think that if a novel has 1800 kanji, that means they must have learned 1800 kanji before they crack the cover of the book.

especially when number 1800 on that list may occur only once in the whole book (and have furigana), while number 1 may be in every other sentence. checking an unknown/unreadable kanji in a dictionary doesn't take long.

yudantaiteki Wrote:

erlog Wrote:The split into domain-specific kanji doesn't really happen until around 2800-3000.

I put it around 1400-1500 -- maybe not "domain specific" but that's around the point when I think it's better to learn kanji from what you're reading rather than an arbitrary list.

(I doubt I know 3000 kanji but I read domain-specific stuff every day.)

I guess there's no point having this discussion again, but you really think that people without a college education know 3000 kanji?

Quote:It's not BS, but it depends on what you want to read. If you're just reading light novels or stuff aimed at children then you probably won't run into it. If you're reading novels aimed at literate Japanese adults that enjoy reading then you definitely will.

To clarify, what I think is BS is the idea that she knew 2000+ kanji, and kanji were the greatest barrier to her reading a mystery novel -- so great that she had to abandon the project and use arbitrary lists of kanji. I read Akutagawa short stories when I was only around 1500 -- of course I didn't know every kanji in the stories but kanji were not an insuperable barrier to using them as study material.

I really just want to break the mental block so many people seem to have where they think that if a novel has 1800 kanji, that means they must have learned 1800 kanji before they crack the cover of the book.

I agree with this. I think Heisig should be done concurrently with learning actual Japanese, and that people should be jumping into native sources as soon as they can.

kitakitsune Wrote:

erlog Wrote:Most Japanese people know around 3000 kanji.

Recognition and a little decoding based on prior knowledge - yes
Production - not by a long shot

What's your point other than needing to smugly point out something that didn't need to be pointed out? It's posts like this one that make me regret that I spend time trying to discuss anything on this forum. No matter how detailed I try to make my posts there's always people who just can't stop themselves from nitpicking dumb stuff.

You're not even bothering to respond to what the actual point of my post was or what this thread is even about. You're basically just trolling, and it feels like over the years this forum has become filled with more and more completely toxic people like you. Not doing anything to further the discussion, just wanting to argue endlessly over minutiae.

I never said Japanese people could write all of them. I'm also not saying people have to be able to write all associated vocab words for each of the kanji out to 3000.

I'm saying people should study them so that they at least have recognition, and that way that most of us here do that on this Heisig forum is via some version of the Heisig method.

I concur with yudantaiteki's assessment of 1400-1500 and I concur that it is BS that 2000 kanji is not enough to read novels aimed at literate adults.

In the first 2 pages of the first chapter of 魔術はささやく, a mystery novel by Miyabe Miyuki, one of Japan's best-known contemporary mystery writers, there are:

197 instances of kyouiku kanji (elementary grades)
32 instances of jouyou kanji, including several instances of 彼
exactly 1 instance of a non-jouyou kanji, and it included furigana. (訊く).

I think this is pretty typical of the difficulty level of an average contemporary novel. If you think I'm wrong, please feel free to try it with another novel of your choosing.

If less than half a percent of the kanji in a given work are unknown to you, it's not kanji that are posing a barrier to understanding.

Granted that there are a handful of non-jouyou kanji that are common in fiction, like 訊く and 頷く, but it's only a handful and they can be absorbed pretty quickly.

To clarify something Yudanitaiteki asked about, I don't think most non-college educated people know 3000 kanji. I think the number of kanji most people know is definitely closer to 3000 than 2000.

The 2000 number gets tossed around a lot because of the 常用 list, but even that list is 2150 now. There's also a few hundred special kanji used in names of places and people. Those also somewhat overlap with the kanji for animals, plants, trees, fruits, etc. Then you toss in older forms of kanji that are used in literature that people don't think of as being different despite the fact that they are. Toss in random body parts like 顎 or 股 that nobody ever thinks about.

By the time you get done tallying up all these things you're left with a number somewhere between 2800-3000 kanji. If you start getting into domain specific stuff or people who like literature that number definitely goes to 3000+.

I know that this is true for a fact because my entire history of Japanese study is documented in Anki, and I've kept track of the kanji statistics for my decks over time. I also know what Japanese people tell me they can read.

Right now the kanji stats for the seen cards in my deck say 2,534. I've only added the cards from RTK3 that I've encountered in the wild so far. Even knowing that many kanji I still encounter kanji I don't know all the time in very common places. I know that my kanji knowledge is still less than that of my adult friends. So the only conclusion I can come to is that they probably know more than 2,500 kanji. I'm guessing the number's somewhere between 2800-3000.

Japanese people, and I think you too, know more kanji than they think they do. Japanese people don't really keep track, and higher level students don't either. I've kept track very meticulously in Anki as I've gone from intermediate to past advanced level. I think if you had kept track you would have found something similar.

I think the key place people have a blind spot is kanji for names and places. It never occurs to them that they should count those too. There's also a lot of words that aren't commonly written in kanji unless people are trying to be fancy, but magically lots of Japanese people can read those too. 林檎 and 薔薇 come to mind.

Fillanzea Wrote:Granted that there are a handful of non-jouyou kanji that are common in fiction, like 訊く and 頷く, but it's only a handful and they can be absorbed pretty quickly.

If by a handful you mean a few hundred then yes you are correct.

Welp, that's that. Financialwar is permanently banned on this forum.

financialwar Wrote:

erlog Wrote:Most Japanese people know around 3000 kanji.

Your credibility went down the drain with this comment. I swear people on this forum just pull bullshit out of their ass and state them as facts

I received my Kanken certificate and report two weeks ago, the pass rate for 準１級 (約３０００漢字） is 15.4%, 15.4% is the pass rate for highly educated Japanese who bothers to study for this test, the pass rate for general population is probably much lower.

Kanken is testing whether people can write them off the top of their head and correlate them to words and situations they're used in. I'm not saying Japanese people can do that. I'm saying Japanese people "know" as in, they have recognition and can read around 3000 kanji.

I'll admit that my using the word, "know," was imprecise.

erlog Wrote:I think the key place people have a blind spot is kanji for names and places.

The Tokyo Subway springs to mind - Many unusual Kanji that you need to be a native to understand.

@erlog
that's all highly debatable, but at any rate just because people may eventually know closer to 3000 kanji doesn't mean that it's a good idea or necessary to go out of your way to learn them all in a deliberate (RTK style) manner, especially not at the start.

After all RTK1 typically takes people a good couple of months of daily review. Double that if you add RTK3. It's a big chunk of time to spend and you're not going to see the returns until you start reading a lot. And you can't read much at first anyway because you don't know words/readings/grammar and have no experience putting it all together.

honyakujoshua Wrote:The Tokyo Subway springs to mind - Many unusual Kanji that you need to be a native to understand.

you could just look them up if/when you go to tokyo. Station names in major cities are almost always displayed in hiragana too.

nadiatims Wrote:@erlog
that's all highly debatable, but at any rate just because people may eventually know closer to 3000 kanji doesn't mean that it's a good idea or necessary to go out of your way to learn them all in a deliberate (RTK style) manner, especially not at the start.

After all RTK1 typically takes people a good couple of months of daily review. Double that if you add RTK3. It's a big chunk of time to spend and you're not going to see the returns until you start reading a lot. And you can't read much at first anyway because you don't know words/readings/grammar and have no experience putting it all together.

I'm really confused by your post here. It's kind of non sequitur. I didn't say these things that you're arguing against.

I never said people should put everything on hold in order to do it. Please quote where I said anything that would lead you to believe that. I even said in one of my responses to Yudanitaiteki that people should be doing RTK concurrently with actual Japanese study + get into native material as soon as possible.

erlog Wrote:Kanken is testing whether people can write them off the top of their head and correlate them to words and situations they're used in. I'm not saying Japanese people can do that. I'm saying Japanese people "know" as in, they have recognition and can read around 3000 kanji.

I'll admit that my using the word, "know," was imprecise.

Agreed, I just thought by stating that most Japanese know 3000 Kanji will mislead people into study more than they otherwise need to.

Unless if you just want to. There's nothing wrong with that.
I probably have learned way more than I'll ever practically use, but I do encounter rare Kanji in all the odd stuff that I read. So, there is an advantage to being a Kanji guru. You also get to learn a lot of bad ass words.

The main reason I was posting was in response to this kind of thinking:

Miyumera Wrote:Well it covers kanji that you would need if you wanted to be able to read a novel apparently since that's the whole reason it was done when one student said she couldn't get through a simple mystery novel with just the basic 2000 kanji.

("simple" mystery novel? What's not "simple" then?)

This is evidently based on this quote from the RTK 3 introduction:

Quote:Having read mystery novels to polish my reading in other languages,
I was disappointed to find that the “essential” or “general-use” charac-
ters were simply not enough to gain entry into the Japanese thriller.
After just a few chapters, my maiden voyage ended on the rocks. So
much for “basic literacy,” I thought to myself. And so was born the idea
for this book.

This is what I was referring to as "bullshit"; Tanya Sienko may have had trouble reading a Japanese mystery novel, but I can't believe that it was because she hadn't learned enough kanji before starting it. This intro is misleading because it seems to say that without learning additional kanji beyond 2000 from a book like RTK 3, you will not even be able to "gain entry" to a Japanese novel because 2000 doesn't even represent "basic literacy".

I'm going to miss FinancialWar </3

Nadia's earlier comments about having some perspective on the language and what to prioritize reminded me of the idea that doing a full RTK at the outset isn't ideal. If learners instead get the very basics of the language down (using a small group of kanji) before starting RTK, they'll have a better understanding of how kanji work and a better ability to access and use information about kanji and words.

They'll also have a better sense of what is required post RTK and whether they even like learning this language. It's a pretty huge upfront investment when some people have no idea even what they're getting into. And it's a long-term project for folks with busy lives. And some people are actually doing RTK even though they want to focus on speech for an upcoming trip! Given that many people apparently quit after RTK, it'd be great if AJATT blogger would re-evaluate his advice in light and include an alternative order.

In the meantime, I was wondering if it'd be worthwhile to put together some sort of 'Intro to Kanji' for those who embark on RTK1 with zero knowledge about the language. It could just be a thread with links to wiki articles and other info on the writing system, kanji, word formation, Joyo, search dictionaries, IME, learning methods, etc. in one place. Does something like this already exist? Or do you think most people seek out this information while doing RTK? (I kinda had the impression that some people were finishing RTK without much understanding of how kanji are read, but I have no sense how prevalent that is.)

I completely agree with you yudan. I think it's bullshit too to have to know that much just to be prepared to read a novel. It's definitely got to deal with a vocab problem.

I'll take one of my books for example.

Tokyo-jima by Natsuo Kirino

Here are the Kanji on the first page

東、京、島、夫、決、籤、引、行、清、子、早、起、下、黒、小、石、覆、入、江、南、洋、思、見、陰、鬱、突、出、大、岩、両、端、挟、圧、迫、感、海、水、盛、上、口、塞、壁、閉、込、増、好、浜、破、隆、何、果、五、年、前、時、嵐、中、狂、喜、度、脱、眺、日、送、素、裸、浸、注、意、必、要、波、揺、底、踏、温、顔、洗、主、役、綺、麗、呟、識、笑、洩、誰、機、嫌、取、奪、合、三、十、民、女、一、人、髪、手、櫛、間、浮、状、藻、四

Let's see, there are 106 Kanji on this page. Only 籤、綺、櫛、呟、洩 are non-general use characters. This is probably harder than other reads, but you can definitely be fine with at least 2000 if not 1500 to get by in this book.

In some ways I feel like an advanced learner can have a better kanji knowledge than a native speaker because native speakers can "cheat" so much more easily than foreigners because of their knowledge of the language. One thing I've noticed about the kanji knowledge of (educated) Japanese people is that their out-of-context kanji knowledge is surprisingly low compared to their in-context knowledge. I think this tends to be because most native speakers typically do not "study kanji" after high school. They learn new kanji, of course, but I think they tend to make a much stronger connection between the symbols and actual language than foreigners sometimes do.

I've sometimes been rather surprised at the kanji that native speakers fail to recognize out of context -- I know that they would absolutely know the kanji if it were in a sentence but when they see it alone, the link isn't there. 籤 is a good example; I imagine the example you saw was either 御籤 or 籤引き -- both of those are pretty common words that almost any native speaker would know, and I think that almost any educated native speaker would be able to read those words in context even without furigana. But most native speakers certainly could not write that character, and I think you might be surprised if you showed people that kanji by itself and asked what the reading or meaning is. Some might know it, but I don't think everyone would.

Of course, this isn't a big problem because there's not a particularly good reason to know kanji out of context. I guess my point is that you can learn to read better without actually studying symbols, although for a foreigner you may have to do that more.

Thora Wrote:it'd be great if AJATT blogger would re-evaluate his advice in light and include an alternative order

Somehow I don't think this is going to happen – if anything, we will see more and more people coming to this forum with AJATT ignited mission.

@yudantaiteki
These days I'm toying with the approach where I try to read without prior kanji study with Heisig and from what I was through so far I can say that Hesig is not necessary for reading. For example, when I see the word 被害 I can easily recognise it without any need to analyse its kanji. I think Hesig experience is helpful in distinguishing the kanji from each other but otherwise detailed knowledge (keywords determination) is not needed. In fact, I haven't studied 被 so my knowledge of this character is limited to knowing that it stands for ひ in 被害 and I would have problems, if it was appearing in some other compound or out of any context.

[Katsuo]

@Imabi. In your example a reader with 2,000 kanji under their belt will still be hindered no less than five times in a single page, whereas a reader who has also completed Ms. Sienko's text will sail right through (籤、綺、櫛、呟、洩 are all in RTK3).

Katsuo Wrote:@Imabi. In your example a reader with 2,000 kanji under their belt will still be hindered no less than five times in a single page, whereas a reader who has also completed Ms. Sienko's text will sail right through (籤、綺、櫛、呟、洩 are all in RTK3).

This is a very misleading way of putting things, though, because it seems to imply that the only thing that determines whether you will be "hindered" or "sail right through" is what frame number you have reached in Heisig. The benefit of the "hindered" version is that at least you know those 5 new kanji you're learning are useful to you, and you may be able to bypass them if you have a good enough grammatical/vocab understanding to pick up some of the meaning from the context.

Of course it's better to know more kanji, the question is always whether it's worth your time to learn them from RTK 3.

erlog Wrote:

Fillanzea Wrote:Granted that there are a handful of non-jouyou kanji that are common in fiction, like 訊く and 頷く, but it's only a handful and they can be absorbed pretty quickly.

If by a handful you mean a few hundred then yes you are correct.

No, I mean maybe a few dozen. But I'm talking about things that you're going to encounter so often that it's going to be a huge barrier to understanding if you don't know them -- and I'm speaking from the perspective of someone who used to wade through novels knowing maybe 500-600 kanji.

From a large sample of fiction texts, the most common non-Joyo kanji are:
伊　云　訊　之　彦　吾　智　坐　廻　阿　嬉　乃
噂　也　此　貰　筈　溜　槍　濡　卿　這　狼　喰　喋
祐　辰　蒼　撫　其　喧　庄　桐　揃　浩　或　逢　馴
杖　傘　鷹　巳　綾　翠　嘩　惚　蓮　綺　李　栗　於
圭　瑞　恰　哉　溢　辻　嘉　庵　磯　薩　弘　馳　劉
茜　勿　昌　汝　蝶　輔　蔭　雀　蘇　耶　尖　紐　桂
柴　聡　鞭　蘭　篇　宋　冴　斐　朋　閃　淋　駕　昂

Out of those, the ones I'd consider important for reading contemporary novels, outside of proper names, are:
訊　廻　坐　噂　貰　溜　濡　這　狼　喰　喋　蒼　撫
喧　揃　逢　馴　傘　嘩　惚　綺　於　溢　蝶　雀　蘇
紐　淋

The others, mostly, are common only for their use as proper names, or in certain domain-specific contexts, or they're common words that are written in kanji much, much less frequently than they used to be (於 is probably one of those, but I think I've seen it recently in more contemporary fiction.)

Should you learn them? Yes. Do you need to study hundreds of extra kanji because you're going to need some of them someday? I wouldn't really think so.

[Katsuo]

yudantaiteki Wrote:Of course it's better to know more kanji, the question is always whether it's worth your time to learn them from RTK 3.

Of course. And that will depend on various factors including an individual's situation and interests.

Ms. Sienko says that lack of kanji knowledge was impeding her progress in other aspects of the language, hence her decision to learn them en masse. I do not think it is reasonable to refer to her preface as "bullshit".

Fillanzea Wrote:Should you learn them? Yes. Do you need to study hundreds of extra kanji because you're going to need some of them someday? I wouldn't really think so.

Also don't forget some of those might have furigana. I basically agree with your "important" list although I would probably add 蔭 and 勿 (for 勿論).

Katsuo:

Quote:Ms. Sienko says that lack of kanji knowledge was impeding her progress in other aspects of the language, hence her decision to learn them en masse. I do not think it is reasonable to refer to her preface as "bullshit".

Although I have no way of knowing, I have a feeling that she either gave up too quickly or had other problems beyond kanji but blamed it all on kanji. I can buy that she found it frustrating to not know some of the kanji in the novel, and that may have caused her to lose her desire or motivation to try to continue through the novel. I can appreciate her desire to learn those kanji on her own. What I cannot believe is this -- that someone would pick up a mystery novel, and be unable to read it solely (or even primarily) because they "only" know 2000 kanji. That's what her introduction says, and Miyumera is not the first person I've seen to take her introduction that way. Taken at face value, what that introduction says is that if you learn only the Joyo kanji, you will not be able to read actual Japanese books, and so you'll need an additional textbook to learn extra kanji you'll need. That is simply false.

Yes the word was 籤引き. In the entire book, there aren't that many more non-常用漢字. However, they are there and there are no readings provided. I can see a learner stumbling on these words, but granted that, I think that it will be the vocabulary and not the Kanji that trips someone up. I know what you say is true yudan. Once I tried seeing if my Japanese friend knew any complicated Kanji. I even asked her if she could spell 憂鬱 out, and she couldn't. However, when written out she automatically recognized it. That's how I am with some words. I can't think of how to spell them out of context. But, if I'm going to take the Kanken 1 one day, I'm going to have to also overcome that to the best of my ability.

@Fillanzea: I feel that all of those Kanji are important to learn because I frequently see them. Now, that may be just because of what I read.

流転輪廻
輪廻転生

One of the easiest books I have on my shelves is 幸せの風が吹いてくる. Even it sometimes used words like 輪廻.

伊 Is seen a lot in the news as it is the symbol used to stand for Italy. It's easy to read at least. I think words like 閃く、訊く、馳せる、etc. would more than likely have readings provided. So, you could just take your chances and hope that the author is nice enough to help you.

[Katsuo]

yudantaiteki Wrote:Although I have no way of knowing, I have a feeling that she either gave up too quickly or had other problems beyond kanji but blamed it all on kanji.

I think her preferred learning style is relatively sensitive to lack of kanji.

Some people who have achieved a moderate level in a language like to progress by reading for pleasure. The interest keeps them motivated and they naturally pick up grammar, vocabulary, idioms, etc. along the way. If they have to pause frequently to check a dictionary the pleasure goes and the method doesn't work.

Ms. Sienko speaks of "polishing" her reading in the same way she did in other languages by reading mystery novels. I take that to mean doing the above.

My inference is that she had to stop to check unknown kanji so often that she could not sustain interest with her usual method. Of course it would be still be possible to read through the book more slowly using dictionaries and picking up kanji, vocabulary, etc. in that way, and some people would prefer that approach. But Ms. Sienko preferred to learn more kanji in advance so she could apply her usual technique.

Quote:My inference is that she had to stop to check unknown kanji so often that she could not sustain interest with her usual method. Of course it would be still be possible to read through the book more slowly using dictionaries and picking up kanji, vocabulary, etc. in that way, and some people would prefer that approach. But Ms. Sienko preferred to learn more kanji in advance so she could apply her usual technique.

Yes, but... I prefer to read for pleasure as a learning method. I prefer to pick up grammar and vocabulary through repeated exposures in context while reading for fun. I prefer to read with very little dictionary lookup. And I have been doing so since I knew quite a small number of kanji.

(Probably 500-600 when I started reading Norwegian Wood, which I read with quite a lot of dictionary lookup, and 700-900 when I read Kitchen and Sputnik no Koibito and a long serious of light novels, which I read with very little dictionary lookup. I may have a very high tolerance for uncertainty, but that's the basis I'm speaking from when I say it's silly to postulate that you would need 3000.)

Obviously I don't know which book(s) she was trying to read, but my own experience -- from having read quite a lot of books by quite a lot of different authors -- is that it's very rare to come across a contemporary novel where a person who knew all the Jouyou kanji would see enough unknown kanji to experience frustration when reading, even if their uncertainty tolerance was quite a bit lower than mine. (They might feel frustration when reading, but that would be more related to lack of vocabulary knowledge than lack of kanji knowledge.)