yudantaiteki Wrote:

Quote:How the word is pronounced is actually of secondary importance.

Unless you want to talk to Japanese people.

Heisig's system is actually based around the whole idea that pronunciation of kanji is secondary. I think the testimony from elated RTKers has been overwhelmingly in favor of this concept.

Btw, are you Japanese per chance? I seem to think you mentioned you were Japanese a while ago, though I could be smoking banana peels.

I am not Japanese.

Heisig's system is based around learning to write kanji first, but it doesn't really have anything to do with whether pronunciation of kanji is secondary or not.

Really? Are we talking about the same meaning of the word "secondary", as in "it's best learn the meaning of kanji first, pronunciation second"?

"The reasons for isolating the writing of the kanji from their pronunciation follow more or less as a matter of course from what has been said. The reading and writing of the characters are taught simultaneously on the grounds that one is useless without the other. This only begs the basic question of why they could not better, and more quickly, be taught one after the other" - J. Heisig

It seems to me it has everything to do with pronunciation being secondary.

"One has only to look at the progress of non-Japanese raised with kanji to see the logic of the approach. When Chinese adult students come to the study of Japanese, they already know what the kanji mean and how to write them. They have only to learn how to read them. The progress they make in comparison with their Western counterparts is usually attributed to their being “Oriental.” In fact, Chinese grammar and pronunciation have about as much to do with Japanese as English does. It is their knowledge of the meaning and writing of the kanji that gives the Chinese the decisive edge. My idea was simply to learn from this common experience and give the kanji an English reading. Having learned to write the kanji in this way—which, I repeat, is the most logical and rational part of the study of Japanese—one is in a much better position to concentrate on the often irrational and unprincipled problem of learning to pronounce them.

In a word, it is hard to imagine a less efficient way of learning the reading and writing of the kanji than to study them simultaneously. And yet this is the method that all Japanese textbooks and courses follow. The bias is too deeply ingrained to be rooted out by anything but experience to the contrary." - J.Heisig

I would suppose Yudan is saying that pronunciation is not secondary in importance.

It's always seemed to me (I might be wrong, of course) that listening (and then speaking) is actually of primary importance, and writing is actually of secondary importance. I've never heard anyone speak in kanji, I must have missed something.
When I see 地獄 I know it is じごく, and it means, well... じごく.
When I see 古里 I know it is ふるさと, and it means, well...ふるさと, not a tombstone and a computer.

EDIT
Any successful strategy of learning written forms of kanji boils down to this: a. dissect kanji into components recurring in many kanji b. name the components and use them as building blocks to remember new kanji c. learn stroke order rules.

buonaparte Wrote:I've never heard anyone speak in kanji, I must have missed something.

And I've never seen someone write with sound waves. Maybe I missed something.

Taishi Wrote:That is indeed what they had in mind when writing old Chinese, from what I understand it is almost completely separate from the spoken Chinese at the time and the characters were used solely for their meaning.

That doesn't even make sense. Why would they go through the trouble to develop a writing system (which by definition records a spoken language), make the characters in all their phonetic glory, and then not use it to record a language, but instead invent some written code that bears no relationship to the spoken language? Of course Old Chinese (which refers to the spoken language of the time, by the way) is reflected in the writings of the time. Scholars believe they pretty much wrote the way they spoke, with some stylistic self-edits appropriate to the formality of writing something down (just like we do today). It wasn't until a good bit later (post-Han dynasty) that the written and spoken languages diverged, because the classical writings were seen as the gold standard while the vernacular had moved on.

And the characters were used for their sounds just as much as for their meanings, if not more. This is why you see some characters substituted for others which have a completely unrelated meaning. They sounded identical or similar, and maybe the author couldn't remember the "right" one, so he used the one he knew. Characters were also borrowed to write similar-sounding words that don't have characters of their own yet. It happened all the time (and happens now too; in fact it ones of the main principles at work in written Cantonese). Some of these borrowings became so common that they were accepted as the correct character (來 is an example; it originally meant 麥). These are now generally called 假借字.

Chinese characters have never been "divorced" from sound as so many people seem to believe. They were developed as a coherent, logical system, both semantically and phonetically. The changes in pronunciation over the past few thousand years and their adoption into other languages do not change that fact.

JimmySeal Wrote:

buonaparte Wrote:I've never heard anyone speak in kanji, I must have missed something.

And I've never seen someone write with sound waves. Maybe I missed something.

But everybody writes/types with 'sound waves'! When you want to type/write 地獄 you have to say it in your mind somehow, it is probably じごく or jigoku, anyway it is closer to to the spoken way than to kanji.
Even Chinese hazi had both a semantic and phonetic hints.

So there you are. Anyway, language is a system of interdependent sub-systems.

bflatnine Wrote:

Taishi Wrote:That is indeed what they had in mind when writing old Chinese, from what I understand it is almost completely separate from the spoken Chinese at the time and the characters were used solely for their meaning.

That doesn't even make sense. Why would they go through the trouble to develop a writing system (which by definition records a spoken language), make the characters in all their phonetic glory, and then not use it to record a language, but instead invent some written code that bears no relationship to the spoken language? Of course Old Chinese (which refers to the spoken language of the time, by the way) is reflected in the writings of the time. Scholars believe they pretty much wrote the way they spoke, with some stylistic self-edits appropriate to the formality of writing something down (just like we do today). It wasn't until a good bit later (post-Han dynasty) that the written and spoken languages diverged, because the classical writings were seen as the gold standard while the vernacular had moved on.

And the characters were used for their sounds just as much as for their meanings, if not more. This is why you see some characters substituted for others which have a completely unrelated meaning. They sounded identical or similar, and maybe the author couldn't remember the "right" one, so he used the one he knew. Characters were also borrowed to write similar-sounding words that don't have characters of their own yet. It happened all the time (and happens now too; in fact it ones of the main principles at work in written Cantonese). Some of these borrowings became so common that they were accepted as the correct character (來 is an example; it originally meant 麥). These are now generally called 假借字.

Chinese characters have never been "divorced" from sound as so many people seem to believe. They were developed as a coherent, logical system, both semantically and phonetically. The changes in pronunciation over the past few thousand years and their adoption into other languages do not change that fact.

I'm not an expert on classical Chinese, but from what you're saying it seems I was referring to the post-Han dynasty that I was referring to. Call me unknowledgable (I would), but almost 2000 years ago is old enough to be called old in my book.

And what I meant was that from what little I've read about classical Chinese, it was like an abbreviated from of normal speech, but from what you're saying, that's not the case, or at least, wasn't always the case. I won't contest your knowledge on old Chinese, but I felt like you took what I said and inflated it. I didn't mean that the written system had no connection to the spoken language, of course the readings of the characters are based on the spoken language. I just meant that you wrote in a highly concise manner.

buonaparte Wrote:

JimmySeal Wrote:

buonaparte Wrote:I've never heard anyone speak in kanji, I must have missed something.

And I've never seen someone write with sound waves. Maybe I missed something.

But everybody writes/types with 'sound waves'! When you want to type/write 地獄 you have to say it in your mind somehow, it is probably じごく or jigoku, anyway it is closer to to the spoken way than to kanji.
Even Chinese hazi had both a semantic and phonetic hints.

So there you are. Anyway, language is a system of interdependent sub-systems.

Of course 地獄 means hell just as much as じごく, I'm not contesting that. But haven't you ever heard a Japanese person explain the meaning of a word by explaining the kanji making up the words?

Sure you can learn Japanese just fine without learning or taking any advantage of kanji whatsoever, but some things just don't make as much sense that way. Sort of like using romaji instead of kana. Ka turning into ga or tsu to zu doesn't make as much sense as か to が and つ to づ, but that doesn't mean you're all helpless and can't learn Japanese, it just means you can't take advantage of all the logical connections that exist.

In short, yes I agree that written and spoken language (kanji text vs sound only) are two entities that are both just as Japanese as the other, that can be learned separately, but that doesn't meant it's the best idea.

buonaparte Wrote:But everybody writes/types with 'sound waves'!

Not people who type with four-corner codes.

And besides, no, the only people who type with sound waves are people using ASR software. Everyon else types with keys. People "speak with kanji" just as much as they "type with sound."

JimmySeal,
you must be pulling both of my legs.

Let's say you want to write シールさんはつらいよ. How on earth can you do it without saying it to yourself in your mind?
Even you don't know any Japanese and start learning Japanese/kanji the Heisig way, you have to mutter under your breath (in English!!!) something like 'scary clown' and write by hand let's say 習 or whatever.

Yes, indeed, people visualize sounds in their head when they write, just like the way literate Japanese speakers visualize kanji in their heads when they speak. You started all this off with your facetious "I've never heard anyone speak in kanji" comment and I am demonstrating that that same logic can be applied in the opposite direction.

I'm thinking what's derailing this thread a bit is that some people are talking from the perspective of kanji learners and others - of kanji users.

There is a fundamental difference depending on whether you are recognizing 地獄 as a familiar word with a familiar reading versus trying to acquire 地獄 as new vocab.

If you already know the word you're not going to analyze the meaning of the kanji. In fact, there's been research of Japanese dementia patients who had their ability to process meaning impaired but they could still read kanji fluently. The researchers said their findings "suggest the existence of an independent orthography-to-phonology transcoding procedure for kanji words". This may be something Heisig explored in RTK vol. 2 where "signal primitives" could be that kind of transcoding procedure. I'm actually more inclined to think that we simply do an OCR (optical character recognition) of familiar shapes and retrieve pronunciation via a shortcut. Again, there is a difference between familiar and new vocab where "signal primitives" are helpful for correct pronunciation of unfamiliar kanji.

If you're feeling particularly nerdy, you can check out a research paper titled "Semantic context effects when naming Japanese kanji,but not Chinese hànzì". In short, they concluded that "Chinese hànzì are likely read via the direct route from orthography to phonology" (i.e. without parsing meaning, straight into pronunciation) but for kanji they detected an "activation from the semantic system" (i.e. the Japanese subjects had to pause think about the kanji meaning to produce correct pronunciation). The researchers explain it by saying that "Chinese and Japanese logographs differ in that Japanese kanji often have more than one pronunciation whereas Chinese hànzì generally have a single pronunciation."

Taishi Wrote:I'm not an expert on classical Chinese, but from what you're saying it seems I was referring to the post-Han dynasty that I was referring to. Call me unknowledgable (I would), but almost 2000 years ago is old enough to be called old in my book.

You do realize that's like saying you can attach any term to anything you like just because it's that way "in your book", right? Old Chinese is a term with a definition (that definition is not "Chinese that is old", and I was correcting your use of it. No need to defend your error. This is a Japanese board, so I don't expect everyone to know these things. Just trying to inform a bit.

Quote:And what I meant was that from what little I've read about classical Chinese, it was like an abbreviated from of normal speech, but from what you're saying, that's not the case, or at least, wasn't always the case. I won't contest your knowledge on old Chinese, but I felt like you took what I said and inflated it. I didn't mean that the written system had no connection to the spoken language, of course the readings of the characters are based on the spoken language. I just meant that you wrote in a highly concise manner.

The written language was more concise because the spoken language was more concise. The spoken language was more concise because the phonology of the language was richer, which means there was less need for multisyllabic words.

yudantaiteki Wrote:

Quote:How the word is pronounced is actually of secondary importance.

Unless you want to talk to Japanese people.

Sorry, but that's not correct. Kanji will greatly improve your ability to Japanese people. Here's why:

Kanji is a huge shortcut to amassing vocabulary, because it links words that otherwise would appear unrelated. Instead of trying to memorize seemingly unrelated sounds--like car, bicycle, and train--you learn the kanji and the sounds. The sounds help you communicate, but the kanji helps you make sense of everything, and it's much easier to remember things that make sense.

It's the difference between a description of a map, and a map.

Think you're going to remember several thousand words without writing anything down? Of course not. So you can either write down a bunch of sounds that bear no relation to one another, or write down the sounds and the kanji, so you can understand the relationships.

It may seem like more work at first. That's because it is. But kanji is the one thing about the language that makes sense. Drawing a map takes a bit more time, but it makes things much clearer.

Anyone trying to learn Japanese without using kanji is making a huge, fundamental mistake. As you increase your vocabulary, you're merely compounding the problem. With kanji, you've got a built-in system that makes sense of everything. Without it, you've just got a random collection of words.

Kanji can be thought of as a place-holder (or a variable, if you're a programmer). Let's say you write a sentence in English: "The woman was attacked by a dog."

And then you think, No, that's not very interesting. Instead of "dog," I'll say "mongrel," "cur," "canine," or "hound." Right there--that's what kanji is for. With kanji, you use 犬 as a place-holder for all those other words. 犬 means "dog," but how you pronounce it--mongrel, cur, canine, or hound--doesn't matter. It's all still "dog."

That one place-holder then serves as a foundation upon which to build other words--puppy, big dog, whatever you want.