I wanted to start learning Kanji, but there is a thought I can't get out of my head. I really only want to be able to read Japanese, and for that I don't see any reason to actually know how the kanji are spoken in Japanese. Onyomi and Kunyomi shouldn't be a problem, same for Radicals. So, would it be viable to translate only once, from the kanji to English? (I'm guessing it's not a good idea, but this would keep bothering me if I wouldn't ask)

生 - life
生ビール - draught beer
生ごみ - compost
生む - to give birth
先生 - teacher
生きる - to live
生かす - resuscitate
異生 - An unenlightened person (Buddhism)

Take that as you will.

Yes, I know that one kanji can have different meanings with kana behind it or together with other kanji, but I don't understand how knowing the pronunciation would be of help.

If you're just interested in reading you don' need to pronounce it correctly, but you're gonna need to assign it some sort of pronunciation. So deciding that 生 is pronounced "mark" is fine, as long as you don't plan on interacting with any form of the language aside from text. the problem with this is that the use of kana is not consistent. so 生 is sometimes written as なま and some times as ナマ and finally as 生. You can pronounce them all as "mark", however if you gonna need to learn vocabulary anyway, there's no real point in arbitrarily assigning pronunciations. Go for something close enough. so if you reading something and you call 富士山 with an English accent that's fine, however kanji and english don't translate well into each other.

I think what RandomQuotes is trying to show through an example that kanji aren't actually the Japanese you want to read, they are signs used to represent it. So you can't read them without knowing what they represent, it would be like learning an averbal language. Concretely you can't read umu (生む) and ikasu (生かす) if you don't know how they sound. Well maybe you can for a few dozens like some sort of highway code and assign only a meaning, but I doubt you can for the few thousands that will constitute your vocabulary.

edit: ninja'd

I feel like when readings come together to form a complete word, they're acting as an additional reference point in my brain. I'm still a beginner so take this with a massive grain of salt, but being able to speak out loud has definitely helped me to remember both the meanings of certain words and how they are written. In fact a few readings have stuck better than their corresponding Heisig keywords. I can't imaging how arduous it would be to read something and only be able to decipher the meaning by essentially adding up English meanings. Learning to do it sounds like a lot more trouble than it is worth. Again, I'm new to this so maybe I'm talking rubbish.

Nanatsu Wrote:I wanted to start learning Kanji, but there is a thought I can't get out of my head. I really only want to be able to read Japanese, and for that I don't see any reason to actually know how the kanji are spoken in Japanese. Onyomi and Kunyomi shouldn't be a problem, same for Radicals. So, would it be viable to translate only once, from the kanji to English? (I'm guessing it's not a good idea, but this would keep bothering me if I wouldn't ask)

I'm not sure exactly what you mean by translating kanji directly to english. But I can say that there just isn't a one to one correlation between kanji and english. Many kanji don't have a single pronunciation, and as RandomQuotes illustrated, they mean different things in different contexts. If you want to see what I mean, try machine translating japanese and see if it makes any sense. So translating kanji to english doesn't really have any meaning unless you mean translating "words written in kanji" to "words written in english" which you're going to have to do anyway if you plan on reading Japanese.

In order to make this easier, many people find it useful to assign a name to each kanji which is probably as close as you're going to get to "translating kanji into english". It could be any name, but many RTK keywords provide hints to what a kanji means in a compound. Many people make up their own keywords which is fine. But RTK is a widely used book, so using the same keywords as thousands of other people has the benefit of being able to talk about them with other people.

I hope one of us answered your question.

I haven't read the replies.

I must say it's one of the most intriguing questions I have ever come across (no irony).
Asking questions is a good thing, of course, but seeing what happens is sometimes better.
Try to read your English way ANY Japanese text, even the simplest one.

EDIT
OK, here's an example. Everything's ready, even a recording.
http://www.coscom.co.jp/newsweather/weat...nki-c.html

If you read 僕 as "I", but you don't know the japanese pronounciation, how can you read the same word when it is written in kana? If you want to read you want to know the japanese word to assign to kanji or compound because sometimes the same word is written in kana and knowing only the kanji equivalent does not help in this.

And as JKS87 said, knowing the pronounciation of a word helps in learning/understanding/remembering other words which share the same kanji.

If you know that "conditions" in japanese is "jijou", and "information" is "jouhou" your life is much easier when you encounter those two words written as kanji compunds, because the "jou" part is written with the same kanji for both the words, 情: 事情 and 情報.

If you instead limit yourself to attribuite english meanings to 事情 and 情報, when you encounter those compounds while you're reading, you'll have a hard time recalling the english meaning because your kanji to english attribution is sort of arbitrary in fact of pronunciation.

It is true that at first you have the added difficulty of learning the japanese pronunciation, but with time and more words on your belt you'll see that some kanji have only one or two common pronounciations in compounds, so...

In other words you're doing something more difficult with much less gain as a result.

Example for 場 here. As you can see it reads almost allways as "ば (ba)" or "じょう (jou)".

Look a particular example from the link: "運動場" that in english is "playground", while "運動" is "(physical) exercise". So while the first compound include the second compound, their english pronounciation is not correlated. But is you read them in japanese, the first word is "undoujou" while the second is "undou". And the same is for "運動会 (undoukai)" etc..

Another thing is that the same word need to be translated in various ways in different contexts if you're reading directly in english, like the previous example "事情", when you encounter this word how do you read it? Situations, circumstances, conditions etc..? I assure you that based on where it appears one translation would be not suitable, so you cannot stick on a single english word for a kanji compound. Instead if you read it in japanese, you read it allways as "jijou" and let your brain interpret the word subconsciously, the same way as it happens when you encounter the words "like" or "lie" in english. But let's pretend the two words are ideogrammatics and not phonetical. What happens if I assign the italian word "come" to the english word "like (I'm just like you)", when I encounter the same word in a context where it means "piacere (I like that girl)"?

Thanks for all the answers, I think I'm able to abandon that idea now. I don't know how often kanji are written as kana, but that could definitely be problematic. Then there are furigana...

I think it will take a while until I really understand these correlations though. Knowing even both single kanji of a word written with only two kanji still doesn't give me the meaning of that word, so I still would have to memorize that word the same way as I would throwing something like "fish" and "ocean" together.

I think one problem is your assumption that translation to English is a necessary part of reading Japanese -- you seem to be asking why we should go kanji->pronunciation->English rather than just cutting out the middle man and doing directly kanji->English. However, you don't want to be translating into English -- your goal is to think and process everything as Japanese, not as something to translate into English.

yudantaiteki Wrote:I think one problem is your assumption that translation to English is a necessary part of reading Japanese -- you seem to be asking why we should go kanji->pronunciation->English rather than just cutting out the middle man and doing directly kanji->English. However, you don't want to be translating into English -- your goal is to think and process everything as Japanese, not as something to translate into English.

Exactly, there's no reading and understanding Japanese fluently without overcoming the constant translating.

Kanji may seem like confusing glyphs that need to be professionally deciphered to you now, but further on it will be more and more like reading naturally. Just in a different language.

You can't read Japanese text (and understand it) just from knowing the Kanji, anymore than you can read this: "Oli aikoinaan suorastaan vavahduttava kokemus, kun osaamieni verbien lukumäärä kaksinkertaistui kertaheitolla" just because you know all the letters. Kanji aren't ideograms. You need to learn the language to understand them.

yudantaiteki Wrote:I think one problem is your assumption that translation to English is a necessary part of reading Japanese -- you seem to be asking why we should go kanji->pronunciation->English rather than just cutting out the middle man and doing directly kanji->English. However, you don't want to be translating into English -- your goal is to think and process everything as Japanese, not as something to translate into English.

That seems reasonable considering I also don't have to translate from English to German anymore. Now that I think about it, that would rather be an annoyance.

So, according to the previous replies, a deaf person trained to communicate only using the sign language and having no access to oral speach could not possibly learn to *read* Japanese?…
Interesting question, though.

You don't need a kana-to-sound mapping to learn to read, it's just a convenience for us hearing people. You could just as well map them to Yubimoji. (I used romaji because the English Wikipedia article is better than the Japanese one).

jmignot Wrote:So, according to the previous replies, a deaf person trained to communicate only using the sign language and having no access to oral speach could not possibly learn to *read* Japanese?

I imagine that a deaf person (someone who can't even hear with a hearing aid) would learn the mapping between Kanji writing and Kana writing, in a way that's similar to how a hearing person has to learn the mapping between Kanji writing and Kana reading. The only difference is that the deaf person doesn't really know what the Kana sound like. But he can (and needs to) memorize them all the same. And, assuming Japanese is his second language, he can map the Kana with the characters used to write his first language (same way we start out), so he doesn't have to start from scratch.

The only significant difference, for a deaf person learning his second language, would be the lack of audio. Basically, he would face the same challenge a hearing person who refuses to use audio (or a hearing person living outside of Japan, before the invention of the radio) would('ve) face(d).

It would be more challenging, but still nowhere near as challenging as ignoring the Kana writing of Japanese words. There is a lot of information in the Kana writing (especially in the Yamato words, but the Kango as well), that you can't get from just looking at the Kanji. Plus, I imagine that if a deaf person went through the trouble of learning Japanese, they would probably want to be able to also lip read to some extent.

Helen Keller, a deafblind person, is said to have learned foreign languages.
She was able to understand speech by touching people's throats and mouths.

Vempele Wrote:You don't need a kana-to-sound mapping to learn to read, it's just a convenience for us hearing people. You could just as well map them to Yubimoji. (I used romaji because the English Wikipedia article is better than the Japanese one).

Oh yeah, well せ you too.

jmignot Wrote:So, according to the previous replies, a deaf person trained to communicate only using the sign language and having no access to oral speach could not possibly learn to *read* Japanese?…
Interesting question, though.

People that are deaf from birth often do have a difficult time learning to read, since the writing systems are typically based on speech rather than a sign language. I believe there are various methods for teaching deaf people to read involving "chaining" via sign language (which later gets dropped).

Relevant?
https://answers.yahoo.com/question/index...035AAyRoCq

Inny Jan Wrote:Relevant?
https://answers.yahoo.com/question/index...035AAyRoCq

VERY relevant, in particular the sentence between parentheses at the end of the first sentence:
"Hearing people often "hear" a voice in the back of their head that typically sounds like their own voice, while thinking or even reading (especially if their native language is phonetic like English as opposed to a non-phonetic language like Mandrin Chinese).
The statement is a bit awkward (Mandarin Chinese certainly cannot be termed "a non-phonetic language"), but it suggests the interesting idea that our need to rely on pronunciation for learning to read a foreign language may be influenced by the phonetic nature of the latin alphabet we have been educated in.
Perhaps someone with a different background could find it more natural to associate directly the written material to a mental representation not necessarily conveyed by a sound.

jmignot Wrote:

Inny Jan Wrote:Relevant?
https://answers.yahoo.com/question/index...035AAyRoCq

VERY relevant, in particular the sentence between parentheses at the end of the first sentence:
"Hearing people often "hear" a voice in the back of their head that typically sounds like their own voice, while thinking or even reading (especially if their native language is phonetic like English as opposed to a non-phonetic language like Mandrin Chinese).
The statement is a bit awkward (Mandarin Chinese certainly cannot be termed "a non-phonetic language"), but it suggests the interesting idea that our need to rely on pronunciation for learning to read a foreign language may be influenced by the phonetic nature of the latin alphabet we have been educated in.
Perhaps someone with a different background could find it more natural to associate directly the written material to a mental representation not necessarily conveyed by a sound.

It has also been confirmed through research that when reading, hearing people will "subvocalize" by moving their vocal chords -- this doesn't mean literally sounding out the words with your lips, though.

The great reliance of reading on sound and speech isn't really related to the writing system or the language -- it's the fact that virtually all non-deaf people learn to speak their native language before they learn to read. By the time you are learning to read, "language" as far as your brain is concerned, is speech. There's a persistent myth that this is different for languages using Chinese characters, but that notion has no support. Chinese people aren't learning to read by associating Chinese characters with abstract meanings -- they learn by seeing how the written symbols represent the spoken language they already know.

yudantaiteki Wrote:The great reliance of reading on sound and speech isn't really related to the writing system or the language -- it's the fact that virtually all non-deaf people learn to speak their native language before they learn to read. By the time you are learning to read, "language" as far as your brain is concerned, is speech. There's a persistent myth that this is different for languages using Chinese characters, but that notion has no support. Chinese people aren't learning to read by associating Chinese characters with abstract meanings -- they learn by seeing how the written symbols represent the spoken language they already know.

I fully agree… well, at least 95%
Yet I remember reading a report on studies showing that different areas are activated in the brains of people reading texts written in either phonetic or ideographic scripts. This seems to imply that, despite the fact that reading abilities heavily rely on prior command of oral speech, the type of script may still have an influence. After all, most children learn to read at an age when their brains still have considerable flexibility in comparison to adults.

Coming back to the OP's point, one example now comes to my mind.

I remember reading one of those XIXth century Russian novels in which the characters all had complicated (to me) names and patronymics. I am pretty sure that, after reading 500 pages of the book, I stil had no idea of how they would be pronounced. I just kind of took a mental snapshot of how the word pattern looked like and associated it to that particular character for the rest of my reading.

In the case of kanji words, it should actually be much more effective because the characters often carry a meaning by themselves which is related to that of the word where they occur.

As I struggle to read a novel in Japanese which is somewhat above my level and thus contains a lot of unknown words, I feel that I tend to do the same: trying to remember a word from its kanji components to recognize it the next time it will pop up in the text, sparing the extra effort of memorizing its pronunciation.

Whether this is good or not for the learning process on the long term is not clear to me.