A book to learn Kanji through real etymologies

Hi there.

I've been studying kanji for more than 7 years and I've come up with a book to teach kanji through real etymologies, which I think is pretty relevant to this board.

In order to publish the book I've launched a Kickstarter campaign and we are on track but still there is a long way left. Many people have shown very good feedback but the project doesn't get much exposure outside the Kickstarter community, so I'm sharing it with people that may have interest in such a book.

Feel free to check the project page here and see if you like the book and think it's for you!

In case you wonder, I'll sum up what's different compared to other kanji learning books out there:

The foundation of the book is the deep etymological research I've made. The basic (indivisible) compounds used in the Chinese character formation are actually pictograms, and we have access now to those pictograms in their bronze script and oracle-bone script forms. I include this glyph besides each component in order to show the real (original) form and how they suggest a meaning and work in relation to other characters. This way you can depict quite clearly a quite comprehensive system of character formation: you start to naturally learn characters and see the logic behind it. The compound characters are deductible from the components you've learn before (although in each new entry the character is entirely explained through its components one more time). You can see an example of this in the Kickstarter page itself.

In addition to that and related to it, is the order in which the characters are learnt. I'll explain you with an example:

• First you learn 145 radicals at once (although they are already grouped by topics)
• Then the book gets divided in levels of difficulty based on school grades, from level 1 to 6 and then secondary school. Each level is a different chapter.
• We go with the chapter 1 (level 1). The chapter first gets divided into topics (human world, natural world and man-made world).
• Lets start with the human world. There we have the first phono-semantic compund: 交 (garde 1), then we have the character 校 (also grade 1), and then the character 効 (grade 5)
• The characters 絞 and 郊 also share the component 交 but they are of the secondary school level, so they will appear in the level 7 (secondary school grade chapter)

In this way you can learn the 2136 joyo kanji in a very efficient and smooth way. I think the book not only makes a good learning system but also a valuable academic resource.

If you are skeptical and think the examples on the page are cherry-picked or don't show the whole system you can ask for any kanji's etymology (within the joyo list) and I'll explain it to you here according with the content of the book. I have not gone further with examples in the page because kanji are teaching gradually and you need to know some components first in order to smoothly learn new kanjis, in any case, in a board like this I have more room to write at length.

I'm up to any suggestion as well!

A Sample of the book:

頁 is only a radical? Since when?

Mike Cash said:

頁 is only a radical? Since when?

Click to expand...

It is if you stick with the joyo kanji list and the current usage. In contemporary kanji 頁 as a standalone only could be used for the meaning of "page", and in that case it's seldom used as nowadays it's commonly written in kana. [頁 is somehow preserved in jimena kanji (kanji for proper nouns), but I don't address this category in the book since many of them are outside the joyo list.

I'm still rather skeptical now that the organization of the book is more clear.

The Kickstarter page praises it as being "the easiest way to learn kanji" and even "the ultimate solution", and therefore presumably has all foreign learners of Japanese as its target audience, even those who aren't interested in etymology per sé. However:

• The book denies the similarity of modern characters like 父 and 交 just because they weren't similar 3000 years ago. This doesn't improve learning efficiency - it reduces it.
  
• It teaches similar-looking characters like 校 and 絞 in different chapters - thereby once again reducing learning efficiency - just because Japanese schoolchildren happen to learn them in different grades. The target audience for this book consists of non-Japanese who are in a completely different environment; they aren't under any pressure to learn the most common characters first.
  
• It provides mnemonics that are based on how the characters looked 3000 years ago, not on how they look today. In fact, it doesn't take long for it to go against its main premise of "etymologically sound mnemonics are better than random ones": the explanation for 天 first gives the real origin, and then immediately gives a random mnemonic instead because it's easier to remember... Thereby admitting that the original one isn't suitable anymore.

I learned the bulk of my kanji from a free online resource about three years ago. It provided mnemonics that didn't just incorporate the character's meaning, but also its readings. It grouped characters by visual similarity regardless of school grade or etymology, and even provided hyperlinks to radicals and subcharacters (something a book can never do).

Looking at the example pages of this book now, I can't say I would've preferred to learn from that instead. Sorry.

Speaking of which, are there any testimonials from people who've used this book? Are there people who've studied kanji using both (a preview version of) your book and other resources, and can attest that the etymology approach indeed works better?

alexadler said:

jimena kanji (kanji for proper nouns)

Click to expand...

Sorry for nitpicking, but it's jinmei kanji (人名漢字, or more correctly 人名用漢字 jinmeiyō kanji), not jimena.

I can't see any point whatsoever for learning 145 radicals at the beginning.

lanthas said:

I'm still rather skeptical now that the organization of the book is more clear.

The Kickstarter page praises it as being "the easiest way to learn kanji" and even "the ultimate solution", and therefore presumably has all foreign learners of Japanese as its target audience, even those who aren't interested in etymology per sé.

Click to expand...

Yes you are right, that's the target audience. I believe that etymology, besides being an academically interesting topic, it can help students in the case of learning the graphic forms of kanji. They function as a very strong mnemonic that stick together most of the characters and make associations and retention easier.

However:

• The book denies the similarity of modern characters like 父 and 交 just because they weren't similar 3000 years ago. This doesn't improve learning efficiency - it reduces it.

Click to expand...

父 and 交 are only graphically similar, you would have to make up an story in order to match those characters, but that match is not as beneficial as knowing that 交 are "crossing legs" and 父 doesn't have anything to do with it. I don't know where it reduces efficiency, since making a story up to match does characters just take you more time and effort. Knowing the etymologies (I'm not telling you to learn the ancient form, the glyphs are there to have a quick reference) helps you making the associations between characters natural in a way that at the end of the day everything makes sense.

• It teaches similar-looking characters like 校 and 絞 in different chapters - thereby once again reducing learning efficiency - just because Japanese schoolchildren happen to learn them in different grades. The target audience for this book consists of non-Japanese who are in a completely different environment; they aren't under any pressure to learn the most common characters first.

Click to expand...

The book is designed in a way that goes to the pace of learning Japanese language as well. 校 (school) is a kanji that you are going to use very early on. But if you learn 絞 (wringing) early on, it is very possible you end up forgetting it, because in the early stages of learning Japanese language you are not going to find or use that kaji very much if at all. It's better to learn those less frequent characters once you have already learnt the chore of the kanji so know you can enjoy reading more Japanese texts. At that point you'll already know the component 交, so learning 絞 later on shouldn't be a problem, while learning these unfrequent characters early I think its yet unnecessary.

• It provides mnemonics that are based on how the characters looked 3000 years ago, not on how they look today. In fact, it doesn't take long for it to go against its main premise of "etymologically sound mnemonics are better than random ones": the explanation for 天 first gives the real origin, and then immediately gives a random mnemonic instead because it's easier to remember... Thereby admitting that the original one isn't suitable anymore.

Click to expand...

You are wrong with your first statement since how characters look today is completely related on how characters looked originally. The case of 天 is exceptional since it is an standalone character very rarely used as a component. In these exceptional cases I just give another mnemonic option, but in the majority of phono-semantic compounds the etymologies are plain straightforward once you know them. You can't throw an entire system using just one example.

I learned the bulk of my kanji from a free online resource about three years ago. It provided mnemonics that didn't just incorporate the character's meaning, but also its readings. It grouped characters by visual similarity regardless of school grade or etymology, and even provided hyperlinks to radicals and subcharacters (something a book can never do).

Click to expand...

There is also many people that learn the characters just when they find them or need them and they are fine with it. Others use plain repetition and so on and so for. Everybody's different, however what I offer is something that simply didn't exist yet.

Looking at the example pages of this book now, I can't say I would've preferred to learn from that instead. Sorry.

Speaking of which, are there any testimonials from people who've used this book? Are there people who've studied kanji using both (a preview version of) your book and other resources, and can attest that the etymology approach indeed works better?

Click to expand...

Nothing to be sorry haha. What I find a little bit frustrating is that it's quite hard to explain and show the system of the book in just one page. I think people will start to appreciate it more once it becomes published (if it does).

Unfortunately the book is not published yet (that's why I run the campaign) so I don't have the kind of testers you've mention, you could be the first one haha!

Thank you in any case for that critical analysis, it's always good for me to revise and trying to better explain my book

Toritoribe said:

Sorry for nitpicking, but it's jinmei kanji (人名漢字, or more correctly 人名用漢字 jinmeiyō kanji), not jimena.

Click to expand...

Right, I believe the corrector betrayed me >_<

Mike Cash said:

I can't see any point whatsoever for learning 145 radicals at the beginning.

Click to expand...

It's a shortcut. Since the vast majority of characters are composed by one of those radicals, if you learn them at the beginning you just have set the pathway. Besides they also serve as a good introduction to the book system and kanji in general

alexadler said:

父 and 交 are only graphically similar, you would have to make up an story in order to match those characters, but that match is not as beneficial as knowing that 交 are "crossing legs" and 父 doesn't have anything to do with it.

Click to expand...

But that's the thing: when you're reading Japanese text, the graphic is all you have. There's no reason why you couldn't see 父 as a pair of crossed legs as well.

Matching characters is beneficial because it reduces learning effort; and knowing that the combined character (交) is unrelated to its components (父 and 亠) speaks for itself, because this is almost always the case. 熊 has nothing to do with 能 or 火. 暗 has nothing to do with 日 or 音, and so on and so forth. You're almost always going to end up with a weird, far-fetched, arbitrary mnemonic.

Even if the book doesn't connect the stories though, should it not at least point out the visual similarity between the modern characters ("watch out: 父 also looks like crossed legs but is missing the person's head") so that students don't risk mixing them up?

alexadler said:

I don't know where it reduces efficiency, since making a story up to match does characters just take you more time and effort.

Click to expand...

The point is not to have students make up their own stories, but to provide stories that make learning as easy and efficient as possible. And it's more efficient to learn a character as a combination of other characters than to learn it from scratch, regardless of whether that combination is "canon" or not.

alexadler said:

Knowing the etymologies (I'm not telling you to learn the ancient form, the glyphs are there to have a quick reference) helps you making the associations between characters natural in a way that at the end of the day everything makes sense.

Click to expand...

Are they truly better for making the association though? Do they really still make sense at this point in time? I took a look at some of the other examples:

右: "the right hand is the one that's used for grabbing food." That's nice, were it not that the hand has moved to the left in the meantime. What students need is a way to tell 右 and 左 apart based on the item that the hand is grabbing, not on the hand itself. For example: katakana ロ for Right and エ for lEft.

百: "words coming out of the protruding tongue (白), counting to a hundred." Wait, doesn't 白 mean "white" these days? And who counts with their tongue sticking out anyway?
Better mnemonic: turn the character 90° counterclockwise and you get "|-口口", which bears a striking resemblance to the number it represents.

九 doesn't look anything like an outstretched arm anymore. What it does look like is 力. And you know what the scouter says about his power level, right? Indeed, it's over nine thousand.

Then there are cases where the original mnemonics overlap. 天 and 頁 have the same one: a person with an enlarged head. And yet, one receives the roundabout "that means this person is aware of the heavens" treatment while the other does not.

alexadler said:

The book is designed in a way that goes to the pace of learning Japanese language as well. 校 (school) is a kanji that you are going to use very early on. But if you learn 絞 (wringing) early on, it is very possible you end up forgetting it, because in the early stages of learning Japanese language you are not going to find or use that kaji very much if at all. It's better to learn those less frequent characters once you have already learnt the chore of the kanji so know you can enjoy reading more Japanese texts. At that point you'll already know the component 交, so learning 絞 later on shouldn't be a problem, while learning these unfrequent characters early I think its yet unnecessary.

Click to expand...

For students who do read Japanese text while they're studying kanji (which is indeed the best way to make them stick), this is fair enough. However, when similar-looking characters are not introduced together, it's all the more important to point them out. Looking at e.g. 九 vs 丸, this book doesn't appear to do that, which results in situations like:

"Wait, 士? Wasn't that 'earth'? I could swear I've seen it before... (Leaf through pages) Oooh, that was 土 which looks almost exactly the same. Couldn't they have told me?"

alexadler said:

You are wrong with your first statement since how characters look today is completely related on how characters looked originally. The case of 天 is exceptional since it is an standalone character very rarely used as a component. In these exceptional cases I just give another mnemonic option, but in the majority of phono-semantic compounds the etymologies are plain straightforward once you know them. You can't throw an entire system using just one example.

Click to expand...

That's fair enough - I know nothing about kanji etymology, so all my comments are based on what I saw in the sample pages. And what I saw there is that 天 is definitely not the only character that has evolved beyond recognition (see list above), and that the mnemonics for compound characters are just as arbitrary as the ones I made myself when I was learning.

境:
TRKW: "The land limit from which convicted people are sent to in exile"
Mine: "If you stand over there, you can see all the way to the border of the land. Sample word 環境 (かんきょう): That sounds cool! Can (かん) we go there today (今日)?"

alexadler said:

Unfortunately the book is not published yet (that's why I run the campaign) so I don't have the kind of testers you've mention, you could be the first one haha!

Click to expand...

Alas, I've already learned my jouyou kanji, so I wouldn't be able to give feedback from the perspective of a true newbie. I'll be interested in seeing what other people say when they try out the book though, especially if they also compare it to other resources.