Fellow Koohii forum-goer @bflatnine (twitter, best post) is a co-founder of a linguistics startup that publishes drop-dead fascinating analyses of the evolution of various hanzi (Chinese characters) and components over the last few millennia. I'm not sure if their paleographic deep-dives into Chinese characters helps my kanji skills in any meaningful way, but it hugely contributes to how interested and engaged I continue to be with kanji practice.

Anyway, they're publishing a new electronic Chinese dictionary that breaks down characters into what seems to be a historically-rigorous format: each character is broken down into "functional components" (~primitives) and each primitive is then labeled: does it contribute to the sound, meaning, or form of the entire character, or does it not contribute anything to it? Apparently it's going to be a Big Deal in Chinese pedagogy.

They're running a Kickstarter, with a couple of days left. I backed it because, besides being interesting, they're hoping to do a Japanese version sometime down the road, which would be totally badass.

(Full disclosure: I backed their Kickstarter, but have no other ties to the company or product.)

Hey, thanks for the shout-out!

We're planning to do a kanji edition next. I live in Tokyo now and am learning Japanese, so that project is particularly important to me.

I'd be happy to answer any questions here!

As an aside, the dictionary will be distributed through Pleco, which is probably the best dictionary app ever. If you are not learning Chinese, the awesomeness of Pleco alone is a good reason to start learning it .

So is this like Remembering the Hanzi, only leaving out invented component names (Mr. T (not in RTK, I know) etc.) and instead replacing them with the "correct" component names? Or am I simplifying things too much?

Edit:
After looking at the links there seems to be a lot more to it than the above, however I feel that a RTK-ish approach is not really possible, since a lot of components seem to represent the same thing. For instance 人 and 大 (and probably many more) seem to represent "a person" in a lot of characters.

I really appreciate the effort and I might even back the project to get it closer to its almost reached goal; however I feel that for the most part it is too late for me personally to fully profit from the project as my reading and writing is "good enough" and the little time I can put in my Chinese is better invested trying to improve my abysmal listening/speaking.

Is this on the idea of Heschall's approach (Guide to Remembering Japanese Characters)?

I enjoyed watching the 5 videos on youtube "How Chinese Characters Actually Work." I learned something this afternoon.

The approach seems to combine lots of methodologies together...still trying to get my hands around it. Ash said there are some "secret weapons" to be disclosed in the near future.

How does this compare to using something like Kanjigen?

Why u do dis when money so tight?!

I would really like to back this though, it's very relevant to where I want to take my linguistics work on Chinese characters.

God, I'm really torn on this. I wish this was ending a few weeks later...

HerrPetersen Wrote:After looking at the links there seems to be a lot more to it than the above, however I feel that a RTK-ish approach is not really possible, since a lot of components seem to represent the same thing. For instance 人 and 大 (and probably many more) seem to represent "a person" in a lot of characters.

Well, a strictly RTK/RTH-type approach isn't compatible, but mnemonic techniques certainly are. In fact, we think using mnemonics in conjunction with our dictionary will be even more powerful, because you'll understand the real logic behind the characters (rather than Heisig's made-up stories), which will help you to recall the characters long after your mnemonics have fallen away.

Quote:Is this on the idea of Heschall's approach (Guide to Remembering Japanese Characters)?

No, it's based on much more up-to-date research than Henshall's book. Although his book does surprise me with some of the things it gets right, it also contains quite a few inaccuracies.

Quote:"secret weapons"

Not least of which is a set of simple formulas which explain sound variation among characters sharing the same sound component. For instance, why can 各 gè represent both ge/ke/he and lu/luo? Why does 尚 shàng represent shang/chang/zhang/dang/tang? Etc.

Quote:How does this compare to using something like Kanjigen?

Are you referring to this site? At the risk of sounding cheeky, it simply links you to a bunch of inaccurate explanations.

We've put a ton of research into this. Our library contains nearly 3000 books on this stuff, and we've spent years in grad school in Taiwan acquiring the training to be able to do this kind of research correctly. We're not simply rehashing what's already been done. We're taking the latest research, the vast majority of which isn't available in English — and certainly not in resources for learners — and distilling it into something that will allow people to learn more effectively.

Quote:God, I'm really torn on this. I wish this was ending a few weeks later...

Well, we'll be making it available for pre-order on our website soon. It won't be quite as good of a deal, but we get that not everyone can spring for something like this on short notice.

I ended up backing against my better financial judgement. Use my 115AUD well please :')

Wow, thanks!

@bflatnine Nope, sorry about that. I meant the books & apps you get when you Google 漢字源. It's one of the major kanji dictionaries in Japanese. I compared some of the sample hanzi with the entries in the version of Kanjigen I have, and it looks relatively on par with it (besides the lack of Japanese-specific info)

The version of Kanjigen I have (I think it's the very first 1998 edition, the EPWING version I have doesn't have an edition number) lists the 大+羊 origin explanation for 美 which is apparently incorrect nowadays according to your app demo, so props for that. Kanjigen may have fixed it in on of the many later editions though. (Does anyone with a later edition mind looking up 美?)

Looks like you did a good job making something like Kanjigen for English speakers! If the pricing for the Japanese version ends up being cheaper than the latest Kanjigen app version (which I currently have on my wishlist) costs, I might buy it instead, to use as a second reference after looking things up in my older version of Kanjigen. Will the Japanese version users still need to install a Chinese learning app to use it though?

Richard Sears of http://chineseetymology.org and Rick Harbaugh of http://zhongwen.com would probably love to hear about this app. Let them know if you haven't already.

Noob question: Is it possible to back this project (or Kickstarter in general) via PayPal? It only gives me the option to pay via Credit Card.

Just for reference, this is an example of what 漢字源 looks like. This entry is relatively short compared to some:

樹

16画　木部　［六年］
区点=2889　16進=3C79　シフトJIS=8EF7

《常用音訓》
ジュ

《音読み》　
ジュ／ズ (呉) ／シュ (漢)〈shù〉

《訓読み》　
き／たてる（たつ）／たつ／うえる（うう）

《名付け》　
いつき・き・しげ・たつ・たつき・な・みき・むら

《意味》

❶｛名｝き。たってはえているき。たちき。▽切ったきを材という。「植樹」「落葉樹」
❷｛名｝ついたて。たてて中を見えなくする小塀コベイ。「邦君樹塞門＝邦君ハ樹モテ門ヲ塞グ」〔→論語〕
❸｛動｝たてる（タツ）。たつ。うえる（ウウ）。⊥型にじっとたてる。また、たつ。木をうえる。▽去声に読む。〈類義語〉→豎ジュ。「建樹（たてる）」「十年之計、莫如樹木、終身之計、莫如樹人＝十年ノ計ニハ、木ヲ樹ウルニシクハナシ、終身ノ計ニハ、人ヲ樹ツルニシクハナシ」〔→管子〕

《解字》→



会意兼形声。右側の字は、太鼓タイコまたは豆（たかつき）を直立させたさまに寸（手）を加えて、⊥型にたてる動作を示す。樹はそれを音符とし、木をそえた字で、たった木のこと。

《単語家族》

豎ジュ（たてる、たて）逗トウ（じっとたち止まる）などと同系。

《類義》

→木・→建

《熟語》

【樹人】
ジュジン・ヒトヲタツ ①人材を育てる。〔→管子〕②賢人を官吏としてあげ用いる。

【樹子】
ジュシ ①諸侯の嫡子チャクシ。「誅不孝、無易樹子＝不孝ヲ誅ストモ、樹子ヲ易フルコト無カレ」〔→孟子〕②木の実。

【樹功】

ジュコウ・コウヲタツ てがらをたてる。『樹勲ジュクン』

【樹立】
ジュリツ ①設備などを設ける。②しっかりとうちたてる。「煌煌太宗業、樹立甚宏達＝煌煌タリ太宗ノ業、樹立甚ダ宏達ナリ」〔→杜甫〕

(~trimmed for space)

《熟語》 →下付・中付語

→三珠樹　サンシュジュ
→仏樹　ブツジュ
→公孫樹　コウソンジュ
→塚樹　チョウジュ
→冢樹　チョウジュ
→刀山剣樹　トウザンケンジュ

(~trimmed for space)

《熟語》 →故事成語

【樹欲静而風不止】キシズカナラントホッスレドモカゼヤマズ　木が静かにしていようとしても、風が吹いて、思いどおりにならない。子が親を養い孝行しようとしても、そのときにはもう親は死んでしまっているものだというたとえ。〔→韓詩外伝〕

Herr. I don't believe so. I've never seen such an option in all my years backing projects.

bflatnine Wrote:We're planning to do a kanji edition next. I live in Tokyo now and am learning Japanese, so that project is particularly important to me.

YEAH!, that is super-awesome to hear!!! Hopefully a post on Koohii counts as being "on the record"

Also, how did you find out so quickly that I'd written about this here?! I posted it in the middle of the night here and didn't tell anyone ?.

bflatnine Wrote:We've put a ton of research into this. Our library contains nearly 3000 books on this stuff, and we've spent years in grad school in Taiwan acquiring the training to be able to do this kind of research correctly. We're not simply rehashing what's already been done. We're taking the latest research, the vast majority of which isn't available in English — and certainly not in resources for learners — and distilling it into something that will allow people to learn more effectively.

So inspiration ?! You guys deserve the best for your years of incredibly difficult hard work.

Sorry for the delay in replying! Things have been crazy.

HerrPetersen, unfortunately Kickstarter takes cards only, sorry!

aldebrn, I noticed some traffic from this page on our site. Just happened to check shortly after you posted, I guess! Thanks for the kind words.


As for 漢字源, yes it seems similar. I don't read Japanese well enough to understand all that, but it looks to be citing examples of usage in classical Chinese (論語、管子 etc.). We won't do that sort of thing, because this is a tool for second language learners, not already literate native speakers, but the idea is similar.

Managed to dig up my credit card and go for the 40$ package. Only 42$ -9$ missing to reach 90000. Pretty cool. Question about the workbook: Will there be a beginner's tutorial teaching the first x charachters in a somewhat Heisig-style approach? (That is: indroducing certain components in the right order to help a beginner learn the most frequently used hanzi?)

Other question: How many hanzi are you able to include per day? Is your workflow worked out by now? Can you give a very brief summary on how the process from a given charachter to a full dictionary entry works?

Mega-congrats to bflatnine and the rest of his team!!! Hope you all have a well-earned stiff drink or ten, and a sleep-filled weekend!

Sorry for the delay.

Thanks for the support and kind words, everyone!

Quote:Other question: How many hanzi are you able to include per day?

That remains to be seen. In theory we're planning on 15 or so per day, but that will depend on how smoothly everything goes in practice.

Quote:Is your workflow worked out by now?

More or less. We still have to hire a few people (research assistants and data entry personnel) before we'll be able to actually see how it works in real life though.

Quote:Can you give a very brief summary on how the process from a given charachter to a full dictionary entry works?

Once everything is in place, this is essentially how it works: I'll tell our team that I'm planning on doing characters X, Y, and Z next week. Then they'll digitize everything we have in our library (3000+ books) about those characters and enter it into our database. When I'm ready for character X the following week, I'll be able to pull it up in our database and have all the relevant research in front of me rather than looking it up manually in our books. There are certain references we prioritize over others due to the quality and breadth of their scholarship, so I'll start with those and move on from there. Once I've got a good handle on the current state of research on that character, I'll start writing the entry. Naturally, not every scholar will agree with each other all the time, so there are some choices that need to be made based on the evidence available, and we of course reserve the right to put forth our own conclusions if we feel that the evidence doesn't support what others have said so far (though this is pretty rare). We will cite all of our sources in the dictionary so that people can check the research for themselves.

That's basically the first pass. Once that version is released, we'll start combing through more specialized journal articles and update the entries that we find need updating. Those changes will mostly be in the "Expert" info, naturally, as they'll have mostly to do with the historical forms and phonology. One of the benefits of releasing this thing digitally is that it will be fairly easy to keep it up-to-date in the long term, and that's what we plan to do.

Hopefully that answered your questions!

Thank you very much for the detailed answers; very interesting . It seems like all of you have a pretty stressfull year in front of yourselves. Good luck (and hopefully) have fun.