I have been searching around, but all the sources give completely different answers ranging from 2,000 to 50,000. So my question is how many Kanji characters that have ever existed since the dawn of time? Does this include unofficial characters?

  • 1
    Idk but ik that we learn 1006 kanjis in primary school - sorry if this doesn't help much... :(
    – user17967
    Sep 20 '16 at 20:20

An authoritative classic, the Kāngxī dictionary, lists over 47,000 characters. The Hanyu Da Zidian, a more modern reference, has over 54,000 characters; the Dai Kan-Wa Jiten, the Japanese equivalent, has over 50,000. Even more recently, the Zhōnghuá Zìhǎi has over 85,000 characters, but apparently many of those are variants.

Of course, such counting is more-or-less academic. In Japan, there are only 2,136 Jōyō kanji (lit. commonly-used kanji), which are the ones taught in school, though literate people usually know more. The equivalent list in Chinese is the Xiàndài Hànyǔ Chángyòng Zìbiǎo, which has about 3,500 characters.


50,000 is usually the number given for the number of Kanji characters since the dawn of time.

2,000 is roughly the number than comprises compulsory education.

5,000 is often assigned to particularly well-read persons (e.g. university professors).

I remember reading a newspaper article about one of these "living national treasures", who was supposedly able to read 10,000 characters. (I have yet to find a reference.) One does have to wonder, however, how they determined the number.

  • 8
    There are 75,963 kanji currently encoded in Unicode available today, and another roughly 5,000 scheduled for the next Extension F update. And there will be more after that.
    – Dono
    Apr 18 '13 at 7:36
  • 2
    I'd like to see sources for this answer
    – ssb
    Apr 18 '13 at 7:39
  • 4
    Perhaps 10,000 if we are talking about Chinese hanzi, but for Japanese kanji, all of the Kanji Kentei level 1 test (漢字検定) comprises "only" 6355 characters. Which is every character you can represent in Japanese on a standard computer. I'd think even the most well-read of Japanese people wouldn't recognise even 5000 characters unless they studied specifically for that purpose. Apr 18 '13 at 8:52
  • 1
    Sorry, I had misremembered hearsay. This is how legends start =) I edited my answer a little.
    – Earthliŋ
    Apr 18 '13 at 9:57
  • 1
    @Dono, yes, but many of those separate encodings are effectively duplicates, so I don't think that number can be used meaningfully. For instance, 鉄, 铁, and 鐵 are all separately-encoded Unicode characters, but they are all the same single kanji character (as used in Japan, Simplified Chinese, and Traditional Chinese writing, respectively). Not all character form differences are encoded separately in this way (画 for instance will simply look different in Japanese vs Simplified Chinese fonts), but a large number are. May 26 '20 at 8:49

For what it's worth, according to wikipedia, the current largest compendium of Chinese characters, 異体字辞典(Yitizi Zidian), has 106,230 entries, which includes all forms (including alternate versions) of each character.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.