Reading kanji in cursive script is devilishly difficult

« previous post | next post »

Here's why:

Now, multiply that by thousands of similarly seeming characters that, when written in regular script, are already hard enough to keep straight, yet become all the harder to distinguish when written in cursive.  To read texts written in cursive, unless you are a master of that type of script, you have to have a pretty good idea of what the subject matter is about.  Failing that, all you can do is struggle to recognize one of the characters in the given passage, then grab hold of it like it's a life raft.  Then build out from that life raft and look around among the adjacent or nearby characters and hope that you recognize one that somehow fits together with the one you think you already know.  After you have a phrase or two figured out, then you might start to have an idea about what the passage, or at least a part of it, is about.

Struggle some more, keep struggling, and eventually it's possible that you can discover some more threads.  But don't be too stubborn.  Face reality.  You might just have to give up before you comprehend what the entire passage is about.

Roughly speaking, that may be the only way to deal with a stubbornly refractory text in cursive.  On the other hand, thankfully, most texts in cursive consist partly, largely, or wholly of previously existing verbiage.  If the text is a completely new creation, the sole individual who knows what it says may be the author him/herself.


Selected readings




  1. Laura Morland said,

    October 18, 2022 @ 6:22 pm

    Wow, the level of difficulty is mind-blowing.

    And I thought Hebrew cursive was hard! See here for a nice example:

  2. Carl said,

    October 18, 2022 @ 7:18 pm

    Yes, although now there is an AI that can read it and spit out the regular characters. See

  3. John Swindle said,

    October 19, 2022 @ 6:41 am

    I gather that Japanese, like Chinese, has varying levels of cursive script, some closer to regular or printed script than this.

    The characters he chose are common in Chinese too. How would a Chinese reader fare with his cursive forms? Is there anything about them that looks Japanese?

  4. Thomas said,

    October 19, 2022 @ 9:49 am

    As a Chinese learner, I must wonder: is there an error in the regular script or how did tip 尖 morph into 先 in Japanese?

  5. VVOV said,

    October 19, 2022 @ 1:02 pm is a cool resource that shows how different kanji look in different typefaces and handwriting styles.

    For example here's the page for 先:

  6. Stephen L said,

    October 19, 2022 @ 5:43 pm

    It's not full cursive, but I came across someone's free translation of a semi-cursive pen calligraphy book from Chinese here - . The biggest thing I hadn't expected is that there are many many exceptions to the stroke order rules Going to practice more with regular script until I have it down better, then semi-cursive, then maybe some day on to proper cursive… .

    I find it very interesting how the chinese script adapts itself to different writing instruments depending on whether or not lines can be tapered/width can be varied, or the like. And I've specifically come to really enjoy writing without the ability to taper width (with a thin-nibbed gel pen, say), because then writing the characters becomes a lot more a pure exercise in layout and proportion, which is fun.

  7. Guy_H said,

    October 20, 2022 @ 12:22 am

    In Chinese, we call this kind of writing "cao shu" ("grass script"). It is a much older form of writing compared to the regular script ("kai shu") that we see in books nowadays. Most Chinese cannot write in this script, as it really is an art form compared to calligraphy. In fact, a lot of Chinese calligraphy is derived from grass script. It would be like asking a German to read using Gothic or Fraktur script, it takes practice.

  8. John Swindle said,

    October 20, 2022 @ 2:50 am

    @Thomas: Good question. The regular and cursive forms in the video are for 先 .Take a look at Wiktionary for 先 and scroll down to its Japanese meanings. It seems to be a matter of extending the meaning out to the tip (the front, the part that arrives first) of the arrow, not a change in the meaning of 尖 .

    @Guy_H: I sounded out Germans about Fraktur text in another forum the other day. Their conclusion: Fraktur easy to read—some did mention the long "s" or say it takes a little practice—but Kurrent, the handwritten form used during approximately the same time period, difficult. Daunting.

  9. Jonathan Smith said,

    October 20, 2022 @ 12:12 pm

    to beat a dead but periodically reincarnated horse, characters don't have glosses, lists like the one in the video are mind poison. It's a question of the complex semantics of the in-this-case Japanese words to which the various characters are applied。 one to which "先" is applied seems to be saki 'tip; front; first; rest; etc.' dunno if "尖" does write / has ever written this word also…

  10. Tom Dawkes said,

    October 20, 2022 @ 3:19 pm

    On Fraktur/Kurrent, the Wikipedia article on Sütterlinschrift is interesting:ütterlin.

  11. John Swindle said,

    October 21, 2022 @ 7:05 am

    @Jonathan Smith: My response like that of some others was biased toward Chinese, and your reminder that writing is secondary to spoken language was highly appropriate. Sino-Japanese characters are nonetheless perceived as having meanings or ranges of meanings, in Japanese as well as in Chinese and other languages in which they've been used. That's part of the rationale or certainly the mystique of these characters, and it's what allows a Japanese word for "tip" to be read as a phonetically and historically unrelated Chinese word for something vaguely similar in meaning. Otherwise you might as well just use a syllabary (or two syllabaries for the same language, or as many as desired) or an alphabet.

  12. Jonathan Smith said,

    October 21, 2022 @ 7:12 pm

    @John Swindle:
    My reflexive response to presentations of this kind — which admittedly relates to my own interests and is tangential to the subject of the post — is that character-oriented study throws up stubborn barriers to real learning. To give but one practical example, productive Taiwanese guidance for Mandarin-speaking performers must be aural and/or phonetic script — giving Chinese-character texts amounts to saying "please never, ever learn this dialogue properly." In studying the southern Chinese languages, character texts *literally* scare me…

    Not sure how relevant this is or isn't for some of more theoretical suggestions in your comment. To take just one, "Sino-Japanese characters are […] perceived as having meanings or ranges of meanings" is not IMO a precise characterization of what is going on. Lists like that shown in the video are fundamentally Western; Chinese/Japanese presentations simply show the words/morphemes actually written by the character(s) in question, issue there being that native speakers tend not to draw a clear ontological distinction between words/morphemes and their associated written characters (they are fluent already, so no harm no foul.)

  13. Chris Button said,

    October 22, 2022 @ 7:56 am

    Isn’t a kun-yomi effectively a gloss? Hence the flexibility when you go beyond prescribed readings.

RSS feed for comments on this post