Question about all the irregular readings of kanji that seem to pop up plentiful in kanji dictionaries. Take for example the kanji 丞. According to the entry on kanji.jitenon, this kanji has following readings:


In what instance would 丞ける ever be used? I can't find any concrete examples.

or 奄:


奄がる yields zero examples on Google.

or 柑:


As far as I'm aware こうじ is written 柑子 and みかん is written 蜜柑, not just 柑 on its own.

What is the purpose of these obscure readings? The only thing I could think of is actual names of persons e.g. 亨 may be read as とおる but is the spelling 亨る for the actual verb ever actually used? Can anybody shed some light on this?

1 Answer 1


What is the purpose of these obscure readings?

It is a question that would lead to the entire Begriffsgeschichte of kun'yomi, but in a quick understanding, those kun readings are a different notion from those in discussion of Japanese orthography.

Kanji dictionaries are conceptually a form of Chinese-Japanese dictionary. Thus as what I wrote in another answer,

Technically, kanji were foreign notions in Japanese; on'yomi was the pronunciation, and kun'yomi was its definition in Japanese

in the context of those dictionaries. The readings are intended to be at best Japanese glosses, not that you are recommended to substitute the kanji for an ordinal word or phrase of Japanese. They can even contain on'yomi words as in your こうじ or みかん. 字訓 of 大漢和辞典 (which sometimes are hilariously long) are a periodically viral internet meme (cf. a Miku song).

Quite a few of those are indeed used in the kanbun kundoku practice i.e. word-by-word translation of Classical Chinese into Japanese. Still, not many of them are employed when you spontaneously write a Japanese text.


  • 1
    Thank you for the elaborate answer! So if I understand it right those readings are basically more a definition than anything?
    – Boolicious
    May 8, 2021 at 12:14
  • 1
    @Boolicious Yes, or I prefer "gloss" since they're Japanese replacement during recitation rather than a full explanation. May 8, 2021 at 14:18
  • 1
    So basically you have equivalences, explanations, and readings in a more traditional 漢和辞典. It so happens that when you stumble across, say, 犬 いぬ, the latter part fulfills all three roles at once without being marked as such. May 8, 2021 at 16:31
  • 1
    @JohnFrazer "Equivalences" and "readings" in your words are not distinct in my argument, if I read your comment correctly. "Kun'yomi" listed in a modern 漢和辞典 usually doubles those functions, besides another more dictionary-like "explanation". In olden times, there were averagely much more kun'yomi per kanji, and those are glosses to a character (="equivalence"?). As time went on, a small number of them started to be so closely tied with certain words in Japanese that became spellings of a word (="reading"?), and this is what somebody learns as "kun'yomi" when they learn Japanese now. May 8, 2021 at 18:43
  • As an addendum -- the word 音【おん】読【よ】み (on'yomi) literally means "sound reading", in reference to the sound or pronunciation of the character as borrowed from (usually) Middle Chinese. The word 訓【くん】読【よ】み (kun'yomi) literally means "meaning reading", in reference to the meaning of the character as translated into Japanese. We see a similar dual-reading classification for Korean hanja, called 음훈 or eumhun, where eum = 音 and hun = 訓. Feb 28, 2023 at 19:37

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .