As far as I understand 明日(あした) is jukujikun (also called semantic ateji), so, as with for example 今日, I would expect for the "reading" to be considered over the whole word, instead of being split into the individual kanji readings as is done with normal words.

But in a lot of places (dictionaries, etc) I see the reading of 明日 split into 明(あ)日(した).

I can see where that might come from, since あ is a normal Kun-reading for 明, but the again した is no valid reading for 日.

  • 明日 is not ateji but jukujikun. Ateji means assigning kanji based on their readings (ignoring their meanings), whereas jukujikun means assigning kanji based on their meanings (ignoring their readings). – naruto Jan 1 at 9:52
  • @naruto I'm aware of that, but, as far as I understand, the term "ateji" is also used for jukijikun, and more commonly known/understood (Basing this on a previous comment/answer somewhere here in the SE, where the distinction was discussed). If you think the correct term should be used instead, tell me, or feel free to make an edit. My intention was merely to comply with common usage. – ーーー Jan 1 at 14:47
  • Ateji and jukujikun are indeed confusing to some, but credible resources never make such a mistake, and I recommend you distinguish them correctly, too. Furigana for ateji like 亜米利加 and 出鱈目 can be easily split into individual kanji readings, so your problem happens only with jukujikun. – naruto Jan 2 at 5:35
  • @naruto Understood. Makes complete sense; I find it odd myself, to call both ateji and jukujikun "ateji", but apparently it is common and not directly considered a mistake to conflate these two concepts under the one term (This is mentioned on both the Japanese and English Wikipedia entries for ateji). Of course in a linguistical context, you would keep these terms clearly separated. – ーーー Jan 2 at 14:35
  • Can you give an example of a dictionary that does this, for reference. – Leebo Jan 2 at 14:51

There’s no really “correct” or “incorrect” when it comes to ateji like 明日{あした} or 大人{おとな}, you just have to pick something or least bad.

With jukujikun (kanji picked purely for meaning and not readings) it’s even worse since you may have not enough kana for the kanji. Some examples from Wikipedia:

kera (啄木鳥, woodpecker), gumi (胡頽子, silver berry/oleaster), and Hozumi (八月朔日, a surname)

I think the best choice in such cases is simply center the reading above the whole word and do not align kana to individual kanji.

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.