When I use an automatic Japanese text analyzer, it provides me with a list of words and also short information about what kind of word it is. Japanese (和), Chinese (漢), Foreign (外) or mixed (混). Is there any dictionary that also provides this information? Note that what I'm asking is not whether it is kun'yomi or on'yomi, but rather the word origin. For example, a word like 誤魔化す is on'yomi, but probably not of Chinese origin.

I own Daijisen and it does not seem to offer it. Since the program offers me such information I guess it's from some kind of online database.

  • This question is technically off-topic, since we don't do resource questions. I wonder if a moderator could move this to Meta? Apr 21, 2022 at 19:28
  • 1
    @EiríkrÚtlendi 5 close-votes from regular users can also move posts to meta.
    – Eddie Kal
    Apr 21, 2022 at 19:55

1 Answer 1


So far as I'm aware, there is no one website that provides comprehensive etymologies for all Japanese words.

  • The best I've been able to find is the 日本国語大辞典【にほんこくごだいじてん】 (NKD) dictionary from Shogakukan. They have a HUGE version with extensive notes, which I don't have. I do have an electronic copy from 1998 when Shogakukan and Microsoft teamed up on their Bookshelf product; that said, later versions were progressively worse, abridging more and more detail. And due to changing IT policies, I can no longer install my old one (admin permissions got yanked so we can't install anything, and my personal machines haven't run Windows for donkey's years).
    There is a halfway decent edition of the NKD available online for free via Kotobank, such as we see here in the entry for 食. Do note that the Kotobank version lemmatizes strangely (i.e. how it files individual entries under headwords). If you search for たべる in kana, you will not easily find the NKD entry, as it is lemmatized under the 食 spelling, without okurigana. Some words are even harder to find, as they are lemmatized under spellings that include multiple kanji variants, such as the かたわり entry, filed under 徒渡・歩渡・歩行渡 with three possible kanji spellings.
    There are also odd gaps in the entries. I've noticed that any word for a plant or animal often gives a short encyclopedic description of the organism, and very little information about the word -- no etymology or derivation, such as this entry for あひる. It's like there was a separate editorial team working on biological terminology, and they had inconsistent ideas about what to include in the entries.
    That said, the Kotobank edition is not as extensive as my old 1998 one, and I've learned from other users here that the huge hard-copy version is even more comprehensive. If you are willing to pay money, I understand that it may be possible to purchase electronic access.

As a bit of a tech nerd, I am quite curious how your "automatic Japanese text analyzer" functions. I suspect it might be backed by a database (maybe even just a text file) that correlates certain Japanese spellings with specific readings and term categories. Without knowing more about how the analyzer functions, it is hard to say anything useful about finding replacements that could offer accuracy of similar, or better, quality.

  • 1
    Thank you for a great answer. Unfortunately I don't have the program on my new Mac but I think it is this one: github.com/wareya/analyzer. The analyser itself makes a lot of errors, but it offers more information about every entry than other parsing programs I have encountered. As you say, it most certainly has parsed its information from somewhere else.
    – timseb
    Apr 21, 2022 at 19:42

You must log in to answer this question.

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