Kanji (漢字), characters adopted from Chinese, are easily one of the most challenging aspects of the Japanese language. I think it’s fiar to say the love or hate relationship a learner has with kanji can determine whether they continue learning the language or give up.
Personally, while I acknowledge kanji take a lot of time and effort to learn, I really appreciate how they add an extra dimension of meaning in written language not present in English, or perhaps any other language. In English, since a word’s pronunciation generally conveys its meaning (except for rare cases where like sounds map to two differently-spelled words), you don’t get much extra meaning from actually reading written language as opposed to listening to an audio narration.
When first learning kanji, oftentimes we learn both the common meanings of certain kanji (ex: 食 = eating/food), as well as compounds that contain them (食べ物 = “food”). But one of the great things about kanji is that just by knowing the meanings of each kanji in a word, we can guess the word’s overall meaning, and maybe even its pronunciation. For example, take these three kanji:
- 目 (eye)
- 医 (medicine)
- 者 (person)
Now if we see the word 目医者 we can guess that it means “eye doctor” (which happens to be the correct meaning). As for the word’s reading, while there are few permutations (since, for example, 目 can be read as “me” and “moku“), the correct pronunciation of this word is consistent with what we would expect: me + i + sha = meisha.
Now, after a long introduction now we can finally talk about what the term ateji (当て字) means. This word doesn’t really have a perfect equivalent in English, so rather than translate it I’ll explain what it means and give a few examples.
Ateji is when one or more kanji in a compound are chosen not because of their meaning, but because of their sound. Let’s look at a very common word in Japanese that uses ateji:
- 多分（たぶん): probably
In order to see how this is ateji, let’s look at the meanings of the kanji characters involved:
- 多 [ta]: many
- 分 [fun/bun]: part, degree, proportion, minute
As you can see, putting together these meanings don’t lead in any obvious way to the idea of “probably”, though if you think hard enough you can probably invent some roundabout connection. However, the readings of these kanji (listed above in brackets) totally match up with たぶん. So even if the meanings don’t match, the readings (pronunciation) do.
Let’s look at another example of ateji:
- 兎に角 (とにかく): Anyway
This one is somewhat rare in that there is a hiragana in-between two other kanji.
There are many other examples of ateji, including 出鱈目 (detarame), 滅多 (metta), and 沢山 (takusan).
As you can see, what defines ateji is somewhat straightforward, and to be honest as long as know a word’s meaning and reading, whether it contains ateji or not is not critically important.
However, when writing Japanese it is important to be able to differentiate between the nuances of writing a word in hiragana, katakana, or kanji (there may be several options including ateji), or even a combination of these. The way a word is written affects both its meaning as well as its nuance and tone, and this is one of the things that makes Japanese writing so expressive.
There are no hard and fast rules for which way to write a word, but using kanji often imparts a feeling of being more difficult, stiff, or literary. If the word involves a kanji the reader is not familiar with, it may even impact their understanding of the text. Katakana, on the other hand––often used with loanwords and onomatopoeia––can emphasize or put focus on the sound of the word itself, somewhat like italics.
By the way, earlier I said there was no translation for ateji in English, but strictly speaking you can use terms like “phonetic equivalent” or “substitute character”. However, neither of these terms are used commonly in English (unless you are a linguist).
One thing I’d like to emphasize here is that there is nothing inherently “wrong” with words written in ateji. Recently I had a discussion with someone who claimed a word was more “correct” when written in hiragana because it was ateji (the specific word was 可哀想 vs かわいそう), and I’d like to point out that isn’t the case. Just in the examples I gave above, some of them are used very frequently in daily life (多分), whereas others are relatively rare (兎に角). Also, what is considered as the “correct” way to write a word depends on the domain and context.
In order to confirm my understanding here, besides asking a native speaker I also posted on Oshiete Goo, which you can see here. While it is not a black and white discussion, from the several answers there I feel confident in saying that there is nothing “less correct” about ateji. (To be fair, I do agree that かわいそう is more common than 可哀想, at least in everyday use, and it’s better to use hiragana when teaching beginning students words like this)
Finally, in case you are curious about the origin of the word “ateji”, it comes from the word “ateru“, which can be written several ways including 当てる or 充てる. You may be familiar with the most common meanings of “ateru” including “hit” or “guess”, but it also has the meaning “assign” or “allot”, which is why it is used in the word “ateji” (the compound literally meaning “assign character(s)”).
(Note: the picture of a Japan cityscape at night was taken from Pexels.com)