To say that Japanese has a convoluted writing system is a rather huge understatement. More accurately described as three writing systems in an orthographical trenchcoat, filtering out which kanji to use, teach, and write in Japanese has posed as a persistent challenge since its first mention in the Meiji period. Previously, we have seen the time when transliteration with the Latin alphabet was proposed as a possible solution to the problem, and to some extent, has survived, notably in passport pages and identity documents, where non-speakers of Japanese could read the romanisation of the individual’s names, for example. But with the intention of abolishing kanji at some point in the future, the Japanese government rolled out something different in 1946. The Tōyō kanji.

Writing reforms were a rather recent thing in Japan at that time. The earliest known efforts date back to the late-Meiji period, but perhaps the first major reform was rolled out by the 臨時國語調査會, or the Select Committee on the Study of the Japanese Language in 1922, in the late-Taisho period. There were 1 962 characters selected and approved by this committee for everyday use, and it was only the year after when the committee reformed kana usage to modernise it.

It was these 1 962 characters that formed the basis for Tōyō kanji, or “general use kanji“. This new list, officiated by the Japanese Ministry of Education, contained 1 850 characters to be taught in its education system. Of course, with many more characters being used in Japan, this new list was not exhaustive; it only contained the kanji one is most likely to encounter in everyday contexts. But this was created with the eventual goal of abolishing kanji from Japanese.

In addition to this reduction of characters in the list, reforms were made to systematically standardise the form of kanji to be used and taught. For example, some characters may have both a traditional form (kyūjitai) and a simplified form (shinjitai), much like the traditional and simplified Chinese characters. Without these standardisations, both traditional and simplified forms would have been used in writing Japanese. Further character revisions for shinjitai were rolled out in 1949.

Furthermore, kanji readings were reformed for some characters, but some were criticised as being too restrictive. For example, take 魚. This is most commonly read as sakana, but under the Tōyō kanji, it only accommodated the readings gyo and uo. While the first table of pronunciations was released in 1948, it was not until 1972 when these readings were revised and updated.

Now what happens when you have a compound kanji word, with one character included in the Tōyō kanji, but the other excluded? This creates a situation where the character excluded from the Tōyō kanji could be replaced with the corresponding kana, hence giving the phenomenon called mazegaki (交ぜ書き) or “mixed writing”. While not enforced, mazegaki became very common in television broadcasts and newspapers.

And in 1956, to ease implementation of Tōyō kanji, the Japanese Language Council laid out some characters that may substitute for words that contained characters excluded from the Tōyō kanji list. This became known as kakikae (書き換え or 書き替え) or “writing substitutes”. The most common example is the use of 才 (sai) instead of 歳 when writing ages. You might find them on toy packaging, for example.

Nevertheless, these two phenomena have attracted great controversy, from the traditionalist perspective that mazegaki erodes traditional culture, to the perspective that such a limitation of Tōyō kanji could introduce ambiguity in written Japanese. When hiragana or katakana is used to replace a character that does not belong to Tōyō kanji, there could be some ambiguity made when determining if the kana in question has a grammatical function, or part of a verb or adjective inflection. Sometimes, the use of mazegaki could cause ambiguity in determining the intended pronunciation of the phrase.

Perhaps the main issue with such substitutions is, these are mainly phonetic substitutions. This could form part of the argument that mazegaki and kakikae eroded Japanese tradition, as it would have encouraged or legitimised what could be seen as a rather sloppy form of writing Japanese. Going by the readability route again, the original meaning of the affected words would more likely be misconstrued. What would happen to words that combine characters that only use the sound component [think of words like 寿司 (sushi)], or words that involve only the meaning of the characters [think about words like 大和 (yamato) or even 今朝 (kesa)]? For words involving ateji, which use the sound component of the constituent characters, the main guideline, though not enforced, is to write them in kana.

Over time, more revisions were made to the Tōyō kanji, including some mentioned earlier. But something else was underway. There were plans to phase out Tōyō kanji, in favour of a new system to standardise written Japanese, which is set to be in place for the remainder of the 20th century, and several decades thereafter.

In 1981, Japan made reforms to the kanji list again, replacing the Tōyō kanji with the Jōyō kanji (regular use kanji), a list of 1 945 characters that every student must learn in school. This was revised up to 2 136 characters in 2010, and the number remains to this day. Today, 1 026 kanji characters are taught in elementary schools in Japan, while a further 1 110 characters are taught in middle (and maybe high) schools.

Similarly, for the Japanese Language Proficiency Test or JLPT, the 2 136 kanji characters are split into three grades, the N3, N2, and N1, with 1 136 kanji being assessed in N1 alone. Would there be further reform to this list of Jōyō kanji? Only time will tell.