Correlating what is spoken with what is written — that is the long-standing challenge faced by many writing systems across the world. Some use the alphabet, and using certain letter combinations to represent more sounds, while others use logographic or ideographic writing systems to express more along the lines of ideas and things rather than just concrete sounds. But Japanese uses three writing systems stretching two or three kinds of writing systems — syllabaries, and logograms and/or ideograms. These writing systems have been used in combination for centuries, but this correlation still poses a problem to this day.

Japanese has an astounding small syllable inventory — just around 100 syllables in total. But this is slightly expanded by its pitch-accent system, which can distinguish different words which would sound the same without this pitch distinction. But how does all of these translate to its writing systems? Japanese has two syllabaries which largely represent all of the syllables it uses, but do not represent the pitch-accent system at all. But it is the kanji system that poses a huge challenge, with some characters having many different ways to pronounce, and a convoluted system of on’yomi, kun’yomi, ateji, gairaigo, nanori (for names) and other special pronunciations. Correlating between what is spoken with what is written here poses an extremely huge challenge compared to the relatively “simpler” hiragana and katakana, and this problem started to gain attention in the Meiji period. Known as the kokugo kokuji mondai (国語国字問題), or the National Language and Script Problem, several writing reforms have been made in the Japanese language, including choosing which kanji to be used, dropped, and taught in Japanese schools and language programs.

However, there are some who advocated for the easy way out — to completely ditch the Japanese writing system for the romanisation of Japanese, or rōmaji. This movement was not new; it was proposed by some even during the Meiji era, when the kokugo kokuji mondai was brought up. With this movement, brought along the Nihon-shiki romanisation, joining the Hepburn system in the existing romanisation systems for Japanese. This movement grew to such a magnitude, that several texts have actually been published entirely in rōmaji. However, this failed to gain sufficient momentum, and this fell by the wayside. That was, until the end of World War II.

If you paid attention in history classes, you might have learned that World War II ended in 1945 with an Allied victory. That is, if you ignore the pedantry surrounding the details between Japan and the then Soviet Union about the Kuril Islands. With the surrender of the Empire of Japan, came the occupation of Japan by the Allies from 1945 to 1952. Led by the United States, this period marked the only time in Japan’s history when it was occupied by foreign powers.

The American Education Delegation was invited to Japan by the Supreme Commander of the Allied Powers (then held by General Douglas McArthur) in 1946, which published a report surrounding the difficulties of kanji use, and advocated for the use of rōmaji, harkening back to the movement in the Meiji period we mentioned earlier. This proposal was justified by the convenience of the Latin alphabet. As a result, the prioritisation for Japanese script reforms moved towards abolishing the use of kanji, and the use of rōmaji. Even though this policy failed to some extent, the movement to gradually abolish kanji largely stayed, and the Japanese government went on and made a standardised rōmaji system for use in schools anyway.

You might wonder, what is the problem with rōmaji?

Japanese is spoken with a rigid set of syllables, about a hundred for the entire language as you might recall. In the kana syllabaries, each character represents a syllable in the language, with additional diacritics to fill in the rest, like voicing of consonants. Rōmaji, instead, uses individual letters to represent consonant and vowel sounds, and the macron diacritic for long vowels. However, there are certain Japanese sounds that are not really easy to romanise, for example, the yotsugana which we have covered here. Certain prefectures might fuse certain sounds, while other would distinguish these syllables. Additionally, there are at least three systems used to romanise Japanese, starting with the Hepburn system and the Nihon-shiki romanisation, and the Kunrei-shiki romanisation. While the Kunrei-shiki system is taught in elementary schools to this day, the Hepburn system is popular among foreign learners of Japanese as they try to learn the three writing systems used in Japanese. Thus, there could be some disagreement in how certain sounds are represented in the Latin alphabet, as with the representation of long vowels.

Secondly, Japanese has a lot of homonyms. Homonyms are words that generally sound extremely similar, or identical, but mean different things, sometimes almost entirely. Take, for example, “deer” and “dear”. While these two words generally sound the same in English, they carry different meanings altogether. The use of kanji functioned as a sort of distinguishing system for different sets of homonyms, with the meaning packaged into the characters that represent the said homonyms. Each character can carry a different meaning, which is why kanji is often termed as a logographic writing system. Consider this romanisation of actual Japanese words, seiki. What does this translate to? There are several things this could mean, from “century” to “vitality”. Now, let us switch back to kanji. The word seiki has different ways of writing, including 世紀 (century), 正規 (formal, legal, normal), and 性器 (genitals). Without context, this distinction would be way more difficult to make.

This is not to say that the romanisation of Japanese is to be entirely frowned upon. As Japan became more international, the need to adapt to international standards also arose. One such example is the romanisation of Japanese names to be used in passports and other identification documents to be read by say, immigration officers, who are unlikely to know how to read Japanese names written in kanji. Additionally, website addresses are normally romanised, making it more readable compared to the Unicode adaptation for the same address, but unromanised. Keyboard inputs in Japanese follow two systems as well — one kana input, and the other, rōmaji input.

Thus, while the romanisation of Japanese generally failed to dominate the way Japanese is written in Japan, it still has spurred further writing reforms in how kana and kanji are used, and which kanji is used in Japanese. The three writing systems are still iconic features of the Japanese language, but rōmaji still remains as a means for Japanese to be more international, and digitally adapted, in addition to its use by beginning learners of the language. To say that Japanese now has a fourth writing system would be a little inaccurate, nonetheless, since most writing is still done using kana and kanji. It would be similarly inaccurate to say that since Mandarin Chinese predominantly uses Chinese characters as a writing system, and uses hanyu pinyin to help pronunciation of these characters, Mandarin Chinese uses two writing systems.

To this day, achieving this correlation between sound and character, or characters, poses a challenge to the Japanese language. Which other reforms have been put forward by the Japanese government to modify their writing systems? This would take a couple more posts to explain, so stay tuned.