The Hidden Structure of Original Japanese

The origins of okurigana

Students of Japanese learn the word tomo-dachi (friend) in their very first class. A bit later, they will learn tomo-ni (together). At some point they will encounter tomo-nau (accompany). But no one ever tells them, and they very likely do not even notice, that all these words have tomo in common.

In fact, the commonality does not occur to native Japanese either, apparently. At least, not according to a survey of native speakers I conducted, where none of the respondents identified any connection between these words (OK, it was actually just one person).

Part of the reason is that these words are usually written with different kanji. The tomo of tomo-dachi would be 友; the tomo of tomo-ni would be 共、and the tomo of tomo-nau would be 伴. And there are even more kanji read tomo, such as 供, meaning companion or servant. In the minds of the average Japanese, that means they are all “different” words.

Examples of this abound. For example, the same informant mentioned earlier insisted that there was no connection between tsuna (rope, 綱) and tsunagaru (connect, 繋がる). When you learned these words, dear reader, did you make the “connection”?

That one was sort of obvious, but what about mura (village, 村), which is obviously related to mureru (flock, 群れる)? What is a village other than a “flock” of people? And murasaki (purple) is the dye from the murasaki plant, so named because it blooms (咲き/saki) in clumps (群生/gunsei). When used to describe the visual appearance of marble, mura refers to the accumulation of blotchy spots and streaks.

These examples are the tip of the iceberg of the densely interconnected lexicon of “original” Japanese, sometimes called “Yamato words” (Yamato-kotoba), or wago (和語). These are one of the three vocabularies of modern Japanese, the other two being Chinese loanwords and foreign loanwords (gairaigo, 外来語). They are the words in the proto-Japanonic language believed to have been brought to Japan from the Korean peninsula several centuries before the start of the modern era by the rice cultivators that founded the Yayoi culture. That is the language in which the Manyoshu and Genji Monogatari was written. It is the foundation, and glue, of contemporary Japanese.

So how, then, did a word like tomo and its cognates, which served perfectly well in spoken language for hundreds of years before the advent of writing, come to be written in three or four different ways when using kanji, based on nothing more than subtle gradations of meaning? How did tomaru come to be written using three or four different kanji? How did toru come to be written using as many as a dozen different kanji (取撮摂撮執採)?

Because of this, poor Japanese children are forced to waste time learning to distinguish between 細い (hosoi, thin, like Komako's nose) and 細かい (koma-kai, fine). Japanese high schoolers commute to “juku” cram schools to learn the correct characters and okurigana in order to pass college entrance examinations. People like us learning Japanese are forced to try to distinguish between okonatta (performed) and itta (went), both of which are written 行った. The Japanese government was forced to issue not one, not two, but three different sets of rules throughout the last century governing the correct use of okurigana.

We take this for granted now, but it did not have to be this way. Some well-regarded Japanese scholars say that there was no need to ever write Japanese verbs and adjectives using kanji in their kun (native Japanese) reading; one went so far as to lament that this approach led to what he called “which character disease”!

This system not only imposes an unreasonable burden on the learner, but perhaps more importantly it actively conceals and obfuscates the beauty of the underlying structure of the original Japanese language.

How did we end up here? What misguided soul, at what point in time, came up with this horrible idea of trying to write native Japanese words using Chinese characters, requiring the entire machinery of okurigana to write their conjugations, in order to draw distinctions in meaning which had not been necessary in the spoken language since the origins of Proto-Japonic?

Actually this entire cumbersome scheme of writing Japanese native words using Chinese characters with kana to conjugate them slipped into the language from an entirely different direction. That was the so-called kanbun approach to marking up classical Chinese texts in such a way that they could be more or less translated in real time into Japanese with correct vocabulary and word order and particles. Here's an example:

Take a look at the fourth column from the right towards the bottom. You can see that the Chinese character for stop 止 has been marked up with okurigana マル (maru). This was an indication to the reader that this word should be read in Japanese as tomaru.

After the Meiji Restoration, as more and more texts began to be written in Japanese instead of classical Chinese, it was natural for the learned class who had experience with classical Japanese to write words like tomaru in a parallel way, namely the Chinese character followed by the okurigana. It is no accident that the initial governmental guidance on okurigana was issued in the middle of the Meiji Period.

It's unfortunate that MacArthur's staff, when they were designing Japanese postwar structures, instead of making the unworkable proposal of getting rid of kanji altogether and moving to romaji, did not simply propose that kanji be eliminated as a way to write native Japanese words, at least declinable verbs and adjectives. It could actually have happened.