Review: Kanji and the Japanese people

高島俊男（2000）『漢字と日本人』文春新書
Takashima Toshio (2000) Kanji and the Japanese people, Bunshun Shinsho

Although some people are prone to exaggeration about it, the present-day writing system of Japanese is a complex hybrid of phonetically transparent elements (the kana syllabaries), and largely phonetically opaque ones (kanji, or Chinese characters). The linguistic, social and political factors that led to this situation are fascinating, and Kanji and the Japanese people is Takashima's introduction to some of them. After taking this book almost at random from a library shelf, at first I almost abandoned it altogether. I found the tone off-putting, but the subject matter itself was enough to keep my interest. The facts themselves are fascinating, but the author spends almost as much time on his own pet theories, likes and dislikes as he does outlining what actually happened and giving reasoned arguments about it.

Just as Takashima has trouble doing justice to the subject in one book, I can't hope to address more than a few isolated points here, but I will try to give some highlights as well as my various criticisms of the book. My reading of the book might seem too demanding for a popular (as opposed to academic) book, but I found it in an academic library, and I don't think in any case that there's any reason to let it off the hook too easily. I must excuse myself for not supplying page numbers for everything -- I really wanted to add them all but have realised that if I put it off much longer this review will never get posted.

The book's subject matter is organised chronologically, from the first use of Chinese characters (kanji) in Japan, to their development into kana, the massive coining of Sino-Japanese words during Japan's modernisation, and the complex and sometimes contradictory movements for language reform in the period of Japan's modernisation (late 19th and early 20th century) and again post-WWII.

Although the author quotes often from key observers at different times in history, and from primary linguistic sources, such as the reflections of Shinmura Izuru and the Man'yooshuu, he does not even hint at the vast amount of serious scholarship that has been done on the topics he is writing about. For instance, he speculates about the dominance of new Sino-Japanese words -- that is, words such as 社会 (shakai, "society") coined in the modern era using Sino-Japanese morphemes and written using kanji -- in recent Japanese, but doesn't seem aware of substantial corpus research, such as the government-sponsored Vocabulary and Chinese characters in ninety magazines of today (『現代雑誌九十種の用語用字』) and its descendants, which do a very careful job of quantifying this very question.

Takashima makes clear his animosity towards a number of targets, including Japanese Industrial Standards, the former Japanese National Language Council, people who pronounce the word for 'ten minutes' as juppun instead of jippun, and even his own editor (perhaps playfully -- I find his humour hard to detect). The last of these is grumbled at for asking the author to give more concrete explanations for his assertions that, for example, Japanese has no word corresponding to the English "brother" and "sister" (see pp.178-9 for the explanation, and pp.249-50 for the grumbling). This particular argument is deployed in service of a strong linguistic relativist position: "... the truth of the world differs according to what language we speak. Language is thought itself." (「…話すことばによって世界のありようがことなる。言語は思想そのものなのだ。」p.178). As far as the author is concerned, this is a self-evident fact that goes without saying (「これは言うまでもない当然のことだ…」p. 233); presumably, this is why he is frustrated at being asked to try and articulate it.

The author also asserts a deep unity of language and nation, without a hint of how problematic an idea this actually is. I suppose it is obvious that I do not share his views on this, although he is hardly alone in thinking this way. What is more unusual is that the author is interested in applying linguistic relativism and the language-nation bond not just to language as it is spoken, but to writing as well. That is, to borrow a familiar circle of reasoning: the Japanese are Japanese because they speak Japanese and write Japanese. Here, 'writing Japanese' means using kanji when necessary, and although the author is aware of the problems they bring, this is his way of arguing against the abolition or simplification of kanji in Japanese. However, he is also very clear that the usefulness of kanji is primarily for writing Sino-Japanese words, and another of his many targets of disapproval is those people who believe kanji to be the correct means of writing every Japanese word.

Takashima believes in avoiding kanji for writing natively Japanese words (as opposed to Sino-Japanese ones), and obviously makes some effort to do so in his own text. Elsewhere, he makes a very interesting observation (which he returns to frequently) that the problem of kanji in Japanese was in fact worsened by the modern mass-coinage of new Sino-Japanese words. This is because these words were coined based on characters' meaning alone, resulting in many homophones (try looking up words pronounced こうとう kootoo in a Japanese dictionary, for example). The confusion can be greatly exaggerated, of course; we are unlikely to confuse kootoo "spoken answer" (口答) and kootoo "high tower" (高塔), for example. But it is true that there lots of cases where words such as kagaku "chemistry" (化学) and kagaku "science" (科学) have the potential to create real confusion. Unlike earlier borrowings from Chinese, Takashima's claim is that this new breed of Sino-Japanese word can only survive usefully with the underlying support of the characters used to write it, and that seems intuitively correct. However, confining kanji for use for Sino-Japanese words (as Takashima suggests doing) would in fact undermine these words completely. The fact that, for example, the Sino-Japanese morpheme koo can be interpreted as meaning "high" in kootoo "high tower" (高塔) is precisely because it shares a written representation (高) with the Japanese word takai, which also means "high". This is by no means an original observation, but it does not seem to have occurred to Takashima. The continued use of kanji with their kun (native Japanese) readings is a crucial underpinning of the whole system.

The modern coinages mentioned above (kagaku 科学 化学, shakai 社会, etc.) receive a lot of attention in the text. Takashima makes a very interesting sketch of the irony of their existence. On one hand, modernisation led to a need to name a great number of new things and concepts, and mass-coinage of new words was a preferred solution. On the other, intellectuals in Meiji Japan and post-WWII felt that in order to be a suitable tool of a modern society, the Japanese language, particularly its written form, needed improvement, and that part of this was the eventual abolition of kanji. The result: the force that was rejecting kanji was the same one that led to the coinage of the very class of words that would make it near-impossible to eradicate kanji because of their dependence on a written form in kanji.

As well as the larger political and philosophical discussions, the text is dotted with lots of interesting facts and observations that I found very educational. For example, now I finally understand (a simplified explanation of) why some characters like 合 and 雑 have pairs of Sino-Japanese (on) readings, in these cases here goo (ごう) and gaku (がく), and zoo (ぞう) and zatsu (ざつ) respectively. These paired on-readings are explained as the result of two different process of sound change operating on a single original Chinese syllable.

This has been an unfocussed review -- I think the book sometimes suffers from the same problem. But my conclusion is that although the use and the politics of kanji in Japanese are fascinating topics, I would rather find a better book to recommend as an introduction to it all. Part of my problem with the text is that I had real difficulty accepting the author's authority on the subject -- once the flaws in his arguments (and his bald assertions) became apparent, I started to wonder why I should trust him on any of it. But I think these various reflections have demonstrated that Takashima's text gives the critical reader a lot to respond to, which we could certainly see as a virtue too.