The written form of Japanese is a gloriously (horribly) complex and multilayered system (mess) loaded with ambiguity and uncertainty. The focus here will be on one part of that writing system, Kanji. Kanji are the "characters" the Japanese adopted from the Chinese. The two characters at the top of the page, which are called Han(4)zi(4) [the numbers indicate tone] in Mandarin Chinese and Kanji in Japanese, mean "Han (Chinese) characters."
In dealing with Kanji I will especially concentrate on how they are put together, and how they do (or do not!) transmit sounds and meanings. Other parts of the Japanese writing system will be given brief reference as necessary.
Note: At this time I will not have much to say about a couple of very fascinating aspects of the Kanji/character system, namely, the esthetic power of calligraphy, and the differences in such things as brain activity, speed of reading, etc., that there may be between reading a sound-based, alphabetic or syllabary system on the one hand and a "character" based system on the other.
I have spent a total of little more than a month in Japan
between 1957 and 2001, and have never formally studied the language.
And although I spent much of my youth in Guangzhou (Canton) and
Hong Kong, I am not a specialist on the Chinese language (having
forgotten much of the not-too-much that I did once know). This
work reflects responses to a rekindled interest in things Asian--especially
of writing systems based on Chinese characters--and some of the
questions/puzzles I had to deal with in attempting to be able
to derive some meaning from Japanese writing during a couple of
recent trips to that country. It is based primarily on experience
attempting to learn enough Kanji to facilitate travel in
Japan, and on a variety of excellent sources available in books,
articles, and the web (click here
for these sources).
The complexity and difficulty that will be illustrated can be disheartening for a prospective learner. It IS a very difficult system to learn well--even for a native-speaker of the language. And the difficulty is not just in the need to memorize thousands of separate signs (as will be shown). Reasonably well educated, intelligent and motivated speakers of Spanish or English can read amost anything in their language well before they are 10 years old. Equally motivated, educated and intelligent Japanese students at the age of 10 cannot. And Japanese teachers often cannot read the names of their students the first day of class! (Click here to see why this might be the case.) But let this dishearten only the faint-hearted. Learning a few Kanji is not impossible, and is a BIG help in traveling (for one thing, learning the Kanji for places to be visited is essential for reading many train or ferry schedules).
These are the regularly-used sub-systems in Japanese writing:
1. KANJI The Kanji includes about 2000 "characters" derived from Chinese starting more than 1500 years ago, with more than 5,000 possible pronounciations [a single character may have from one to more than six or seven] and covering at least 5000 possible "meanings" or "semantic fields" which form the heart of the writing system. Most of these would be recognized by most educated Japanese readers. Several thousand more Kanji were also adopted for use by the Japanese (and have "standardized" Japanese pronounciations), although they are not "general use" characters, and most would not be recognized by most educated Japanese readers.
Because Kanji were developed from Chinese characters, and are still linked (however imperfectly) to Chinese meanings and sound systems, I will draw a lot of material from sources on Chinese writing as well as Japanese writing (and may eventually include some material on Korean and Vietnamese use of Chinese characters). It is important to recognize that Japanese on the one hand, and the Chinese languages such as Mandarin and Cantonese and Hakka (sometimes called "dialects") on the other are not very close. In fact, though they now share Kanji/characters and a lot of vocabulary (Japanese has borrowed thousands of spoken words from Chinese) they are not "related" in the traditional linguistic sense. The Chinese languages are part of the Sino-Tibetan language family. Japanese, and perhaps Korean, may be a part of the Altaic language family (which includes Turkic languages)--or maybe the Austronesian language family (whose members range from off the coast of Africa to Pacific Islands and even Tiawan). Linguists are not agreed as to the exact relationship of Japanese and Korean to each other or to other language families--except that they are NOT part of the Sino-Tibetan family. To put this in context, Japanese is no more related (in the sense that linguists use the term) to the Chinese languages than it is to English--and maybe even less so if some of the higher order links between language families some linguists (e.g. Nostraticists) are speculating about turn out to be "real" (see Ruhlen for one view on language families and relationships).
Although having origins in pictographic writing, only a few Chinese characters are now pictographic in any meaningful sense. Nor are they purely ideographic (symbols representing ideas), or logographic (symbols representing specific words) or phonetic (symbols representing specific sounds as in the graphemes used in alphabets and syllabaries). To avoid getting into further linguistic difficulties, let me put it this way: most (but not all) Chinese characters and Kanji include both meaning elements and sound elements and could therefore be called "phonetic-ideographic" or "semasio-phonetic" (see Henshall xv-xix for this and for some discussion of millenial-old Chinese classifications of the various "types" of characters used in that language).
The Japanese also invented a number of new Kanji. Some of them were accepted back into the Chinese character system; others are considered "Japanese-only." Click here to view some Japanese-made Kanji.
2. KANA Some time after adopting Chinese characters, the Japanese developed two syllabaries (KANA), partially derived from Kanji. Syllabaries are writing systems in which whole syllables, rather than discrete sounds (phonemes), are represented by signs which cannot be broken down into their constituent sound elements.
2a. Hiragana In this syllabary verb endings and other grammatical elements--not present in the Chinese character system--and other features of the Japanese language are reflected in writing. Hiragana appears to be more cursive ("rounder") than katakana.
2b. Katakana This syllabary, which covers exactly the same syllable sound-system, is used primarily for writing out the sounds of borrowed words (especially of non-Asian origins) and for emphasis. Katakana is written in a more angular, linear form than is hiragana.
2c. Furigana These are smaller than usual syllable signs (usually hiragana) that are written alongside or above a Kanji primarily to indicate its correct pronounciation (remember that any Kanji may have several pronounciations or readings which may or may not alter its meaning).
3. Romaji is the Roman alphabet. Most literate Japanese are familiar with romaji (even if not any of the other languages usually expressed by that alphabet).
4. ARABIC NUMERALS Most of the time Japanse writing utilizes Arabic numerals.
5. CHINESE NUMERALS These are actually characters, or Kanji. They may be used alone or in combination with Arabic numerals (as in the example immediately below). Note that while the Chinese languages and the Japanese languages use a base-ten numerical system (as does most of the rest of the world), the character/Kanji system has independent signs for 100, 10,000 and so on (not just the ten signs in Arabic numbers running from zero to nine).
Following is a single sentence in Japanese that is written using all of the above subsystems (Kanji, hiragana, katakana, Arabic numerals and a Chinese number) except for romaji and furigana (click here for an example of furigana use). It is accompanied by a transliteration (writing the sounds of one language in the writing system of another) into English, a word by word translation, and an overall English translation:
You can buy all kinds of hot and cold concoctions in cans in Japanese vending machines, and it may also be true that the Japanese love visiting Hawaii, but I don't think they've made a drink out of that Hawaiian favorite, poi. And this wouldn't be it, anyway. The letters that look like "POI" in the photo are actually the katakana form of the syllables, A, RO, and E, which are a transliteration of "aloe." (There are no "L" sounds in Japanese, so when they occur in a word to be put into Japanese they are usually transliterated into an "R" sound). The transcription below aroe is YO-GU-T'-TO, which you, thirsty (but maybe not THAT thirsty) reader, have already figured out to be that Turkish word for a fermented milk product.
In terms of merely representing the sounds of Japanese, the Kanji make up an incomplete system. That is, the correct SOUNDS of very few complete and grammatical Japanese sentences can be written entirely in Kanji. For one thing, verb forms require hiragana elements for completion (as in the verb form, kaeru, at the end of the sentence given above). On the other hand, an accurate and complete sound representation of almost any complete and grammatical Japanese sentence could be written with hiragana alone (and some books and magazines ARE thus written) or katakana alone, or even romaji alone. If the Japanese wished to join the rest of the world system, to ease their use of typewriters (do they still exist?) and computers, they COULD (it might seem) use the Roman alphabet. If they wished to simplify the incredible burden on learning and machine use, but at the same time retain something of their own distinctive system, they COULD use hiragana and/or katakana.
Some Japanese might prefer one of these solutions to the incredible complexity of their system. And those of us who value diversity and differentness for its own sake, are worried that this might happen. It all seems so simple. After all, in the case of the Roman alphabet, many Japanese can already read their language in that form. Why be burdened with all this other stuff, including the need to memorize about 2000 Kanji? (see Abolishing Kanji: The Dream and the Reality for a more detailed discussion of these issues.)
Several points to be made here. In the first place, calligraphy is a commonly appreciated art form in China, as well as in Japan. The esthetics of the writing system are culturally much more on the surface for the Japanese than they are for most users of the Roman alphabet. Just as important, perhaps, is that both Japanese and Chinese (but not Korean) have a more limited SOUND SYSTEM than does, say, English or French. To be more specific, there are far fewer syllables in these sound systems than is the case in English. According to material cited in DeFrancis (1984:42) Japanese uses 113 different syllables, Mandarin Chinese uses 1277 (about 400 if tones are not included) and English uses more than 8,000. The Japanese syllabary systems, hiragana and katakana, have only 51 basic syllable symbols (all syllables contain one of these). To these can be added a diacritical element in both hiragana and katana to indicate voicing: e.g. to change ka to ga ; or to to do. Palatization (adding a "y" sound between an initial consonant and a vowel: e.g. to change ka to kya; no to nyo) is added with three more diacritical marks. To all this can be added indications of the lengthening of vowel sounds (which makes a difference--click here for an example). Thus a total of under under 60 graphic symbolic elements are used to represent the slightly more than one hundred syllables used in Japanese (in both hiragana or katakana).
Anyway, the point is this. Japanese (and to a lesser extent, Chinese) is much less rich than English in syllables. Chinese is sometimes thought of as monosyllabic (Japanese is not). All this means is that each Chinese character has a reading that is only a syllable long (most "words" however are be made up of two or more character/syllables, thus the language in NOT monosyllabic). A Japanese reading of a Kanji/character may be more than one syllable long--most are one or two syllables, but many have three or more.
In both Chinese and Japanese, rather that stretching things out to accommodate a more limited sound system, there is the use of the same sound for the carrying of many multiple meanings (homophones). Chinese takes care of some of this over-lap of sounds by using four (Mandarin) or more (Cantonese) tone distinctions.
Thus, for example, in Japanese, the same sound, "shin" can carry the meanings/semantic fields of new, god, mind, heart, body, the dragon (of the zodiac), advance, quake, (and about 15 more!). Talk about opportunities for double entendre, in the spoken language--we have to deal with possible multiple quintipule entendre! So ... Multiple Kanji, all carrying the same sound (well, it ain't even that simple, but make the assumption for argument's sake, for now) can have their meanings distinguished in a couple of ways: e.g. Context (which has to operate in the spoken form of the language--but which still may lead to some ambiguity); Combination (most "words" are, in fact, represented by Kanji combinations rather than a single Kanji--this would be a special form of context); and Kanji form (the form of the Kanji indicates clearly which meaning(s) of the sound are being communicated. Thus, if all Japanese were written in romaji or katakana or hiragana, the precision (in meaning--such as it is) of the language would be tremendously reduced. That is, with only sound representation, the meaning-transmission system would be weakened.
And ... By using Chinese derived characters, Japanese people can at least get some idea of what they might have the opportunity to read in Chinese ... And many Chinese languages (Mandarin and Cantonese, for example) are mutually unintelligible. A monolingual Mandarin speaker cannot communicate orally with a monolingual Cantonese speaker. However, because the various Chinese languages (sometimes called dialects) are close enough grammatically, speakers of the different languages are able to communicate to some extent each other through writing.