Japanese itself had no written form. Eventually a writing system called manyogana (used in the ancient poetry anthology Manyoshu) evolved that used a limited set of kanji for their phonetic value alone, not for their semantic value, which was necessary for writing Japanese poetry. Manyogana written in highly cursive style became hiragana, a writing system that was accessible to women (who were denied higher education). Major works of Heian-era literature by women were written in hiragana. Katakana emerged via a parallel path: monastery students simplified manyogana to a single constituent element. Hiragana and katakana are referred to collectively as kana.
As the Japanese system of writing matured and expanded, kanji began to be used to write certain parts of speech, such as nouns, adjectives and verbs, while kana were used to write verb endings, uniquely Japanese words, and foreign words (but note that this usage developed much later - originally foreign words were all written phonetically using kanji).
Types of kanji
While some kanji and Chinese hanzi are mutually readable, many more are not. In addition to characters that have different meanings in Japanese, and characters that have identical meanings but are written differently, there are also characters peculiar to Japanese known as kokuji (国字; literally "national characters"). Kokuji are also known as wasei kanji (和製漢字; lit. "Chinese characters made in Japan"). There are hundreds of kokuji, and although some are rarely used, many others have become important additions to the written Japanese language. These include:
- 國 国 kuni (country)
- 號 号 gō (number)
- 變 変 hen, ka(waru) (change)
Kyū-jitai were used before the end of World War II, but after the war the government introduced the simplified shin-jitai. Some of the new characters are similar to simplified characters used in the People's Republic of China, but the two are essentially different things.
There are also Chinese characters that are only used phonetically in Japanese (当て字 ateji), and many Chinese characters that are not used in Japanese at all.
A kanji character may have several (in rare cases ten or more) possible pronunciations, depending on its context, intended meaning, use in compounds, and location in the sentence. These pronunciations, or readings, are typically categorized as either on'yomi or kun'yomi (often abbreviated on and kun).
The on'yomi (音読み) of a kanji (also called its on reading or Chinese reading) is based on the Japanese approximation of the original Chinese pronunciation of the character at the time it was introduced. Some kanji were reintroduced from different parts of China at different times, and so have multiple on'yomi (and often multiple meanings as well). Contrariwise, wasei kanji typically have no on'yomi at all.
For example, the kanji for light or next (明) may be pronounced myō, from an early (c. 5th–6th century) borrowing from southern China, or mei, from a later (c. 7th–9th century) borrowing from northern China. However, the kanji 込 was not derived from any Chinese character, and thus lacks any on'yomi.
On'yomi are phonologically characterized by their tendency toward single-syllable readings, since each character expressed a single Chinese syllable. However, tonality aside, most Chinese syllables (especially in Middle Chinese, in which final stop consonants were more prevalent than in most modern dialects) did not fit the largely-CV (consonant-vowel) phonotactics of classical Japanese. Thus most on'yomi are composed of two moras (syllables or beats), the second of which is either a lengthening of the vowel in the first mora (this being i in the case of e and u in the case of o, due to linguistic drift in the centuries since), or one of the syllables ku, ki, tsu, chi, or syllabic n, chosen for their approximation to the final consonants of Middle Chinese. (In fact, palatalized consonants before vowels other than i (written as y in consonant clusters and the consonants ch, sh and j in these environments), as well as syllabic n, were likely added to the Japanese phonotactic system to better simulate Chinese; none of these features occur in words of native Japanese origin.)
On'yomi primarily occur in multi-kanji compound words (熟語 jukugo), many of which are the result of the adoption (along with the kanji themselves) of Chinese words for concepts that either didn't exist in Japanese or could not be articulated as elegantly using native words. This borrowing process is often compared to the English borrowings from Latin and Norman French, since Chinese-borrowed terms are often more specialized, or considered to sound more erudite, than their native counterparts.
The kun'yomi (訓読み) of a kanji (also called its kun reading, Japanese reading, or somewhat misleadingly its native reading) is a reading based on the pronunciation of a native Japanese word, or yamatokotoba, that closely approximated the meaning of the Chinese character when it was introduced. Again, there can be multiple kun readings for the same kanji, and some kanji have no kun'yomi at all.
For instance, the kanji for east, 東, has the on reading tō. However, Japanese already had a word for east, pronounced higashi (or sometimes azuma). Thus, the kanji character 東 had the latter pronunciations grafted onto it as kun'yomi. However, the kanji 寸, denoting a Chinese unit of measurement (slightly over an inch), had no native Japanese equivalent; thus it has only its on'yomi, sun.
Kun'yomi are characterized by the strict (C)V syllable structure common to yamatokotoba, passingly similar to that of the nearby Polynesian languages. Most noun or adjective kun'yomi are two to three syllables long, while verb kun'yomi are more often one or two syllables in length (not counting trailing hiragana called okurigana, although those are usually considered part of the reading).
Some kanji also have lesser-known readings called nanori, which are mostly used for people's names, and are generally closely related to the kun'yomi. Place names sometimes also use nanori (or, occasionally, unique readings not found elsewhere).
Gikun (義訓) are readings of kanji combinations that have no direct correspondence to the characters' individual on'yomi or kun'yomi, but are instead connected by the meaning of the written and spoken phrases. For example, the compound 一寸 might na´vely be read issun, meaning "one sun", but it is more often used to write the indivisible word chotto, "a little". Gikun also feature in some Japanese family names.
Many ateji (kanji used only for their phonetic value) have meanings derived from their usage: for example, the now-archaic 亜細亜 ajia was formerly used to write "Asia" in kanji; the character 亜 now means Asia in such compounds as 東亜 tōa, "East Asia". From the written 亜米利加 amerika, the second character was taken, resulting in the semi-formal coinage 米国 beikoku, lit. "rice country" but meaning "United States of America".
When to use which reading
The division between on'yomi and kun'yomi can seem arbitrary and unnecessarily difficult to the learner of Japanese. Words for similar concepts, such as east (東), north (北) and northeast (東北), can have completely different pronunciations: the kun readings higashi and kita are used for the first two, while the on reading tōhoku is used for the third. However, the situation is actually no less coherent than the similar mixture of pronunciations in English which resulted from similar borrowings from other languages.
To complicate the matter, there are two basic guidelines for determining the pronunciation of a particular kanji in a given context. First, and most simply, kanji occurring in compounds are almost always read using on'yomi. These sorts of words are sometimes called jukugo (熟語). For example, 情報 jōhō "information", 半月 hangetsu "half-moon", and 革命家 kakumeika "[a] revolutionary" all follow this pattern.
Secondly, kanji occurring in isolation -- that is, written adjacent only to kana, not to other kanji -- are typically read using their kun'yomi. Together with their okurigana, if any, they generally function either as a noun or as an inflected adjective or verb: e.g. 月 tsuki "moon", 情け nasake "sympathy", 赤い akai "red", 建てる tateru "to build". The rare kanji compounds that also have okurigana, such as 空揚げ karaage "fried" and 名無し nanashi "nameless", also fall into this category.
There are numerous exceptions to both rules. 赤金 akakane "copper", 日傘 higasa "parasol", and the famous 神風 kamikaze "divine wind" all use kun'yomi despite being simple kanji compounds. Fortunately, most exceptions to the second rule are simple nouns: 愛 ai "love", 禅 Zen, 点 ten "mark, dot".
The situation is further complicated by the fact that many kanji have more than one on'yomi: witness 説明 setsumei "explanation" versus 灯明 tōmyō "light offered to a god".
There are even kanji compounds that use a mixture of on'yomi and kun'yomi, known as jūbako (重箱) words. The word jūbako itself is an example: the first character is read using on'yomi, the second kun'yomi. Other examples include 金色 kin'iro "golden" (on-kun) and 影法師 kagebōshi "silhouette" (kun-on-on).
Some famous place names, including those of Tokyo (東京 Tōkyō) and Japan itself (日本 Nihon or rarely Nippon) are read with on'yomi; however, by far the vast majority of Japanese place names are read with kun'yomi (e.g. 大阪 Ōsaka, 青森 Aomori, 箱根 Hakone). Family names are also usually read with kun'yomi (e.g., 山田 Yamada, 田中 Tanaka, 鈴木 Suzuki). Personal names, although they are not typically considered jūbako, often contain mixtures of kun'yomi, on'yomi, and nanori, and are generally only readable with some experience (e.g., 大助 Daisuke [on-kun], 夏美 Natsumi [kun-on]).
Because of the ambiguities involved, kanji will often have their pronunciation for the given context spelled out in ruby characters known as furigana (small kana written above the character) or kumimoji (those printed to the side). This is especially true in texts for children or foreign learners and manga (comics). It is also used in newspapers for rare or unusual readings and for characters not included in the officially recognized set of essential kanji (see below).
Orthographic reform and kanji lists
In 1946, following World War II, the Japanese government instituted a series of orthographic reforms. Some characters were given simplified glyphs, called 新字体 (shinjitai). The number of characters in circulation was reduced, and formal lists of characters to be learned during each grade of school were established. Many variant forms of characters and obscure alternatives for common characters were officially discouraged. This was done with the goal of facilitating learning for children and simplifying kanji use in literature and periodicals. These are simply guidelines, so many characters outside these standards are still widely known and commonly used.
In 1946, the Japanese government created a list of 1850 "general use kanji" (tōyō kanji 当用漢字). This list of kanji was expanded to 1945 characters in 1981 and called the jōyō kanji (常用漢字).
Official kanji lists include:
- Essential kanji : 881 characters, to be learned by 6th grade.
- General-use (tōyō kanji 当用漢字): 1850 characters, to be learned by the end of high school: this includes the essential kanji. This list was replaced by the introduction of the daily use list in 1981.
- Daily-use kanji chart (常用漢字 jōyō kanji): 1945 characters, taught during primary and secondary school in Japan. In publishing for the general public, characters outside this category are often given ruby.
- Name kanji (jinmeiyō kanji 人名用漢字): 2232 characters as of 2004. Over the years, the Minister of Justice has increased the size and has a plan for further addition in response to requests from parents.
- Grade Breakdown Kanji Allotment Chart (学年別漢字配当表; Gakunen-betsu kanji haitōhyō)
- JIS Kanji: defines what characters should be available for use on computers. The JIS standard has been through numerous revisions; JIS X 0208:1997 includes 6,335 kanji.
The ideographic iteration mark (々) is used to indicate a plural meaning; it is pronounced as though the kanji were written twice in a row.
The Japanese government provides the Kanji kentei (日本漢字能力検定試験 Nihon kanji nōryoku kentei shiken; "Test of Japanese Kanji Aptitude") which tests the ability to read and write kanji. The highest level of the Kanji kentei tests about 6000 kanji.
Gaiji (外字), also known as "external characters," are rare kanji that are not represented in existing Japanese encoding systems. These include variant forms of common kanji that need to be represented alongside the more conventional glyph in reference works, and can include non-kanji symbols as well.
Gaiji can be either user-defined characters or system-specific characters. Both are a problem for information interchange, as the code-point used to represent an external character will not be consistent from one computer to another (in the former case) or from one operating system to another (in the latter).
Gaiji were nominally prohibited in JIS X 0208-1997, while JIS X 0213-2000 actually used the range of code-points previously allocated to gaiji, making them completely unusable. Nevertheless, they persist today with NTT DoCoMo's "iMode" service, where they are used for pictorial characters
Unicode allows for something similar to gaiji with its private-use area.