One kanji may have more than one way to read it. For example the character 木, meaning "tree" may be read as moku or ki. Japan adopted its writing system from China from the fifth century. The Japanese language is unrelated to Chinese, and the two languages had no common vocabulary originally, but Japanese had no written form. Chinese pronunciations such as moku were introduced to Japan together with the character. However, Japanese already had a word for "tree", ki. Both the Chinese and the Japanese pronunciations ended up being used to mean "tree".
The readings like moku, which come from Chinese, are called on-yomi (音読み), and readings like ki, which are original to Japanese, are called kun-yomi (訓読み). The on-yomi readings tend to be used in words made from more than one kanji, called compounds or jukugo (熟語), for example mokuzai (木材), meaning "wood".
Most Chinese characters came to Japan between the fifth and ninth centuries. The on-yomi is an approximation of the Chinese pronunciation of the character at the time it was introduced. Some kanji were introduced more than once from different parts of China at different times, and so have multiple on-yomi readings. For example the 行 kanji has on-yomi readings gō, gyō, and an.
The most common form of on-yomi readings is Kan'on (漢音), which come from the pronunciation during the Chinese Tang Dynasty in the 7th to 9th centuries. Go'on (呉音) readings, common in Buddhist terminology such as gokuraku (極楽) "paradise", came to Japan during the fifth and sixth centuries, from the pronunciation of the historical Wu region of China, in the vicinity of modern Shanghai. The less-common Tō'on (唐音) readings, which occur in some words such as isu "chair" or futon come from the pronunciations of later Chinese dynasties, such as the Song and Ming Dynasties, covering readings adopted from the Heian era (794-1185) to the Edo period (1603-1867). In addition, there are Kan'yō-on (慣用音) readings which are mistakes which became established.
In addition to the variety of Chinese-derived readings, a kanji may have several native Japanese kun-yomi readings when a kanji used for more than one meaning. For example, 行 has two kun-yomi readings, iku and okonau. Different kun-yomi can usually be distinguished by kana placed after the kanji, called okurigana. Iku is written 行く, with the kanji followed by く, and okonau is 行う, with the same kanji followed by う. See What is okurigana? for more on the way this is done.
Here are some examples of the various readings:
On-yomi mostly occur in words which are compounds of several kanji (jukugo). Many of the jukugo were adopted along with the kanji themselves from Chinese words. Jukugo sound more erudite or formal than native words. For example shinsei (申請) is a more formal word for "application" than the native Japanese mōshikomu (申し込む).
In surnames, a kanji like 藤 may appear with either kun-yomi, fuji as in 藤原 "Fujiwara", or the on-yomi, dō, as in 工藤, "Kudo". Male personal names often use on-yomi; female personal names tend to use kun-yomi. See How do Japanese names work? for more details of Japanese names.
Most of the kanji invented in Japan do not have on-yomi, but some do, such as 働 'to work', which has the kun-yomi hataraku and the on-yomi dō, or 腺 'gland', which has the on-yomi sen. See Which kanji were created in Japan? for more about the made-in-Japan Chinese characters.
Copyright © 1994-2020 Ben Bullock