There are several kinds of Chinese characters, including a handful of pictograms (象形; xiángxìng) and a number of ideograms (指示; zhǐshì), but the vast majority are phono-semantic compounds (形聲; xíngshēng). Although Chinese characters are often called ideograms, only a handful fit this category in any sense, and sinologists and linguists discourage referring to Chinese characters as ideograms.

The different classes of Chinese characters

Traditional Chinese lexicography divides characters into six categories, which are described below. This classification system dates back to Xú Shěn's second century dictionary, the Shuōwén jiězì. Although this categorisation is no longer the focal point of modern lexicographic practice, it is fairly simple to understand and remains useful.

The earliest Chinese characters were written by oracles on turtle shells and cattle bones for use in scapulomancy . These ancient characters are called jiǎgǔwén (甲骨文). They are generally pictograms which amount to stylised drawings of the things they refer to, or in some cases ideograms created with more conceptual intentions. They have changed drastically in shape, usage and meaning in the intervening millenia.

At present, more than 90% of all Chinese characters are phono-semantic compounds, constructed out of elements intended both to hint at its meaning and its pronunciation. However, as both meaning and pronunciation have changed over time, these components are no longer good guides either to meaning or to pronunciation. Reconstructing Old Chinese phonetics from the clues present in characters and other sources is a part of diachronic linguistics with a long tradition in China. In Chinese, it is called yīnyùnxué (音韻學).

Pictograms

象形 xiángxìng "form imitation".

Roughly 600 Chinese characters are pictograms - characters which are stylised drawings of the things they represent. These are generally among the oldest characters in Chinese. A few, indicated below with their earliest forms, date back to the 16th century BCE and are found on stone tablets and in bone and shell engravings used in scapulomancy . (Scapulomancy is the practice of writing a question on a bone or a shell, putting it into a fire, and devining the answer by interpreting the cracks that appear in the medium.)

These characters (called "Seal Script") were used up to the 2nd century BCE, evolving through that period. They are still used today for certain formal and artistic purposes, such as personal seals and Chinese calligraphy.

Simplified Chinese characters, in contrast, have only been formalised since 1958, and are not used by all Chinese users. Simplified characters are generally well attested as having been used since ancient times as shorthands and as variants of traditional forms, although a few have been simply invented in modern times.

The table below summarises the evolution of a few Chinese pictograms. Where no simplified form is provided, it is identical to the traditional character.

N.B.:

• 女 nǚ is a stylised drawing of a woman kneeing in deference. The oldest pictograms draw her from the front, but the profile view was ultimately adopted in Seal script.
• 子 zǐ is a child being carried in the traditional manner, with the legs out of view.
• 川 chuān, "river" is also a radical used in modern characters in a different form, as 巛 or 巜.
• 水 shuǐ, "water" originally represented a rapids surrounded by eddies. Although similar to the character for water, it is different in etymology.

Note that modern characters can only use a fixed number of possible stroke forms. Curved lines are not allowed in regular Chinese writing. There are only 24 (or by some counts, 21) fundamental strokes used today.

Pictograms have long been classed into eight groups - body, man, travel, village, brush, dragon, jade and yellow - based on semantic and cultural relationships between their referents. 214 pictograms are used as radicals - graphic elements shared by different characters used in indexing - and each character must contain at least one of them.

Simple Ideograms

指示 zhǐshì "indication"

Ideograms are intended to express a relatively abstract idea by means some non-arbitrary sign or by modifying an existing pictogram. This most often means pictograms with added dots or lines to indicate what part or action is intended. In the examples below, abstract notions like numbers are represented by a matching number of strokes and the parts of trees are represented by marking them on a pictogram of a tree.

N.B.:

• 本 běn, "root" - a tree (木 mù) with the base indicated by an extra stroke.
• 末 mò, "apex" - the reverse of 本 (běn), a tree with the top highlighted by an extra stroke.

Composed Ideograms

會意 huìyì "joined meaning"

Two or more characters are juxtaposed to indicate a new meaning associated with the conjunction of those elements. Every Chinese character must be written within a unit square, so these elements will change shape and size to fit, often reduced to a less complex version of the original form. Some examples:

• 人 "human" → 亻
• 水 "water" → 氵
• 艸 "grass" → 艹

Several reduced forms may exist for a single character.

These ideograms are recognised by their use of more than one element to indicate a different meaning related to all of them but different from any of them. The results are often a sort of Rebus puzzle, and need not be obvious at all. The character 明 - composed of the characters for the sun and the moon - means "bright". One might just as well have expected the juxtaposition to mean "shadow".

A few further examples:

There is more information on composed ideograms in the article for radicals.

Phono-semantic compound characters

形聲 xíngshēng "form and sound"

By far the bulk of Chinese characters - over 90% - were created by linking together a character with a related meaning and another character to indicate its pronunciation. These constructs came into being in part because of the sharp reduction the number of distinct syllables in Mandarin Chinese compared to the earlier forms of Chinese. In general, Chinese characters have single syllable pronunciations, but the standard form of Mandarin Chinese allows only roughly 2000 distinct syllables, and even many of those distinctions are eliminated in context. This means that a great many characters have homophones or near homophones.

For example, the verb meaning "to wash one's hair" is pronounced mù, which sounds the same as the character for "tree". So, the character used to indicate washing one's hair is composed of the character for "tree", because it sounds the same, and the character for "water" (水, shuǐ), because "water" is semantically related to "washing".

This practice appeared relatively late in the development of Chinese writing. Ancient Chinese had a much richer set of phonemes, and considerably less homophony, and as the spoken language dropped phonological distinctions, the written langauge did not keep up. Thus, both "grove" and "to pour" were long written with the same character - 林 - and only context distinguished them.

The phonetic element of a compound character is often chosen to some extent with the meaning in mind. For example, 認 rèn "to know (someone)" contains three radicals:

• 言 yán "speech"
• 刀 dāo "knife" (Notice the stroke across the knife, indicating that 刃 rèn "blade" is the intended character bearing the pronunciation information.)
• 心 xīn "heart"

The choice of 刃 rèn as the phonetic part of the character is not arbitrary. One might poetically interpret the character as to know someone, you must cut their heart with your words, indicating intimacy.

菜 cài ("vegetable") is much simpler to interpret. The pictogram for "grass" is used, in conjunction with 采 cǎi ("harvest") as its pronunciation. But of course, here too the combination is not arbitrary. A vegetable is a plant that one harvests.

This type of decomposition should not necessarily be taken at face value. These poetic interpretations were generally constructed long after the characters were invented, and they form the basis for many Chinese word games. Furthermore, this system is complicated because the Chinese language has changed a great deal since the era when many of these phono-ideographic characters were first coined. Consequently, the phonetic portion of a character may no longer reflect the modern pronunciation and the semantic portion may not relate to its meaning.

False friends

假借 jiǎjiè "false borrowing"

These are characters that are written using pictograms with unrelated meanings, but with the same pronunciation. Many grammatical particles are written this way, but also the verb 來 (lái; "to come"), which is identical to its homonym in ancient Chinese "wheat" (now pronounced mái and written 麥).

This process is to some extent still productive. The character 后 traditionally means "queen, empress", but has long been used as shorthand for its homophone "behind" (traditionally written 後). Character simplification has simply formalised this usage, using 后 exclusively for "behind".

Derived characters

轉注 zhuǎnzhù "reciprocal meaning"

This group of characters generally covers words with similar meanings and often the same etymological root, but which are pronounced differently and usually have somewhat different meanings. The English words chance and cadence, for example, have the same Latin root word: cadentia, cadentiam, meaning "fall". If English was written the way Chinese is, these two words would likely have similar characters.

The characters 老 lǎo ("old") and 考 kǎo ("a test") are examples of derived characters, which come from a common etymological root but differ in that one part is changed to indicate a different pronunciation and meaning.

See also

• Radical
• Chinese writing
• Chinese character
• Chinese calligraphy
• Seal script

External links

• This page draws heavily on the French Wikipedia page Classification des sinogrammes , downloaded 12 April 2005.