Posted 2017/5/15
Excerpt from a 1436 primer on Chinese charactersThe earliest known Chinese texts, in the Oracle bone script, display a fully developed writing system, little different functionally than modern characters. It can only be assumed that the early stages of the development of characters were dominated by pictograms, which were the objects depicted, and ideograms, in which meaning was expressed iconically. The demands of writing full language, including words which had no easy pictographic or iconic representation, forced an expansion of this system, presumably through use of rebus.
The presumed methods of forming characters were first classified c. 100 AD by the Chinese linguist Xu Shen, whose etymological dictionary Shuowen Jiezi (说文解字) divides the script into six categories, the liùshū (六书). While the categories and classification are occasionally problematic and arguably fail to reflect the complete nature of the Chinese writing system, this account has been perpetuated by its long history and pervasive use.
Four percent of Chinese characters are derived directly from individual pictograms, though in most cases the resemblance to an object is no longer clear. Others are ideograms, compound ideograms, where two ideograms are combined to give a third reading, or rebus. But most characters are phono-semantic compounds, with one element to indicate the general category of meaning and the other to suggest the pronunciation. Again, in many cases the suggested sound is no longer accurate.
Pictograms
象形字 xiàngxíngzì
Contrary to popular belief, pictograms make up only a small portion of Chinese characters. While characters in this class derive from pictures, they have been standardized, simplified, and stylized to make them easier to write, and their derivation is therefore not always obvious. Examples include 日 (rì) for "sun", 月 (yuè) for "moon", and 木 (mù) for "tree"....
There is no concrete number for the proportion of modern characters that are pictographic in nature; however, Xu Shen (c. 100 AD) estimated that 4% of characters fell into this category.
Ideograms
指事字, zhǐshìzì
Also called simple indicatives or simple ideographs, these characters either modify existing pictographs iconically, or are direct iconic illustrations. For instance, by modifying 刀 dāo, a pictogram for "knife", by marking the blade, an ideogram 刃 rèn for "blade" is obtained. Direct examples include 上 shàng "up" and 下 xià "down". This category is small.
Ideogrammic compounds
會意字/会意字 huìyìzì
Translated literally as logical aggregates or associative compounds, these characters symbolically combine pictograms or ideograms to create a third character. For instance, doubling the pictogram 木 mu "tree" produces 林 lin "forest", while combining 日 rì "sun" and 月 yuè "moon", the two natural sources of light, makes 明 míng "bright".
Xu Shen estimated that 13% of characters fall into this category.
Some scholars flatly reject the existence of this category, opining that failure of modern attempts to identify a phonetic in a compound is due simply to our not looking at ancient "secondary readings", which were lost over time. For example, the character 安 ān "peace", a combination of "roof" 宀 and "woman" 女, is commonly cited as an ideogrammic compound, purportedly motivated by a meaning such as "all is peaceful with the woman at home". However, there is evidence that 女 was once a polyphone with a secondary reading of *an, as may be gleaned from the set 妟 yàn "tranquil", 奻 nuán "to quarrel", and 姦 jiān "licentious".
Adding weight to this argument is the fact that characters claimed to belong to this group are almost invariably interpreted from modern forms rather than the archaic forms, which as a rule are quite different and often far more graphically complex. However, interpretations differ greatly between sources.
Phono-semantic compounds
形声字 xíngshēngzì
By far the most numerous category are the phono-semantic compounds, also called semantic-phonetic compounds or pictophonetic compounds. These characters are composed of two parts: one of a limited set of pictographs, often graphically simplified, which suggests the general meaning of the character, and an existing character pronounced approximately as the new target word.
Examples are 河 (hé) river, 湖 (hú) lake, 流 (liú) stream, 沖 (chōng) riptide (or flush), 滑 (huá) slippery. All these characters have on the left a radical of three dots, which is a simplified pictograph for a water drop, indicating that the character has a semantic connection with water; the right-hand side in each case is a phonetic indicator. For example, in the case of 冲 (chōng), the phonetic indicator is 中 (zhōng), which by itself means middle. In this case it can be seen that the pronunciation of the character has diverged from that of its phonetic indicator; this process means that the composition of such characters can sometimes seem arbitrary today. Further, the choice of radicals may also seem arbitrary in some cases; for example, the radical of 貓 (māo) cat is 豸 (zhì), originally a pictograph for worms, but in characters of this sort indicating an animal of any sort.
Xu Shen (c. AD 100) placed approximately 82% of characters into this category, while in the Kangxi Dictionary (AD 1716) the number is closer to 90%, due to the extremely productive use of this technique to extend the Chinese vocabulary.
This method is still sometimes used to form new characters, for example 钚 ("bu", meaning "plutonium") is the metal radical 金 plus the phonetic component 不 ("bu"), described in Chinese as "不 gives sound, 金 gives meaning". Many Chinese names of elements on the periodic table were formed this way.
Transformed cognates
转注字 zhuǎnzhùzì
Characters in this category originally didn't represent the same meaning but have bifurcated through orthographic and often semantic drift. For instance, 考 (kǎo) to verify and 老 (lǎo) old were once the same character, meaning "elderly person", but detached into two separate words. Characters of this category are rare, so in modern systems this group is often omitted or combined with others.
Rebus
假借字 jiǎjièzì
Also called borrowings or phonetic loan characters, this category covers cases where an existing character is used to represent an unrelated word with similar pronunciation; sometimes the old meaning is then lost completely, as with characters such as 自 (zì), which has lost its original meaning of nose completely and exclusively means oneself, or 萬 (wan), which originally meant scorpion but is now used only in the sense of ten thousand.
