Traditional Shuowen Jiezi scheme

The Shuowen Jiezi was a hugely influential character dictionary written by the scholar Xu Shen c. 120 CE. In the dictionary's postface, Xu analyses what he sees as all the methods by which characters are created. This work introduced a categorisation scheme which would later become known as the liùshū (六書; 六书; 'six writings'). Mature formulations of this scheme stated that every character belonged to one of six categories, each mentioned with varying emphasis in the Shuowen Jiezi. For nearly two millennia afterwards, this framework would serve as the traditional lens through which characters were analysed throughout the Sinosphere.^[10] Xu based most of his analysis on examples of Qin seal script that were written down several centuries before his time—these were usually the oldest forms available to him, but Xu stated that he was aware of the existence of even older forms.^[11]

Modern scholars agree that the theory presented in the Shuowen Jiezi is problematic, failing to fully capture the nature of Chinese writing, both in the present, as well as at the time Xu was writing.^[12][13] However, the model has proven resilient and pervasive; it continues to serve as a guide for those studying writing systems that use Chinese characters. One of the most important features of the Shuowen Jiezi was its grouping of characters by radical: a component within a character that is generally considered to be of particular import. The Shuowen Jiezi recognised over 500 radicals—this number would be reduced substantially in future dictionaries, but the concept itself would remain ubiquitous.

Pictograms

Graphical evolution of pictograms

日 ('Sun')

山 ('mountain')

象 ('elephant')

Presented as a fundamental class upon which the rest of the writing system depends, a relatively small number of characters are pictograms, representational pictures of physical objects.^[14][15] In practice, their forms are highly stylised and simplified from centuries of iteration: examples include 日 ('Sun'), 月 ('moon'), and 木 ('tree').^{[upper-alpha 1]} Xu Shen placed approximately 4% of all characters into this category.

Over time, pictograms became increasingly stylised, simplified, and standardised, in order to make them easier to write. As character forms developed, distinct depictions of various physical objects within pictographs became reduced to instances of a single written component.^[16][17] As such, what a pictogram is depicting is often not immediately evident. For example, within a given character the radical ⼝ 'MOUTH' often carries a meaning related to mouths, but within 高 ('tall')—a pictogram of a tall building—it instead depicts a window, ultimately lending to the character's meaning of 'tallness'. In another instance, the same 'mouth' radical depicts the lip of a vessel in the modern form of the pictogram 畐 ('full').^{[upper-alpha 2]}

Pictograms have often been extended from their original concrete meanings to take on additional layers of metaphor and synecdoche, which sometimes even displace the pictogram's original, literal meaning. Over time, this process sometimes creates excess ambiguity between graphically or phonetically similar characters, which is then usually resolved through adding additional components to disambiguate the characters in question. This can result in new pictograms, but usually results in other character types instead.^[18]

Compound ideographs

Also translated as logical aggregates or associative idea characters, characters in this class are formed by combining two or more pictographs or ideographs to suggest a new, synthetic meaning. The canonical example is 明 ('bright'), often interpreted as the juxtaposition of the two brightest objects in the sky: 日 ('sun'), and 月 ('moon'), together expressing their shared quality of brightness. Though the historicity of this particular etymology has been contested in recent scholarship, it is definitively a canonical reading: for example, the common compound word 明白 means 'understanding', touching on the derived association of 明 with 'illumination'. The addition of the abbreviated 艹 'GRASS' radical on top results in the compound ideograph 萌 ('to sprout'), alluding to the heliotropic behaviour of plant life. Other commonly cited examples include 休 ('rest'), composed of pictographs ⼈ 'MAN' and ⽊ 'TREE', and 好 ('good'), composed of ⼥ 'WOMAN' and ⼦ 'CHILD'.^{[upper-alpha 3]}

The compound character 好 illustrated as its component characters 女 and 子 repositioned side by side

Xu Shen placed approximately 13% of characters in this category, but many of his examples are now believed to be phono-semantic compounds, whose origin has been obscured by subsequent changes in their form.^[20] Peter Boodberg and William Boltz go so far as to deny that any of the compound characters devised in ancient times were of this type, maintaining that now-lost "secondary readings" are responsible for the apparent absence of phonetic indicators,^[21] but their arguments have been rejected by other scholars.^[22]

In contrast, associative compound characters are common among kokuji, kanji originally coined in Japan. An example of a modern compound ideograph in the Chinese language is 砼 ('concrete'), combining the ⼈ 'MAN', ⼯ 'WORK', and ⽯ 'STONE' radicals.^{[upper-alpha 4]}

Contemporary schemes

The traditional Shuowen Jiezi schema presupposes either a phonetic or semantic purpose for every character component.^[38][39] More recently, using the lens of modern semiotics, many components have been identified as not functioning in either role: they are purely signs, or "pure forms". Basic examples of pure form characters are found with the numerals beyond four, e.g. 五; 'five' and 八; 'eight', whose forms do not give visual hints to the quantities they represent.^[40] From this structural point of view, various systems have been proposed by modern scholars, with a straightforward example being seven categories of:^[41][42]

1. Semantic characters made of only semantic components,
2. Phonetic characters made of only phonetic components,
3. Pure form characters made of only pure form components,
4. Semantic–phonetic characters made of both semantic and phonetic components,
5. Semantic–form characters made of semantic and pure form components,
6. Phonetic–form characters made of phonetic and pure form components, and
7. Semantic–phonetic–form characters with all three component types.

According to Yang,^[43] of the 3,500 frequently used characters in contemporary Standard Chinese, semantic characters are the rarest, accounting for about 5% of the lexicon, followed by pure form characters with 18%, and semantic–form and phonetic–form together accounting for 19%, with the remaining 58% being semantic–phonetic characters—loosely analogous to the traditional category of phono-semantic compounds.

Old Chinese

Excerpt from a 1436 primer on Chinese characters

Words in Old Chinese were generally monosyllabic; as such, each character denoted an independent word.^[48] Affixes could be added to form a new word, which was often written with the same single character. In many cases, the pronunciations then diverged due to the systematic sound changes caused by the affixes. For example, many additional readings in modern varieties reflect the Middle Chinese 'departing tone', the major source of the 4th tone in modern Standard Chinese. Many scholars now believe that this Middle Chinese tone is the reflex of an Old Chinese derivational suffix /*-s/ called the qusheng 去聲 that served a range of semantic functions—possibly the only example of inflectional morphology extant in the otherwise analytic language.^[49][50] For example:

Another common sound change occurred between voiced and voiceless initials, though the phonemic voicing distinction has disappeared in most modern varieties. This is believed to reflect an Old Chinese de-transitivising prefix, but scholars disagree on whether the voiced or voiceless form reflects the original root. Note how the pairs of readings below reflect opposite transitivity from one another.

Vernacular Chinese

Multi-syllable words began entering the language during the Western Zhou period; it is estimated that between 25% and 30% of the vocabulary used in Warring States period texts is polysyllabic. The process has accelerated over the centuries as phonetic change has increased the number of homophones.^[56] The most common process of Chinese word formation after the Classical period has been to create compounds of existing words. Words have also been created by appending affixes to words, by reduplicating words, and by borrowing words from other languages.^[57] While polysyllabic words are generally written with one character per syllable, abbreviations are occasionally used.^[58]

Many compound words are composed from two near-synonymous characters words, creating a new, less ambiguous form that is often used in variation with one of its component characters, depending on context. For example:

Equally as common are nouns composed from a root and a particle suffix possessing no particular meaning, such as 子 (zǐ). These constructions serve to create a disyllabic word with the same meaning as the root character. As above, the root word usually, though not always, remains independent, in variation with the compound word.

Morphemic characters that have fallen out of use as independent words, and are now used only in compounds, are called bound forms.

Large-scale surveys by the PRC's Ministry of Education and State Language Commission have shown strong distribution patterns in the use of characters and words. This form of analysis is essential to the quantitative research of the Chinese language, with applications in pedagogy, publishing, and information processing.^[59]

The number of characters used in modern Chinese is stable, hovering around 10,000 in recent decades. Contrastingly, 80% of Chinese-language text is composed of just 590 characters, with 90% coverage achieved with 960 characters, and 99% with 2,400.^[60]

Legendary origins

Several works of classical Chinese literature indicate that, prior to the invention of characters, knotted cords were used to keep records.^[62][63] The practice had some similarities to the Inca technique of quipu.^[64] Works that reference the practice include chapter 80 of the Tao Te Ching^[65] and the "Xici II" chapter within the Yijing.^[66]

According to tradition, Chinese characters were invented by Cangjie, a mythical figure said to have been a scribe to the legendary Yellow Emperor during the 3rd millennium BCE. Frustrated by the limitations of knotting, and inspired by his study of the animals of the world, the landscape of the earth, and the stars in the sky, Cangjie is said to have invented symbols called 字 (zì)—the first Chinese characters. The legend relates that on the day the characters were created, grain rained down from the sky and that night the people heard ghosts wailing and demons crying because the human beings could no longer be cheated.^[67]

Neolithic

In recent decades, a series of inscribed graphs and pictures have been found at Neolithic sites in China, including Jiahu (c. 6500 BCE), Dadiwan and Damaidi from the 6th millennium BCE, and Banpo (5th millennium BCE). Often these finds are accompanied by media reports that push back the purported beginnings of Chinese writing by thousands of years.^[68][69] However, because these marks occur singly without any implied context and are made crudely, Qiu Xigui concludes that "we do not have any basis for stating that these constituted writing nor is there reason to conclude that they were ancestral to Shang dynasty Chinese characters."^[70] However, they do demonstrate a history of sign use in the Yellow River valley from the Neolithic through to the Shang period.^[69]

Oracle bone script

Ox scapula inscribed with characters recording the result of divinations

The earliest known examples of writing directly ancestral to modern characters are a body of inscriptions made on bronze vessels and oracle bones during the late Shang dynasty (c. 1250 – 1050 BCE),^[71][72] with the very oldest dated to c. 1200 BCE.^{[73][74]: 108} Oracle bones and the script they bore were first documented by modern scholars in 1899, after examples were discovered being sold as "dragon bones" for medicinal purposes, with the symbols carved into them identified as being Chinese writing. By 1928, the source of the bones had been traced to a village near Anyang in Henan, which was excavated by the Academia Sinica between 1928 and 1937. To date, over 150,000 such fragments have been found.^[71]

Oracle bone inscriptions are records of divinations performed in communication with royal ancestral spirits.^[71] The inscriptions range from a few characters in length at their shortest, to around 40 characters at their longest. The Shang king would communicate with his ancestors by means of scapulimancy, inquiring about subjects such as the royal family, military success, and weather forecasting. The interpreted answers would be recorded on the divination material itself.^[71]

Oracle bone script is a well-developed writing system,^[75][76] suggesting that the Chinese script's origins may lie earlier than the late second millennium BCE.^[77] Although these divinatory inscriptions are the earliest surviving evidence of ancient Chinese writing, it is widely believed that writing was used for many other non-official purposes, but that the materials upon which non-divinatory writing was done—likely on wood and bamboo—were less durable than bones and shells, and have since decayed away.^[77]

Zhou scripts

The Zhou-era Shi Qiang pan, a bronze ritual basin dated c. 900 BCE. Long inscriptions on the surface describe the deeds and virtues of the first seven Zhou kings.

The traditional notion of an orderly procession of scripts, with each suddenly invented and displacing the one previous, has been conclusively superseded by modern archaeological finds and scholarly research.^[78] More often, it was the case that two or more scripts coexisted in a given area, and that scripts evolved gradually. As early as the Shang dynasty, oracle bone script coexisted as a simplified form alongside the normal script in bamboo books—preserved in bronze inscriptions—as well as the elaborate pictorial forms, often clan emblems, found on many bronzes.

Based on studies of these bronze inscriptions, it is clear that the mainstream script evolved in a slow, unbroken fashion from the Shang to the Zhou dynasty, until assuming the form that is now known as small seal script in the state of Qin, without any sudden shifts.^[79][80] Meanwhile, other scripts had evolved during the late Zhou, especially in eastern and southern regions. These include decorative scripts such as the bird-worm seal script, and the regional 'ancient' forms of eastern Zhou states, preserved as variant forms in the Han-era Shuowen Jiezi.

Qin unification and small seal script

Part of a series on
Calligraphy
By region Arabic (Islamic) Chinese Cyrillic Georgian Indian Japanese Korean Mongolian Persian Serbian Tibetan Turkish Vietnamese Western

Small seal script, which had evolved conservatively in the state of Qin during the Eastern Zhou, became standardised as the orthographic convention used throughout all of China by the imperial Qin dynasty. However, more than one script was in use at the time: a little-known, rectilinear, 'vulgar' form of the characters had coexisted alongside the more formal seal script for centuries in the Qin state; the popularity of this vulgar form grew as the practice of writing itself became more widespread.^[81] An immature form of clerical script called "early clerical" or "proto-clerical" had already developed by the Warring States period in the state of Qin^[82] based upon this vulgar form, with influence from seal script as well.^[83] The coexistence of the three scripts—small seal, vulgar and proto-clerical, with the latter evolving gradually into clerical script—runs counter to the traditional belief that the Qin dynasty only used one script, and that the clerical script was suddenly invented during the early Han.

Han clerical script

The proto-clerical script matured gradually, and by the early Han period its sophistication was comparable to small seal script.^[84] Recently discovered bamboo slips show the emergence of mature clerical script by the end of Emperor Wu of Han's reign in 141–87 BCE.^[85]

As in previous eras, multiple scripts were in use during the Han,^[86] although mature clerical script—also called 八分 (bāfēn)^[87]—was dominant. An early type of cursive script was also in use at least as early as 24 BCE,^{[lower-alpha 13]} incorporating cursive forms popular at the time, as well as elements from the vulgar writing that originated in Qin state.^[88] By the time of the Jin dynasty, this Han cursive style became known as 章草 (zhāngcǎo), sometimes known in English as 'clerical cursive', 'ancient cursive', or 'draft cursive'. Some believe this name, which uses the character 章 ('orderly'), arose because the style was considered by the Jin to be a more orderly form^[89] than what would become the modern form of cursive, called 今草 (jīncǎo),^[90] which had first emerged during the Jin, and is still used today.

Wei to Jin

Regular script

Main article: Regular script

A page from a printed Song publication in a regular script typeface, which resembles the handwriting of Tang-era calligrapher Ouyang Xun

The design of regular script has been credited to Cao Wei calligrapher Zhong Yao (c. 151 – 230), often called the "father of regular script". However, some scholars observe that one person could not have unilaterally developed a new script which went on to see universal adoption, but could only have been a crucial contributor to the style's gradual formation. The earliest surviving manuscripts written in regular script are copies of Zhong Yao's work, including at least one copied by Wang Xizhi, often called the "Sage of Calligraphy". Regular script developed out of a neatly written form of early semi-cursive, with the addition of a 'pause' (顿; 頓; dùn) technique to end horizontal strokes, plus heavy tails on strokes which are written the downward-right diagonal.^[95] Thus, early regular script emerged from a neat, formal form of semi-cursive, which had itself emerged from neo-clerical, a simplified, convenient form of clerical script. It matured further during the Eastern Jin in the hands of Wang Xizhi and his son Wang Xianzhi. However, it had not yet achieved widespread use, with most writers continuing to use the earlier neo-clerical and semi-cursive styles for daily writing,^[95] with the conservative clerical script also remaining in use on some stelae.^[96]

Maturation of regular script

It was not until the Northern and Southern period that regular script acquired a dominant status.^[98] Nevertheless, it continued to evolve stylistically, only reaching full maturity during the early Tang dynasty. Some credit Ouyang Xun with producing the first examples of a mature regular script. After this point, though developments in calligraphy as an art form, as well as in the simplification of character forms would continue, there would not be another major stylistic shift for the Chinese family of scripts.

Calligraphy

Chinese calligraphy of mixed styles by Song poet Mi Fu

Chinese calligraphy is usually done with ink brush, and was considered one of the four arts to be mastered by Chinese scholars. The set of rules is deliberately minimalist, but each character has a set number of brushstrokes. Strict regularity is not required, since strokes may be accentuated for dramatic effect of individual style. Calligraphy was considered a means by which scholars could artfully express their thoughts and teachings.^[110]

Printing and typefaces

'Song' typefaces (宋体; 宋體; sòngtǐ)—also called 'Ming', especially in Japan, Taiwan, and Hong Kong—are named for the respective periods whose printed styles are being imitated, considered to be periods during which woodblock printing flourished in China. Ming and sans-serif are the most popular in body text.

Sans-serif typefaces, called 'black form' (黑体; 黑體; hēitǐ) in Chinese and 'Gothic' (ゴシック体) in Japanese, are characterised by simple lines of even thickness for each stroke, akin to sans-serif styles in Western typography.

Typefaces that emulate regular script are also common, but not as common as Ming or sans-serif typefaces in body text. Most typefaces in the Song dynasty were regular script typefaces, which resembled a particular calligrapher's handwriting, while most modern regular script typefaces tend toward general-purpose use.

Use with computers

The first four characters of the Thousand Character Classic in different typefaces and historical styles. From right to left: seal script, clerical script, regular script, Ming, and sans-serif

Even before the advent of computers, the very first electromechanical input/output and text encoding methods to be designed were done so for use with alphabet-based writing systems, exemplified by the design of typewriters and the Morse code and ASCII standards. Adaptation of these technologies for use with a logography of thousands of characters was non-trivial.^[111]

Like English and other languages, Chinese characters are output on printers and screens in different fonts.^[112] In addition to the international system of measuring with points, Chinese characters are also measured by a unit called zihao (字号), first invented for Chinese printing in 1859.^[113]

Japanese

Japanese writing
Components
Kanji Stroke order Radicals Kyōiku kanji Jōyō kanji Jinmeiyō kanji Hyōgai kanji List of kanji by stroke count
Kana Hiragana Katakana Hentaigana Man'yōgana Sōgana Gojūon
Typographic symbols Japanese punctuation Iteration mark
Uses
Syllabograms Furigana Okurigana Braille
Transliteration
Rōmaji Hepburn (colloquial) Kunrei (ISO) Nihon (ISO transliteration) JSL (transliteration) Wāpuro (keyboard)
Cyrillization Polivanov system

In the Japanese writing system, Chinese characters used are known as kanji. Japanese historically borrowed many words from Chinese, which were written with their original characters, while native Japanese words were also written with orthographic borrowings of Chinese characters with similar meanings. Most kanji arrived via both borrowing processes, and thus have both native Japanese readings, known as kun'yomi, as well as Chinese-original readings, known as on'yomi. Moreover, Chinese words were often borrowed multiple times from different varieties and at different times, resulting in several distinct on'yomi readings for the same character.^[124] Modern Japanese uses kanji for most word stems, as well as hiragana and katakana, a pair of syllabaries collectively known as kana. Hiragana are used to write words, including grammatical inflections and particles, and katakana are used for transcribing non-Chinese loanwords, as well as for emphasis of native words, similar to how italics are used in languages written with the Latin script.^[125] The syllabaries were derived by simplifying Chinese characters selected to represent Japanese syllables; they differ from one another in part because each selected different characters for certain syllables, in addition to the different strategies employed to reduce the characters for easy writing. The angular katakana were obtained by selecting a smaller component from each character, while the curving hiragana were based on the cursive form of the entire character.^[126]

Due to Japanese being a synthetic language, many words consist of multiple syllables, and as such many kanji have multi-syllable pronunciations. For example, the kanji 刀 has a native kun'yomi reading of katana. In different contexts, it can also be read with the on'yomi reading tō, such as in the Chinese loanword 日本刀, nihontō, 'Japanese sword', whose pronunciation descends from the Chinese pronunciation at the time of borrowing. (In contemporary Standard Chinese, the word is pronounced rìběndào.) While modern loanwords from languages outside of the Sinosphere are usually written with katakana, loanwords prior to the Meiji era were typically written with unrelated kanji whose on'yomi had the same pronunciation as the syllables in the loanword. These spellings are called called ateji: for example, 亜米利加 was written for modern アメリカ, Amerika, 'America', 歌留多 or 加留多 for modern カルタ, karuta, 'card', 'letter', and 天婦羅 or 天麩羅 for modern テンプラ, tenpura, 'tempura'. Only some ateji spellings are still in common use, such as 缶, kan, 'can'.

Korean

As early as the Gojoseon period, Classical Chinese was the dominant form of written communication in Korea. Although the hangul alphabet was invented by the Joseon king Sejong in 1443, it was not taken up by Korean literati, and did not come into widespread use until the late 19th century.^[127][128] Even today, much of the Korean vocabulary, especially in areas of science and sociology, comes directly from Chinese. However, due to the lack of tones in the Korean language, many dissimilar Sino-Korean words took on identical pronunciations, and as such are spelled identically in hangul.^[129] For example, the phonetic dictionary entry for 기사, gisa yields more than 30 different entries. In the past, this ambiguity had been efficiently resolved by parenthetically displaying the associated hanja. While hanja is sometimes used for Sino-Korean vocabulary, native Korean words are rarely, if ever, written in hanja.

When learning to write hanja, students are taught to memorise both native and Sino-Korean Korean pronunciations for each hanja. The collation of hanja is similar to if the word water were listed as 'water; aqua', 'horse; equus', or 'gold; aurum', as hybridisations of the English and Latin terms.^[130] Examples of listings include:

Okinawan

Chinese characters are thought to have been first introduced to the Ryukyu Islands in 1265 by a Japanese Buddhist monk.^[137] After the Okinawan kingdoms became tributaries of Ming China, especially the Ryukyu Kingdom, Classical Chinese was used in court documents, but hiragana was mostly used for popular writing and poetry. After Ryukyu became a vassal of Japan's Satsuma Domain, Chinese characters became more popular, as well as the use of kanbun. In modern Okinawan, which is labelled as a dialect of Japanese by the Japanese government, katakana and hiragana are mostly used to write Okinawan, but Chinese characters are still used.

Vietnamese

The first two lines of the classic Vietnamese epic poem The Tale of Kiều, written in both chữ Nôm and the Vietnamese alphabet

  Borrowed characters representing Sino-Vietnamese words

  Borrowed characters representing native Vietnamese words

  Invented chữ Nôm representing native Vietnamese words

Until the early 20th century, Literary Chinese (Hán văn) was used in Vietnam for all official and scholarly writing. The chữ Nôm script began to be developed around the 13th century to record folk literature in the Vietnamese language. Chinese characters, called chữ Hán (𡨸漢), chữ Nho (𡨸儒), or Hán tự (漢字), are now limited to ceremonial use in Vietnam.

The oldest written Chinese text found in Vietnam is an epigraphy dated to the year 618, erected by local Sui officials in Thanh Hóa.^[138] Similar to Zhuang sawndip, some chữ Nôm characters were created by combining semantic character components with phonetic components that resembled Vietnamese syllables.^[139] This process resulted in a highly complex system whose use was limited to a small portion of the Vietnamese population, never more than 5%.^[140] The oldest chữ Nôm written alongside Chinese is a Buddhist inscription dated to 1209.^[139] Before 1945, the library of the French School of the Far East (EFEO) in Hanoi collected a total of around 20,000 Chinese and Vietnamese epigraphy rubbings from throughout Indochina.^[141] The oldest surviving extant manuscript in Vietnamese is a late 15th-century bilingual copy of the Buddhist Sutra of Filial Piety, currently kept by the EFEO. It features Chinese text in larger characters, and an Old Vietnamese translation in smaller characters glossing the text.^[142] Every Hán Nôm book in Vietnam after the Phật thuyết is dated between the 17th and the 20th centuries, with most being hand-copied works, and few printed texts. By 1987, the library of the Institute of Hán-Nôm Studies in Hanoi had collected a total of 4,808 Hán Nôm manuscripts.^[143]

A page from a bilingual copy of the Sutra of Filial Piety, with Classical Chinese alongside an early form of chữ Nôm, representing the Old Vietnamese pronunciation. Sometimes, pairs of characters are used to represent the consonant clusters present in Old Vietnamese

Classical Chinese and chữ Nôm fell out of use during the French colonial period, and were gradually replaced with the Vietnamese alphabet, which uses Latin characters and remains the primary writing system for Vietnamese.^[144][145] Contemporaneous use of chữ Hán in Vietnam is often connected with traditional culture, such as the practice of calligraphy.

Vietnamese imperial edict in Classical Chinese

Other languages

Several minority languages of South and Southwest China were formerly written with scripts based on Chinese characters, but also included many locally created characters. The most extensive is the sawndip script used to write the Zhuang languages of Guangxi, which is still in use despite efforts to encourage the writing of Zhuang with a Latin-based alphabet. Other languages written with such scripts include Miao, Yao, Bouyei, Mulam, Kam, Bai, and Hani.^[146] All these languages are now officially written using Latin-based scripts. According to surveys, traditional sawndip script has twice as many users as the official Latin script.^[147]

The foreign dynasties that ruled northern China between the 10th and 13th centuries developed scripts that were inspired by Chinese characters but did not use them directly: the Khitan large script, Khitan small script, Tangut script, and Jurchen script—though Chinese characters were used to phonetically transcribe the language of the Jurchen people, renamed the 'Manchu' after the founding of the Qing dynasty. Other scripts within China that have adapted a few Chinese characters but are otherwise distinct include the Geba script, Sui script, Yi script, and the Lisu syllabary.^[146]

Transcription

Main article: Transcription into Chinese characters

Excerpt from the Secret History of the Mongols featuring Chinese characters used to write Mongolian, with glosses to the right of each row

Along with the Persian and Arabic scripts, the Mongolian language was also written with Chinese characters phonetically transcribing Mongolian sounds. Notably, the only surviving copies of The Secret History of the Mongols were written in such a manner.

According to the 19th century missionary John Gulick:

"The inhabitants of other Asiatic nations, who have had occasion to represent the words of their several languages by Chinese characters, have as a rule used unaspirated characters for the sounds g, d, b. The Muslims from Arabia and Persia have followed this method ... The Mongols, Manchu, and Japanese also constantly select unaspirated characters to represent the sounds g, d, b, and j of their languages. These surrounding Asiatic nations, in writing Chinese words in their own alphabets, have uniformly used g, d, b, etc., to represent the unaspirated sounds."^[148]

Received forms

From left to right: the regional forms for the character 次 in the Noto Serif CJK typeface family, as used in mainland China, Taiwan, and Hong Kong (top), as well as in Japan and Korea (bottom)

With the use of woodblock printing, there was a considerable consolidation in forms prior to the standardisation efforts of the 20th century, especially during the Ming. These orthodox forms are in turn well-represented in touchstone reference works throughout the modern era, such as the 1716 Kangxi Dictionary and the 1915 Zhonghua Da Zidian.

Simplified characters

One of the earliest proponents of character simplification was Lufei Kui, who proposed in 1909 that simplified characters should be used in education. In the years following the Xinhai Revolution and its associated May Fourth Movement, many anti-imperialist Chinese intellectuals sought ways to modernise China as quickly as possible. Traditional culture and values were challenged and subsequently blamed for societal and economic problems. Soon, people in the movement began pointing to the traditional Chinese writing system as an obstacle to the modernisation of China, proposing that it should either be reformed or abolished entirely. Lu Xun, a renowned 20th century author, stated 'If Chinese characters are not destroyed, then China will die'.^[157]

The first standardised list of simplified forms, introduced in 1935 and consisting of 324 characters^[158]

During the 1930s and 1940s, discussions on character simplification took place within the Kuomintang government, and a large number of the intelligentsia maintained that character simplification would help boost literacy throughout the country.^[159][160] In 1935, a table of 324 simplified characters collected by Qian Xuantong was introduced as the first official batch of simplified characters; however, it was rescinded in 1936 due to fierce opposition within the party.

Traditional (們)

Simplified (们)

Comparison of strokes between character forms,^{[lower-alpha 18]} showing systematic simplification of the component ⾨ 'GATE'

Although most closely associated with the PRC, the modern process of character simplification began well before 1949. Cursive script were the source of inspiration for many of the simplified forms, while others were already used in print, albeit not for most formal works. With the goal of increasing functional literacy, a major concern at the time, discussions on character simplification took place among Chinese intelligentsia and within the Kuomintang (KMT) government during the Republican period.^[161] This earlier initiative to simplify the Chinese writing system was later inherited and implemented by the Communists after its subsequent abandonment by the KMT.

The use of traditional versus simplified characters varies greatly, and can depend on both the local customs and the medium. Before official reforms, character simplifications were not officially sanctioned and generally took the form of vulgar variants and idiosyncratic substitutions. Unofficial, often simplified forms would be used in everyday writing or for quick notes. Since the 1950s, the PRC has officially encouraged the use of simplified characters on the mainland. Along with the Republic of China, Hong Kong and Macau—at the time still under colonial rule—were not affected by the reform. There is no firm rule for which characters to use, and often it is determined by tastes and inclinations of the audience and writer.

In other Sinophone countries, the use of simplified characters is generally more common among younger people, while many older generations literate in Chinese still use traditional forms. Outside of China, Chinese-language shop signs are also often written using traditional characters.

Traditional characters

Regional allographs of 漢 in Chinese, Japanese, Korean, and Vietnamese styles

Kanji

After World War II, the Japanese government also instituted a series of orthographic reforms. Some characters were given simplified forms called shinjitai; the older forms were then labelled the kyūjitai. The use of numerous variant forms was discouraged, and lists of characters to be learned during each grade of school were created: first the 1850-character tōyō kanji list in 1945, and then the 1945-character jōyō kanji list in 1981, with a 2136-character revision in 2010. The Japanese government restricts characters that can be used in names to the jōyō kanji plus an additional list of 983 jinmeiyō kanji historically prevalent in names.^[178][179] While these lists serve as a guideline, unlisted characters are still widely used by native Japanese speakers, such as the kyūjitai form of 'dragon' (龍) alongside the shinjitai form (竜).

Contractions and abbreviations

In certain cases, compound words and set phrases may be represented by single-character contractions. Some of these can be considered logograms, where characters represent whole words rather than syllable-morphemes, though these are generally considered as non-standard ligatures or abbreviations instead—similar to scribal abbreviations such as an ampersand for the digraph et, or an ñ for the digraph nn. These usually see use in handwriting or decorations, but sometimes in print as well. These ligatures are called 合文; héwén, 合书; 合書; héshū or 合体字; 合體字; hétǐzì in Chinese; in the special case where two characters are combined, they are known as 'two-syllable characters' (双音节汉字; 雙音節漢字; shuāngyīnjié hànzì).

A commonly seen example is the 'double happiness' character 囍, formed as a ligature of 喜喜 and referred to by its disyllabic name 双喜; 雙喜; shuāngxǐ.^{[upper-alpha 10]} In handwriting, numbers are very frequently squeezed into one space or combined—common ligatures include 廿; niàn; 'twenty', normally read as 二十; èrshí, 卅; sà; 'thirty', normally read as 三十; sānshí, and 卌; xì; 'forty', normally read as 四十; sìshí in Standard Mandarin,^{[upper-alpha 11][8]} though other Chinese varieties may differ. For example, 廿 is given a monosyllabic reading of jaa⁶ in Cantonese.^[184] Calendars often use numeral ligatures in order to save space, and in modern printings of the traditional Chinese calendar, the use of 廿 is standard. Thus, one would generally write "21 March" as 三月廿一.

Examples of modern contractions include characters sometimes used to represent SI units. In Chinese these units are disyllabic and usually written with two characters, as with 'centimetre' (厘米; límǐ) from 厘; 'centi-' and 米; 'metre', or 'kilowatt' (千瓦; qiānwǎ). However, in the 19th century these were often written via compound characters, pronounced disyllabically, such as 瓩 for 千瓦 or 糎 for 厘米—some of these characters were also used in Japan, where they were pronounced with borrowed European readings instead. These have now fallen out of general use, but are occasionally seen. Less systematic examples include 图; 圕 (tuān), a contraction of 图书馆; 圖書館 (túshūguǎn; 'library').^{[upper-alpha 12]} Since polysyllabic characters are often non-standard, they are often excluded from character dictionaries.

The use of such contractions is as old as the characters themselves, and they have frequently been used in religious or ritual contexts. In the oracle bone script, personal names, ritual items, and even phrases are commonly contracted into single characters, such as 受又 (shòu yòu; 'receive blessings') becoming 祐 (yòu). A dramatic example is found in medieval manuscripts, where 'bodhisattva' (菩萨; 菩薩; púsà) is sometimes contracted to a single character composed of four 十 arranged in a 2×2 grid—derived from the 艹 'GRASS' radicals present in the original characters.^[8] For the sake of consistency and standardisation, the Chinese government has sought to limit the contemporary use of polysyllabic characters in public writing.^[8]

Conversely, with the erhua phenomenon in Mandarin varieties, expressed via the fusion of the diminutive 儿; ér suffix, some monosyllabic words may be written with two characters, such as in huār (花儿; 'flower').

Multi-syllable morphemes

Chinese characters are primarily morphosyllabic, meaning that there is usually a one-to-one correspondence between Chinese morphemes and spoken Chinese syllables, and therefore written Chinese characters. However, in modern Chinese varieties most common words are disyllabic and therefore dimorphemic. In modern Standard Chinese, 10% of morphemes are bound forms, only appearing in compound words. However, a few morphemes are disyllabic, some of which even date back to Classical Chinese.^[185] Excluding loanwords, these are typically words for plants and small animals, usually written with a pair of phono-semantic compounds sharing a common radical. Examples are 蝴蝶 (húdié; 'butterfly') and 珊瑚 (shānhú; 'coral')—the first character of 'butterfly' and the second character of 'coral' each have 胡 for a phonetic component, with the ⾍ 'INSECT' and ⽟ 'JADE' radicals as their respective semantic components, also present within the other character of each word. Neither of the aforementioned hú characters exist as independent morphemes, except as poetic abbreviations of the disyllabic words.

Rare and complex characters

Often a rare character, complex, or antiquate variant will appear in personal or place names. As many computer encoding systems have historically included only the most common characters, this can create problems. As a representative example, the name of Taiwanese politician Yu Shyi-kun contains the rare character 堃 (kūn); printing this character is often nontrivial. Newspapers covering Yu that encountered difficulty with his name dealt with this problem in various ways, including using software to combine two extant characters into a visually similar amalgam, embedding a picture of the character instead of encoding it as text, substituting a homophonic character with the expectation that the reader would make the correct inference. Generally, printed materials in Taiwan will annotate such a character with bopomofo. Japanese newspapers often replace obscure characters with katakana instead, as is accepted practice in Japanese style guides.

There are also extremely stroke-rich characters, which tend to be rare. A notable example is 𪚥 (zhé; 'verbose'), which fell out of use by the end of the 5th century, containing 64 strokes. This character may not necessarily be seen as the most complex or difficult, as it simply requires writing the 16-stroke character 龍 (lóng; 'dragon') four times within the space allotted for one. Another 64-stroke character created in the same manner is 𠔻 (zhèng), composed of the character 興 (xīng, xìng; 'flourish') in quadruplicate.

One of the most complex characters found in modern Chinese dictionaries is 齉 (nàng; 'snuffle') with 36 strokes.^{[upper-alpha 13]} Other stroke-rich characters include the triplicated 靐 (bìng) with 39 strokes, and the quadruplicated 䨻 (bèng) with 52, both meaning 'the loud noise of thunder'—however, these are not commonly used. As an example, the most complex character that can be input with a representative IME^{[lower-alpha 19]} is 龘 (dá; 'appearance of a dragon in flight'). It is composed of the ⿓ 'DRAGON' radical in triplicate, having a total of 16 × 3 = 48 strokes. Among the most complex characters presently in common use are 籲 (yù; 'to implore') with 32 strokes, 鬱 (yù; 'luxuriant', 'lush', 'gloomy')—also the character in the jōyō kanji list having the most strokes, with 29—豔 (yàn; 'colourful') with 28, and 釁 (xìn; 'quarrel') with 25. Also occasionally in modern use is 鱻 (xiān; 'fresh'), a variant of 鮮 with 33 strokes.

In Japanese, an 84-stroke kokuji exists: , normally read taito. It is composed of the 'cloud' character 䨺 atop the aforementioned triple-'dragon' character, also possessing the meaning of 'appearance of a dragon in flight': it has readings おとど, otodo, たいと, taito, and だいと, daito.^[186]

• 
  Zhé, 'verbose'
• 
  Zhèng, meaning unknown
• 
  Nàng, 'snuffle'
• 
  Taito, 'appearance of a dragon in flight'
• 
  Alternative form of taito
• 
  Biáng, a kind of noodle from Shaanxi

Lect-specific variants

In addition, there are a number of 'dialect characters' (方言字; fāngyánzì) that are not generally used in formal written Chinese but represent colloquial terms in various spoken varieties of the language. In general, it is common practice to use standard characters to transcribe previously-unwritten words in Chinese dialects when obvious cognates exist. However, when no obvious cognate exists due to factors like irregular sound changes or semantic drift over time, or an origin in a non-Chinese language, like a substratum or loanword, then characters to transcribe it are borrowed according to the rebus principle, or invented in an ad hoc manner. These new characters are generally phono-semantic compounds, e.g. Min Nan 侬 ('person'), although there are examples of compound ideographs, e.g. northeast Mandarin 孬 ('bad').

There may be several ways to write a dialectal word—often, one that is etymologically correct, and one or several that are based on the word's pronunciation—e.g. the etymological 觸祭 versus the phonetic 戳鸡 (⁷tshoq¹ci) in Shanghainese, meaning 'eat'. Speakers of a dialect will generally recognise a dialectal word if it is transcribed according to pronunciation, while the etymologically correct form may be more difficult to recognise. For example, few Gan speakers would recognise the character 隑 as meaning 'to lean' in their dialect,^{[upper-alpha 14]} because this sense of the character is now archaic in Standard Mandarin.

As an exception, written Cantonese is widespread in Hong Kong, even for certain formal documents, due to the former British colonial administration's recognition of Cantonese for use for official purposes. In Taiwan, there is also a body of semi-official characters used to represent Taiwanese Hokkien and Hakka. An example of an Hakka vernacular character is 㓾 (cii¹¹, 'kill').^{[upper-alpha 15]} Other varieties of Chinese with a significant number of speakers—like Shanghainese Wu, Gan Chinese, and Sichuanese Mandarin—also have their own series of characters, but these are not often seen, except on advertising billboards directed toward locals and are not used in formal settings except to give precise transcriptions of witness statements in legal proceedings. Standard Chinese is the preferred written language within every region of mainland China.

Kanji

Kanji for 剣道, kendo, with equivalent Korean pronunciation kumdo and equivalent Standard Chinese pronunciation jiàndào^{[lower-alpha 20]}

The Japanese Ministry of Education has designated 2,136 jōyō kanji to be taught in primary and secondary school. Today, a well-educated Japanese person may know upwards of 3,500 characters.^[191] The kanji kentei tests a speaker's ability to read and write kanji. The highest level of the kanji kentei tests according to the full JIS X 0208 list, which includes over 6,000 kanji.^[192]

Hanja

The South Korean Basic Hanja for Educational Use is a set of 1,800 characters standardised in 1972, with the first 900 hanja taught to middle school students, and the rest taught to high school students.^[155]

In March 1991, the Supreme Court of Korea published the 2,854-character Table of Hanja for Use in Personal Names.^[193] The list expanded gradually: by 2015 there were 8,142 hanja, including the set of basic hanja, permitted for use in Korean names.^[194]

Large character lists

Ostensibly, Chinese characters can be created and used arbitrarily, though they are unlikely to gain widespread use or inclusion in official character sets.^[195] Counting the entries within major Chinese dictionaries is a viable means of estimating the growth of the character inventory over time.

Estimates of the total number of characters in modern use can be sourced from encoding schemes and dictionaries:^[196] according to sources from mainland China, Taiwan, Hong Kong, Japan, and Korea, this number is likely around 15,000.^[197] For comparison, Unicode encodes over 90,000 CJK Unified Ideographs.^[198]

Even the Zhonghua Zihai does not include characters in the Chinese family of scripts created to represent non-Chinese languages, except the unique characters in use in Japan and Korea. Characters formed by Chinese principles in other languages include the roughly 1,500 Japanese-made kokuji given in the Kokuji no Jiten,^[214] the Korean gukja, the over 10,000 sawndip characters still in use in Guangxi, and the almost 20,000 Nôm characters historically used in Vietnam.^[215] More divergent descendants include the Tangut script, which created over 5,000 characters with visually similar strokes to Chinese characters, but different principles of formation.

Modified radicals and new variants are two common reasons for the ever-increasing number of characters. There are about 300 radicals, with 100 being in common use. Creating a new character by modifying the radical is an easy way to disambiguate homographs among picto-phonetic compounds (xíngshēngzì). This practice began long before the Qin standardisation of Chinese script. The third-person personal pronoun 他 (tā) written with the ⼈ 'MAN' radical, traditionally used regardless of the target's gender or animacy, illustrates modifying signifiers in order to form new characters. In modern written Chinese, further graphical distinctions have been made between 她 ('she') with the ⼥ 'WOMAN' radical, 牠 ('it') with the 牜 'COW' radical, 它 ('it') with the ⼧ 'ROOF' radical, and 祂 ('He') with the 礻 'SPIRIT' radical, though all are pronounced identically as tā. One consequence of modifying radicals is the fossilisation of rare and obscure variant logographs, some of which are not even found in Classical Chinese texts. For instance, 和 (hé; 'harmony', 'peace'), which combines the ⽲ 'GRAIN' radical with the ⼝ 'MOUTH' radical, has variants 咊 with the components' positions reversed, and 龢 with ⿕ 'FLUTE' replacing ⼝ 'MOUTH'.

Citations

Works cited

 This article incorporates text from Chinese Recorder and Missionary Journal, vol. 3, a publication from 1871, now in the public domain in the United States.

Works of historical interest

Online references and databases

• Unihan Database – Official Unicode site on Chinese characters and Han unification, with reference glyphs, readings, and meanings for all characters encoded in the standard
• Chinese Text Project Dictionary – Comprehensive character dictionary, including data for all Chinese characters within Unicode, and exemplary examples of use in Classical Chinese texts
• zi.tools – Character lookup by component description, character etymology, phonology, orthography, and dictionary
• Chinese Etymology by Richard Sears
• Chinese text computing – Statistics regarding the use of Chinese characters, by Jun Da

Character history and construction

• Excerpt from Visible Speech: The Diverse Oneness of Writing Systems by John DeFrancis, published in 1989 by University of Hawai‘i Press, reproduced with permission
• Evolution of Chinese Characters at Omniglot