History of Writing
Japanese writing, employing a combination of kanji and kana, originated with the adoption and adaptation of Chinese characters in the 5th-7th centuries CE, to be applied to the Japanese (spoken) language, a language with completely separate origins.
Though a massive portion of Japanese vocabulary today derives from Chinese terms, and adaptation of Chinese pronunciation (e.g. kanji for the Chinese hanzi 漢字), Japanese grammar, word order, and sentence structure differs entirely, and initially at least the nuanced connotations and meanings of words were different, with characters being chosen in a great many cases for concepts (words) that were not perfectly comparable. Chinese characters were also used solely for their sounds, in order to represent native Japanese words (see man'yôgana); these later evolved into hiragana and katakana - two syllabaries of phonetic symbols that represent only sounds and not inherent meanings.
Over the centuries, Japanese writing of course took myriad forms running the full gamut from works solely in kana to those solely in kanji, being used to produce everything from poetry and literature, to letters and other forms of communications, to formal records. Forms using kanji exclusively, or almost exclusively, are known as kanbun, while other forms are known as wabun.
Writing was done chiefly, if not exclusively, with brushes for many years, until the advent of the pen in modern times. Pencils using lead or red ochre were imported from the Dutch East India Company for a time, and were treasured by the shogunate and daimyô as exceptionally rare luxury items; graphite pencils came into use in Europe only first beginning in the 17th century, and did not become common in Japan until the late 19th.
Woodblock printing in the Edo period (17th-19th centuries) became a major mode of publication of written materials; though Japan briefly experimented with movable type in the late 16th century, it was only in the late 19th century that moveable type re-emerged and replaced woodblock as the chief mode of publication. The advent of moveable type, along with the shift from brushes to pens/pencils, brought about significant changes in the aesthetic forms of Japanese characters, and of Japanese writing as a whole. Pen & pencil produce much sharper lines, with sharper contours and consistent thickness of line; moveable type and, later, typewriters and modern electronic printing & displays, allowed for the language to become considerably standardized, such that the flowing and cursive forms of the past have now become relatively stark and standard in form.
The earliest object with written characters on it to be uncovered in Japan dates all the way back to 57 CE. This is a golden seal granted to the Japanese state of Na by Emperor Guangwu of the Han Dynasty. Inkstones dating to the Yayoi period have also been found, suggesting that writing was being practiced by at least some people, in some regions of the Japanese islands, far earlier than was previously believed. However, research on those inkstones and their implications remains in very early stages. According to the standard historical understanding, writing was not believed to have been truly adopted by Japanese elites in any significant way until the 5th century CE, and it was only beginning in the 7th century that materials began to be read and written more widely by the capital elites, including Buddhist and Confucian texts, and other Chinese materials.
The Nihon shoki and Kojiki suggest that figures such as Wani or Achiki came to Japan from Paekche during the reign of Emperor Ôjin, and introduced the Analects of Confucius and other documents at that time. Though the dating of the events of the Nihon shoki and Kojiki would place this in the 3rd-4th centuries, scholars today believe that if the introduction of the Analects and the Thousand Character Classic by Wani and Achiki actually took place, it more likely took place in the early 5th century.
Outside of the golden seal of Na, another of the oldest objects found in Japan bearing inscriptions is a 4th century seven-branched sword, associated with Isonokami Shrine in Tenri City, Nara prefecture, and believed to have been produced in Korea. It is associated with the myths of Empress Jingû, who according to the Nihon shoki is said to have invaded Korea in 364, and to have been given a seven-branched sword and a seven-branched mirror by the King of Paekche in 372. Indeed, the sword associated with Isonokami Shrine is inscribed with a date corresponding to 369 in the era names of China's Eastern Jin Dynasty; the Jin are believed to have given the sword to Paekche as a gift.
The oldest objects with written inscriptions believed to have been made in Japan include a sword found in the Eta Funayama kofun, dating to the 5th century, a sword found in the Inariyama kofun, dating to 471, and a mirror dating to 443 or 503, associated with Sumida Hachiman Shrine. The Inariyama sword, found in the Inariyama kofun in Gyôda City, Saitama prefecture, has 57 characters on one side, and 58 on the other. It is believed to date to the reign of Emperor Yûryaku. The mirror, meanwhile, associated with Sumida Hachiman Shrine in Sumida-machi, Hashimoto City, Wakayama prefecture, has 48 characters inscribed upon it, and is believed to have been made in Japan by an immigrant from Paekche. The date is given according to the sexegenary cycle, and could correspond to either 443 or 503 CE.
Buddhist texts are believed to have been first conveyed to Japan from Paekche in the 6th century. It is believed that it was sometime around this time, in the 5th-6th centuries, that Japanese first began to adopt, and adapt, Chinese characters in earnest. Characters were employed, sometimes for their meaning, sometimes just for their sound, in order to represent native Japanese words (yamato kotoba). This was the beginning of what is known as man'yôgana. One of the simplest examples of this can be seen in the character 「安」, meaning "peaceful" or "calm" in Chinese. The Japanese already had a word for this concept, yasuraka, and began using this character to represent that word. They also took from it its Chinese pronunciation, whatever that may have been at the time (today, in modern Mandarin, it's pronounced ān), so 安 can be pronounced in Japanese as either yasu- or an. But, at this early stage, the Japanese also used the character purely for its sound, not for its meaning, simply to represent the sound 'a'. In some of the earliest documents, the full character 安 is used precisely in that way, merely to represent the sound 'a'. Over time, it later became simplified into the kana forms あ and ア, which today are phonetic characters which contain no inherent meaning, but only convey the sound 'a'.
Japanese continued to use kanji in this way, merely for their sound and not for meaning, all the way down until the Meiji period (late 19th century), albeit to a lesser and lesser extent over the centuries, as terminology became standardized, and kana came to be used ever more exclusively for representing sounds. Terms which continued to use kanji merely for their sounds and not for their meaning are known as employing ateji (当字). Placenames are a prominent example of this, with the characters used to write Nara (奈良), for example, having no particularly relevant meaning. Foreign names, places, and terms also continued to be spelled out in kanji (that is, in ateji) into the Meiji period, with 亜米利加 (a-me-ri-ka) as one such example. The Chinese used characters simply for their sounds sometimes as well, since ancient times, especially in adapting words from foreign languages, such as Buddhist terminology.
As kanji began to be applied to Japanese words, there were many cases in which a single Japanese word covered many different aspects of a single concept, represented already in China by different characters. Thus, for example, the Japanese word au, meaning "to meet" or "to come together," came to be represented by a whole series of characters including 会 (to meet up with people), 合 (to fit together), 逢 (to rendezvous with, romantically, often clandestinely), 遭 (to meet with disaster, to encounter difficulties), all pronounced au.
One of the oldest surviving inscriptions written in Japanese language is found on the statue of Yakushi Buddha in the Kondô of Hôryû-ji. It is written entirely in kanji but with distinctively Japanese turns of phrase, word order, and character choice. An inscription in similar style is found on a stone stele in Gunma prefecture, dated to 681, and considered the oldest stone inscription in Japan.
The Nihon shoki (720) and Kojiki (712) are generally considered the earliest surviving major-length works in Japanese. Other documents from the Shôsôin Imperial Repository of similar age should similar linguistic forms. While the Nihon shoki was written almost entirely in kanji, the Kojiki employed a more thoroughly indigenous (non-Sinic) form. Buddhist texts written in Chinese began to be notated, or re-written, in various ways at this time to become legible as Japanese, marking the beginning of some of the earliest forms of kundoku.
Katakana began to be used in combination with kanji in the 11th century, if not earlier. This took place in three different ways: (a) as furigana pronunciation guides alongside the characters in kanbun texts, (b) as okurigana, interspersed along with kanji to provide prepositions, particles, verb & adjective conjugations, and so forth in wabun texts, or (c) as the dominant mode of writing, with only a more limited distribution of kanji. One famous early example of this is seen in a famous 13th century handscroll manuscript copy of the Hôjôki, which uses katakana (and no hiragana) extensively throughout the text, and which most often serves today as the model for katsuji (modern type-printed) versions of the text.
In Heian and early medieval kanji-heavy texts, where kana were used directly in the text (i.e. not as furigana to the side of a kanji), they were generally written smaller, and to one side, within the column of characters. By the 13th century, however, they were being written full-size and centered within the column. To a certain extent, katakana came to be used to give a text a more Chinese feel, while hiragana was reserved for texts with a more Japanese feel, including collections of waka poetry and monogatari tales.
Up until the late Heian period, it was common for columns of characters to shift towards the right, as one moved down the page. Wherever a character ended towards the right side, the writer would often simply continue from there, beginning with the left side of the next character. Over the course of the 12th-13th centuries, however, it became more common for writers to return from the right side of the column to the left before starting the next character, resulting in straighter columns (which did not veer off to the right), and a pattern of slash-mark "returns" often appearing in between characters. This remained a common feature in medieval and early modern calligraphy, down to today.
Another change visible in calligraphy over the course of the 12th-13th centuries is a shift from more graceful, sweeping entrances and exits to the more deliberate, and halting, calligraphic mode still employed today, in which the brush is placed powerfully onto the page and held there for a moment before moving into the first stroke of a character, and similarly held for a moment before removing the brush from the page at the end of each character. Hiragana documents from the Heian period and earlier do not evidence this same kind of halting deliberation, but rather suggest that the calligrapher placed the brush onto the page with a sweeping motion, as he or she wrote the first stroke.
At the same time, the myriad cursive or calligraphic forms used to write the same character, or the same sound, settled down in the Muromachi period into a relatively standard set of forms which continued to be used through the Edo period, and which remain standard among calligraphers today. This means that hentaigana or kuzushiji dictionaries, listing the half dozen or so typical ways of writing any given character, can be used effectively to decipher most documents from the late medieval and early modern periods, and also that variation in character style can be used to date, with some accuracy, documents from earlier periods.
In the realm of kanbun, Keian Genju (1427-1508), founder of the Satsunan school of Japanese Neo-Confucianism, developed in the 15th century his own system of kundoku notation for reading classical Chinese as Japanese. This system, called Keian-ten, was later adapted by Satsunan leader Nanpo Bunshi (1555-1620), whose Bunshi-ten texts were then distributed more widely by his student Tomari Jochiku (1570-1655), representing perhaps a significant antecedent to the modern systems of marking kanbun for Japanese readers.
While hiragana are standardized today into a single form for each kana, e.g. あ being the only hiragana character for the sound "a", up until the Meiji period there was no singular standard character. While many of the kana standard today were in use in the pre-modern period, they were used alongside calligraphically abbreviated forms of a number of other kanji as well, with no particular preference given to the kana we now take as standard. For example, と, the standard kana for the sound "to" today, is derived from an abbreviation of the kanji 止 (tomaru, "to stop"). But in pre-modern and early modern texts, this と is used interchangeably with calligraphically abbreviated forms of the kanji 登、東、斗、度、土、and 刀. It was only from the Meiji period onwards that と developed any special prominence over these other ways of conveying the sound "to."
Hiragana was used extensively up through the early modern period, as it still is today, for particles, okurigana, and furigana; however, up through the Meiji period, it was quite common to use katakana for these purposes as well. Today, by contrast, katakana is reserved largely for foreign words and onomatopoeia. One of the few exceptions is seen in the use of the katakana "ke" (ケ) to denote the sound "ka" when counting, or in certain placenames, where it derives from the use of "ga" as a possessive particle, for example in the placename Ichigaya 市ヶ谷 (lit. "valley of the market"), or in the phrase sankagetsu 三ヶ月 ([a length/duration of] three months).
In the 1940s, the government undertook two major spelling reforms. They established new, simplified standard forms, known as shinjitai ("new character forms") for many of the kanji, eliminating the old character forms (kyûjitai) from standard usage. To give just a few examples, the characters for "country" (kuni), "etiquette" (rei), and "body" (karada) changed from 國、禮、and 體 to 国、礼、and 体, respectively. They also eliminated a few kana, and their corresponding sounds, from the language, including the ye (ゑ or ヱ, as in Yedo, now Edo, though still used by Yebisu Beer), wi (ゐor ヰ, still used today by Nikka Whiskey), and kwa and gwa (as in Kwannon and Hongwan-ji, now Kannon and Hongan-ji).
- Nakashima Takashi, Ogawa Yasuhiko, Unno Keisuke, lectures, Wahon Literacies symposium/workshop, UCLA & UC Santa Barbara, 31 Aug to 4 Sept, 2015.
- Cynthia Viallé, "In Aid of Trade: Dutch Gift-Giving in Tokugawa Japan," Tokyo daigaku shiryôhensanjo kenkyû kiyô 16 (2006), 75n43.
- Nakamura Shunsuke, "2,000-year-old tool offers new proof of Japan’s earliest writing," Asahi Shimbun, 14 Nov 2017.
- Takatsu Takashi, “Ming Jianyang Prints and the Spread of the Teachings of Zhu Xi to Japan and the Ryukyu Kingdom in the Seventeenth Century,” in Angela Schottenhammer (ed.), The East Asian Mediterranean: Maritime Crossroads of Culture, Harrassowitz Verlag, 2008. 255-260.
- Kasama eiin sôkan kangyôkai, Jiten kana: shahon wo yomu tanoshimi 字典かな～写本をよむ楽しみ, Kasama shoin, 2010, 33-34.