[Journal of the Simplified Spelling Society, J19, 1995/2, p27-29.]

The 20th Century Japanese Writing System: Reform and Change.

Christopher Seeley.

Dr Seeley is Head of the Department of Asian Languages at the University of Canterbury, New Zealand. He has written articles on the Japanese writing system, and a book on the history of the Japanese script from the earliest times down to the present (see 'Further Reading' below). His research also encompasses the history of the Chinese script in China.

Explanatory note.

The Chinese script was originally adopted in Japan early in the first millennium AD because at that period the Japanese had no writing system of their own. Modern Japanese is written predominantly in a combination of Chinese characters or kanji (about 3,000-3,500 in texts of a general nature such as newspapers), and kana (Japanese syllabic signs [about 92 in total in two syllabaries], each such sign representing a short syllable of Japanese such as a or ki), with an admixture in some texts of Roman letters and/or Arabic numerals.

There is significant fluctuation in usage, but Chinese characters are used to represent in writing many nouns and verb and adjective stems, while kana (hiragana - 'flowing' in appearance) are used for inflectional suffixes, grammatical particles, some nouns, etc. A second variety of kana (katakana - angular in appearance) is used for representing Western loanwords (largely from English), for onomatopoeic expressions, etc. Thus, for instance, in the Japanese sentence

Example of a japanese sentence

(Ano hito wa senkyuuhyakunanajuunen kara Igirisu ni sunde imasu 'That person has been living in Britain since 1970'), the noun hito 'person' and the verb stem sum- 'live' are typically written in kanji characters, the grammatical particle wa and the -te imasu verb form are written in hiragana, as also the demonstrative 'that', while the loanword Igirisu is represented in katakana.


In considering simplified Chinese characters in China and Japan, one needs to distinguish between non-centralized or informal use on the one hand, and centralized official adoption of such characters on the other. In China, simplified characters have been used in some texts in an informal way for several thousand years, though it was not until the 1950s that the Mainland China government moved towards adoption of such characters. In Japan, examples of variant characters, including simplified forms, can be found in texts dating from about the 5th century AD onward. Dictionaries featuring numerous simplified forms were compiled in China from the 8th century onwards; in Japan, many such dictionaries were compiled during the Edo period (ca, 1600-1868). In Japan, as in China, the force of conservative tradition was such that simplified characters were not officially adopted until about the middle of the 20th century (1946 onwards, to be precise).

In this article, we will focus on attempts at reform, actual reform, and change in Japan affecting Chinese characters and kana signs from about 1900 onwards.

The period 1900-45.

In Japan of the pre-modern period (before the Meiji Restoration of 1868), writing was invested with values which meant that it was perceived as far more than a mere utilitarian tool. In order to read and write anything more than the simplest and most basic text, one needed a knowledge of many thousands of Chinese characters. To make matters worse, the written language was typically not just a visual rendition of everyday spoken Japanese, but a convoluted entity which reflected the influence of Chinese linguistic forms and older Japanese forms. As if this were not bad enough, it was necessary - other than in informal documents - to write using the often more intricate orthodox forms of characters rather than simplified equivalents, though perhaps it should be noted that this complexity of structure was compensated for to a small extent in the Edo period through the extensive use of a cursivized way (o-ie-ryuu) of writing Chinese characters which meant that some of the strokes which were separate when the characters were written in the slow and formal way ('block script' or 'model script') became joined up or even omitted.

Even from the time before Japan opened its doors to the full influence of the West in 1868, some Japanese intellectuals could not help but note the simplicity of the Roman alphabet compared with the cumbersome writing system for Japanese and Chinese. There were progressive individuals who argued for script simplification from a very early date, e.g. the noted educationalist Fukuzawa Yukichi (1834-190l), but they had to fight an uphill battle against the traditionalists, who saw writing in general and Chinese characters in particular as a kind of intellectual showcase for exhibiting erudition on the part of the writer.

From about 1900 onwards, script reform became a subject of often intense debate among Japanese intellectuals. Discussion and proposals regarding reform, and reform itself as it eventuated from 1946 onwards, focussed on the two areas of the number of different characters for general use, and the shape (form) of individual characters.

When it came to defending the case for retention of use of a very large range of characters and for adhering to orthodox (non-simplified) forms, the Japanese military establishment had been a staunch advocate. Ironically, though, during the Pacific War circumstances led the Army Ministry to implement restrictions on the number of different characters used in certain types of texts. The motivation for this was that the educational level of new recruits had fallen to the point where they were not able to read properly the instructions on weapon-related texts, a situation which resulted in various accidents involving weapons.

The period 1945 onwards.

Following the defeat of 1945, the Japanese nation was receptive to change in various areas relating to daily life, including the writing system.

It appears that the Japanese themselves probably made the decisions at this time regarding modifications to their writing system, and that such changes were not imposed on them by the Occupation Forces, as has been suggested. The eventual romanization of Japanese texts appears to have been considered after the War, this being reflected in, for instance, an increase in the number of romanized primary school textbooks around that time. However, this option for reform was overtaken by the more readily acceptable simplification of the conventional writing system.

The main focus of the early postwar orthographic reforms was on reducing the number of different characters in daily use. To this end, a list of 1295 characters was drawn up, but this was considered too restrictive for general use, yet too extensive for teaching during the period of compulsory education. Consequently, a new list was compiled, consisting of 1850 characters. Known as the Tooyoo kanji (Interim Use Characters) List, this group of characters was to form the backbone of the new simplified orthography for Japanese for the next 35 years. Following promulgation of the Tooyoo kanji (TK) List in November 1946, it was adopted by government offices and in school textbooks. The TK List has been misunderstood by some as a character list which set limits not only as noted above, but which also aimed to restrict the range of characters used by individuals; however, this was not the case. If an individual such as a writer wrote an essay which was selected for inclusion in a primary school textbook, then the orthography of the text selected was modified so as to employ only characters in the TK List, words written in non-TK characters typically being rewritten in kana; but if the same writer was writing a letter to a friend, then he/she was free to use any Chinese characters whether they were in the TK List or not.

The TK List of 1946 contributed substantially towards making the Japanese writing system less complicated in everyday life, including education. The List's success was made possible in part by selective but quite extensive modifications to language and script - for instance, many scientific and technical terms were changed so that it was easier to write them using less obscure characters, or alternatively in other cases the same term was retained but was then represented in writing using more common characters. For the modern generation, of course, the revised forms simply represent the norm rather than a recent revision which deviates from an earlier standard.

Despite some movement in the direction of simplification, there were still several aspects of the TK List which merited attention, not least the form of characters. As a result of the 1946 reform, a number of simplified forms had been officially adopted in place of the more complicated orthodox equivalents, for instance
However in the 1946 list many other orthodox forms of complicated structure still remained, for instance
Artistic accomplishment.
In order to help the simplification process further, in 1949 a substantial number of additional simplified characters were officially adopted, replacing their orthodox equivalents in the TK List of 1946.

Taking an overall view, the TK List appears to have been accepted by the majority as not being excessively restrictive. However, there was one area in which the TK List did impinge on the individual's use of characters in a specific way, namely official registers of births. Until the TK List was put into effect, in choosing how to write the names of newborn children, parents had been free to choose any characters (the largest dictionary of Chinese characters published in Japan contains about 48,000) and could assign to them any reading (way of pronunciation), however bizarre or inexplicable in terms of the conventional readings for the characters concerned. This practice gave rise to considerable inconvenience in everyday life, and so to overcome this from November 1946, in deciding on how to write the names of newborn children, parents were restricted to selecting from within the 1850 characters in the TK List (alternatively, kana could be used). This abrupt change was seen by many as an excessive restriction on the rights of the individual, and so to counteract such criticism an additional 92 characters were selected and permitted for use from 1951 specifically for writing given names, over and above the TK List characters.

Apart from the sheer number of Chinese characters in use and the complexity of shape of many of them, another source of difficulty with the prewar writing system had been the fact that a given character often had a large number of words or morphemes (conventionally known as 'readings') associated with it. The character The character for respond.) for instance had readings which included ō 'respond', kotaeru (id.), irae '(an) answer', and masa ni 'on the point of', the reading to employ in a particular case being determined by context. This aspect of Japanese orthography was simplified in 1948, at least in some texts, when a comparatively restricted list of officially approved readings for TK List characters was put into effect in government publications, etc.

Changes to usage of kana (1900 onwards).

The focus of the above has been firmly on Chinese characters, but mention needs to be made also of the kana syllabaries. Traditionally, the usage for kana was based on historical principles. Thus, for example, in early documents written in kana, the form corresponding to the modern Japanese verb iru 'to be' is written The character for 'to be' reflecting Old Japanese pronunciation wiru. Similarly, the word for 'face' (modern Japanese kao) is written The character for face. in early kana texts, reflecting the Old Japanese pronunciation kaFo. In principle, historically correct kana spellings such as those noted above were taken as the basis for historical systems of kana usage, though some spellings (a small proportion) that were proposed were in fact incorrect in terms of the historical principle.

Historical kana usage is in fact rather more convoluted than the above examples might suggest. This is because there are in Japanese a number of syllables for which there are not just two alternative kana spellings but up to five or six (!), the question of the 'correct' spelling being linked to the particular word or morpheme concerned, rather like I and eye in English. A good example of this is the syllable ō which in historical kana spelling was written differently in each case, depending on whether it meant 'parrot', 'concave', 'respond', 'press', 'ruler', or 'old man'.

Around the beginning of this century, such orthographic complexity understandably caused schoolchildren serious difficulty, especially in early years of primary school, because they had as yet learned only a small number of kanji. In consequence the children used a high proportion of kana (many to write words which in texts for adults would normally be written in kanji). To help overcome this difficulty, a pronunciation-based form of kana spelling was put into effect in primary school texts, on a restricted basis, for a few years from about 1900. There was, though, an inherent weakness in this limited reform, in that it was only applied to words belonging to a certain layer (Sino-Japanese) of the vocabulary, not to the vocabulary as a whole. This subtlety was difficult for young pupils to grasp, and in the face also of strong opposition from traditionalists, in 1908 the Japanese Education Ministry abandoned this attempt at pronunciation-based kana spelling.

In 1946, in the new political and social climate after the War, the historical kana usage was replaced by a simpler, pronunciation-orientated type of kana spelling at the same time as the TK kanji reforms came in.

The 1980s and 1990s.

In 1981, the character list that had held sway since 1946 (the TK List) was replaced by another list, the Jooyoo kanji (General Use Chinese Characters) List (JK List) after a lengthy process of deliberation and consultation.

The JK List was a modestly expanded version of the earlier list, containing all 1850 TK List characters together with a further 95. These had been selected on the basis of their usefulness for representing words which had become more prominent in everyday life in Japan since 1946. An important point to note about the new List is that it was intended to be less prescriptive, being essentially a 'guide' to the principal characters for use in the general life of society. This 'guide' status was the subject of much controversy, with opposition from those who feared that the new List would lead to a sudden increase in the range of characters used.

In fact, from about the mid 1980s on there does appear to have emerged something of a trend towards use of a greater number of different characters, though this has been brought about not so much by the replacement of the TK List by the more liberal JK List, but through changes in writing technology in Japan.

Until the late 1970s, using a typewriter for Japanese had meant employing a manual machine. This was of necessity very cumbersome and slow because of the need to handle a large character set (over 2,000 characters) plus kana, etc. Round about 1978, though, a technological development took place that was a milestone in writing technology - the invention of a word-processor for Japanese. Now, at the touch of a button, the user had access to more characters than on manual typewriters for Japanese, and could print out Japanese text in the conventional character-kana orthography much more quickly than in the past.

By about 1984, word-processors were being marketed as an affordable and convenient tool for the ordinary consumer, and since then their use has become very widespread. It is precisely because of the ease of use of word-processors for Japanese that users have tended to end up using more Chinese characters than if writing the same text by hand, since all that is needed to produce a Japanese text, at the level of characters, is passive recognition of the characters concerned, not active recall.

Since word-processors for Japanese have encoded in them about 6,300 characters (the characters having been determined in the late 1970s as a Japan Industrial Standard), the use of the word-processor in Japan is something which has major implications for education and script use in society in general. The changes which the word-processor has brought to Japanese writing habits have yet to be reconciled to Japanese script education in the schools. Change could well take the form of establishing a smaller set of common Chinese characters for active acquisition, and a second set of less widely-used characters for passive acquisition. Whatever steps are taken, change in this area represents an urgent task for the Education Ministry and other groups, and needs to be attended to without further delay.

Further reading.

Gottlieb, Nanette (1993) 'Written Japanese and the Word Processor' in Japan Forum Vol.5, No.1 (April) pp115-32.

Seeley, Christopher (1991) A History of Writing in Japan, Leiden: E J Brill.

Twine, Nanette (1991) Language and the Modern State: the reform of written Japanese, London/New York: Routledge.

Back to the top.