Most Japanese characters are made up of a part that has a relevant meaning for the character in question (the ‘signific’) and a part that has a relevant pronunciation (the ‘phonetic’). Characters that are made that way are called 形声文字 keisei moji in Japanese¹ and might account for as much as 90% of all the characters. In practice the percentage of 形声 keisei² characters that one encounters is less that 90% because some of those outside that group are more frequent. However, most characters do contain a phonetic element, and it turns out it is possible to fool around with these phonetics quite a bit.
Both in China and in Japan there have been attempts at simplifying characters, and especially the part that plays a phonetic role has been targeted for that purpose. The idea behind it is really simple: if the symbol that a given character uses as a phonetic is rather complicated, why not use another, simpler phonetic that has the same pronunciation? In fact this has been done for millennia, in both an ad hoc fashion and more systematic ways, and occasionally canonized by the authorities. After the second world war both in Japan and in the PRC a number of characters have been officially simplified using this method.
In Japan simplified forms are officially limited to characters that are taught in primary and secondary school, but the public is free to apply rules of simplification more broadly if they want to, and in fact Asahi Shimbun (a large newspaper) has done so in their publications systematically. But what I find really interesting is that new ways of simplifying characters keep on emerging spontaneously, and are not limited to using only the traditional forms and abbreviations of Chinese characters.
In Japan the Latin alphabet is used together with Chinese characters and the national syllabics scripts. Especially in abbreviations Latin characters are quite common, and get a pronunciation that is derived from English, but modified to suit the phonetic structure of Japanese. See for example ＯＬ ōeru, which is a common word for a female office worker (form English ‘office lady’) or ＧＷ jīdaburyū, which is an abbreviation for ゴールデンウイーク Gōruden Uīku (‘Golden Week’, a holliday period in Japan). The interesting thing is that in hand writing, the individual Latin letters can be used as substitutions for complex phonetic parts of characters just as traditional abbreviations. For example, 慶應 Keiō is the name of an educational institution. Since the Japanese pronunciation of ‘Ｋ’ is kei [ke:] the character 慶 can be simplified as 广 + Ｋ and likewise 應 can be simplified as 广 + Ｏ. If your browser supports it, that would look like: 广K广O.
To be fair, this idea of mixing Latin characters with Chinese characters probably got inspired by earlier usage whereby letters from the national syllabic scripts (仮名 kana) where used as substitutes for phonetics. For example, 魔 or 摩 would then be written as 广+マ (with マ being the syllabic sign for ma). Simplifications like these surely baffle Chinese, since the kana symbols are purely Japanese. However, even in the official simplifications one can find modifications that are incompatible with Chinese. For example, 圍 (i) has been simplified to 囲, using the character 井 as a simpler phonetic. However, for that the Japanese assume the Japanese pronunciation of 井 (which is i in Japanese) not its Chinese (sino-japanese) pronunciation (which is jō, as in 天井 tenjō ‘ceiling’). In Chinese 圍 and 井 are pronounced very differently (wéi versus jǐng) therefore 囲 might not constitute a completely unambiguous simplification outside of Japan.
1. 形声文字 keisei moji are similarly but shorter called 形聲字 or 形声字 (xíngshēngzì) in Chinese.
2. The English equivalent for this word varies a lot; some possibilties are: phono-semantic, semantic-phonetic, semasio-phonetic, phonetic-ideographic, signific-phonetic.
I think to many Japanese the 井 represents ’[w]i’ which was the Man’yogana character representing that phonetic sound (a 借訓仮名＝Shakkun kana (based on a Japanese reading)), which preceded and was converted into the now-unused Katakana ヰ([w]i)...
The native Japanese word for ’well’ is ’ido’... perhaps its use as a man’yogana originally derived from the fact that it meant ’well’ in Chinese 井 and ’well’ was ’i / ido’ in Japanese... and perhaps more people recognised that character for ’well’ and read it as ’i’ than people who would dig out the Chinese word for such a common daily word... ”ido he ikitekuru” (before they clenched it to ”ido ni ittekuru”)...
I think any kanji converted to 形声文字 based on Shakkun kana 借訓仮名 would not be recognisable to Chinese readers. Only kanji abbreviated using Shakuon kana (借音仮名)... but even then it may not be clear, because on’yomi also have their divisions depending on when it was imported from China (呉音/漢音/唐音/慣用音) and what was read as ’gyou’ (e.g. 行), when imported in the 6th century, might have been read as ’an’ when imported in the 10th-11th century. (now read as ’hang’ or ’xíng’ in Chinese...?).
Questions come to mind.. are there any 形声文字 that use characters outside of the Man’yogana? Is the use of katakana or hiragana or the use of man’yogana a useful measure of when these kanji were abbreviated?
There’s a chart here of all the characters that formed the katakana (see History in Wikipedia: Katakana) and a more comprehensive chart of Man’yogana on Wikipedia.
I say thanks.. for taking my mind off -na adjectives for a few hours... :-)
Best regards
Oliver
