💾 Archived View for hoseki.iensu.me › posts › one-or-two-dots.gmi captured on 2023-03-20 at 17:43:29. Gemini links have been rewritten to link to archived content
⬅️ Previous capture (2022-03-01)
-=-=-=-=-=-=-
Maybe it's an apparent case of Stockholm syndrome after spending a considerable part of my early twenties learning them, but I love Chinese characters. I love writing them, I love looking at them and I love reading about them. In [George Atsuji's lecture series] there's one lecture dedicated to [the contemporary use of kanji (Chinese characters) in Japan] (I also found a [newer version of the lecture]). What's great about the lecture is that it touches upon something I found very confusing when learning kanji, namely why the same radical differs between kanji.
Most kanji are phono-semantic compounds (形声文字) where one part of the character is called the /radical/ indicating a general semantic area and the another part indicating a phonetic value for the character. For instance:
I find it fascinating that so many of these phonetic indicators still work after more than 2000 years of language change and in both Japanese and Chinese.
The lecture mentioned above talks about the radical /Shin'nyou/ (辵部) indicating /road/ which usually appears as 辶, and the fact that you encounter it in two variants: one-dotted (道, 追, 込, 進) and two-dotted (邂, 逅, 邁). It appears that the reason for this is pretty interesting and has to do with the digitalization of the Japanese language. Historically the preferred way to write or print /shin'nyou/ has changed several times.
In the [Ganlu Zishu (干禄字書)] from the 8th century Tang dynasty all examples use the one-dotted variant. This book was used when preparing for the [Imperial examination (科挙)] so can be seen as fairly authoritative. Skip forward a 1000 years to 1716 and the first publication of the [Kangxi dictionary (康煕字典)] and suddenly all /shin'nyou/ appear as two-dotted. The Kangxi dictionary was used as reference by all countries using Chinese characters which meant that the two-dotted variant was used in print in China, Japan, Korea and Vietnam. All was well and good (kanji-wise) until Japan's defeat in 1945 and [the allied occupation of Japan] during which the allied forces wanted to punish but also transform Japan into a "modern nation" by issuing several social and cultural reforms. One of the goals of these reforms was to abolish the use of Chinese characters in Japan and replace it with either Latin characters or /kana/ (hiragana and katakana). Spoiler alert: they didn't succeed and Japan still uses one of the most complicated writing systems seen since Akkadian combing both Chinese characters, hiragana, katakana and "western" letters.
The idea was to incrementally limit the number of kanji publishers were allowed to use by first limiting them to 1850 /necessary kanji/ (当用漢字) and then over time decrease the number until only kana remained. Apart from restricting the allowed kanji, it also standardized some simplifications such as using the one-dotted /shin'nyou/ instead of the two-dotted one which had been the standard from the 18th century. As for words containing characters outside of the designated set publishers were advised to either use only kana, partially rewrite it in kana or replace the word with a synonym. For instance the word 探偵小説 (detective novel) was to be written either as たんていしょうせつ, 探てい小説 or replaced with 推理小説 reasoning or deduction novel. In the end the plan was scrapped and the set of /necessary kanji/ was renamed to /regular-use kanji/ (常用漢字) and expanded to 1945 characters. This is the set of Chinese characters taught in Japanese schools today. There is no restriction on what characters a publication can use, but it is customary to add pronunciation indicators to words containing characters not in the regular-use kanji set using so called furigana or ruby. Now for the messy part...
As all of this was taking place computers became more and more prolific and the need to use them for Japanese became pressing. The Japanese Industrial Standards committee (JIS) set out to develop an encoding system for Japanese and came up with the JIS encoding. The JIS encoding separates kanji into different levels (水準) depending on their occurrence frequency where the first level contains the 2965 most frequently used characters and the second level contains an additional 3384 characters. From the beginning all /shin'nyou/ kanji in level 2 and above use the two-dotted variant. The inconsistencies lies in the characters that belong to level 1, but are not part of the 1945 regular-use kanji. Some of these kanji, for example 辻 and 迂, can appear as either one or two-dotted depending on which version of the JIS encoding is used. All devices using a JIS encoding of 0213:2004 or later use the two-dotted variant, whereas earlier versions had kanjis of both types.
So it turns out that it really is an encoding issue, and not something pertinent to the character itself. When you learn kanji you tend to use your phone to look up characters and that will make you believe that this distinction is important when in reality it isn't. And it's not only /Shin'nyou/, the same goes for /shoku-hen/ 飠(eat) as well: 飲, 飯, 飾 vs 餌, 餅, 饅. George Atsuji's recommendation is to use the version you prefer when writing, since that has always been at the heart of Chinese character culture.
Wow, this post turned out a lot longer than I first intended. The reason I found this topic so interesting is that it shows how kanji and our perception of them are changing even during modern times. I also noticed that as I was typing this on Linux all examples in the text appear as one-dotted, I wonder if it's a font issue or an encoding issue... apparently this requires further investigation :)
George Atsuji's lecture series
the contemporary use of kanji (Chinese characters) in Japan
the allied occupation of Japan
イェンス - 2022-01-07