In all occasions (half-point, half-eye etc. in CN, JP, KR) where the editors wanted to print the character 半 (Chin.: bàn half) they wrote a wrong character. The first two (inclined) strokes are wrongly turned by 90 deg., resp. left and right strokes are exchanged.
I.e., the inclined characters look like in 尘 chén (dust / dirt / earth) while the vertical/horizontal strokes remain the same.The funny thing is, that such a character does not seem to exist in Chinese (and I suspect neither in Japanese, I checked “2001 Kanji” by De Roo). The question is, how do you produce a Kanji that does not exist? (not with Unicode, I guess).
Correction: I have several fantizi-jiantizi conversion tables.
None shows above character. Two 40 year old Chinese having had their university education in China (proving their literacy) told me that they would not know this character. However, a British, teaching Chinese history in Leiden told me that it does exist (although being archaic, obsolete?).
- Indeed my Taiwanese copy of Matthews’ Chinese-English dictionary of 1931 (sic) shows this character with the different inclinations (and in all other composites too).
Similarly, these two books show this particular form of the character:
- “Analysis of Chinese Characters” by G.D. Wilder & J.H. Ingram, Dover 1974, unabridged republication of 2nd ed. of College of Chinese Studies in China 1934, 1st ed. 1922 (sic) and
- "Chinese Characters - Their origin, etymology, history, classification and signification " by Dr. L. Wieger (or. French), Dover 1966, unabridged and unaltered republication of 2nd ed. publ. by Catholic Mission Press in 1927 (1st ed. 1915)
– tderz, 2005, SL; in discussion of the book Contemporary Go Terms