Using Kanji characters

To permit the use of Kanji characters in atoms, variable names, predicate names, messages and comments you need to set the environment variable QUINTUS_KANJI_FLAG to true before starting Prolog.

When the Kanji flag is set, an 8-bit character codes (that is, a character code between 128 and 255) is assumed to be part of a multi-byte sequence representing a Kanji character. Such characters are treated, for the purposes of Prolog's syntax, as if they were lower-case alphabetic characters. The Kanji flag does not affect character I/O, or the conversion between atoms and lists of character codes using name/2 or atom_chars/2. In the case of I/O, two (or more) character codes will need to be read or written for every Kanji character. Similarly, an atom made up of Kanji characters will be transformed by name/2 or atom_chars/2 into a list of twice (or more than twice) as many 8-bit character codes.

(Japanese Industry Standard) is the Japanese analogue to ASCII. This includes two standards: JIS-X0201, which defines half-size Katakana characters, represented in one byte and JIS-X0208 (called "JIS" code) that defines 6349 Kanji and 453 non-Kanji characters stored in two bytes, using the lower seven bits of each byte.
Is JIS code shifted so that the high bit is used (so it doesn't interfere with some MSDOS control codes and special characters).
Is Sun's Japanese Language Environment. For each codeset this defines exactly one possible character set:
          codeset                  byte(s) representation
          0:      ASCII           0xxxxxxx
          1:      JIS-X0208       1xxxxxxx 1xxxxxxx
          2:      JIS-X0201       10001110 1xxxxxxx
          3:      Gaiji           10001111 1xxxxxxx 1xxxxxxx

Codeset 1 is JIS code with the hit bits set; Gaiji is a user-defined character set.

This is a standard used by Digital. All I know about this is that character codes 0-127 indicate a one byte ASCII character, and the codes 128-255 indicates a two byte Kanji character.