For the record, I did some hunting and came across a page which contains modified source files for Python which enable Japanese to be handled nicely by the string and token handlers. After re-compiling Python 1.52 using the substitute source files (stringobject.c, compile.c, tokenizer.c) I found that Python readily accepts Japanese characters. Upon sending a test message or three to my temporary mailing lists, I found that Mailman also processes the characters perfectly now. I am now a happy camper. ;-)

The page is listed below, though I warn you it is entirely in Japanese:

http://hp.vector.co.jp/authors/VA010598/python/pylha32.htm