This is documentation for Mathematica 6, which was
based on an earlier version of the Wolfram Language.
View current documentation (Version 11.1)


specifies the default raw character encoding to use for input and output functions.
  • The setting $CharacterEncoding=None takes all special characters to be represented externally by printable ASCII sequences such as \[Name] and \:nnnn.
  • Typical examples include:
"AdobeStandard"Adobe standard PostScript font encoding
"ASCII"full ASCII, with control characters
"EUC"extended Unix code for Japanese
"ISOLatin1"ISO 8859-1 standard
"ISOLatin2"ISO 8859-2 standard
"ISOLatin3"ISO 8859-3 standard
"ISOLatin4"ISO 8859-4 standard
"ISOLatinCyrillic"ISO 8859-5 standard
"MacintoshRoman"Macintosh roman font encoding
"PrintableASCII"printable ASCII
"ShiftJIS"shift-JIS encoding of JIS X 0208-1990 and extensions
"Symbol"symbol font encoding
"Unicode"raw 16-bit Unicode (with +1 byte ordering)
"UTF8"Unicode transformation format
"WindowsANSI"Windows standard font encoding
"ZapfDingbats"Zapf dingbats font encoding
  • With $CharacterEncoding="encoding" characters that are included in the encoding can be input in their raw 8- or 16-bit form, and will be output in this form.
  • Unencoded characters can be input and will be output in standard \[Name] or \:nnnn form.
  • When using a text-based interface, resetting the value of $CharacterEncoding has an immediate effect on standard input and output in a Mathematica session.
  • When using the notebook front end, raw character encodings are normally handled automatically based on the fonts you use. Only raw 16-bit Unicode is ever sent through the MathLink connection to the kernel.
  • $CharacterEncoding affects the input and output of all characters, including those in symbol names and comments.
New in 3