XHTML Validation

From what I know a XHTML document without a XML declaration can only be
UTF-8 or UTF-16.  So perhaps the parser should test for and assume one of
these encodings if the encoding is unspecified.  Also if the previous
statement is true perhaps it should be considered an error or a warning if a
XHTML document without declaration is declared to be something other then
UTF-8 or UTF-16 by other means.