It doesn't quite make sense to say "display the first 19 bytes", since characters are displayed, not bytes. Bytes are simply used to encode the characters. Do you mean "display the first 19 characters"?

The number of bytes that a character requires depends on the character encoding. In most encodings for Japanese characters, yes, 1 character will require more than one byte --but not necessarily 2. For example in UTF-8, some could take 3 bytes.

Are you referring to the fact that some Japanese characters are "wide" when displayed, and may take as much room as two Roman characters? That is a separate issue, and has nothing to do with encoding or bytes.

Why bother with bytes at all? If the String is Japanese, simply display the first 19 characters, and if not, display 38 -- is that what you really want?

String toDisplay = ...;
boolean stringIsJapanese = ...;
if (stringIsJapanese) {
  if (toDisplay.length() > 19) {
    toDisplay = toDisplay.substring(0, 19);
  }
} else {
  if (toDisplay.length() > 38) {
    toDisplay = toDisplay.substring(0, 38);
  }
}