FWIW the Charset I think affected is (untested pseudo code) :

Locale.setDefault(new Locale("tr_TR"));
Characterr.valueOf('L').toLowerCase() != 'l';
Characterr.valueOf('L').toLowerCase() == '\u0131';
// but with Locale.ENGLISH we still get ASCII output ?
Character.valueOf('L').toLowerCase(Locale.ENGLISH) == 'l'; // untested claim

Charset charset = Charset.forName("ISO-8859-9");