If your script is in latin-1, decode('iso-8859-1',$string) will work too. As far as I know decode() will always upgrade to utf8 (or ascii, which is a byte-compatible subset of utf-8)

If $string = 'ρα'; is a literal in a utf-8 encoded script, you should use the utf8 pragma to set the utf-8 markers correctly on literals. And then decode('iso-8859-1') probably won't work correctly on it. But utf8::upgrade() will still work.

By default, there is a fundamental asymmetry in Perl's unicode model: implicit upgrading from byte strings to Unicode strings assumes that they were encoded in *ISO 8859-1 (Latin-1)*, but Unicode strings are downgraded with UTF-8 encoding. This happens because the first 256 codepoints in Unicode happens to agree with Latin-1.

I don't know what "Unicode strings are downgraded with UTF-8 encoding" means. Also the line below that paragraph in perlunicode says

If you wish to interpret byte strings as UTF-8 instead, use the "encod
+ing" pragma:

      use encoding 'utf8';
[download]

Don't believe it. You should use utf8; instead. use encoding 'utf8' is broken.