diff --git a/Lib/email/charset.py b/Lib/email/charset.py
index ad56c58..dddaa76 100644
--- a/Lib/email/charset.py
+++ b/Lib/email/charset.py
@@ -209,7 +209,7 @@
                 input_charset = unicode(input_charset, 'ascii')
         except UnicodeError:
             raise errors.CharsetError(input_charset)
-        input_charset = input_charset.lower()
+        input_charset = input_charset.lower().encode('ascii')
         # Set the input charset after filtering through the aliases and/or codecs
         if not (input_charset in ALIASES or input_charset in CHARSETS):
             try:
diff --git a/Lib/email/test/test_email.py b/Lib/email/test/test_email.py
index 358b6a7..4aac500 100644
--- a/Lib/email/test/test_email.py
+++ b/Lib/email/test/test_email.py
@@ -3140,6 +3140,13 @@
             'attachment; filename*="iso-8859-1\'\'Fu%DFballer.ppt"',
             msg['Content-Disposition'])
 
+    def test_encode_unaliased_charset(self):
+        # Issue 1379416: when the charset has no output conversion,
+        # output was accidentally getting coerced to unicode.
+        res = Header('abc','iso-8859-2').encode()
+        self.assertEqual(res, '=?iso-8859-2?q?abc?=')
+        self.assertIsInstance(res, str)
+
 
 # Test RFC 2231 header parameters (en/de)coding
 class TestRFC2231(TestEmailBase):