diff --git a/Lib/csv.py b/Lib/csv.py
index 0a627ba7a512fa8..c593e635aa5ca1e 100644
--- a/Lib/csv.py
+++ b/Lib/csv.py
@@ -282,6 +282,9 @@ def _guess_quote_and_delimiter(self, data, delimiters):
         """
         import re
 
+        # Remove lists from the data, see issue 119123
+        data = re.sub(r'\[.*?\]', '', data)
+
         matches = []
         for restr in (r'(?P<delim>[^\w\n"\'])(?P<space> ?)(?P<quote>["\']).*?(?P=quote)(?P=delim)', # ,".*?",
                       r'(?:^|\n)(?P<quote>["\']).*?(?P=quote)(?P<delim>[^\w\n"\'])(?P<space> ?)',   #  ".*?",
diff --git a/Lib/test/test_csv.py b/Lib/test/test_csv.py
index 9aace57633b0c64..c6340d8e1b70f3f 100644
--- a/Lib/test/test_csv.py
+++ b/Lib/test/test_csv.py
@@ -1331,11 +1331,22 @@ class TestSniffer(unittest.TestCase):
 ghi\0jkl
 """
 
+    sample15 = dedent("""\
+                        id;field1;group;sub_group
+                        1;False;['345'];['10']
+                        2;False;['112', '112'];['765', '766']
+                     """)
+
     def test_issue43625(self):
         sniffer = csv.Sniffer()
         self.assertTrue(sniffer.has_header(self.sample12))
         self.assertFalse(sniffer.has_header(self.sample13))
 
+    def test_issue119123(self):
+        sniffer = csv.Sniffer()
+        dialect = sniffer.sniff(self.sample15, ",;")
+        self.assertEqual(dialect.delimiter, ';')
+
     def test_has_header_strings(self):
         "More to document existing (unexpected?) behavior than anything else."
         sniffer = csv.Sniffer()