diff --git a/ckan/logic/converters.py b/ckan/logic/converters.py
index 540ee5a07e6..18a2ad50619 100644
--- a/ckan/logic/converters.py
+++ b/ckan/logic/converters.py
@@ -1,18 +1,14 @@
 from ckan.lib.navl.dictization_functions import Invalid
 from ckan.lib.field_types import DateType, DateConvertError
-
-
+from ckan.logic.validators import tag_length_validator, tag_name_validator
 
 def convert_to_extras(key, data, errors, context):
-
     extras = data.get(('extras',), [])
     if not extras:
         data[('extras',)] = extras
-
     extras.append({'key': key[-1], 'value': data[key]})
 
 def convert_from_extras(key, data, errors, context):
-
     for data_key, data_value in data.iteritems():
         if (data_key[0] == 'extras'
             and data_key[-1] == 'key'
@@ -33,3 +29,34 @@ def date_to_form(value, context):
         raise Invalid(str(e))
     return value
 
+def convert_to_tags(vocab):
+    def callable(key, data, errors, context):
+        tag_string = data.get(key)
+        new_tags = [tag.strip() for tag in tag_string.split(',') if tag.strip()]
+
+        if not new_tags:
+            return
+
+        # get current number of tags
+        n = 0
+        for k in data.keys():
+            if k[0] == 'tags':
+                n = max(n, k[1] + 1)
+
+        # validate
+        for tag in new_tags:
+            tag_length_validator(tag, context)
+            tag_name_validator(tag, context)
+
+        # add new tags
+        for num, tag in enumerate(new_tags):
+            data[('tags', num+n, 'name')] = tag
+            data[('tags', num+n, 'vocabulary')] = vocab
+
+    return callable
+
+def convert_from_tags(vocab):
+    def callable(key, data, errors, context):
+        pass
+    return callable
+
diff --git a/ckan/tests/logic/test_converters.py b/ckan/tests/logic/test_converters.py
new file mode 100644
index 00000000000..9e9e06337bc
--- /dev/null
+++ b/ckan/tests/logic/test_converters.py
@@ -0,0 +1,20 @@
+from ckan import model
+from ckan.logic.converters import convert_to_tags
+from ckan.lib.navl.dictization_functions import unflatten
+
+class TestConverters:
+    def test_convert_to_tags(self):
+        def convert(tag_string, vocab):
+            key = 'vocab_tag_string'
+            data = {key: tag_string}
+            errors = []
+            context = {'model': model, 'session': model.Session}
+            convert_to_tags('vocab')(key, data, errors, context)
+            del data[key]
+            return data
+
+        data = unflatten(convert('tag1, tag2', 'vocab'))
+        for tag in data['tags']:
+            assert tag['name'] in ['tag1', 'tag2']
+            assert tag['vocabulary'] == 'vocab'
+