diff --git a/lib/matplotlib/category.py b/lib/matplotlib/category.py
index a1ca001fe587..8b64c7d6cb98 100644
--- a/lib/matplotlib/category.py
+++ b/lib/matplotlib/category.py
@@ -4,15 +4,35 @@
 """
 from __future__ import (absolute_import, division, print_function,
                         unicode_literals)
-
 import six
 
 import numpy as np
 
-import matplotlib.cbook as cbook
 import matplotlib.units as units
 import matplotlib.ticker as ticker
 
+# np 1.6/1.7 support
+from distutils.version import LooseVersion
+import collections
+
+
+if LooseVersion(np.__version__) >= LooseVersion('1.8.0'):
+    def shim_array(data):
+        return np.array(data, dtype=np.unicode)
+else:
+    def shim_array(data):
+        if (isinstance(data, six.string_types) or
+                not isinstance(data, collections.Iterable)):
+            data = [data]
+        try:
+            data = [str(d) for d in data]
+        except UnicodeEncodeError:
+            # this yields gibberish but unicode text doesn't
+            # render under numpy1.6 anyway
+            data = [d.encode('utf-8', 'ignore').decode('utf-8')
+                    for d in data]
+        return np.array(data, dtype=np.unicode)
+
 
 class StrCategoryConverter(units.ConversionInterface):
     @staticmethod
@@ -25,7 +45,8 @@ def convert(value, unit, axis):
         if isinstance(value, six.string_types):
             return vmap[value]
 
-        vals = np.array(value, dtype=np.unicode)
+        vals = shim_array(value)
+
         for lab, loc in vmap.items():
             vals[vals == lab] = loc
 
@@ -81,8 +102,7 @@ def update(self, new_data):
         self._set_seq_locs(new_data, value)
 
     def _set_seq_locs(self, data, value):
-        strdata = np.array(data, dtype=np.unicode)
-        # np.unique makes dateframes work
+        strdata = shim_array(data)
         new_s = [d for d in np.unique(strdata) if d not in self.seq]
         for ns in new_s:
             self.seq.append(ns)
diff --git a/lib/matplotlib/tests/test_category.py b/lib/matplotlib/tests/test_category.py
index 83847e3150bb..6e5c43d76fb9 100644
--- a/lib/matplotlib/tests/test_category.py
+++ b/lib/matplotlib/tests/test_category.py
@@ -3,8 +3,6 @@
 from __future__ import (absolute_import, division, print_function,
                         unicode_literals)
 
-from distutils.version import LooseVersion
-
 import pytest
 import numpy as np
 
@@ -14,11 +12,6 @@
 import unittest
 
 
-needs_new_numpy = pytest.mark.xfail(
-    LooseVersion(np.__version__) < LooseVersion('1.8.0'),
-    reason='NumPy < 1.8.0 is broken.')
-
-
 class TestUnitData(object):
     testdata = [("hello world", ["hello world"], [0]),
                 ("Здравствуйте мир", ["Здравствуйте мир"], [0]),
@@ -28,14 +21,12 @@ class TestUnitData(object):
 
     ids = ["single", "unicode", "mixed"]
 
-    @needs_new_numpy
     @pytest.mark.parametrize("data, seq, locs", testdata, ids=ids)
     def test_unit(self, data, seq, locs):
         act = cat.UnitData(data)
         assert act.seq == seq
         assert act.locs == locs
 
-    @needs_new_numpy
     def test_update_map(self):
         data = ['a', 'd']
         oseq = ['a', 'd']
@@ -87,7 +78,6 @@ class TestStrCategoryConverter(object):
     def mock_axis(self, request):
         self.cc = cat.StrCategoryConverter()
 
-    @needs_new_numpy
     @pytest.mark.parametrize("data, unitmap, exp", testdata, ids=ids)
     def test_convert(self, data, unitmap, exp):
         MUD = MockUnitData(unitmap)