diff --git a/spacy/tests/en/__init__.py b/spacy/tests/en/__init__.py
new file mode 100644
index 000000000..e69de29bb
diff --git a/spacy/tests/tokenizer/conftest.py b/spacy/tests/en/conftest.py
similarity index 100%
rename from spacy/tests/tokenizer/conftest.py
rename to spacy/tests/en/conftest.py
diff --git a/spacy/tests/en/tokenizer/__init__.py b/spacy/tests/en/tokenizer/__init__.py
new file mode 100644
index 000000000..e69de29bb
diff --git a/spacy/tests/tokenizer/sun.txt b/spacy/tests/en/tokenizer/sun.txt
similarity index 100%
rename from spacy/tests/tokenizer/sun.txt
rename to spacy/tests/en/tokenizer/sun.txt
diff --git a/spacy/tests/tokenizer/test_contractions.py b/spacy/tests/en/tokenizer/test_contractions.py
similarity index 100%
rename from spacy/tests/tokenizer/test_contractions.py
rename to spacy/tests/en/tokenizer/test_contractions.py
diff --git a/spacy/tests/tokenizer/test_exceptions.py b/spacy/tests/en/tokenizer/test_exceptions.py
similarity index 100%
rename from spacy/tests/tokenizer/test_exceptions.py
rename to spacy/tests/en/tokenizer/test_exceptions.py
diff --git a/spacy/tests/tokenizer/test_indices.py b/spacy/tests/en/tokenizer/test_indices.py
similarity index 100%
rename from spacy/tests/tokenizer/test_indices.py
rename to spacy/tests/en/tokenizer/test_indices.py
diff --git a/spacy/tests/tokenizer/test_prefix_suffix_infix.py b/spacy/tests/en/tokenizer/test_prefix_suffix_infix.py
similarity index 100%
rename from spacy/tests/tokenizer/test_prefix_suffix_infix.py
rename to spacy/tests/en/tokenizer/test_prefix_suffix_infix.py
diff --git a/spacy/tests/tokenizer/test_punct.py b/spacy/tests/en/tokenizer/test_punct.py
similarity index 96%
rename from spacy/tests/tokenizer/test_punct.py
rename to spacy/tests/en/tokenizer/test_punct.py
index f6e8a0293..b6ae9224d 100644
--- a/spacy/tests/tokenizer/test_punct.py
+++ b/spacy/tests/en/tokenizer/test_punct.py
@@ -6,12 +6,13 @@ from __future__ import unicode_literals
 
 import pytest
 
-from ... import util
-from ...language_data import TOKENIZER_PREFIXES
+from ....util import compile_prefix_regex
+from ....language_data import TOKENIZER_PREFIXES
 
-en_search_prefixes = util.compile_prefix_regex(TOKENIZER_PREFIXES).search
 
 
+en_search_prefixes = compile_prefix_regex(TOKENIZER_PREFIXES).search
+
 PUNCT_OPEN = ['(', '[', '{', '*']
 PUNCT_CLOSE = [')', ']', '}', '*']
 PUNCT_PAIRED = [('(', ')'),  ('[', ']'), ('{', '}'), ('*', '*')]
diff --git a/spacy/tests/tokenizer/test_tokenizer.py b/spacy/tests/en/tokenizer/test_tokenizer.py
similarity index 99%
rename from spacy/tests/tokenizer/test_tokenizer.py
rename to spacy/tests/en/tokenizer/test_tokenizer.py
index 31bff9b14..8b34c5ec2 100644
--- a/spacy/tests/tokenizer/test_tokenizer.py
+++ b/spacy/tests/en/tokenizer/test_tokenizer.py
@@ -4,7 +4,7 @@ from os import path
 
 import pytest
 
-from spacy.util import utf8open
+from ....util import utf8open
 
 
 def test_tokenizer_handles_no_word(en_tokenizer):
diff --git a/spacy/tests/tokenizer/test_whitespace.py b/spacy/tests/en/tokenizer/test_whitespace.py
similarity index 100%
rename from spacy/tests/tokenizer/test_whitespace.py
rename to spacy/tests/en/tokenizer/test_whitespace.py