diff --git a/wordfreq/data/large_de.msgpack.gz b/wordfreq/data/large_de.msgpack.gz
new file mode 100644
index 0000000..fb717c2
Binary files /dev/null and b/wordfreq/data/large_de.msgpack.gz differ
diff --git a/wordfreq_builder/wordfreq_builder/config.py b/wordfreq_builder/wordfreq_builder/config.py
index 595c191..4eac43b 100644
--- a/wordfreq_builder/wordfreq_builder/config.py
+++ b/wordfreq_builder/wordfreq_builder/config.py
@@ -41,9 +41,11 @@ CONFIG = {
         'subtlex-en': ['en'],
         'subtlex-other': ['de', 'nl', 'zh'],
         'jieba': ['zh'],
-        'reddit': [
-            'de', 'en', 'es', 'fr', 'it', 'ja', 'pt', 'sv'
-        ]
+
+        # About 99.2% of Reddit is in English, but there are pockets of
+        # conversation in other languages. These are the languages that seem
+        # to have enough non-spam comments to actually learn from.
+        'reddit': ['de', 'en', 'es', 'sv']
     },
     # Subtlex languages that need to be pre-processed
     'wordlist_paths': {
@@ -63,7 +65,7 @@ CONFIG = {
         'jieba-dist': 'dist/jieba_{lang}.{ext}'
     },
     'min_sources': 2,
-    'big-lists': ['en', 'fr', 'es', 'pt']
+    'big-lists': ['en', 'fr', 'es', 'pt', 'de']
 }
 
 
diff --git a/wordfreq_builder/wordfreq_builder/tokenizers.py b/wordfreq_builder/wordfreq_builder/tokenizers.py
index cea3283..d8bfd12 100644
--- a/wordfreq_builder/wordfreq_builder/tokenizers.py
+++ b/wordfreq_builder/wordfreq_builder/tokenizers.py
@@ -116,19 +116,17 @@ def tokenize_by_language(in_filename, out_prefix, tokenizer):
     Produces output files that are separated by language, with spaces
     between the tokens.
     """
-    out_files = {}
+    out_files = {
+        language: open('%s.%s.txt' % (out_prefix, language), 'w', encoding='utf-8')
+        for language in KEEP_THESE_LANGUAGES
+    }
     with open(in_filename, encoding='utf-8') as in_file:
         for line in in_file:
             text = line.split('\t')[-1].strip()
             language, tokens = tokenizer(text)
-            if language != 'un':
+            if language in KEEP_THESE_LANGUAGES:
+                out_file = out_files[language]
                 tokenized = ' '.join(tokens)
-                out_filename = '%s.%s.txt' % (out_prefix, language)
-                if out_filename in out_files:
-                    out_file = out_files[out_filename]
-                else:
-                    out_file = open(out_filename, 'w', encoding='utf-8')
-                    out_files[out_filename] = out_file
                 print(tokenized, file=out_file)
     for out_file in out_files.values():
         out_file.close()