diff --git a/Objects/unicodeobject.c b/Objects/unicodeobject.c
index ddd8d53..92d1777 100644
--- a/Objects/unicodeobject.c
+++ b/Objects/unicodeobject.c
@@ -80,8 +80,9 @@
 
 /* --- Globals ------------------------------------------------------------
 
-   The globals are initialized by the _PyUnicode_Init() API and should
-   not be used before calling that API.
+NOTE: In the interpreter's initialization phase, some globals are currently
+      initialized dynamically as needed. In the process Unicode objects may
+      be created before the Unicode type is ready.
 
 */
 
@@ -98,18 +99,30 @@
    Another way to look at this is that to say that the actual reference
    count of a string is:  s->ob_refcnt + (s->state ? 2 : 0)
 */
-static PyObject *interned;
+static PyObject *interned = NULL;
 
 /* Free list for Unicode objects */
-static PyUnicodeObject *free_list;
-static int numfree;
+static PyUnicodeObject *free_list = NULL;
+static int numfree = 0;
 
 /* The empty Unicode object is shared to improve performance. */
-static PyUnicodeObject *unicode_empty;
+static PyUnicodeObject *unicode_empty = NULL;
+
+#define _Py_RETURN_UNICODE_EMPTY()                      \
+    do {                                                \
+        if (unicode_empty != NULL)                      \
+            Py_INCREF(unicode_empty);                   \
+        else {                                          \
+            unicode_empty = _PyUnicode_New(0);          \
+            if (unicode_empty != NULL)                  \
+                Py_INCREF(unicode_empty);               \
+        }                                               \
+        return (PyObject *)unicode_empty;               \
+    } while (0)
 
 /* Single character Unicode strings in the Latin-1 range are being
    shared as well. */
-static PyUnicodeObject *unicode_latin1[256];
+static PyUnicodeObject *unicode_latin1[256] = {NULL};
 
 /* Fast detection of the most frequent whitespace characters */
 const unsigned char _Py_ascii_whitespace[] = {
@@ -214,7 +227,7 @@
 
 #define BLOOM_MASK unsigned long
 
-static BLOOM_MASK bloom_linebreak;
+static BLOOM_MASK bloom_linebreak = ~(BLOOM_MASK)0;
 
 #define BLOOM_ADD(mask, ch) ((mask |= (1UL << ((ch) & (BLOOM_WIDTH - 1)))))
 #define BLOOM(mask, ch)     ((mask &  (1UL << ((ch) & (BLOOM_WIDTH - 1)))))
@@ -479,10 +492,8 @@
     if (u != NULL) {
 
         /* Optimization for empty strings */
-        if (size == 0 && unicode_empty != NULL) {
-            Py_INCREF(unicode_empty);
-            return (PyObject *)unicode_empty;
-        }
+        if (size == 0)
+            _Py_RETURN_UNICODE_EMPTY();
 
         /* Single character Unicode objects in the Latin-1 range are
            shared when using this constructor */
@@ -528,10 +539,8 @@
     if (u != NULL) {
 
         /* Optimization for empty strings */
-        if (size == 0 && unicode_empty != NULL) {
-            Py_INCREF(unicode_empty);
-            return (PyObject *)unicode_empty;
-        }
+        if (size == 0)
+            _Py_RETURN_UNICODE_EMPTY();
 
         /* Single characters are shared when using this constructor.
            Restrict to ASCII, since the input must be UTF-8. */
@@ -1393,15 +1402,11 @@
 
     /* Decoding bytes objects is the most common case and should be fast */
     if (PyBytes_Check(obj)) {
-        if (PyBytes_GET_SIZE(obj) == 0) {
-            Py_INCREF(unicode_empty);
-            v = (PyObject *) unicode_empty;
-        }
-        else {
-            v = PyUnicode_Decode(
-                    PyBytes_AS_STRING(obj), PyBytes_GET_SIZE(obj),
-                    encoding, errors);
-        }
+        if (PyBytes_GET_SIZE(obj) == 0)
+            _Py_RETURN_UNICODE_EMPTY();
+        v = PyUnicode_Decode(
+                PyBytes_AS_STRING(obj), PyBytes_GET_SIZE(obj),
+                encoding, errors);
         return v;
     }
 
@@ -1421,12 +1426,11 @@
     }
 
     if (buffer.len == 0) {
-        Py_INCREF(unicode_empty);
-        v = (PyObject *) unicode_empty;
+        PyBuffer_Release(&buffer);
+        _Py_RETURN_UNICODE_EMPTY();
     }
-    else
-        v = PyUnicode_Decode((char*) buffer.buf, buffer.len, encoding, errors);
 
+    v = PyUnicode_Decode((char*) buffer.buf, buffer.len, encoding, errors);
     PyBuffer_Release(&buffer);
     return v;
 }
@@ -8323,10 +8327,8 @@
     Py_ssize_t nchars;
     size_t nbytes;
 
-    if (len < 1) {
-        Py_INCREF(unicode_empty);
-        return (PyObject *)unicode_empty;
-    }
+    if (len < 1)
+        _Py_RETURN_UNICODE_EMPTY();
 
     if (len == 1 && PyUnicode_CheckExact(str)) {
         /* no repeat, return original string */
@@ -10056,8 +10058,6 @@
 
 void _PyUnicode_Init(void)
 {
-    int i;
-
     /* XXX - move this array to unicodectype.c ? */
     Py_UNICODE linebreak[] = {
         0x000A, /* LINE FEED */
@@ -10071,14 +10071,12 @@
     };
 
     /* Init the implementation */
-    free_list = NULL;
-    numfree = 0;
-    unicode_empty = _PyUnicode_New(0);
-    if (!unicode_empty)
-        return;
+    if (!unicode_empty) {
+        unicode_empty = _PyUnicode_New(0);
+        if (!unicode_empty)
+            return;
+    }
 
-    for (i = 0; i < 256; i++)
-        unicode_latin1[i] = NULL;
     if (PyType_Ready(&PyUnicode_Type) < 0)
         Py_FatalError("Can't initialize 'unicode'");
 
@@ -10123,15 +10121,11 @@
 {
     int i;
 
-    Py_XDECREF(unicode_empty);
-    unicode_empty = NULL;
+    Py_CLEAR(unicode_empty);
 
-    for (i = 0; i < 256; i++) {
-        if (unicode_latin1[i]) {
-            Py_DECREF(unicode_latin1[i]);
-            unicode_latin1[i] = NULL;
-        }
-    }
+    for (i = 0; i < 256; i++)
+        Py_CLEAR(unicode_latin1[i]);
+
     (void)PyUnicode_ClearFreeList();
 }
 
@@ -10250,8 +10244,7 @@
             "mortal/immortal\n", mortal_size, immortal_size);
     Py_DECREF(keys);
     PyDict_Clear(interned);
-    Py_DECREF(interned);
-    interned = NULL;
+    Py_CLEAR(interned);
 }