Clarify initializer documentation

Make it clear that there is only one initializer in the whole grammar.
The previous formulation could have been understood to mean that there
can be an initializer for every rule in the grammar.

Fixes #82.

README.md

@@ -182,14 +182,14 @@ written as a JavaScript string between the name and separating equality sign.
 Rules need to be separated only by whitespace (their beginning is easily
 recognizable), but a semicolon (“;”) after the parsing expression is allowed.
 
-Rules can be preceded by an *initializer* — a piece of JavaScript code in curly
-braces (“{” and “}”). This code is executed before the generated parser starts
-parsing. All variables and functions defined in the initializer are accessible
-in rule actions and semantic predicates. The code inside the initializer can
-access the parser object using the `parser` variable and options passed to the
-parser using the `options` variable. Curly braces in the initializer code must
-be balanced. Let's look at the example grammar from above using a simple
-initializer.
+The first rule can be preceded by an *initializer* — a piece of JavaScript code
+in curly braces (“{” and “}”). This code is executed before the generated parser
+starts parsing. All variables and functions defined in the initializer are
+accessible in rule actions and semantic predicates. The code inside the
+initializer can access the parser object using the `parser` variable and options
+passed to the parser using the `options` variable. Curly braces in the
+initializer code must be balanced. Let's look at the example grammar from above
+using a simple initializer.
 
     {
       function makeInteger(o) {