diff --git a/spec/app/syntax-spec.coffee b/spec/app/syntax-spec.coffee
index 943868fc4..6f5ece9df 100644
--- a/spec/app/syntax-spec.coffee
+++ b/spec/app/syntax-spec.coffee
@@ -23,8 +23,8 @@ describe "the `syntax` global", ->
       expect(syntax.selectGrammar("file.js").name).toBe "JavaScript" # based on extension (.js)
       expect(syntax.selectGrammar("/tmp/.git/config").name).toBe "Git Config" # based on end of the path (.git/config)
       expect(syntax.selectGrammar("Rakefile").name).toBe "Ruby" # based on the file's basename (Rakefile)
-      expect(syntax.selectGrammar("curb").name).toBe "Plain Text"
-      expect(syntax.selectGrammar("/hu.git/config").name).toBe "Plain Text"
+      expect(syntax.selectGrammar("curb").name).toBe "Null Grammar"
+      expect(syntax.selectGrammar("/hu.git/config").name).toBe "Null Grammar"
 
     it "uses the filePath's shebang line if the grammar cannot be determined by the extension or basename", ->
       filePath = require.resolve("fixtures/shebang")
@@ -37,7 +37,7 @@ describe "the `syntax` global", ->
       expect(syntax.selectGrammar("dummy.coffee", fileContent).name).toBe "CoffeeScript"
 
       fileContent = '<?xml version="1.0" encoding="UTF-8"?>'
-      expect(syntax.selectGrammar("grammar.tmLanguage", fileContent).name).toBe "Plain Text"
+      expect(syntax.selectGrammar("grammar.tmLanguage", fileContent).name).toBe "Null Grammar"
 
       fileContent += '\n<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">'
       expect(syntax.selectGrammar("grammar.tmLanguage", fileContent).name).toBe "Property List (XML)"
diff --git a/src/app/text-mate-grammar.coffee b/src/app/text-mate-grammar.coffee
index a9f5090b6..95d52db1c 100644
--- a/src/app/text-mate-grammar.coffee
+++ b/src/app/text-mate-grammar.coffee
@@ -43,12 +43,10 @@ class TextMateGrammar
     contents = fsUtils.read(path) if not contents? and fsUtils.isFile(path)
 
     if syntax.grammarOverrideForPath(path) is @scopeName
-      Infinity
-    else if @matchesContents(contents)
       3
-    else if @matchesPath(path)
+    else if @matchesContents(contents)
       2
-    else if @isTextGrammar()
+    else if @matchesPath(path)
       1
     else
       -1
@@ -78,9 +76,6 @@ class TextMateGrammar
       pathSuffix = pathComponents[-fileTypeComponents.length..-1]
       _.isEqual(pathSuffix, fileTypeComponents)
 
-  isTextGrammar: ->
-    @scopeName is 'text.plain'
-
   tokenizeLine: (line, ruleStack=[@initialRule], firstLine=false) ->
     originalRuleStack = ruleStack
     ruleStack = new Array(ruleStack...) # clone ruleStack