diff --git a/packages/compiler/src/ml_parser/lexer.ts b/packages/compiler/src/ml_parser/lexer.ts
index 57d12756d7..713b054e7e 100644
--- a/packages/compiler/src/ml_parser/lexer.ts
+++ b/packages/compiler/src/ml_parser/lexer.ts
@@ -461,12 +461,14 @@ class _Tokenizer {
     let tagName: string;
     let prefix: string;
     let openTagToken: Token|undefined;
+    let tokensBeforeTagOpen = this.tokens.length;
     const innerStart = this._cursor.clone();
     try {
       if (!chars.isAsciiLetter(this._cursor.peek())) {
         throw this._createError(
             _unexpectedCharacterErrorMsg(this._cursor.peek()), this._cursor.getSpan(start));
       }
+
       openTagToken = this._consumeTagOpenStart(start);
       prefix = openTagToken.parts[0];
       tagName = openTagToken.parts[1];
@@ -483,10 +485,10 @@ class _Tokenizer {
       this._consumeTagOpenEnd();
     } catch (e) {
       if (e instanceof _ControlFlowError) {
-        // When the start tag is invalid, assume we want a "<"
+        // When the start tag is invalid (including invalid "attributes"), assume we want a "<"
         this._cursor = innerStart;
         if (openTagToken) {
-          this.tokens.pop();
+          this.tokens.length = tokensBeforeTagOpen;
         }
         // Back to back text tokens are merged at the end
         this._beginToken(TokenType.TEXT, start);
@@ -528,6 +530,10 @@ class _Tokenizer {
   }
 
   private _consumeAttributeName() {
+    const attrNameStart = this._cursor.peek();
+    if (attrNameStart === chars.$SQ || attrNameStart === chars.$DQ) {
+      throw this._createError(_unexpectedCharacterErrorMsg(attrNameStart), this._cursor.getSpan());
+    }
     this._beginToken(TokenType.ATTR_NAME);
     const prefixAndName = this._consumePrefixAndName();
     this._endToken(prefixAndName);
diff --git a/packages/compiler/test/ml_parser/lexer_spec.ts b/packages/compiler/test/ml_parser/lexer_spec.ts
index 897ff19654..533c90f5d3 100644
--- a/packages/compiler/test/ml_parser/lexer_spec.ts
+++ b/packages/compiler/test/ml_parser/lexer_spec.ts
@@ -7,7 +7,6 @@
  */
 
 import {getHtmlTagDefinition} from '../../src/ml_parser/html_tags';
-import {InterpolationConfig} from '../../src/ml_parser/interpolation_config';
 import * as lex from '../../src/ml_parser/lexer';
 import {ParseLocation, ParseSourceFile, ParseSourceSpan} from '../../src/parse_util';
 
@@ -378,6 +377,18 @@ import {ParseLocation, ParseSourceFile, ParseSourceSpan} from '../../src/parse_u
         ]);
       });
 
+      it('should report missing closing single quote', () => {
+        expect(tokenizeAndHumanizeErrors('<t a=\'b>')).toEqual([
+          [lex.TokenType.ATTR_VALUE, 'Unexpected character "EOF"', '0:8'],
+        ]);
+      });
+
+      it('should report missing closing double quote', () => {
+        expect(tokenizeAndHumanizeErrors('<t a="b>')).toEqual([
+          [lex.TokenType.ATTR_VALUE, 'Unexpected character "EOF"', '0:8'],
+        ]);
+      });
+
     });
 
     describe('closing tags', () => {
@@ -552,6 +563,31 @@ import {ParseLocation, ParseSourceFile, ParseSourceSpan} from '../../src/parse_u
         ]);
       });
 
+      it('should parse start tags quotes in place of an attribute name as text', () => {
+        expect(tokenizeAndHumanizeParts('<t ">')).toEqual([
+          [lex.TokenType.TEXT, '<t ">'],
+          [lex.TokenType.EOF],
+        ]);
+
+        expect(tokenizeAndHumanizeParts('<t \'>')).toEqual([
+          [lex.TokenType.TEXT, '<t \'>'],
+          [lex.TokenType.EOF],
+        ]);
+      });
+
+      it('should parse start tags quotes in place of an attribute name (after a valid attribute) as text',
+         () => {
+           expect(tokenizeAndHumanizeParts('<t a="b" ">')).toEqual([
+             [lex.TokenType.TEXT, '<t a="b" ">'],
+             [lex.TokenType.EOF],
+           ]);
+
+           expect(tokenizeAndHumanizeParts('<t a=\'b\' \'>')).toEqual([
+             [lex.TokenType.TEXT, '<t a=\'b\' \'>'],
+             [lex.TokenType.EOF],
+           ]);
+         });
+
       it('should be able to escape {', () => {
         expect(tokenizeAndHumanizeParts('{{ "{" }}')).toEqual([
           [lex.TokenType.TEXT, '{{ "{" }}'],