diff --git a/pod/perldelta.pod b/pod/perldelta.pod
index 57c8746e8c44..4088f20c147b 100644
--- a/pod/perldelta.pod
+++ b/pod/perldelta.pod
@@ -393,6 +393,11 @@ consisted of only ASCII characters.  The real upper limit was as few as
 Chinese or Osage.  Now an identifier in any language may contain at
 least 255 characters.
 
+=item *
+
+Fixed parsing of array names starting with a digit in double-quotish
+context under C<use utf8;>.
+
 =back
 
 =head1 Known Problems
diff --git a/t/op/sub_lval.t b/t/op/sub_lval.t
index bfb17a1ff1e8..982044ebef89 100644
--- a/t/op/sub_lval.t
+++ b/t/op/sub_lval.t
@@ -5,7 +5,7 @@ BEGIN {
     require './test.pl';
     set_up_inc('../lib');
 }
-plan tests=>213;
+plan tests=>215;
 
 sub a : lvalue { my $a = 34; ${\(bless \$a)} }  # Return a temporary
 sub b : lvalue { ${\shift} }
@@ -1041,6 +1041,14 @@ sub else119797 : lvalue {
 eval { (else119797(0)) = 1..3 };
 is $@, "", '$@ after writing to array returned by else';
 is "@119797", "1 2 3", 'writing to array returned by else';
+
+{   # Being in UTF-8 used to break this
+    use utf8;
+    eval { (else119797(0)) = 1..3 };
+    is $@, "", '$@ after writing to array returned by else';
+    is "@119797", "1 2 3", 'writing to array returned by else';
+}
+
 eval { (else119797(1)) = 4..6 };
 is $@, "", '$@ after writing to array returned by if (with else)';
 is "@119797", "4 5 6", 'writing to array returned by if (with else)';
diff --git a/toke.c b/toke.c
index 0d7d9dc45923..d201be5b7729 100644
--- a/toke.c
+++ b/toke.c
@@ -3697,10 +3697,7 @@ S_scan_const(pTHX_ char *start)
              * (@foo, @::foo, @'foo, @{foo}, @$foo, @+, @-)
              */
         else if (*s == '@' && s[1]) {
-            if (UTF
-               ? isIDFIRST_utf8_safe(s+1, send)
-               : isWORDCHAR_A(s[1]))
-            {
+            if (isDIGIT_A(s[1]) || isIDFIRST_lazy_if_safe(s+1, send, UTF)) {
                 break;
             }
             if (memCHRs(":'{$", s[1]))