It was a wrong assumption that a <> 1 and a <> 2) is equal to (a, a) <> (1, 2). Fixed.
a <> 1 and a <> 2)
(a, a) <> (1, 2)
#239 (closed)