YES proof: The input problem is infeasible because [1] the following set of Horn clauses is satisfiable: f(f(x, y), x) = x f(f(x, y), f(z, w)) = f(f(x, z), f(y, w)) f(b, a) = f(a, b) ==> true__ = false__ true__ = false__ ==> \bottom This holds because [2] the following E does not entail the following G (Claessen-Smallbone's transformation (2018)): E: f(f(x, y), f(z, w)) = f(f(x, z), f(y, w)) f(f(x, y), x) = x t1(f(a, b)) = false__ t1(f(b, a)) = true__ t2(false__) = true__ t2(true__) = false__ G: true__ = false__ This holds because [3] the following ground-complete ordered TRS entails E but does not entail G: f(f(x, y), w) = f(f(x, z), w) f(Y2, f(Y1, Y3)) -> f(Y2, Y3) f(f(f(Y1, X1), Y3), Y2) -> f(Y1, Y2) f(f(x, y), x) -> x t1(f(a, b)) -> false__ t1(f(b, a)) -> true__ t2(false__) -> true__ t2(true__) -> false__ with the LPO induced by a > b > f > t1 > t2 > false__ > true__