We consider the system 482. Alphabet: apply : [o -> o * o] --> o compo : [o -> o * o -> o * o] --> o Rules: compo(/\x.X(x), /\y.Y(y), Z) => X(Y(Z)) apply(/\x.X(x), Y) => X(Y) We use rule removal, following [Kop12, Theorem 2.23]. This gives the following requirements (possibly using Theorems 2.25 and 2.26 in [Kop12]): compo(/\x.X(x), /\y.Y(y), Z) >? X(Y(Z)) apply(/\x.X(x), Y) >? X(Y) We use a recursive path ordering as defined in [Kop12, Chapter 5]. We choose Lex = {} and Mul = {apply, compo}, and the following precedence: apply > compo With these choices, we have: 1] compo(/\x.X(x), /\y.Y(y), Z) > X(Y(Z)) because [2], by definition 2] compo*(/\x.X(x), /\y.Y(y), Z) >= X(Y(Z)) because [3], by (Select) 3] X(compo*(/\x.X(x), /\y.Y(y), Z)) >= X(Y(Z)) because [4], by (Meta) 4] compo*(/\x.X(x), /\y.Y(y), Z) >= Y(Z) because [5], by (Select) 5] Y(compo*(/\x.X(x), /\y.Y(y), Z)) >= Y(Z) because [6], by (Meta) 6] compo*(/\x.X(x), /\y.Y(y), Z) >= Z because [7], by (Select) 7] Z >= Z by (Meta) 8] apply(/\x.X(x), Y) >= X(Y) because [9], by (Star) 9] apply*(/\x.X(x), Y) >= X(Y) because [10], by (Select) 10] X(apply*(/\x.X(x), Y)) >= X(Y) because [11], by (Meta) 11] apply*(/\x.X(x), Y) >= Y because [12], by (Select) 12] Y >= Y by (Meta) We can thus remove the following rules: compo(/\x.X(x), /\y.Y(y), Z) => X(Y(Z)) We use rule removal, following [Kop12, Theorem 2.23]. This gives the following requirements (possibly using Theorems 2.25 and 2.26 in [Kop12]): apply(/\x.X(x), Y) >? X(Y) We use a recursive path ordering as defined in [Kop12, Chapter 5]. We choose Lex = {} and Mul = {apply}, and the following precedence: apply With these choices, we have: 1] apply(/\x.X(x), Y) > X(Y) because [2], by definition 2] apply*(/\x.X(x), Y) >= X(Y) because [3], by (Select) 3] X(apply*(/\x.X(x), Y)) >= X(Y) because [4], by (Meta) 4] apply*(/\x.X(x), Y) >= Y because [5], by (Select) 5] Y >= Y by (Meta) We can thus remove the following rules: apply(/\x.X(x), Y) => X(Y) All rules were succesfully removed. Thus, termination of the original system has been reduced to termination of the beta-rule, which is well-known to hold. +++ Citations +++ [Kop12] C. Kop. Higher Order Termination. PhD Thesis, 2012.