I am asked to show this following argument is valid.
There is a politician who does not tell the truth. Every politician says we should tell the truth. Therefore, there is a politician who says we should tell the truth, but does not tell the truth.
So, What I've done until now is below
The domain is the set of politicians.
T(x) : x tells the truth
S(x,y) : x says y should tell the truth.
∃x¬T(x) = There is a politician who does not tell the truth. ∀x(x,y) = Every politician says we should tell the truth. ∴∃x(¬T(x) ∧ S(x,y)
1.∃x(¬T(x)) : Pemises
2.c∧¬T(c)) : Existential instantiation 1
3.∀xS(x,y) : Premises
4.S(c,d) : Universal instantiation 3
- c : simplification 2
6.¬(T(c)) : Simplification 2
7.¬T(c) ∧ S(c,d) : Addition 4,7
8.∃x(¬T(x) ∧ S(x,y) ) Existential generalization 7
My question is whether I need to define "y" or not && this processing is fine. The reason I use y without definition is I've seen a lot of solutions using y without definition. To be honest, English is not my first language, so there might be something ambiguous you guys can not understand. Let me know. I want you to be clear.