I have the given formula $$A = P(a) \wedge \forall x(P(x)\rightarrow (\exists y Q(x,y)))\wedge (\exists x\neg Q(x,a))$$, and have to show that it has no Herbrand model.
As far as I know, i have to convert it in Skolem Normal Form, then get the Herbrand Expansion and use Resolution to show that the formula is unsatisfiable.
My Skolem Normal Form looks like this: $$\forall x(P(a)\wedge (\neg P(x)\vee Q(x,f(x)))\wedge \neg Q(c,a))$$
But I have no clue how to show that this is unsatisfiable.
Are my steps so far even correct, and what can I do next?