On the logical circularity of the notion of rational equilibrium