A rational agent revises its goals if something changes in its mental state. In this paper, we propose (i) a general framework based on classical propositional logic, to represent changes in the mental state of the agent after the acquisition of new information and/or after the arising of new desires; (ii) fundamental postulates that the function which generates the goal set must obey; and (iii) properties that this function must have to guarantee both the agent's maximal satisfaction and the consistency of the goal set. 1