Negation is present in all human languages and it is used to reverse the polarity of some part of statements that are otherwise affirmative by default. Negation has been largely ignored in recent efforts of representing text with semantic roles. A negated statement often carries positive implicit meaning but to pinpoint the negated part from the positive part is rather difficult. This paper aims at thoroughly representing the semantics of negation by revealing implicit positive meaning. The method relies on focus detection. For this, new annotation over PropBank and a learning algorithm are proposed.