The purpose of this article is to show that long-established insights into the close relation between predicate structure and information structure in Mandarin Chinese can account for a number of concrete observations once they are formalized. In the course of the discussion, I will develop formal definitions of the principle I refer to as the Predicate-Comment Mapping Hypothesis and of the copula and comment marker shi. After discussing how they apply to simple assertive clauses, I will show that these definitions allow us to derive the correct predictions about the differences between three different types of polarity questions—the so-called ma questions, shi-bu-shi questions and A-neg-A questions.