<p>While direct assertions of hateful propositions by means of declarative utterances may be the examples of hate speech that come to mind most readily, hate speech is not restricted to such utterances. To contribute to the understanding of non-declarative types of hate speech this paper discusses hateful questions. The paper examines how various types of written utterances with interrogative marking may be used to communicate hateful content by exploiting pragmatic dynamics. A corpus harvested from the comment threads of the Danish pages of a global social media platform serves as the source of empirical examples. Through analysis of a broad variety of question types the paper shows that a wide range of different pragmatic dynamics contribute to how questions communicate hateful content. Hateful questions may communicate derogatory content by virtue of presuppositions or implicatures but may also be categorized based on various other characteristic properties, e.g., as questions with slurs or biased questions. Understanding fully how questions may be used for hate speech requires an understanding of the distinct pragmatic properties of all these different question types.</p>