In this paper I discuss the consistency and accuracy of Husserl’s sketch of a theory about non-declarative sentences in the last chapter of Logical Investigations. Whereas the consistency is acknowledged, the accuracy is denied, because Husserl’s treatment of non-declarative phrases such as questions or orders implies that those phrases contain, in some way, a declarative sentence and an objectifying act. To construct a question like »is A B?« as being equivalent to a declarative sentence such as »I ask whether A (...) is B« is, indeed, a false phenomenological analysis, because to ask or to order or to beg is not to assert. I turn, then, to John Austin’s theory of performative utterances and illocutionary acts in order to find a more accurate approach to the logical-semantic content of non-declarative sentences. Eventually, I show how this Husserlian theory of non-declarative sentences has a negative impact on the phenomenological theory of social acts and communication. (shrink)