In the course of ordinary communication, people transmit messages (i.e., say things) which may involve the application a truth predicate. The receiver of such a message needs to have a method which allows the extraction of non-truth-theoretic information from uses of the truth predicate; such a method can be modeled with an axiomatic system. On close examination, the choice of which axiomatic system to employ can be seen to depend on whether or not the source of the message is considered trustworthy—that is, whether the information in the message can simply be accepted, or if it must first be examined for consistency with previously known information and, on the basis of that determination, possibly be rejected. This paper explores some of the consequences involved in this framework.