We conducted three experiments to examine the effects of the underlying premise of a message on subsequent issue agreement, perceptions of the message and source, subjects' cognitive responses, and attitudes toward related but nonmentioned issues. In all experiments, subjects who read a message based on an acceptable premise were subsequently more in favor of the message recommendation than were subjects who read an identical message with a less acceptable premise. In Experiment 3, this effect also occurred for attitudes toward proposals that were not mentioned in the message. In contrast, premise acceptability did not have a reliable effect on perceptions of the message and source, and had only a small effect on subjects' cognitive responses. This suggests that the effect of premise acceptability on issue agreement is primarily the result of a tendency to maintain consistency between beliefs in a premise and logically related proposals.