Theories of persuasion have long assumed a process which includes comprehension of the message by the recipient. Several hundred undergraduates at Ohio State University and Marshall University (Ohio) participated in six experiments examining persuasion and the use of unintelligible messages. Subjects in individual cubicles of a university language laboratory were told they would hear talks delivered at a United Nations conference. Subjects listened to a taped message in English and in an unintelligible version. In Experiment VI subjects heard an unintelligible version and a no-message version. An equal number of students heard the tape in reverse order. Subjects responded to standard attitude dependent measures: semantic differentials, attitude scales, and cognitive response measures. Manipulations of message length, number of repetitions, and source credibility were added as an additional between-subjects factor in some experiments. Major findings revealed were that: (1) most subjects agreed with and listed cognitive responses to unintelligible communications; (2) persuasion processes with unintelligible communications were similar to persuasion processes using intelligible messages; (3) an unintelligible communication evinced more cognitive responses and more favorable ratings than a nonmessage control; (4) the unintelligible messages produced increased cognitive elaboration as a function of total number of thoughts generated; and (5) student high in Need for Cognition were more persuaded by unintelligible messages than were students low in Need for Cognition. These results challenge the message-comprehension assumption. (NB)