The question in the title of this article sounds like something my undergraduate philosophy teacher would ask. As a philosopher, he was very skeptical about reality. He could prove that he did not exist. I once asked him you think that you do not exist, does not that, in and of itself, show that you do? but he said, No. I could very well be a disembodied mind. Pretty clever. But that is not the kind of thing I am concerned with here. Rather, I am interested in how we think, and the limitations thereof.Tripartite NarrativesSpecifically, I am concerned with an idea of Mortola (i999) that narratives-spoken or written communications-tend to get conceptualized in a tripartite manner, with the first part being equilibrium, possibly some positive event, say, marriage, the second disequilibrium, a threat to that event, say, arguments between the couple, and the third event modified equilibrium, some reconciliation, possibly only partial, of the first two, say, the couple learns some ways to get along. Of course, many di∂erent things could comprise what we are dealing with. The second event could be divorce instead of arguments. The third event would still be some attempt at reconciliation of the disequilibrium, to some extent, such as learning how to live without the other person.I thought Mortola (i999) was probably not correct: our thinking is much superior to being stuck in some tripartite way of conceptualizing things. After all, the way we think about something influences the information we have available. If we are enmeshed in some rigid style of thinking, certain outcomes seem to make sense and others do not. In fact, some outcomes will never even be considered. Of course, there are always limits on the way we process information, due to our biases, styles of thinking, the information available, etc. (Stapel, Reicher, & Spears, i994; Eckes, i994). But the tripartite way of thinking seems extremely limiting, since it forces us to order things into categories that may not be the best way to think. I thought that our thinking is superior to the limitations that Mortola described.My StudyTo test Mortola's (i999) tripartite theory, I had eight students (3 males, 5 females) volunteer as patients in group psychotherapy sessions (Eisenman, 200i). These were undergraduates who volunteered, for extra course credit, to participate in group psychotherapy. Thus, they got extra credit and free psychotherapy as well. The group psychotherapists were graduate students in a Master's level counseling psychology program; the masters level students were learning to be therapists. These graduate students were observed via closed circuit television by their instructor and other graduate students, and received feedback on their therapy skills. If Mortola is correct, when the volunteer patients give narrative accounts, they should speak in the three stages, with the first stage occurring first, the second stage next, and the third stage last. For this study, the 8 student patients received group psychotherapy with two graduate student psychotherapists, one male and one female. Two trained observers had 92 percent agreement on scoring the narratives for the presence or absence of the tripartite style. There were five group psychotherapy sessions.Here is what I found: In the first psychotherapy session, 45 percent of the patient narratives were judged to follow the Mortola (i999) view. In the second session only 20 percent of the narratives followed this pattern. In the third session 60 percent fit the triadic pattern; in the fourth 75 percent, and in the fifth and last 55 percent. Overall, for the five psychotherapy sessions the mean percentage of narratives that fit Mortola's triadic theory was 5i percent.Although there is certainly great variation from therapy session to therapy session, the overall percentage of 5i seems quite high. One could object that these student volunteers were not real patients, but that would not seem to impact the results, because the theory says it applies to narratives in general, so any story telling should fit the tripartite model. …