An old man once told me when I was a student of philosophy at St. Louis University, "Francis, never trust a successful teacher." A very successful college professor of mathematics once told me later, "Most teachers talk too God damned much." I have puzzled over these seemingly strange comments on teaching for the past twenty years. They are the genesis of this paper.

With regard to the "successful" teacher, Herbert Kelman sheds some light on why it is possibly a very bad idea to be successful.1 He identifies three models of social influence on opinion change--compliance, identification, and internalization. Within the classroom perspective I interpret his study as saying, the teacher who demands total compliance must be present to enforce it upon the students. The teacher who is charismatic is also heavily dependent on physical presence; students tend to identify with this teacher primarily and run the risk of parroting whatever he or she says as long as they are close to this person; personal attractiveness is the key. An agent (teacher) who relies neither on the rod nor personal attractiveness is a more likely agent of truly internalized and lasting change. I do not mean by quoting Dr. Kelman, to erect a universal model of good teaching. I do hope to shed some light on the remark about successful teachers quoted at the beginning of this essay.