In view of the practical importance of neurotic disorders (with something like one-third of the population suffering such dis turbances at some time of their lives) and the equally great theoretical importance of types of behaviour that clearly seem to contradict both common sense and the law of effect, one might have expected that psychologists would develop consis tent and testable theories of neurosis and that there would be many textbooks outlining these theories and describing the experiments done to test them. Oddly enough nothing of the kind seems to have happened. There is a dearth of...