The following excerpt is from the remarkable book, Facing Human Suffering, by Ronald B. Miller (2004, pp. 184-185). [Emphases added.]
"People who are difficult to get to know often are that way for very good reason. They are concealing aspects of their lives that they believe it would be dangerous to reveal. The information they possess involves aspects of human relationships that are regarded as shameful and immoral, if not illegal (child physical and sexual abuse, illicit sexual affairs, deviant sexual interests, cheating and dishonesty in business, etc.). There is a qualitative difference between the content of the knowledge that is concealed regarding other people and knowledge that nature conceals in the natural sciences via mystery and complexity. Only in a relationship that contains certain moral features -- trust, safety, respect, and confidentiality -- can these critical features of human existence be revealed and discussed. A psychology that wishes to go beyond the surface of social pretense and masks must provide a methodology for exploring the taboo and forbidden aspects of human relationships (Faberow, 1963). Freud (1920/1966), writing in the Introductory Lectures on Psycho-Analysis, made the point that the data of clinical psychoanalysis -- and, by implication, psychology -- are vulnerable to this sort of distortion:
The talk of which psycho-analytic treatment consists brooks no listener, it cannot be demonstrated (to the audience). A neurasthenic or hysterical patient can of course like any other, be introduced to students in a psychiatric lecture. He will give an account of his complaints and symptoms, but of nothing else. The information required by analysis will be given by him only on condition of his having a special emotional attachment to the doctor; he would become silent as soon as he observed a single witness to whom he felt indifferent. For this information concerns what is most intimate in his mental life, everything that, as a socially independent person, he must conceal from other people, and beyond, that as a a homogeneous personality, he will not admit to himself. (pp. 20-21)
Rollo May (1969) concurred in Love and Will and quoted H.S. Sullivan as expressing a similar position:
But neither these psychologists in their laboratories nor those philosophers in their studies can ignore the fact we do get tremendously significant and often unique data from persons in therapy -- data which are revealed only when human beings break down their customary pretenses, hypocrisies and defenses behind which we all hide in "normal" social discourse. There is also the curious situation that unless we are oriented towards helping the person, he will not, indeed in some ways cannot, reveal the significant data. Harry Stack Sullivan's remark on research in therapy is still as cogent as when he first made it: "Unless the interviews are designed to help the person, you'll get artifacts, not real data." (pp. 18-19)
Whether one wishes to join the more psychoanalytic observers in believing that all clinical problems have at their core such unspeakable elements, or whether one takes the more moderate position that many, if not most, do, it is clear that the clinical method of gaining knowledge in psychology is, for all its limitations, the only game in town."