By the end of graduate medical training, novice internists (collectively known as the housestaff) were initiated into the experience of either having done something to a patient which had a deleterious consequence or else having witnessed colleagues do the same. When these events occurred, the housestaff engaged in social-psychological processes, utilizing a variety of coping mechanisms and in-group practices to manage these mishaps. Three major mechanisms were utilized by the housestaff for defining and defending the various mishaps which frequently occurred: denial, discounting and distancing. Denial consisted of three components: the negation of the concept of error by defining the practice of medicine as an art with 'gray areas', the repression of actual mistakes by forgetting them and the redefinition of mistakes to non-mistakes. Discounting included those defenses which externalized the blame; namely mistakes which were due to circumstances beyond their control. These included: blaming the bureaucratic system outside of medicine; blaming superiors or subordinates within internal medicine; blaming the disease and blaming the patient. When they could not longer deny or discount a mistake because of its magnitude, they utilized distancing techniques. Not withstanding this shared elaborate repertoire of denial, discounting and distancing, it was found that profound doubts and even guilt remained for many housestaff. These troublesome feelings neither easily nor automatically resolved themselves. Interspersed among their defenses were fundamental questions of culpability and responsibility as they vacillated between self and other blame. For many 'the case was never closed', even as they terminated formal training, a point neglected in the medical and sociological literature. Little in their 3 year graduate program allowed them to work through the attendant vulnerability and ambiguity accompanying the managing of mistakes. Hence, there were maladaptive aspects of the collectively acquired defense mechanisms. The whole system of accountability during graduate medical specialty training was found to be a variable, and at times, contradictory process. The housestaff ultimately sees itself as the sole arbiter of mistakes and their adjudication. Housestaffers come to feel that nobody can judge them or their decisions, least of all their patients. As they progress through training even internal accountability cohorts--the Department of Medicine, teaching faculty and peers--are discounted to varying degrees. They have developed a strong ideology justifying their jealously guarded autonomy.(ABSTRACT TRUNCATED AT 400 WORDS)