The past decade has seen many changes in the care for chronic mental patients in the Netherlands. Large-scale institutions, which Goffman aptly called 'total institutions', because of the all pervasive influence they had on behavior and attitudes of their inmates, were gradually 'deinstitutionalized'. The main drift of these changes was in the direction of small-scale psychiatric care units, known as `sheltered homes', where institutional influences are reduced as much as possible. The theoretical notion behind the movement of deinstitutionalization was the belief that typical characteristics of chronic mental patients, such as apathy, lack of initiative, loss of interest, submissiveness, a loss of individuality, did not originate in premorbid or psychiatric handicaps but were mainly due to the restrictive ward-system. We aim to prove this notion, 'the thesis of institutionalism', to be a questionable one. In our view the reverse of the thesis could well be true, namely, certain premorbid characteristics predispose patients to long terms of hospitalization. We use the concept of 'anomy' to describe the premorbid handicaps in question. We do not question the influence of the wardsystem on behavioral changes such as `social withdrawal'. Sheltered homes will certainly reactivate withdrawn patients. Our doubts concern the effectivenes of these experimental care-units on the anomic attitude of many chronic mental patients. In this respect we emphasize the importance of developing new techniques of dealing with anomic attitudes.