Garfinkel claims to have demonstrated the documentary method and its reflexive nature by an experiment conducted in a university department of psychiatry. Students were invited to take part in what was described as a new form of psychotherapy.
They were asked to summarize a personal problem on which they required advice and then ask a counsellor a series of questions. The counsellor sat in a room adjoining the student; they could not see each other and communicated via an intercom.
The counsellor was limited to responses of either ‘yes’ or ‘no’. Unknown to the student, his advisor was not a counsellor and the answers he received were evenly divided between ‘yes’ and ‘no’, their sequence being predetermined in accordance with a table of random numbers.
Despite the fact that the answers he received were random, given without reference to the content of questions and sometimes contradicted previous answers, the student found them helpful, reasonable and sensible. Similar assessments of the counseling sessions were made by the other students in the experiment.