If sanity and insanity exist, how shall we know them?

The question above, the many others like it, and the thoughts it inspires have bothered philosophers, physiatrists, lawyers, poets and angsty teenagers for centuries. There are the questions of cultural relativity, moral ambiguity and just what makes something normal that produce books, films and papers all trying to tackle the subject in various ways. Some, such as R.D Laing’s works, have made their mark and are always references when dealing with this controversial topic – but many are brushed aside or laughed at by those professionals seeking neat and clear definitions.

On being sane in insane places1 is the name of a paper written in 1973 by Dr. David L. Rosenhan, in which he tries to deal with the question of just what makes a person insane in somewhat empirical terms. The specific aspect of the question he was most intrigued with is yet another that can be found in some form in all cultures and all human thought – nature vs. nurture. Are the characteristics that make a person become classified as ‘insane’ held within them, or can they be found in the situations and surroundings in which they are presented?

Rosenhan devised an experiment that was meant to test the hypothesis that psychiatrists cannot accurately tell the difference between sane and insane people. In proving this hypothesis he hoped to address some of the issues in psychiatry that bothered him the most – the classifying of people with mental disorders as ill, the context (psychiatric hospitals) in which they find themselves and the 'stickiness of psychological labels’.

Preparing the experiment

The basic premise of the experiment was to get a number of ordinary people and have them attempt to gain admission into a psychiatric hospital. The idea here obviously being that the psychiatrists should reliably be able to tell that these people were not really mentally ill.

8 people were selected for the experiment:

12 different hospitals in the United States were chosen as the places to try and gain entry to. The hospitals were a mixture of old, new, research, teaching, government funded and privately owned – but all with a good, solid reputation.

Beginning the experiment

Each of the pseudo-patients was told to phone up one of the hospitals and arrange a meeting with one of the psychiatrists there. They were instructed that the only symptom they should complain of was that of hearing a voice. These auditory hallucinations would be of the same sex as the patient and what they heard would be indistinct although all would say they could make out the words 'hollow', 'dull' and 'thud'. These words were chosen because of their similarity to common existential symptoms and because there was no case of them being reported before in any psychiatric literature. Other than this the patients were to behave exactly as they normally would and to be honest with any questions they were asked. This meant that they talked about their own lives exactly as they were, that is, pretty normal, the only exception to this was those who were in the psychiatric field were asked to choose another profession as they might other-wise receive different treatment from the doctors at the hospital.

Every single one was admitted with a diagnosis of schizophrenia, except one who was diagnosed with manic-depression.

Inside the hospitals

The pseudo-patients were instructed that should they gain entry into the hospital they were to stop showing any symptoms immediately and behave exactly as normal. Just as in the case when a patient gets committed for real, they had no idea when they would be released; it was up to them to get out – basically by proving to the doctors that they were fine. On the ward they were to be the models of cooperativeness and politeness and were to be totally asymptomatic. Each patient was to write down in a note-book the way that they were treated by the staff and other patients, as well as details of their medication and general ward life.

During this time the patient was allowed to see 'visitors’, and these friends or family members were asked to note if they saw any changes in manner, behaviour or thought in the people they knew. All of the patients were eventually discharged with a diagnosis of 'schizophrenia/manic-depression in remission’. The length of stay was between 7 and 52 days – with an average of 19.

Phase two

Once this part of the study had been completed a number of times at all 12 of the chosen hospitals, the next stage began. For this part of the experiment, several other hospitals were chosen and the doctors and staff were shown the results of the previous study. Without exception they all insisted that this could never happen at their hospital and that the results must be false. They were then told that during the next three months there would be some more of these pseudo-patients coming to their hospitals to try and gain entry. During this period all doctors and staff should keep a note of the new patients and record which, if any, they thought were faking it.

This, however, was all part of a further act. Rosenhan was not in fact sending any pseudo-patients to their hospitals. However during that time, 41 of the 193 patients who were admitted at the various hospitals were 'confidently judged to be pseudo-patients' by the doctors and staff.

Now – although this experiment was not as shocking in the way of some other studies which have now almost passed into popular fiction – it none the less raised quite a fuss in the psychiatric circles as well as raising some interesting questions for everyday life.

The normal are not detectibly sane

The very fact that every single pseudo-patient gained entry casts a worrying doubt over the validity of these classifications. How is it that a qualified doctor using standard guidelines cannot detect someone who is merely presenting them with false symptoms? In his paper Rosenhan suggests:

"Perhaps doctors are strongly biased towards Type-two errors, i.e. they are more inclined to call a healthy person sick (a false positive) than a sick person healthy (a false negative, Type-one error). It is clearly more dangerous to misdiagnose illness than health, so it is better to err on the side of caution."

However, this was an aspect that worried Rosenhan in relation to psychiatry, as what holds true for medicine may not be so acceptable in the case of the mind. The labels involved in medicine are not as pejorative (this was before the days of AIDS) and do not carry with them such weight. A misdiagnosed cancer – while worrying at the time – would be a cause for celebration and would not affect the person’s life and legal rights in the same way as a misdiagnosed case of MPD or schizophrenia. Indeed, as shown in this study, the label of mental illness is never really removed – the patient was not released with a diagnosis of 'normal', or 'normal: initial diagnosis in error', the diagnosis was 'in remission'.2

The second part of the experiment seems to indicate that the tendency to call sane people insane can be reversed – that in the eagerness to avoid Type-two errors doctors tend to then make many more Type-one errors. Whatever the case may be, it certainly lends its self to proving Rosenhan’s hypothesis:

Any diagnostic process which lends itself so readily to massive errors of this sort cannot be a very reliable one.

And it’s a rather scary thought that something so supposedly reliable is used to make legal, moral and personal decisions on a daily basis. However, in defence of the psychiatrists, another example was put forward by a doctor in 19743:

If I were to drink a quart of blood and, concealing what I had done, had come to the emergency room of any hospital vomiting blood, the behaviour of the staff would be quite predictable. If they labelled and treated me as having a bleeding peptic ulcer, I doubt I could convincingly argue that medical science does not know how to diagnose that condition.

Now, grisly as it sounds, this seems to be a valid argument, until you realize that you can apply all of Rosenhan's worries about 'labels' directly to it. Think about what would happen the next day when you were fine and the tests came back negative?

Just as there are many overlapping behaviours between different types of mental disorders, so there are many types of behaviour present in those that are declared insane and other normal people. It is common for everyone to feel depressed on occasion or become angry without direct provocation – these are human traits and it makes little sense to classify oneself as 'depressive' on the basis of these traits – however this often seems to happen in psychiatric hospitals.

The stickiness of psycho diagnostic labels

As well as determining that once a person has been given a diagnostic label there is little they can do to remove it, it was also noted that the label tended to 'stick' to all aspects of the person – including their thoughts and behaviours. The doctors and staff would view every behaviour of the patient only in terms of their illness. This was very much like Gestalt psychology, which focuses on the meaning given to things based on their context. The pseudo-patients noted that many of their perfectly ordinary behaviours – and those of other patients in their ward – were overlooked or only interpreted in ways that seemed to fit within the already given diagnosis.

A primary example of this was noted with one of the pseudo-patients. When he first went to the meeting with the psychiatrist, despite from the words in his head, he told his life exactly as it was. When he was young he got along well with his mother, but as he grew older he tended to have a closer connection with his dad. He was very happily married and felt that he and his wife were close, and only had occasional moments of friction. He spent a lot of time with his children and there had only been a very few occasions were they had been spanked for being naughty. Now this sounds like a pretty normal guy, nothing strange about it. Here is the profilers report:

“This white 39 year old male…. Manifests a long history of considerable ambivalence in close relationships, which begins in early childhood. A warm relationship with his mother cools during adolescence. A distant relationship to his father is described as becoming very intense. Affective stability is absent. His attempts to connect emotionally with his wife and children are punctuated by angry outbursts and, in the case of the children, spankings.”

It seems quite clear that basic facts have been distorted to fit in with the label already given to him – schizophrenic – because according to the books, those are all common traits present in a case of schizophrenia. The same happened on the ward – as he took notes of the day’s main events, all the nurses reports described him as 'engaging in compulsive writing behaviour', yet no one ever asked him about it. One day a pseudo-patient was walking up and down the corridors, a nurse stopped and asked if he was nervous, he replied that no, he was just bored and decided to stretch his legs. The next day his medication was upped to control this 'compulsive behaviour.'

Indeed one of the most surprising aspects of the experiment was that all the pseudo-patients were suspected of being fake by the other patients on their ward! Despite the pseudo-patients assertions that they had been sick but were now feeling a bit better, many of the other patients claimed that 'You’re not crazy. You’re a journalist, or a professor. You’re checking up on the hospital.'

Another factor noted in most of the real patients was that there would be occasions were one would become upset, or even 'go berserk'. In such cases the doctors and staff would always assume the cause of this to be the patient’s disorder, or sometimes a visit from a family member or friend; it was never thought that the upset might stem from the situation within the hospital – despite the fact that the patients often stated that that was in fact the reason. All behaviour was viewed within the model of the diagnostic label already attached.

The effects of labelling, hospitalization and depersonalization

This was the topic that Rosenhan was most concerned with, and much of the later half of the paper is dedicated to discussing it. The idea that mental illness was something 'for life' and could not really be cured, bur rather just go into remission or be controlled with medication was almost ubiquitous amongst the staff and doctors at the hospitals. It was also well noted by the pseudo-patients that the lack of contact between the staff and patients was very pronounced.

Any questions directed to a staff member were usually ignored or glossed over. For example, the pseudo-patients were to approach different doctors of staff members and ask them questions and see what the reaction was. They always made sure not to be rude, to interrupt, or to keep on with one particular staff member. Instead they would approach one on the ward and ask questions about their medication, when they were expected to go up before the board again or when the next visiting time was. In every case there was no eye-contact and getting the encounter over with as soon as possible, with the least exchange of information would be the main objective of the staff member. Responses to questions about their treatment or life within the hospital were often answered in forms such as 'Now, now, you don’t have to worry about that, we will take care of it.' Or in cases where the pseudo-patient wanted to note the answer, something like, 'Don’t write it down, if you have trouble remembering you can always ask again.' However, most often the response had nothing to do with the question asked and would simply be something like, 'Hello X, how are you today?'

After these results became apparent, Rosenhan conducted some other smaller experiments. A volunteer would go onto a university campus or school and find a senior member of staff who looked as though they may be in a hurry or on the way to somewhere in particular. They would stop them and ask 6 questions, things along the lines of, ‘Where is the library?’, ‘Could you tell me about x course’. Without exception the staff member stopped and answered the questions.

This lack of personal contact as well as the fact that patients had many of their legal rights taken away due to their diagnosis and were restricted in their activities and interactions contributed to an overwhelming feeling of powerlessness and depersonalization. These feelings, along with the label, seemed to remain even if the patients were discharged. All of which is incredibly counter-therapeutic. The study seemed to prove Rosenhan’s point, and it has been used on many occasions since then to highlight the unreliable nature of people’s perceptions within certain contexts. However it should be noted that he never wishes to blame people, but rather bring to their attention the flaws within the system as a whole.

In his paper he stated:

“Our overwhelming impression of the staff was of people who really cared … Where they failed, as they sometimes did painfully, it would be more accurate to attribute those failures to the environment in which they, too, found themselves … In a more benign environment, one less attached to global diagnosis, their behaviour and judgements might have been more benign and effective.”

Hereby coming full circle, to the fact that it is context of behaviour which not just doctors, but people in general, should be very careful to take note of.

Rosenhan’s conclusion

“It is clear that we cannot distinguish the sane from the insane in psychiatric hospitals, which themselves impose a special environment in which the meanings of behaviour can easily be distorted. Patients suffer powerlessness, depersonalization, segregation, mortification and self-labelling

Rather than confessing that we don’t know, or are just embarking on understanding, we continue to label patients as schizophrenic etc., as if in those words we had captured the essence of understanding. But we have known for a long time that diagnoses are often not useful or reliable, and that we cannot distinguish insanity from sanity. How many people have been needlessly stripped of their privileges of citizenship, right to vote and drive and handle their own accounts? How many have feigned insanity in order to avoid the criminal consequences of their behaviour and, conversely, how many would rather stand trail than live intermittently in a psychiatric hospital, but are wrongly thought to be mentally ill.

Finally, how many patients might be 'sane' outside the psychiatric hospital, but seem insane in it; not because craziness resides in them, but because they’re responding to a bizarre setting?”

Being sane today

One of the main counter-points today is the fact that at the time of the study it was the DSM-II that was in use. In the later versions now in use, the criteria are much stricter and much more specific – very much due to studies such as these - which highlighted the inadequacies in the previous models and methods. Today, patients complaining of the same symptoms as those used in the original study would be highly unlikely to be admitted. That makes it no less interesting to wonder what would happen today if the study were to be repeated with the requisite symptoms in place

Another factor is that most patients in psychiatric hospitals nowadays are their by their own admission, involuntary admissions are far less common today than they were in the time of Rosenhan’s study. However there are many recent studies done that still show a remarkable difference between diagnoses – for example, even in the early 90’s the same patient was twice as likely to be diagnosed as schizophrenic in America than he would be in Britain.4

Although for every study done there will be others to counter the claims made previously, Rosenhan’s still makes an interesting case. If mental health professionals cannot distinguish between the mentally ill and the healthy, the question of whether they can distinguish between different types of mental illness seems premature and, perhaps, even pointless.5 And the effects of having ones behaviour and thoughts judged based on preconceived notions is something that just about everyone can relate to in some form or another, a universal desire to be understood, yet a constant feeling of being misjudged, misplaced and misinterpreted.


  1. Paper was first published in Science, 179, pp. 250-8
    All paragraphs within quotes are from the original paper
  2. Neisser (1973)
  3. Kety (1974)
  4. WHO (1973), US-UK Diagnostic Project
  5. Lilienfeld (1995)


  • Dr. Glen Luccini (Psychology)
  • On being sane in insane places by Dr. David L. Rosenhan – can be found online in many places - http://members.aol.com/ahunter3/psych_inmates_libfront/vol_2/Rosenhan/rosenhan.html
  • Key studies in psychology (Third edition) – Richard Gross
  • A-Level psychology notes

Log in or register to write something here or to contact authors.