Obedience to authority
Obedience is where an individual complies in response to a direct request. While there are many examples of obedience in day to day life, it is those acts of destructive obedience that have been of particular interest to psychologists.
Milgram’s studies of obedience
The most influential study of obedience is that of Stanley Milgram (1963). In fact, as Myers (1996) points out: ‘Milgram’s experiments on what happens when the demands of authority clash with the demands of conscience have become the most famous and controversial experiments in social psychology.’ (p. 240)
As you study the details of his procedure, it is interesting to bear in mind that Milgram had a background in the theatre as a writer and playwright. The scenario that he developed, which for most participants was totally convincing, is clearly a product of this creative background!
Conditions that influence obedience
It was relatively straightforward to adapt Milgram’s procedure in order to investigate the conditions that foster obedience. Important factors were shown to be:
- Emotional closeness to the victim. When the victim was remote and could not be seen, obedience levels were highest. They were lowest when the ‘teacher’ had to physically touch the ‘learner’ (by forcing his hand onto the plate). Most compassion is shown when the victim is ‘personalised’.
- Closeness and legitimacy of the authority. Obedience was lowest when the experimenter issued instructions by telephone, and is greatest when the experimenter and ‘teacher’ are physically close. The authority must also be seen as legitimate. When another participant (actually a confederate of Milgram) was asked to take over the authority role because the experimenter had to leave the room, obedience level dropped markedly (80% refused to comply).
- Institutional authority. In Milgram’s original study obedience was reinforced by the prestige of Yale University, where the experiments were conducted. However, this was not crucial. Obedience levels remained robust although somewhat diminished, when the setting was transferred to a commercial building.
- Group influence. Some of the most dramatic reductions in conformity occurred when the participants were paired with confederates who refused to continue with the experiment.
Explanations of obedience
Early explanations of obedience tended to talk, rather mechanistically, about the ‘forces’ acting on the participants. Milgram himself talked about a conflict of opposing demands: external authority and individual conscience. A number of explanations have been suggested to explain participants’ behaviour:
- Personal responsibility is diffused to the ‘experimenter’.
- The perception of legitimate authority leading to a state of mind in which the participant sees himself or herself as an agent of the experimenter (the agentic state).
- The ‘foot in the door’: participants are lead in to the experiment by a series of escalating demands.
More recently, psychologists have re-examined the participants’ own justifications for why they obeyed. These could be broadly divided into three groups (Sabini, 1995):
- The legitimacy of authority. This could not have been the only reason, however. Recall how obedience rates were still high in the variation conducted in a ‘downtown’ office.
- A belief in the ‘fairness’ of the experiment, because the role of teacher and learner were apparently decided by drawing lots. Again however, this can’t be the only reason. When Milgram set out to make the experiment less fair, by having the learner inform the teacher that he had a heart condition, obedience levels dropped, but were still surprisingly high.
- Denying responsibility for shocking the victim, instead passing it to the authority figure. Although research has shown that people in organisations do tend to pass responsibility to superiors, most of us would not accept this as a legitimate excuse.
Whatever we think of their ‘excuses’, the fact is that participants came to believe that the situation was legitimate and fair, and that they could escape responsibility for their actions. Sabini points out that the reason for obedience lies in participants’ interpretation of the experiment. They thought this way because:
‘…this is the interpretation the experimenter offers them. The participants see the experimenter as the authoritative interpreter of the objective moral order. This is where the social influence lies.’ (1995, page 54.)
Evaluation of Milgram’s studies
In addition to the serious (and fairly obvious) ethical concerns raised about Milgram’s research, a number of methodological issues have been raised as well as concerns about the extent to which the results can be generalised to wider settings. Orne & Holland suggest that the main failing of Milgram’s research are that it lacks:
- Experimental validity (or experimental realism): for example were the participants actually taken in by the scenario? Did they actually believe that they were administering real shocks or were they just ‘going along with the experimenter’?
- Ecological validity (or mundane realism): for example, can the results be applied beyond the laboratory setting that Milgram used; were his samples representative, etc?
Zimbardo’s research
In this very well known example of social influence research, Zimbardo was concerned with the extent to which an individual will be conform to a situation: how far they could be persuaded to act out (i.e. obey) a role. The so-called ‘Stanford Prison Experiment’ (Zimbardo et al., 1973) showed that in the context of a mock prison, college students randomly assigned to be ‘guards’ engaged in substantial brutality towards other college students randomly assigned to be ‘prisoners’.
Watch a video of the Zimbardo study at: Google Video or YouTube
One explanation for the behaviour observed in this study was the process of disinhibition. In other words, normal inhibitions against this sort of behaviour were overridden by a number of factors: dehumanisation of the victim, moral drift — the ‘bad guards’ influenced the ‘good guards’ to see escalating brutality as a norm — and deindividuation (see next section). Sherif’s work on group consensus in ambiguous situations could be relevant here, since participants had been placed in a highly unusual situation and would naturally look to others for a lead in how to behave (informational influence).
The main outcome of Zimbardo’s work, is claimed to be that he showed that giving people power can have a great effect on their behaviour, leading them to behave in ways that they previously thought unacceptable. The problem is that the mock warders may simply have been play-acting; in a real situation they may have behaved quite differently.
There are also serious ethical issues with conducting an experiment in this way. Can an experiment be ethically justified when four participants had to be released because of extreme depression, disorganised thinking, uncontrollable crying and fits of rage?
Back to AS Psychology FrontPage