This week sees the start of three trials in the US all related to corporate malfeasance during the bubble years. Bernie Ebbers, formerly of WorldCom fame goes on trial in New York. Dennis Kozlowski, formerly head of Tyco will also enter the dock for his second trial, the first having ended with a hung jury. Down in Alabama, Richard Scrushy, former head honcho of HealthSouth will also be appearing before the courts. Also in court in the US are some more of the soldiers accused of abusing prisoners in the Abu Ghraib prison.
Our own press here in the UK are full of stories of prisoner abuse by UK soldiers. What could connect these events? The answer is that human nature lies at the root of both sets of problems. In order to understand what drives this behaviour we need to outline two diametrically opposed views – the situational and dispositional perspectives. Those who subscribe to the situational viewpoint argue that if you put good people in bad places, they tend to become bad. That is, it is the situation that determines behaviour. Those who follow the dispositional school believe that people’s actions reflect their attitudes, or that it is people’s nature that determines their behaviour.
Fundamental attribution error
We all tend to believe that our own actions are the considered reflective outcome of our analysis of the situation. At the same time we believe that others’ actions are reflections of their underlying disposition. We see ourselves from a situational viewpoint, and others from a dispositional perspective. Psychologists call this the fundamental attribution error (also known as actor/observer bias). Nisbett et al[1] were the first to observe this schism in beliefs. Nisbett and his co-authors asked male students to write four paragraphs on why they had liked the girl they had dated most regularly in the past year or so, and why they had chosen their particular major at university.
They were then asked to repeat the process but pretending to be their best friend. Finally, they were asked to repeat the process pretending to be their best friend being them (i.e. seeing themselves through their best friends eyes). The sentences they had written were then scored on the basis of the number of times situational remarks were made (i.e. she’s a relaxing person – a comment about external factors) and the number of times dispositional remarks were recorded (i.e. I need someone I can relax with – a comment about internal factors or one’s self).
Subjects gave more than twice as many situational reasons compared to dispositional reasons when talking about their girlfriends. Subjects gave four times as many dispositional reasons for their friend’s choice of college course than situational reasons. When pretending to be their best friend commenting upon themselves, subjects duplicated the results for when they were commenting upon their best friend’s choices. Ross et al[2] offer an alternative example.
They asked people to participate in a quiz game. The subjects were divided into three groups, questioners, contestants and observers. Questioners had to compose 10 difficult questions to ask the contestants. The contestants, of course, had to try and answer. The final group watched the interaction. The subjects were asked to rate the general knowledge of the questioners and the contestants. The results are shown in the chart below. The contestants’ ratings are consistent with the fundamental attribution error. That is to say, they rated the questioners as having higher levels of general knowledge than they themselves had.
The observer’s scores were even more pronounced in favour of the questioners. Behaviour was essentially ascribed to dispositional qualities, rather than the subject’s role in the study. That is, the participants who asked the questions were seen as having very good general knowledge, whilst those who answered them were seen as having limited general knowledge – their actions and behaviour were attributed to their underlying nature (dispositional view). Whereas in fact they reflected the random role assignment at the start of the experiment (situational factor).
The fundamental attribution error leads us to believe that the various heads of US corporates who are now on trial are essentially bad people. In a similar fashion, it is tempting to believe that the actions of the prison guards at Abu Ghraib are just the limited actions of a few ‘bad apples’.
However, two classic experiments in social psychology argue strongly for a more situational perspective. That is we shouldn’t be so fast to judge the accused as bad or evil people. We simply cannot be sure that we wouldn’t act in a similar fashion if placed in a similar circumstance.
1 Nisbett, Caputo, Legant and Maracek (1973) Behaviour as seen by the actor and as seen by the observer
2 Ross, Amabile and Steinmetz (1977) Social roles, social controls, and biases in social perception processes, Journal of Personality and Social Psychology, 35
By Dr James Montier
Next: Zimbardo's prison experiment
Summary: Index