I teach research methods courses and when it comes to sessions on ethics we normally go over the classics: Milgram’s experiments with obedience; the Stanford prison experiment; poor Alfred and the white rat [1]. All very startling but on the whole students do not get as excited as I do. I think this is first because most have heard of these stories before and second they know they could never do something as off-the-wall as these researchers. So they don’t always see the point in talking about them. Well the point, as I try to make clear, is that however extreme we find these cases today they did not seem so when they were carried out. For example, Milgram’s experiment in obedience became championed in counter cultural politics of the 1960’s and 1970 but as far as I know few people at the time said, ‘Hey wait a minute you cannot ask people to believe they are inflicting electric shock treatment just because you want to find out if they will do it!’.

So I don’t want to throw out the classics, but I would like to broaden my examples and find something a little less obvious. So it was that the other day I looked up Garfinkel’s reporting of experiments ‘in trust and stable actions’ in the early 1960’s [2]. First thing to say is that Garfinkel’s work was insightful. His career long research theme was the maintenance of order in social interactions and the role that conversation played in keeping a tight hold on the roles we could play and how we played them. To illustrate this passion we had for order in our relationships he asked (and it is not clear whether he merely suggested or told) his students to act in unexpected ways in keeping with the old adage, ‘if you want to understand something then try to change it’.

One of his examples was asking students to enter a store (or shop), to select a customer and treat the customer as a clerk (shop assistant) while giving ‘no recognition that the subject was any other person that the experimenter took him to be and without giving any indication that the experimenter’s treatment was anything other than perfectly reasonable and legitimate’. In the examples he describes the ‘subjects’ (or, let us be honest, the duped customers) becoming ‘nervous and jittery’, one was ‘flushed with anger’ and another ‘stalked out of the shop’. For good measure one volunteer student had a friend ‘a professor emeritus of mathematics at the California Institute of Technology’ (so that is alright then) who ‘begged to be allowed to accompany the student’ and joined in an experiment of his own.

Garfinkel followed these relatively mild and short term experiments with a more well-known one in which students were asked to spend from fifteen minutes to an hour in their home acting as ‘if they were boarders (lodgers)’; by this he meant they should act the role of being ‘circumspect and polite’, using ‘formal address and speaking only when spoken to’. He writes that out of 49 students, five refused to do it and 4 ‘were unsuccessful’ (i.e. they were willing but the circumstances did not seem right). But four fifths of the students did try the experiment and Garfinkel reports that family members were ‘stupefied’ and ‘vigorously sought to make the strange actions intelligible and to restore the situation to normal’. He elaborates that students’ reports were ‘filled with accounts of bewilderment, shock, anxiety, embarrassment, and anger, and with charges by various family members that the student was mean, inconsiderate, selfish, nasty, or impolite.’

There were however  no cases in which the situation was not restorable after the student had explained what they were doing, but family members were angry. As for the students, they did not for the most part did not find the experiment difficult to carry out or particularly taxing.

For Garfinkel these examples worked, i.e. they showed we were made uncomfortable when what was taken for granted was disrupted. But were they ethical? No I don’t think so. They may have been comparatively mild cases but this was reckless for the families and for the students concerned and I would say that family members had it right when they complained that the students were ‘mean, inconsiderate, selfish, nasty, or impolite’.  And this actually makes me question the conclusions to be drawn from the experiments, people may have been angry, not so much because they were over-committed to the maintenance of social order, but because the student volunteers were taking the piss; as one family member said to a student ‘we are not rats’. But my overarching complaint is why would you ever want to do social science in this way? Yes, if you want to understand something then do try to change it, but why not try to change it for the better. For example, there is all manner of research into what goes disastrously wrong in families and relationships and really worthwhile attempts to explain what helps in redressing an imbalance [3]. Ethically this seems where social research should be positioned.

[1] In Milgram’s experiment volunteers were told they were assisting in a learning experiment and told to administer an electric shock every time a learner made a mistake in a test (in fact there was no shock and the ‘learner’ was playing a part). The shock generator was marked from 15 volts (slight shock) to 450 (danger – severe shock). The study was about how far volunteers would go. See:

https://www.youtube.com/watch?v=mOUEC5YXV8U

In the Stanford prison experiment the psychology building was turned into a mock prison and 24 paid male volunteers were assigned roles of guard or prisoner in order to explore the impact of taking on a role on otherwise well adjusted men. Some got carried away in their role. 

https://www.youtube.com/watch?v=F4txhN13y6A

The third example concerned Watson and classic conditioning. Little Albert was around 9 months old and exposed to different stimuli and showed no fear of any of small animals including a white rat. However the next time Albert was exposed to the rat the researcher made a loud noise and thereafter the child associated the noise with the rat and would cry on seeing it.

https://www.youtube.com/watch?v=9hBfnXACsOI

[2] Garfinkel, H. (1963) ‘A conception of, and experiments with, ‘trust’ as a condition of stable concerted actions’, in O.J. Harvey (ed.) Motivation and Social Interaction: Cognitive Approaches, New York: Ronald Press.

[3] As one example, Yoshihama, M. (2002) ‘Breaking the web of abuse and silence: Voices of battered women in Japan’, Social Work, 47, 4: 389-400.