Breaking the Rules of Reality

Garfinkel's breaching experiments are pretty famous in the sociology world, but what exactly are these “breaching experiments”?  Essentially, Harold Garfinkel was a pretty cool sociologist guy who developed endomethodology as a field of inquiry in sociology.  Garfinkel was interested in how social structures are ordinarily and routinely maintained, which most sociologists would examine by selecting some set of stable features of an organisation of activities and look for variables that contribute to their stability.  Garfinkel thought it would be more economical to start with a situation of stable features and ask what can be done to make for trouble (Heritage 1984, p. 78).

Garfinkel decided to conduct a series of breaching experiments, where he broke the normal rules of a situation and wanted to see create a senseless situation.  The simplest experiment was where the experimenter played a game of ticktaktoe (noughts and crosses) with a subject.  The subject made the first move, and then the experimenter erased the subjects first move, moved it to another cell and marked his own, without ever indicating this was unusual behaviour.  There were 253 experimental trials and in 95% of these, the subjects reacted to this action in some way.  75% of subjects objected or demanded an explanation and the rest assumed there was some sort of new game in progress and adapted to the new “rules” or assumed it was a practical joke.  Essentially, the subjects abandoned the “noughts and crosses” framework and adopted a new interpretive framework to make sense of the situation.  In contrast, people who held on to the idea that noughts and crosses was still in play were pretty disturbed.

From this experiment Garfinkel made two significant conclusions; one that when there were behaviours that didn’t comply with the basic rules of the game people tried to treat it as a legally possible event.  The other conclusion he made is that disturbance was increased if people tried to keep the original set of rules.

So this is pretty interesting, and Garfinkel though so too, but he didn’t think he could take his experiments into real life situations.  Games have obvious rules that can be followed or broken, and life is a lot more complicated.  Also, breaking rules in real life might have more serious consequences.

So he didn’t get his hands dirty, Garfinkel got his students to do the experiments.  The most famous experiments, and the ones mentioned in the reading, are when Garfinkel requested his students engage an acquaintance or friend in an ordinary conversation and, without indicating what the experimenter was saying was in any way out of the ordinary, to insist that the personal clarify the sense of his commonplace remarks (Garfinkel 1963, p. 221).

One of my favourite examples is when an experimenter was in a carpool and the subject said they had a flat tyre.  With the simple question of, “What do you mean a flat tyre?” The subject got really upset, saying “What do you mean? What do you mean? A flat tyre is a flat tyre.  That is what I meant.  Nothing special.  What a crazy question!”

Just looking at this experiment it is clear that the interaction broke down incredibly quickly, apparently even Garfinkel was surprised and how rapid and complete the breakdowns were (Heritage 1984, p. 81).  This breakdown was not just because the subject didn’t want to answer the experimenter’s question; it was because if we can’t bring background knowledge to a situation and obey conversational rules, then mutual understanding and the entire shared experience of life could collapse at any moment (Heritage 1984, p. 95)!

As much as people were very upset at the slightest breaking of perceived normality, Garfinkel couldn’t create a situation that was utterly senseless.  People seemed to be less distressed, indignant and less likely to demand an explanation if they could make sense of the situation in some way (Heritage 1984, p. 83).  So Garfinkel found that all actions, whether in a simple game or a social interaction, are perceived to have a constitutive structure and a threat to the normal order of these events is upsetting because if we don’t behave according to the rules social organisation simply disintegrates (Garfinkel 1963, p. 198).

After some YouTube searching, it seems like performing breaching experiments is a pretty common task for sociology students and people who enjoy creating a little chaos.  So I’ll show you two examples that I thought illustrated the idea of what a breaching experiment is quite nicely.  

In this clip it’s interesting that the lecturer and the audience respond by laughing, which suggests they adopted a framework of thinking that this person is doing flips for their entertainment.  They don’t descend into chaos or assume that it was normal lecture behaviour.

This clip is a fantastic breach of normal behaviour as it’s almost as simple as the noughts and crosses example.  Instead of going down the escalator, the person runs up it.  You probably noticed in this clip too that people didn’t all start running up or go crazy, they just moved to the side and probably assumed the girl was playing a joke of some kind rather than believe a world eists where the down escalator wasn’t down any more.

Another famous breaching experiment was conducted by Stanley Milgram in the 1970s.  Milgram instructed students to ride the New York Subway and ask people to give them their seats.  This experiment was interesting, because not only did many people give up their seats when someone asked with no justification, the students experimenting found the task emotionally very difficult (Luo 2004).  In an article on the experiment in the New York Times, interviews with the students many years later show that they found the experiment traumatic and some were physically sick at the idea of violating the rules of behaviour expected on the subway.  This is very interesting as it suggests that breaking the perceived normality is very distressing even when you do have the framework to know what is going on.  It’s important to note that although both experimenters and subjects in the breaching experiments got upset, angry or even felt sick, there was never a situation where people just decided, “OK, this makes no sense and I can’t comprehend this.”

Reference List:

Garfinkel, H. 1963 ‘A Conception of and Experiments with ‘Trust’ as a Condition of Concerted Stable Actions’ in O’Brien, J (ed.) The Production of Reality: Essays and Readings on Social Interaction, vol. 4, California, Pine Forge Press, pp370-81

Heritage, J. 1984, ‘The Morality of Cognition’, in Garfinkel and ethnomethodology, Polity Press, Cambridge, pp75-102.

Luo, M 2004, ‘Excuse Me. May I Have Your Seat?'; Revisiting a Social Experiment, And the Fear That Goes With It’, New York Times, 14 September, accessed 02/09/2012, http://select.nytimes.com/gst/abstract.html?res=F20F15FE3A540C778DDDA00894DC404482