Hard Choices: Why do we do what we do?

Imagine a psychological experiment. You set up a “shop” where the people taking part in the experiment have to purchase a washing power from a choice of 12 available. They have to pay for the washing power with their own money, but they get to keep it after the purchase. The psychologists running the experiment are interested in exploring the impact of four potential factors on human choice: 1) price (there are twelve possible prices in a fairly narrow range); 2) brand (there are 12 possible brand names with associated marketing blurb); 3) visual aspects of the packaging (there are 12 possible colour schemes) 4) display layout (there are 12 possible positions a product can occupy in the display). You then run the experiment with thousands of different individuals, offering each individual a choice where the four factors are combined on a random basis. This generates  a large amount of data about people’s choices.

If we ask the subjects of this experiment, why they made the choices they did, we would certainly expect them to have answers. I imagine these answers would focus on price (“I choose the cheapest”, “I thought the safest choice was the most expensive”) or brand promise (“I chose the one that the won the poll of housewives”, “I bought this one because it contains a new ingredient called “clearon”, “The one I bought is backed by scientific research”) or visuals (“I just liked the look of this one”, “I didn’t know what to choose so I went with my favourite colour – blue”). This is our usual language-game of offering explanations for the sort of choices we are all continually making. It wouldn’t really make sense to ask the people concerned: “how do you know you bought it for the reason you state?”. It would also be a bit odd to ask them for the evidence that justifies their statement. It would certainly be very misleading to suggest that what happens is that they look inside themselves and then give the best possible report of what went on inside them. No, the truth of the matter is that generally we just expect people to be able to say why they did things. This is how we are and this is how the language game works.

The psychologists, however, analyse the data and it may well generate some very interesting results. For example, their analysis may demonstrate that in the particular context of this experiment (perhaps because of the limited range of the prices or the particular nature of the layout) layout was a much stronger predictor of choice than price. So does this mean that some (or all) of the people who said they based their choice on price were wrong? Or let’s suppose that the variations in brand name and marketing blurb had no statistically significant impact on the choices people made. Does this give us the right to contradict the person who cited the marketing blur as the reason for their choice and to claim that in fact the real reason he chose the product he did was because it was in the most prominent position in the layout? Do such results provide further evidence that we don’t know as much about our own actions and choices as we think we do?

Well, what we have here is two very different ways of getting a result and since the processes are different, we shouldn’t really be surprised that the results might vary. Probably, most of us would accept that in this sort of context there are many factors that influence our choices in ways we do not recognise. But it would be eccentric (to say the least) to see such experiments as a reason for giving up our usual language game entirely. We might change our detergent buying practices (“I always decide what brand I am going to buy before I get to the supermarket so I cannot be manipulated by the display layout”), but we are not going to stop generally asking other people why they do things on the grounds that they are blind to the real determining factors. Similarly, if someone asks we why play tennis rather than golf, we are unlikely to reply: “how should I know? the psychologists haven’t yet completed their statistical analysis”.

Imagine one of the subjects of the experiment then goes to see his psychoanalyst and they discuss the choice the analysand made. The person chose the cheapest product available to him and says he only chose on price because there was no other reasonable basis for making a decision. But the analyst doesn’t find this account very convincing. After all, the individual is fairly well off and he doesn’t usually go in for penny pinching. So they talk a bit more. Who usually makes this sort of purchase in the analysand’s household? As it happens, it’s the analysand’s wife. And come to think of it, they recently had a row about this expensive holiday his wife splashed out on that the analysand thinks is too expensive and not really what he wanted to do anyway. So yes he was angry with his wife, although it is no big deal really. However, after the discussion the individual sees his choice differently. He still sees the price as influencing his decision but not because he wanted to save money or get good value, but because he was more angry with his wife than he realised. He now believes that unconsciously his purchase was a way of getting back at his wife and telling her that if she was going to waste money on expensive, selfish holiday choices then she would have to make do with the cheapest (and hopefully worst!) washing detergent.

So now we have a third process for arriving at result and again not surprisingly because the process is different, it can lead to different results. As with the second possibility, this third possibility does not automatically consign our normal language game to the rubbish heap. In some ways it is a less radical challenge to our normal language game than the psychologist’s approach, in some ways, a more challenge. We don’t want to be told that we did things for very different reasons than the ones we think motivated us, and there is a tendency to want to say: “Hang on. These are my actions and my experiences, so my account is definitive”. But in fact the normal language game is not quite as straightforward as I earlier made it seem. If a teacher says, “I am only giving her special help because she is clearly struggling to understand the material”, we may accept the utterance as sincere, but think the teacher is deceiving himself especially when it becomes clear that he only give special help to beautiful female students. Or if someone’s elderly father dies and they say: “I was expecting it, so it had no impact on me at all”, we may notice other aspects of their behaviour that point towards a different conclusion. Indeed, a month (or a year or 10 years) later, the individual himself may say: “at the time I didn’t realise how hard his death hit me”.

So it is not the case that in the normal language game the subject/speaker is beyond challenge and can never be wrong. The game is more sophisticated than that. Having said that, a game of reasons that includes unconscious reasons is a new game – it may build on possibilities in the old game, but if you allow all the players and not just the goalkeeper to use his hands, then you are not playing football any more. But again even if you find it interesting or useful to play the new game, there is no reason why you have to play it all of the time or even most of the time. After all, even psychoanalysts spend most of their time leading fairly normal lives!


  1. [...] Hard Choices: Why do we do what we do? is a post from: Wittgenstein and Freud [...]