Daniel Kahneman's latest work is more than just a book written by a psychologist and behavioural economist on the ways that people think. It reads like a detective story, but with each chapter offering up a discovery that completely changes the plot we will have had in mind up to that moment. Only here, the plot concerns not a crime but us, the readers of his book: we are the main characters of Kahneman's story. We cannot be innocent readers of his narrative; we are immediately drawn in and discover more about ourselves as we turn each page. Moreover, we cannot escape it: those biases and heuristics, ie, simple procedures that help us to find satisfactory, but often imperfect, answers to difficult questions, stick with us. The book's suspense lies precisely in the fact that we may have to surrender to the post-Enlightenment idea of our own imperfection - we are not perfectly rational individuals in the way that most economic theory, for example, wants us (and describes us) to be.
Let me give some telling examples. According to Kahneman, we are prone to an optimistic bias that is a driving force of capitalism. In general, we believe we will succeed with our endeavours, even where statistics tell the contrary. For example, only 35 per cent of new small businesses in the US survive for five years, and only 40 per cent of new restaurants are still in business after three. Not only are we overconfident about our qualities and prospects, but we also see some of our traits as good qualities when they are anything but. Kahneman describes an experiment that tests people's inclination to help others: a man (played by an actor) cries for help as he gives the impression of choking as he has a seizure. Of a group of 15 participants in the experiment, Kahneman recounts, only four rushed to help. The exercise suggests that individuals feel relieved of responsibility when they know that others have heard the same plea for help. There may be other explanations for their behaviour, too, but the fact remains that more often than not, people do not come forward.
These people who fail to help are, of course, us. We need to accept that we may be eager to open our own business and still fail, and that we may also fail to help others in need even though we contend that we obviously would come to their aid. Accepting this insight is difficult, and most of us do not. In another experiment described here, participants are shown pictures of people and are asked to say whether these people would be likely to help a choking person. Participants are inclined to say that these particular people would have done so, despite being told of statistics that indicate the contrary. What is going on here?
Kahneman explains much of our behaviour via two constructs he calls System 1 and System 2. The former represents a bias to believe and confirm, overlook ambiguity and suppress doubt, create coherent stories by associating ideas, operate automatically and quickly, and jump to conclusions. This is "fast" thinking. System 2, on the other hand, is slow. It represents effortful thinking. It monitors System 1 to some extent, but it is lazy and insensitive to the quality of evidence suggested by System 1, and therefore is not always successful in correcting the errors of System 1.
Thinking, Fast and Slow offers an intriguing portrait of how Kahneman and colleagues have helped to redefine not rationality but irrationality. By shedding light on our behaviour and discovering how the human mind works, in many places they expose the purely rational individual as a fictitious benchmark. Taking their discoveries seriously would require the reformation of social institutions and the way we think of human interaction. Yet the irony is that Kahneman removes one fiction by reintroducing another. System 1 and System 2 are tools that help to represent why we behave the way we do. But as with all fictions, and especially "dualistic fictions", while they help us to understand what we do, they hide the insight of who we "really" are. We think, Kahneman seems to suggest, that "we" are System 2. But why do we think this? Why is it that we develop a sense of identity when we exert effort, for example by calculating 17 x 24, but we feel no particular connection (and can even feel alienated) when we hear about the "subversive" System 1 that torpedoes our textbook rationality? Perhaps one day Kahneman will respond to these questions, holding out the prospect of another interesting book.
Thinking, Fast and Slow
By Daniel Kahneman
Allen Lane, 512pp, £25.00
Published 3 November 2011