Last night I had dinner with tonight’s speaker at the RSA, Jonah Lehrer (shameless namedropping alert). First off let me say he is a very friendly and humble guy, but that this persona belies a prodigious, keen and subtle intellect. That’s a rare combination of characteristics in a person and one can see it at work in Jonah’s writings: they are accessible but have depth; they are scientific without being baffling to the layperson; and they are philosophical without being haughty or pretentious.
I’d like to write about something we talked about briefly last night.
There is at the moment a slew of research in different areas that comes on the tail of the collapse of rational choice theory. But there is no new model of agency and decision-making to replace the old one and tie the research together. As Matthew Taylor points out in his blog, to provide such a model (or at least the beginnings of one) is the aim of the Social Brain project.
The following might seem tangential, but bear with me.
Nietzsche famously wrote: “I am afraid we are not rid of God because we still have faith in grammar.” Presumably the idea is that grammar contains things like absolute phrases, such as the apposite clause in this sentence: ‘Gordon Brown surveyed the economy, dread spreading upon him.’ The absolute phrase gives the impression of being spoken by an omniscient voice from nowhere. Hence Nietzsche is at once pointing to a mundane source of the belief in an omniscient being, and showing why we might continue to believe in Him.
Nietzsche said similar things about the ‘Subject’, which finds a ‘firm form in the functions of language and grammar’. The Greek word for subject is 'hypokeimenon' – meaning that which underlies and keeps something in being. The Ancient Greeks derived this concept from that of a grammatical subject (such as ‘Gordon Brown’), which can have things and attributes predicated of it (‘Gordon Brown is very worried about the economy’), but which cannot be predicated of other things (‘The economy is Gordon Brown’) – well, not without breaking the rules of grammar anyway.
The monotheistic conception of the soul is strongly influenced by the idea of a hypokeimenon, and further by the idea of the self-conscious subjects that are often taken to be the grammatical subjects of sentences. So this idea goes very deep in our culture.
One thing we hear a lot about at the moment is how ‘irrational’ 'we' are. Or how ‘we’ don’t make half the decisions we think we do. Rather, the brain makes them for us, via unconscious cognition. Jonah’s book brilliantly spells out the facts that inform these pronouncements.
What Jonah and I talked about briefly last night was this: why do we call it irrational to (say) find immediate rewards gratifying, rather than delay our gratification? Sure, it is irrational to announce we think we should delay gratification and then not in fact delay it, because that is inconsistent. But behavioural economists like Pete Lunn or Robert Shiller will talk about our being irrational just because we are driven by our emotions to want immediate reward. But if ‘we’ are not the always self-conscious, unemotional selves of rational choice theory, this doesn’t make sense.
On the emerging model that Jonah writes about, emotions help constitute rationality (help us decide in light of appropriate reasons), and there are many reasons why ‘we’, now conceived as far less in conscious control of our decisions, might prefer immediate reward: in an unstable and unpredictable world it is perfectly rational (there is good reason) for our brains to have evolved to direct us to think in this way.
I fear we are not giving up rational choice theory because of an ingrained association of the grammatical subject ‘we’ with the idea that self-conscious subjects underlie all decision-making. Moreover, understanding often flows from comparison – so it is by saying emotional cognition is irrational in comparison to such subjects that we come to understand how it is differentiated, and thus how it is defined.
But there may be an even deeper reason. Neuroscientists like Chris Frith have argued that the ‘rational’ and wholly self-conscious self is an illusion, albeit a necessary one, that the brain produces (see Matthew Taylor’s talk on ‘neurological reflexivity’). They argue it is far more effective in evolutionary terms to have a subject that conceives of itself as almost always in self-conscious control. What will be the effect of the new model of decision-making and agency the Social Brain is investigating on that conception?
My guess is it will not be dislodged completely but modified: we will recognise that we are not the selves of rational choice theory, but rather the selves our brain produces through a complex of different cognitive processes. As Jonah argues, we should trust our emotions because they in large part constitute us as authors of our decisions (remembering that the meaning of ‘we’ has changed). But we - meaning the other elements of our selves, the more reflexive self-conscious elements - can still work with the more unconscious elements in order to shape our lives.
The big difference will be in how we think about this shaping process. It is a misstep to think of it as happening outside of emotion - of the rational part of the brain as separated from the emotional part in self-conscious reflection and awareness. I think the right way to think about it is this: we learn when to recognise where we are going wrong and stop and think about what we are doing. But we learn to do that through practice and habit, and what alerts us to the mistakes are emotions - we get the feeling we should stop and think about a particular decision.
And now I'll stop. One problem with this post has been finding the language to write it in!