Readers of my blog posts may feel tired about my recurrent blasts against the dual systems approach. But I think that this is a necessary effort, given the strong impact of these models on behavioural economics and, ultimately, on policies based on this. The problem is not just one of methodology and descriptive and explanatory validity: We also face fundamental issues of ethics of research. These issues are important for our INSOSCI project.

Dual models of mind typically assume that one part produces errors, dysfunctionalities and misbehaviour, whereas the other part would enable us to act in a ‘rational’ way. Thus, the dual models have a very strong normative flavour. Now, the idea is that our ‘better self’ is too weak to overcome the bad influence of its counterpart. Therefore, it needs help! Rescue is there: The scientist who designs certain measures that nudge us to the rights decisions. Therefore, it is easy to see why the dual approach is so fundamental for the theory of ‘libertarian paternalism’: Since the rational solution is part of ourselves, the scientist can claim to know this solution and designs certain interventions that help us to act in our own best interest.

In a unified model of mind, these claims of behavioural economists would simply be groundless: They could only state that people are irrational, and that we should intervene to make them more rational. Thus, in this case the intervention would be clearly authoritarian and anti-liberal. Further, the standard of rationality would be defined by the scientist, which would obviously contradict the fundamental stance of methodological individualism of economics. Therefore, as standard economics builds on a unified model, it was consequential, for example, to explain addiction as rational behaviour, with the policy implication that either you need to punish it if you want to change it (defining it as deviance that is not desired by society) or legalize it, if you think there is no damage to society, or that you can design mechanisms of internalization of all costs. You would not try to change people’s behaviour by stealth. As has been noticed by many observers, a unified model would only justify interventions in the form of giving information or perhaps of persuasion.

A unified model that would move beyond the standard model seems to face considerable troubles in making any normative claims, because it would not axiomatically assume that free people do what is best for them. Of course, many scientists may think that they would not make any normative claims at all. But there is the clinical normativity, that is, normativity based on the concept of health, which, at first sight, is not a normative issue in terms of ethics or societal values. But health is a multidimensional and complex concept, and it includes psychic health which may be influenced by many social factors, and thus raises also ethical normative concerns. In a previous blog post, I discussed the notion of ‘taking care of your self’, which would be built on the idea of a ‘good life’.

Establishing the idea of one’s own ‘good life’ is what distinguishes a cross-disciplinary unified model from the economic one. The difference lies in the role of individualism: Economics takes individualism as the basis for normative arguments, whereas since Aristoteles the idea of ‘good life’ is seen as being about individuals living in communities. That would imply that it cannot be the scientist who defines what is the best for individuals, but the communities in which they live. These are very tricky issues. Take, for example, addiction. Most societies in the Western world have turned against smoking within a very short time, whereas alcohol is still acknowledged as more acceptable. Sugar is not included in the debate at all, but there may be similar issues of public health as in the other cases of problems generated by consuming certain substances, and also behavioural phenomena similar to addiction. Certainly, we have scientific criteria of defining addiction, which do not include sugar, but they change through time, depending on our knowledge, and also on prevailing social standards.

Therefore, I think that scientists need to recognize the complexity and multidimensionality of human notions of ‘good life’. They can point to certain detrimental effects on physical health, but social consequences may be already beyond their conceptual reach, as these are also determined by the cultural and institutional context. They should not try to be our better self, but help us to nurture our capability to take care of our self. They should not make us believe that we are split into two.