(This is a reposted message to new subscribers -- the message was originally sent on 31 May. There have been several exchanges on this issue since then.)

The following is the outline for the talk that Professor Stephen Reed, of Chuo University, Tokyo, gave at the Institute of Social Science on April 20. The talk was intended to provoke discussion, it did, and a resumption of such is most welcome in this forum. Please note that Professor Reed is a subscriber to the forum, so questions can be put directly to him.

-----------------------------------------

Why NOT Rational Choice?

Rational Choice assumes that people are smart and selfish and uses sophisticated mathematics to deduce conclusions from these and a set of other assumptions. This would appear to be a scientific approach to the social sciences. I argue that the path towards more scientific social sciences does not lie with rational choice.

Problems with Rational Choice

1) the underlying assumptions of rational choice are tautological.

2) the assumptions of rational choice are bad psychology. We know that people do not in fact maximize their self-interest and that people make a series of highly predictable errors.

3) rational choice has very few empirical successes to its credit. It does not work. Rational choice explains everything after the fact (because it is a tautology), but nothing before the fact.

The Attractions of Rational Choice

If the above is true, why do so many intelligent people still find it attractive?

1) rational choice is bad science but good math. Economists really deserve those Nobel prizes for math. Physicists have chosen theories that work even when they make no sense. Economists have chosen theories that make sense even when they do not work.

2) rational choice is based on folk psychology, the psychology that most people believe naturally, without reflection, even though psychologists have shown it to be mistaken. In many ways, rational choice is rationalization: human beings have a talent for for making up wonderful reasons for their actions after the fact.

3) rational choice has some attractive normative characteristics. Even if people are not rational, they should be. It has great potential as an application of sophisticated math to problems of political philosophy. Unfortunately, the mixing of normative issues with scientific research is one of the main reasons that the social sciences have failed to become more scientific. Rational choice not only does not help solve this problem, it makes it worse.

4) because rational choice is based on assumptions shared by most people, uses mathematical logic, and has attractive normative characteristics, it is an extremely convincing way to argue. A more scientific approach produces more accurate but less convincing arguments. A physicist trying to influence a Congressman would be well-advised to avoid mentioning the curvature of space and to use out

dated but comprehensible Newtonian physics instead.

-----------------