Paul Cockshott — Competing Theories: Wrong or Not Even Wrong?

Liberal economics has been able to claim scientificity based both on the large and sophisticated mathematical apparatus of neoclassical value theory, and on a vast number of detailed econometric studies. Those who are professionally involved in the subject are expected to be mathematically literate and experi- enced in the analysis of statistical data. These aspects of their training means that their background has in some ways more in common with people who are trained as natural scientists than with other social scientists. There has also been a long tradition of economists borrowing conceptual structures from the natural sciences. Mirowski showed that many of the concepts used in marginalist economics were borrowed directly from classical mechanics during the late 19th century[Mir89]. But there is, I think, a significant difference between the way the natural sciences are taught and the way neo-classical economics is taught, and this difference is significant.

When a student is taught an introductory course in physics or biology, they are both taught theories and told of the crucial experiments that validated the theories. They are told of Galileo's experiment that validated what we would now see as the equivalence of gravitational and inertial mass. They learn of Michelson Morley's experiment on the invariance of the speed of light, that inconvenient fact whose explanation required Special Relativity. Biology students hear of the experiments of Pasteur and Koch that established the germ theory of disease, etc. The function of these accounts in science education is twofold. On the one hand they emphasize to students the reasons why they should give credence to the theory being taught, on the other, these historical examples are used to teach the scientific method.

If one contrasts this with introductory courses in economics one sees that whilst theory is taught, the student gets no equivalent history of crucial economic observations in order to support the theory. This is no accident.

No history of crucial observations is taught, because there is no such history.

Competing Theories: Wrong or Not Even Wrong?
Paul Cockshott | School of Computing Science, University of Glasgow