422. Major and minor science

From Deleuze and Guattari, DeLanda[i] adopts the distinction between ‘major and minor science’. Major science is characterized as a more or less tight deductive, axiomatized system, preferably formalized with mathematics. Minor science lacks that, is looser, less structured, and is more inductive, messy.

DeLanda showed how chemistry, in contrast with physics, used to be minor, with a proliferating population of chemical substances, and became major with the adoption of the Periodic Table, formulae for molecular structure and nomenclature of substances based on that. 

In economics one finds the contrast between mainstream, neoclassical economics, which is highly deductive, axiomatic and mathematical, and more inductive, informal, economics of institutions and organization.

Interestingly, part, and perhaps the crux of the difference, offered by DeLanda, is that major science is oriented at the stable, and minor science towards the dynamic, processes of change. I find this interesting in the light of an experience I had as director of a research/PhD school at the University of Groningen, the Netherlands, in the 1990’s.

I was given the task of aligning the faculties of economics and business in a joint organization. It was an almost total failure, but an interesting one, since it raised the question why this was so.

One feature was, I discovered, that business/management is oriented towards processes, of production and development, while mainstream economics is oriented at equilibrium outcomes.

Mathematization and quantitative, econometric testing were possible for the second  but not for the first. Therefore, neoclassical economics carried the most prestige, and won. The process led not to integration, a coupling between the two faculties but to a take-over by economics, but by that time I had left.

There was a similar outcome concerning a Max Planck Institute for Evolutionary Economics in Jena, which was abolished in favour of a take-over by established, mainstream neo-classical economics.

Evolutionary economics, similarly to organization theory, is process-oriented. Rather than being oriented at equilibrium outcomes, it is oriented at evolutionary processes that may or may not, and in general do not, yield equilibria. As a result, predictions and implications were less clear and unambiguous, depending on details of the evolutionary processes of variety generation, selection and transmission of success. That was less respectable. It was a minor science. 

A way out for process research is computer simulation, enabled by the development of appropriate hard- and software. There, one can model and simulate out-of-equilibrium processes on the basis of what is known as ‘agent-based simulation.[ii]  

The problem there is lack of determinacy, with outcomes sensitive to small changes of parameter settings, and an explosion of complexity of what is going one with an extension of the number of interacting variables. With n variables there are n(n-1)/2 possible binary combinations, so that for ten variables there are 45 possibilities. And to that one must add triple and more interactions, and ranges of the values the variables can take. 

Therefore, to make sense and allow for interpretation, simulations need some anchoring in the use of analytically derived equilibrium outcomes of different settings, as a benchmark to compare the simulation outcomes with.

With that, process study becomes more ‘scientific’, in the sense of determinacy and rigour of interpretation, but it does not thereby become a ‘major’ science in the sense of axiomatic, deductive structure. 
    

---

[i] Manuel DeLanda, 2016, Assemblage theory, Edinburgh University Press.

[ii] I had a PhD project and a postdoc project in that area.