The limits of computer simulations as epistemic tools

Society & HLRS
The limits of computer simulations as epistemic tools

Juan Durán / Universität Stuttgart

The use of computers simulations in laboratory practice has turned out to be of central importance for the scientific activity. How is this dependence understood by philosophers? One possible analysis consist of discussing the epistemological payoff of computer simulation. Ironically, current philosophical positions seems to be stuck in a dilemma that is difficult to solve.

The dilemma of computer simulations centers on the question whether the epistemological payoff of a traditional experiment has greater (or less confidence than a computer simulation. The so called “materiality problem” is in the basis of this dilemma: “in genuine experiments, the same ’material’ causes are at work in the experimental and target systems, while in simulations there is merely formal correspondence between the simulating and target systems (...) inferences about target systems are more justified when experimental and target systems are made of the ‘same stuff’ than when they are made of different materials (as is the case in computer experiments)” (Parker, 2009 “Does matter really matter? computer simulations, experiments, and materiality”). It is the logic behind this characterization that must be rejected: it presupposes that once the “materiality” of computer simulations is settled, it will serve for the determination of the epistemic power. In other words, the ontological characterization of computer simulations will determine its epistemology.

A different approach would be to defend that the epistemic reliability of computer simulations is philosophically detached from the materiality problem. This does not suggest, though, that they are two unrelated issues, but only that they are independent from each other. In fact, they are related insofar the materiality problem becomes, to certain extent, a limiting case for the epistemology of computer simulations.

In this presentation I would like to discuss two claims: firstly, that materiality only restricts computer simulation from “accessing” certain aspects of the world which require a causal story; in other words, materiality draws the boundaries from where experiments become a specific and irreplaceable method for knowing something about the world. Secondly, that computer simulations provide ways of inference that do not depend on its materiality but on its capacity for representing empirical as well as non-empirical systems.