Is simulation a substitute for experimentation?

Download Edit this record How to cite View on PhilPapers
Abstract
It is sometimes said that simulation can serve as epistemic substitute for experimentation. Such a claim might be suggested by the fast-spreading use of computer simulation to investigate phenomena not accessible to experimentation (in astrophysics, ecology, economics, climatology, etc.). But what does that mean? The paper starts with a clarification of the terms of the issue and then focuses on two powerful arguments for the view that simulation and experimentation are ‘epistemically on a par’. One is based on the claim that, in experimentation, no less than in simulation, it is not the system under study that is manipulated but a system that ‘stands-in’ for it. The other one highlights the pervasive use of models in experimentation. It will be argued that these arguments, as compelling as they might seem, are each based on a mistaken interpretation of experimentation and that, far from simulation and experimentation being epistemically on a par, they do not have the same epistemic function, do not produce the same kind of epistemic results.
PhilPapers/Archive ID
PESISA
Upload history
Archival date: 2015-11-21
View other versions
Added to PP index
2012-02-05

Total views
720 ( #7,868 of 2,454,489 )

Recent downloads (6 months)
18 ( #34,534 of 2,454,489 )

How can I increase my downloads?

Downloads since first upload
This graph includes both downloads from PhilArchive and clicks on external links on PhilPapers.