Thought experiments, computer simulations, and real world experiments in scientific knowledge: a comparison

The purpose of this paper is to restate, in a more concise form and taking into account some articles subsequently appeared in the literature, the main point of a previous article regarding the relationship between real world experiments, computer simulations and (empirical) thought experiments. After distinguishing four main families of accounts which have emerged in the literature, it is argued that they, although each contains an element of truth, have failed to distinguish between real experiment, computer simulation and thought experiment. In fact, concerning the empirical intension of the respective concepts, it is a hopeless task to find a qualitative difference which applies exclusively to thought experiments, computer simulations, or real experiments. For every particular characteristic of one of these notions there is a corresponding characteristic in the two others. However, from another point of view, there is between thought experiment and computer simulation on the one hand, and real experiment (or empirical knowledge) on the other, an epistemological-reflective difference which we must not overlook. Unlike computer simulations and thought experiments, real experiments always involve an ‘external’ or impersonal realisation, namely that of what I propose to call an ‘experimental-technical machine’, always in causal-real interaction with the experimenter’s body. 1 Introduction The purpose of this paper is to restate, in a more concise form and taking into account some articles subsequently appeared in the literature, a point made elsewhere (cf. Buzzoni 2016) regarding the relationship between real world experiments (hereafter REs), computer simulations (hereafter CSs) and (empirical) thought experiments (hereafter TEs). After distinguishing four main families of accounts which have emerged in the literature, it is argued that they, although each contains an element of truth, have failed to distinguish between RE, CS and TE (Section 2). In the second part of the paper, I shall briefly outline my own account on this topic. To avoid comparisons that are insignificant or of little importance for the philosophy of science, it will be convenient to compare TEs and CSs with real world experiments (hereafter REs). To take the notion of RE as the basis of comparison between CS and TE will enable us not only to better understand the methodological similarities between CS, TE and RE, but also to find a subtle but important distinction between CS and TE. I shall maintain Models and Representations in Science, edited by Hans-Peter Grosshans. Comptes Rendus de l’Acad´emie Internationale de Philosophie des Sciences 3 (2025). M. Buzzoni, Thought experiments, computer simulations, and real world experiments, pp. 1–17.