34I0091
This course is co-taught with Baptiste Le Bihan.
Science produces knowledge by articulating hypotheses about the world, which are then empirically assessed through observations and experiments. However, these direct means of testing may not be the only way to gain scientific knowledge; in fact, several methods of non-direct theory assessment have been proposed. A first example of non-direct testing are computer simulations, which are centrally used e.g. in cosmology and in the climate sciences. But can we really learn something genuinely new about the world from running these simulations? If so, how should we understand the resulting form of knowledge? Should we think of computer simulations as inferences or experiments? Second, material analogue systems constitute a different way in which a system under study can be simulated. For instance, it has been claimed that we may learn something about black holes from studying water in motion in a tub. But how do we know that those material systems adequately represent relevant properties of black holes, i.e., that black holes really exhibit the kind of universal behaviour that is being emulated in these distinct physical systems? Third, can we test a physical theory by checking its mathematical consistency as it has been claimed by some proponents of string theory? The seminar will address these questions and so aims at a better understanding of the nature and status of the non-direct testing of scientific hypotheses.
This seminar will be in English.