The Monte Carlo method in early particle physics: simulation or approximation?
Historian of Science Arianna Borrelli will join us on Friday May 12th – 11:00 for an in-person talk and discussion on approximation methods in the context of her research on the history of the Monte Carlo Method in physics. To prepare for the discussion you might find it helpful to study her paper Program FAKE: Monte Carlo Event Generators as Tools of Theory in Early High Energy Physics.
As usual, should you want to attend please write to phil@hlrs.de.
Abstract:
Discussing the history of Monte Carlo computations Peter Galison (1997) argued that, after having originated in nuclear weapon research, the Monte Carlo method soon came to be regarded by scientists as an “artificial reality” in which experiments could be virtually performed. Casting doubts on some aspects of this picture, I will show how the view that Monte Carlo computations “simulate” reality was initially not so dominant as would be the case later on, and how the Monte Carlo method could be assimilated not only to experimental practices, but also to theoretical ones. In particular, I will reassess Galison’s discussion of Monte Carlos in particles physics of the late 1950s and early 1960s, showing that, depending on the context, the “same” Monte Carlo computation could be seen as a simulation of physical processes, as a tool to numerically approximate analytical expressions or as a means to represent theoretical models of particle interactions. Only later on did the idea that Monte Carlos “simulate” reality became dominant.