Evaluating a Solution to Old Evidence: Lessons from the Development of Quantum Theory
In this paper, I evaluate a Garber-style solution to the Bayesian problem of old evidence proposed by Stephan Hartmann and Branden Fitelson. I analyse how successful it is in explicating an early stage in the development of quantum theory. I argue that while the solution does provide an appropriate framework for understanding this historical case, Hartmann and Fitelson’s formal assumptions are not sufficient to fully capture the details of the historical situation. I propose some further assumptions that must be introduced in order to give an accurate explication.
Perrin and Planck: Two Cases of Unification
There has been much philosophical discussion on the nature of Jean Perrin’s claim for the reality of atoms as related to his work in determining Avogadro’s constant. I propose that we can better understand this argument by comparing it with a different case, namely, with the argument for quantization in the case of multiple agreeing measurements of Planck’s constant.
Evaluating the Quantum Postulate in the Context of Pursuit
The purpose of my dissertation is to contribute to our understanding of scientific theory pursuit by providing a detailed case study on the development of quantum theory.
In 1900, Max Planck introduced the notion of `energy elements’ into his attempt to account for the observed anomalous blackbody radiation spectrum. Despite the fact that the physical interpretation of this notion was ambiguous, a similar idea was later taken up by several scientists. These investigations eventually led to the formulation of quantum mechanics, one of our most successful physical theories to date. However, the intervening years of study were marked by theoretical uncertainty and inconsistency, during which time scientists had to proceed according to a patchwork collection of principles and heuristics.
I suggest that this period should be considered a case of what I call “piecemeal pursuit” by presenting the historical `quantum conjectures’ that were being used in different contexts. These conjectures gave quite varied interpretations of what quantization might refer to. By comparing these conjectures, I identify a general quantum postulate that captures the underlying assumption that is common to all the cases. I argue that it is possible to consider a general postulate about quantization even when its proper application is ambiguous in a given context, and that the postulate can be separated from various elements of the framework being used to investigate it. I show that the quantum postulate can be deemed promising by analysing the support it gains using a Bayesian framework. I defend the use of such a framework by considering the purported inconsistencies in Planck’s introduction of his quantum conjecture and how we should handle these. I then explicate two cases of support for the postulate. First, I show how we can use a particular solution to the Bayesian problem of old evidence to interpret the support the quantum postulate received by accounting for phenomena that had no previous explanation. Finally, I show that the quantum postulate is also supported by a unification argument, where unification is interpreted as informational relevance between the various domains of inquiry.