Reichenbach: Probability & the A Priori: Has the Baby Been Thrown out with the Bathwater?

Throughout the work of the early 20th century philosopher/physicist Hans Reichenbach there are traces of Kantian philosophy. These traces can also be found within Reichenbach’s mature, logical positivist, writings. The author of this dissertation tracks the Kantian traces from Reichenbach’s early frequentist interpretation of probability in 1916 to Reichenbach’s later work.
In the first part of this dissertation we investigate the early development of the theory of probability. After investigating what role determinism leaves for probability and uncertainty, we analyse the role of the concept of the a priori in the philosophy of science. In the chapters that follow we look at how probability and the a priori have become intertwined in the classical interpretation of probability. We will investigate the relation between probability and determinism in the work of the classical probability theorists PS Laplace and J von Kries. In the classical interpretation, the probability of throwing an even number with a fair die is defined as the ratio between the number of favourable cases (three) and the number of possible cases (six). It is an a priori assumption that the possible cases that go into the classical definition are equally probable.
The second part of this dissertation concerns the role of the a priori in the work of Reichenbach. The clearest (but not the only) trace of neokantianism in Reichenbach’s philosophy is an a priori element within his probability interpretation. The trace runs from Reichenbach’s idea of a ‘continuous probability function’ (1916); via his ‘probabilistic posit’ (1935/49); to the idea that the relation between reality and our observations thereof is a ‘projection’ (1938).
Reichenbach’s frequentist interpretation about which he wrote in 1916 is closely related to the classical theory of probability. Reichenbach replaces the classical a priori that all possible cases are equally probable with the assumption that what Reichenbach calls the ‘probability-function’ is continuous. The new assumption is a priori in a neokantian sense: it is a condition for the possibility of a certain type of knowledge, and therefore must be in place if this knowledge is to be possible. Reichenbach’s later view on probability leaves the classical interpretation further behind (as it is no longer based on a deterministic principle of causality) but retains a Kantian element in the form of the idea that probability statements are posits – neither true nor false but hypothetical. The idea is Kantian because Reichenbach introduces the posit as a condition for probabilistic knowledge. In 1938 Reichenbach applies the idea of the posit in order to defend scientific realism. The neokantian a priori forms a persistent and essential element in Reichenbach’s philosophical views.
The final chapter of this dissertation is in the form of an ‘apologia’. The results of the previous chapters – the identification of the elements in Reichenbach’s philosophy that make it far more subtle than usually understood – are used to show that the traditional criticism levelled against the logical positivists does not apply to Reichenbach.

Summaries in Dutch and Frisian are included at the end of the dissertation.