Law and rule in the Essay towards solving a problem in the doctrine of chances

In his famous Essay towards solving a problem in the doctrine of chances, Bayes calculates probabilities as chances to be in the right in one's guess or as fractions of certainty. He was led to use two schemas. The first one is a monetary schema. But he forwarded a second schema of a square table on which a ball is thrown and the points where it rests are located and recorded. This schema was to him the best possible one to give shape to the problem of ascertaining the degree of trust to be given to a guess when, having at one's disposal only a small number of data about the conjunction of subsequent events, one extrapolates from this small number to a greater number. The purpose of my paper is to examine the nature of the "I" of the "I guess", which must be understood in a transcendental or intersubjective way rather than in a private or individual sense. But then, the right interpretation of the Bayesian "I guess" might well be found in terms of the theory of games, as convincingly expounded by J. C. Harsanyi.