By 1660, Arnauld and Nicole could already deploy a recognisable concept of probability:1 . . . in order to decide what we ought to do to obtain some good or avoid some harm, it is necessary to consider not only the good or harm in itself, but also the probability that it will or will not occur, and to view geometrically the proportion all these things have when taken together. (Arnauld and Nicole, 1996: pp. 273–4) We can discern in this passage several core features of the concept: (i) probability is a mathematical measure of the possibility of the occurrence of events (ii) it is intimately connected with rational decision making (iii) its assignment to an event is not dependent on the actual occurrence of that event. Call this the pre-theoretical conception of probability: it makes no claim about what instantiates or embodies the probability of some event, it simply states that there is a useful notion that has these features. With the rise of classical statistical mechanics, this pre-theoretical conception of probability was given a home in science. Some minor alterations were made, but it was recognisably a precisification of that same pre-theoretical concept that was being used.2 It was at this point that empiricist philosophical...