Dokic and his coauthor,Engel, do a good job of covering the contributions of Frank Ramsey in this book.For instance,Dokic shows how Ramsey's initial, basically Platonic, view of mathematical entities( and his acceptance of the logicist attempt to base all of mathematics on logical principles) evolved and changed,moving toward an intuitionistic(Brouwer)view of the foundations of mathematics.The book is marred by Dokic's blind acceptance of Ramsey's extremely weak attack on J M Keynes's logical theory of probability that is contained in Keynes's masterpiece,the 1921 A Treatise on Probability.Dokic makes the following claims:"Ramsey makes two criticisms against Keynes.The first is that,for practically every pair of propositions,such as "This is Red" and "This is blue", we do not have the slightest idea of the probability relation between them."(Dokic,p.7).Given that Keynes's probability relations are always conditional,Dokic needed to combine tham in some fashion.Suppose we compare the conditional probability of" This is Red,given that That is red" with "This is Red,given that That is Blue".It is obvious that the first conditional probability is greater than the second,although one can not say how much greater.This problem, in the form of the red-blue books problem(just add "book" after Red or Blue) ,was covered by Keynes in chapter III of the TP to illustrate precisely that point.The point went completely over Ramsey's head.What we can't do is say that the first conditional probability is .7 and the second conditional probability is .6.Of course,Ramsey's error is to believe that all probabilities are point estimates.The second criticism of Keynes is that"...contrary to what Keynes holds,we cannot perceive these probability relations through introspection..."(Dokic,p.7).This is a mere assertion.Ramsey never supplied any deductive proof for his claim.Nor has anyone else.Finally,the foundation for Ramsey's entire approach is the claim that there exist" ethically neutral"propositions.These "propositions"are simply postulated to exist and be perceived by the individual decision maker so as to be able to always derive precise numerical probabilities,assuming that he/she is willing to bet on all propositions.This appears to be a hidden assumption of Keynes's principle of indifference,which Ramsey claimed his system of subjective probability did not need.
