The Science of Conjecture: Evidence and Probability before Pascal by James Franklin (review)

152 BOOK REVIEWS that many of essays refer to each other. While an editor's note is sometimes inserted next to footnotes pointing out that the essay there noted is located elsewhere in the volume, this is not done consistently, and the corresponding page number in the volume is never indicated. Finally, the book has no subject index, and the "Case" index omits mention of many of the famous cases discussed in the volume. A select bibliography of important essays taking up the subject of double effect not included in the volume would also have been hdpfol. These weaknesses notwithstanding, the book is an important and significant collection of essays on the and is appropriate for use in advanced undergraduate classes in moral theory, as weH as graduate dasses either in moral theory or on issues of warfare or biomedical ethics. The Catholic University ofAmerica Washington, D.C. JOHN BERKMAN The Science of Conjecture: Evidence and Probability before Pascal. By JAMES FRANKLIN. Baltimore and London: The Johns Hopkins University Press, 2002. Pp 512. $22.50 (paper). ISBN 0801871093. James Franklin's The Science Of Conjecture traces the discovery of rational methods of evaluating evidence and of dealing with uncertainty, methods that have been much used in law, commerce, science, philosophy, and logic in order to get at the truth in cases in which certainty is not attainable. They were in use long before Pascal, and continue to be employed in the evaluation of evidence, whether it to be in legal situations by judges and juries, or in science in the balancing of reasons for and against competing scientific theories, or indeed in ordinary language situations about what is more or less probable, more or less hkdy. Probability may or may not be expressed in numerical terms; it may avoid numbers completely, as is obviously the case in "prove beyond reasonable doubt." Franklin warns us against the easy assumption that in probability numbers are good, words bad. He distinguishes between a Whig history and an Enlightenment one and insists that his is Whig history. An Enlightenment history, he says, is one in which a heap of perfectly formed propositions, previously hidden in darkness, are gradually brought to light. A Whig history is "a story of the Advance of Knowledge as the forc~s of Reason roll back the frontiers of ignorance. As such it does not exactly need a condusion, as it records the gradual discovery of preexisting intellectual terrain in more or less rational order. Generally, a new idea in probability is seen to replace an older one because it is a better idea" (321). Probability is more like law, or psychoanalysis, in which there are BOOK REVIEWS 153 confused conceptions that work reasonably well in practice, and in which progress lies in clarifying those conceptions while keeping them grounded in reality. It is in such cases that there is a need for the historian to set out what the situation was like before and after the transition in ideas and to explain how it occurred. That is what the book is about: a study of how notions and distinctions required for nondeductive reasoning have been teased out. Their development took place over many centuries and in response to many practical demands largely, but not exclusively, in areas of law and of conscience. Franklin explains how predecessors of Pascal and Fermat learnt to handle-without any trace of irrationality-such notions as attaching weight to certain kinds of evidence. They also learned to distinguish such concepts as "suspicion," "simple presumptions," "presumptions of law," and "conjectures." Early writers on probability can therefore be regarded as having made advances if they distinguish between conclusive and inconclusive evidence and if they grade evidence by understanding that it can make a conclusion "almost certain," "more likely than not," and so on. Franklin shows that rationality cannot be restricted to what is demonstrable, and probability cannot be restricted to what can be expressed in numerical terms. One can admire not only the range of topics with which Franklin deals, but also his command of the material extending through Roman Law, evidence as found in mediaeval law, Renaissance Law, ecclesiastical disputes about doubts, the doctrine of...