It is not at all unusual--on the contrary, it is very common--for two (or more) scientific theories to compete as the "best" explanation of a set of empirical evidence. Within the past 30 years, for example, plate tectonics competed with traditional theories about the explanation for the current configuration of the continents. With results of NSF sponsored deep-sea drilling along fault lines, so-called "irrefutable" evidence "convinced" scientists of the 'validity" of plate tectonics. Unfortunately, such "critical experiments" cannot operate in a vacuum. Even if a theory explains some set of "facts" better than another, it does not explain "all" the evidence that a previous one did. Scientists must weigh--in some probabilistic way--the likelihood that a new theory will account for a whole range of evidence better than another. One way that scientists can do this is through a so-called "subjective" or Bayesian probability theory. Dr. Maher is examining the value of using such a subjective approach on a famous historical case in order to determine if in fact subjective theory can contribute to resolution of scientific disputes. Quicklime is produced by strongly heating substances such as limestone. In 1756, Joseph Black published a new theory of this transformation, according to which it consists in the release of a type of "air." Black's experimental support for his theory has been cited ever since as a model of cogent experimental proof. However, Black's work was not widely accepted for several decades after it was published, and for some years, a very different theory, published by Johann Meyer in 1764, was widely accepted in continental Europe. Meyer's theory was a variation on the traditional view that the properties of quicklime are due to matter added from fire. Under this grant, Dr. Maher is comparing the degree to which Black's evidence, and other information at the time, favored Black's theory as against Meyer's. Dr. Maher is using subjective probability theory as the framework that enables judgments about the many different pieces of evidence to be combined into an overall judgment. This study is designed to provide a model for how the strength of evidence for disputed scientific claims may be evaluated--a model equally applicable to contemporary and historical debates.
