ABSTRACT
Philosophers have often emphasized the distinction between discovery and de monstration in science: reflection and cursory investigation of a problem may make the explanation appear, without suggesting a convincing demonstration that it is correct, which may have to be obtained in some different way. On the other hand it may be the case that the discovery and the demonstration are identical. W ith a rather elementary discovery of fact, such as that made by the young Isaac Newton (in which he had been anticipated by Marcus Marci) that blue rays of light are more refracted — in a prism, for example - than red rays are, this is commonly the case: for the demonstration of the effect is simply the repetition of the original experiment, which may be varied infinitely to show that it holds in all cases. Or, to show that oxygen supports combustion better than ordinary air, one repeats some variant of Priestley’s experiment by which this fact was discovered. The real difficulty lies in the demonstration of an explanation, or theory. So, to go back to colour and refraction, Newton’s conviction that refrangibility is an inherent property of the ray, and that the rays are thus as it were characteristically labelled by a definite proportionality as well as by their individual colours, is much harder to demonstrate than the original property it seeks to explain. For it is not rationally necessary that the rays should be thus labelled, nor each existent as a separate, elementary and unchangeable component of ordinary white light, as Newton argued. This he recognized from the first: one can imagine theories of the physical nature of light in which this property does not hold, just as well as other alternative theories in which the property does hold true. Accordingly, the truth of the pro perty inferred by Newton from his original experiment with the prism (p. 268) must be contingent, and made evident from different experiments concerning colour and refraction. In his first optical paper, in which he asserted the het erogeneity of white light, Newton offered no such evidence from further experiments and accordingly certain critics doubted the necessity for the infer ence he had asserted.
