It is eminently desirable that people should have faith in science, but in order to do that they must have the same sort of faith in it that the scientist has. Otherwise they will regard it as a lot of ingenious fancies which are proved false by each succeeding generation. Science is moulting just now and looks queer. The public ought to understand clearly that the process means growth and not disease. There is another reason now for the popularization of the scientific mode of thought. It is beginning to be applied where entirely different conceptions have so far prevailed--to art, ethics, religion, sociology, and the like. This is already arousing a great commotion and will cause more before the process is complete. It will, for example, involve the rewriting and to a large extent the reinvestigation of history. Poincaré has hinted at this in a passage which seems to me of very great significance:
Carlyle has somewhere said something like this: "Nothing but facts are of importance. John Lackland passed by here. Here is something that is admirable. Here is a reality for which I would give all the theories in the world." Carlyle was a fellow countryman of Bacon, but Bacon would not have said that. That is the language of the historian. The physicist would say rather: "John Lackland passed by here. That makes no difference to me for he never will pass this way again."--"Science and Hypothesis", p. 102.
The aim of science is prevision, and I believe that this will eventually be recognized as the true aim of all knowledge. The historian, or let me say rather the antiquarian, for the historian may have the scientific temperament, values facts for their rarity. The scientist values facts for their commonness. A unique fact, if there be such, would have no possible interest to him. The antiquarian goes about looking for things, facts, or furniture, which have been of importance in the past. The scientist is looking only for things that will be of importance in the future.
According to Poincaré, the proper choice of facts is the first duty of the scientist. He must be able to pick out the significant and reject all the rest. "Invention consists in avoiding the constructing of useless combinations and in constructing the useful combinations which are in infinite minority. To invent is to discern, to choose." It is most desirable to bring together elements far distant from one another. Such unions are mostly sterile, but when this is not the case, they are the most fruitful of all. The successful scientist does not, like a shopper, look over one by one all available samples and pick out what he wants. Life is too short. The unsuitable ideas do not even present themselves to his mind. It is as if he were an examiner of second resort who only concerns himself with the candidates who have passed the first test. This preliminary sifting and sorting process is done largely by the unconscious mind, as Poincaré shows by telling how he came to make his first mathematical discoveries: