True science teaches us to doubt and, in ignorance, to refrain.

We must never make experiments to confirm our ideas, but simply to control them.

But while I accept specialization in the practice, I reject it utterly in the theory of science.

As soon as the circumstances of an experiment are well known, we stop gathering statistics. ... The effect will occur always without exception, because the cause of the phenomena is accurately defined. Only when a phenomenon includes conditions as yet undefined,Only when a phenomenon includes conditions as yet undefined, can we compile statistics. ... we must learn therefore that we compile statistics only when we cannot possibly help it; for in my opinion, statistics can never yield scientific truth.

It is impossible to devise an experiment without a preconceived idea; devising an experiment, we said, is putting a question; we never conceive a question without an idea which invites an answer. I consider it, therefore, an absolute principle that experiments must always be devised in view of a preconceived idea, no matter if the idea be not very clear nor very well defined.

Men who believe too firmly in their theories, do not believe enough in the theories of others. So ... these despisers of their fellows ... make experiments only to destroy a theory, instead of to seek the truth.

Man does not limit himself to seeing; he thinks and insists on learning the meaning of phenomena whose existence has been revealed to him by observation. So he reasons, compares facts, puts questions to them, and by the answers which he extracts, tests one by another. This sort of control, by means of reasoning and facts, is what constitutes experiment, properly speaking; and it is the only process that we have for teaching ourselves about the nature of things outside us.