A theory is scientific only if it can be disproved. But the moment you try to cover absolutely everything the chances are that you cover nothing.

Our humanist attitude should therefore throughout be to stress what we all have in common with each other and relegate quarrelsome religion to the private domain where it can do [less] harm.

All science is full of statements where you put your best face on your ignorance, where you say: ... we know awfully little about this, but more or less irrespective of the stuff we don't know about, we can make certain useful deductions.

The test of science is not whether you are reasonable—there would not be much of physics if that was the case—the test is whether it works. And the great point about Newton’s theory of gravitation was that it worked, that you could actually say something about the motion of the moon without knowing very much about the constitution of the Earth.

An observer situated in a nebula and moving with the nebula will observe the same properties of the universe as any other similarly situated observer at any time.

We find no sense in talking about something unless we specify how we measure it; a definition by the method of measuring a quantity is the one sure way of avoiding talking nonsense.

The kind of lecture which I have been so kindly invited to give, and which now appears in book form, gives one a rare opportunity to allow the bees in one's bonnet to buzz even more noisily than usual.