Feynman has another fantastic talk on "What is Science?" [1]. Among other things, at a certain point in that talk, this is how he lays out his "best definition

Feynman has another fantastic talk on "What is Science?" [1]. Among other things, at a certain point in that talk, this is how he lays out his "best definition of science": "What science is, I think, may be something like this: There was on this planet an evolution of life to a stage that there were evolved animals, which are intelligent. I don't mean just human beings, but animals which play and which can learn something from experience--like cats. But at this stage each animal would have to learn from its own experience. They gradually develop, until some animal [primates?] could learn from experience more rapidly and could even learn from another’s experience by watching, or one could show the other, or he saw what the other one did. So there came a possibility that all might learn it, but the transmission was inefficient and they would die, and maybe the one who learned it died, too, before he could pass it on to others. The question is: is it possible to learn more rapidly what somebody learned from some accident than the rate at which the thing is being forgotten, either because of bad memory or because of the death of the learner or inventors? So there came a time, perhaps, when for some species [humans?] the rate at which learning was increased, reached such a pitch that suddenly a completely new thing happened: things could be learned by one individual animal, passed on to another, and another fast enough that it was not lost to the race. Thus became possible an accumulation of knowledge of the race. This has been called time-binding. I don't know who first called it this. At any rate, we have here [in this hall] some samples of those animals, sitting here trying to bind one experience to another, each one trying to learn from the other. This phenomenon of having a memory for the race, of having an accumulated knowledge passable from one generation to another, was new in the world--but it had a disease in it: it was possible to pass on ideas which were not profitable for the race. The race has ideas, but they are not necessarily profitable. So there came a time in which the ideas, although accumulated very slowly, were all accumulations not only of practical and useful things, but great accumulations of all types of prejudices, and strange and odd beliefs. Then a way of avoiding the disease was discovered. This is to doubt that what is being passed from the past is in fact true, and to try to find out ab initio again from experience what the situation is, rather than trusting the experience of the past in the form in which it is passed down. And that is what science is: the result of the discovery that it is worthwhile rechecking by new direct experience, and not necessarily trusting the [human] race['s] experience from the past. I see it that way. That is my best definition." [1] Feynman, R. P., "What is Science?" The Physics Teacher Vol. 7, issue 6, 1969, pp. 313-320 http://www.fotuva.org/feynman/what_is_science.html http://www.fotuva.org/feynman/what_is_science.html