These last remarks will remind us at once of the strongly rhetorical character of the word “science” itself. If there is good reason for placing “progress” rather than “science” at the top of our series, it is only that the former has more scope, “science” being the methodological tool of “progress.” It seems clear, moreover, that “science” owes its present status to an hypostatization. The hypostatized term is one which treats as a substance or a concrete reality that which has only conceptual existence; and every reader will be able to supply numberless illustrations of how “science” is used without any specific referent. Any utterance beginning “Science says” provides one: “Science says there is no difference in brain capacity between the races”; “Science now knows the cause of encephalitis”; “Science says that smoking does not harm the throat.” Science is not, as here it would seem to be, a single concrete entity speaking with one authoritative voice. Behind these large abstractions (and this is not an argument against abstractions as such) there are many scientists holding many different theories and employing many different methods of investigation. The whole force of the word nevertheless depends upon a bland assumption that all scientists meet periodically in synod and there decide and publish what science believes. Yet anyone with the slightest scientific training knows that this is very far from a possibility. Let us consider therefore the changed quality of the utterance when it is amended to read “A majority of scientists say”; or “Many scientists believe”; or “Some scientific experiments have indicated.” The change will not do. There has to be a creature called “science”; and its creation has as a matter of practice been easy, because modern man has been conditioned to believe that the powers and processes which have transformed his material world represent a very sure form of knowledge, and that there must be a way of identifying that knowledge. Obviously the rhetorical aggrandizement of “science” here parallels that of “fact,” the one representing generally and the other specifically the whole subject matter of trustworthy perception.