The Structure of Scientific Revolutions (review)

BOOK REVIEWS 351 The Structure o] Scientific Revolutions. By Thomas Kuhn. (International Encyclopedia of United Science, Vol. II, No. 2; Chicago: University of Chicago Press, 1962. Pp. 172. $4.00.) This is, generally, a very interesting and stimulating volume, full of interesting information and ideas. The chief problem it sets to solve is the problem of the demarcation of science, which came to Kuhn's notice in two striking facts which have greatly impressed him. First, when he became familiar with the history of science he had to reject most of the ideas about science which he had uncritically and unnoticingly absorbed from his environment as a student of physics. Second , when he was established as a historian of science he started mixing with social scientists and was struck by their frequent use of the words "scientific" and "unscientific," which words he had seldom heard physicists use. Whereas the practitioners of an established science like physics seldom bother to notice the problem of the demarcation of science, he noted, practitioners of a pre-science like sociology or psychology are quite obsessed with it. Especially since physicists hold as a matter of course an easily criticizable view of the nature of science, it is remarkable that they can ignore the question: What is the nature of science? This, Kuhn explains, is rooted in the fact that they regularly see a living example, though definitely not a case history, of a science. A science proper has a standard textbook ; a pre-science does not. Kuhn does not refer to the history of the problem of demarcation or to others' studies of it; he does not stress that he is demarcating science from pre-science rather than from superstition (Bacon), from unscientific metaphysics (Kant), or from pseudo-science (Whewell, Popper). The problem of demarcating science from pre-science has a prehistory in the works of Frazer and of his followers in both anthropology and the prehistory of science. Yet, to my knowledge, this is the first explicit, direct, and full presentation of the problem, at least as it appears when one compares established science with modern pre-science. Unfortunately, however, Kuhn returns to the discussion of the demarcation of science from pre-seience only when he wants to present a concrete example of the demarcation of science from non-science in general. Sometimes he gives the impression that he confuses pre-science with non-science, much as others have identified nonscience with superstition or with metaphysics or with pseudo-science, and much as the positivists have confused non-science with nonsense. To make this clear, let me elaborate. According to Kuhn, whereas a pre-science does not have a standard textbook, every branch of science does have one. This is admirably clear cut and simple ; one may now ask certain questions about the textbook and how it comes to be written, but these questions are secondary : The demarcation between pre-sicence and science has already been made, and discussing such questions would only elucidate it. This demarcation, however, will not work between science and petrified science (like contemporary phrenology or astrology). For this, another demarcation can easily be set, very much in the spirit of the present volume : Whereas the textbook of a petrified science is unchangeable, the science textbook is always alterable. (Indeed, yesterday's science textbook, if it survives, becomes today's pseudo-science, superstition, and/or petrified science; at least this holds for Marxist economic theory, astrology, and phrenology.) One may now ask again how a science textbook is changeable, and when does it have to undergo a change in order to escape becoming petrified? This question is subsidiary to the question of demarcating science from petrified science. Can we demarcate science as such? Can we, in other words, demarcate science from all other intellectual activities, or theoretical systems, or whatever else ? I do not know; following Popper's teaching, I would suggest that this cannot be achieved (as it would amount to finding the essence of science). But I will be less than fair to the author if I pretended that when demarcating science from pseudo-science Popper avoids the same pitfalls: I think all authors thus far have...