SHiPS Resource Center ||   BOOK BRIEF

reviewed: 10/99 Beauty and Revolution in Science. James W. McAllister. Cornell Univ. Press (1996/1999). ISBN 0801486254. 248 pp. $16.95pb.

This book is a work of philosophy that concerns two things that are not often linked together: science and aesthetics. Despite the general lack of attention to this link, a number of scientists have written on the importance of aesthetic qualities in making scientific judgments. These include the physicist Subrah-manyan Chandrasekhar in his book Truth and Beauty (1987) and of the chemist Roald Hoffman in an article in the Journal of Aesthetics and Art Criticism (1990). Both these Nobel Prize winners see aesthetic factors as important in keeping scientists at work and influencing what science they do and how they do it. While Chandrasekhar and Hoffman take pains to carefully describe what is so attractive about the work they do and about the ideas in their respective sciences, neither takes the analytical approach of a philosopher, the type of approach that James McAllister uses in Beauty and Revolution in Science.

McAllister holds to a rationalist image of science and admits that fellow rationalists have so far failed to investigate scientists' use of aesthetic criteria in evaluating theories. His book is an effort to fill this gap and to relate the use of aesthetic criteria in science to the development of scientific revolutions. He provides evidence that scientific communities perform two kinds of evaluations of theories: the judgment of empirical performance and of aesthetic appreciation. The aesthetic properties of theories include symmetry, simplicity, innovation of a model, and visualizability; and McAllister goes into each of these in more detail that I can here. He then relates these qualities to the empirical by arguing that theories showing persistent empirical success have aesthetics properties that win favor in the scientific community. In other words, successful theories become more aesthetically attractive, and "scientists' aesthetic evaluations tend in the longer term to swing into line with their empirical appraisals" (p. 66).

McAllister argues that a community of scientists compiles its aesthetic canon at a particular point in time by attaching to each aesthetic property a weight proportional to the empirical adequacy attributed to current theories that exhibit that property. He labels this process aesthetic induction. He contends that new theories are judged in terms of the aesthetic properties of previously successful theories, and thus there is an association of aesthetic properties of theories with expectations of empirical success. If a new theory is not considered aesthetically attractive, it can gain favor as more empirical evidence accumulates in its favor. And by its empirical success, a theory can predispose the scientific community to choose future theories with properties similar to its own. Thus successful theories tend to inhibit revolutions in science, but then, as a revolution progresses, the scientific community loses its commitment to its old aesthetic canon.

To support his claims, McAllister presents various lines of evidence, including Jeremy Bernstein's comment that in the sciences as in the arts, sound aesthetic judgments are usually arrived at only in retrospect; a truly new art form or scientific idea is usually first seen as ugly. McAllister also goes outside of science to build an analogy in support of aesthetic induction. In a very entertaining and fasci-nating chapter, he analyzes the history of the use of cast iron in 19th-century architecture. Like almost every other building material when new, cast iron was first used only on the fringes of architectural activity until its utility was firmly demonstrated. As it came to be used more and more in architecture, cast iron gained greater aesthetic acceptance as a building material. McAllister's point here is that, as with scientific theories, empirical validity preceded aesthetic recognition. In both architecture and science, "the demonstrated practical worth of a work— empirical success in the case of scientific theories, utility in the case of buildings—is capable of reshaping the aesthetic canons on which subsequent work is evaluated and by which the line of progess of the discipline is partly determined" (p. 160).

For a work of serious philosophy, McAllister's book is very readable, and he develops his case carefully and thoroughly. I would recommend this book to anyone interested in the philosophy of science and, particularly, in the scientific aesthetic. I do find his description of the use of aesthetic qualities in scientific judgment rather limited; he writes little about the importance of the aesthetic in the development of theories in the first place. But despite this limitation, which is in part the result of his taking a rationalist stance, his concept of aesthetic induction deserves attention from the history and philosophy of science communities.

—Maura Flannery

  SHiPS helps teachers share resources for integrating history, philosophy and sociology in the science classroom. E-mail us at for our quarterly news.