Review of:


Miller, Arthur I. Insights of Genius: Imagery and Creativity in Science and Art.
xxii + 482 pp., illus., figs., bibl., index.
Cambridge, Mass.: The MIT Press, 2000. $18.95 (paper).


By Christopher D. Green

York University

Toronto, Ontario, Canada


To be published in Isis.

©2002, Christopher D. Green


Miller's book has a number of aims.  Throughout he tries to make a case for the importance of visual imagery in the development of modern science, especially 20th-century physics. Alongside this, however, is an argument about how ideas concerning imagery and intuition fundamentally changed (at least among scientists) at about the time that quantum theory and relativity were first challenging some of the basic ways in which we understand the world. In addition, he repeatedly and vociferously argues for scientific realism, the claim that science aims to give an accurate description of a world that exists apart from our conceptualizations of it, and that the theoretical terms of science (e.g., electron) refer to real, even if not observable, entities.


Miller's understanding of the quantum and relativistic revolutions is impressive. He presents an engaging account of the personalities, cognitive styles, and internal political struggles that underlay the transformation from a Newtonian world-view to the 20th-century view, giving us insights into figures such as Poincaré, Einstein, Planck, Heisenberg, Schrödinger, and Bohr. His historical orientation is largely internalist, but he provide fascinating insights into his subject nonetheless.


Unfortunately his grasp of related issues in philosophy and in cognitive science issues is not so sure.  His selection of materials is often tendentious and occasionally outdated. His interpretations of others' positions are often less astute than they need to be in order to make his criticisms credible. Instead of attacking his opponents' strongest suit, he too often sets out on the less interesting excursion of demolishing a weakened version. For instance, his arguments for scientific realism rapidly descend into a Gross-&-Levitt-inspired (Higher Superstition, Johns Hopkins, 1994) diatribe again "postmodernists," especially "eco-feminists" (he uses both terms derisively), rather than focusing on (or even mentioning) the analytically-trained anti-realists and non-realists who actually mount the greatest challenge to his position -- Bas van Fraassen, Arthur Fine, Larry Laudan, Nancy Cartwright, and the like. (And I write this as one who is relatively sympathetic to scientific realism.)  His account of the mental imagery debate in cognitive science focuses squarely on the quite old argument between Zenon Pylyshyn and Stephen Kosslyn. Pylyshyn's position against mental imagery, counterintuitive as it may be, is handled so dismissively that the central problems he exposed (just what would it be to have a picture encoded directly in the mostly digital neurons of the brain?) are never given a fair hearing. Miller also fails to mention, much less grapple with, even the most basic difficulties with the view that mental images are closely related to visual perception (e.g., Imagine a Star of David. Does it contain a parallelogram? Most people find this a difficult question to answer if they rely solely on their mental image, but give the correct answer quite rapidly if they have a picture before them. Here is clear evidence that mental images are encoded quite differently from normal visual representations.)


In the final chapter he tries to connect his discussions of imagery, creativity, and aesthetics in science with parallel issues in the visual arts. Here his focus is on post-impressionist and modernist painters such as Cézanne, Picasso, Braque, and Kandinsky. His argument does not so much make the case that terms such as "creativity" and "aesthetics" refer to same processes in the arts and the sciences, as it does proceed from the assumption that they do, and then go on to make observations about how they are manifested in these two disparate "cultures."


In the final analysis, the book is worth reading if only for Miller's insights into the early-20th-century revolutions in physics.  It falls far short, however, of providing adequate reasons to accept the broad claims that appear to have driven him to write the book in the first place.

Christopher D. Green

York University, Toronto