This book is a congeries. Not indeed an incongruous congeries, as of congers and costermongers, but withal a congeries to conjure with. For all its slenderness it offers us a philosophy of style, a philosophy of quotation, a philosophy of art, a philosophy of optical illusion, and a philosophy of nature. It is this last that packages the lot and gives it a name. The looseness of the package is in keeping with the philosophy that assembles it; for the doctrine is that there are many worlds, none all-embracing.

There is currently a Leibniz revival that has philosophers luxuriating in a continuum of possible worlds. One sterling virtue of Nelson ‘Goodman’s philosophy is that it is no part of that. Goodman means all his worlds to be actual. Proceeding then to try to penetrate what one hopes is a figure of speech, one finds that where the purported multiplicity really lies is not so much in worlds as in versions: world versions. I cannot quite say versions of the world, for Goodman holds that there is no one world for them to be versions of. He would sooner settle for the versions and let the world or worlds go by.

His doctrine rests partly on an appreciation of the creative component in natural science. Even the most rudimentary of scientific laws is a generalization beyond the instances observed. There are divergent ways of generalizing from the same observations; some ways relatively simple, others arbitrarily gerrymandered.

We settle tentatively on the simplest one, but we may be forced off it by later observations. Even in observation itself there is a creative component: we overlook features irrelevant to our concerns, we perceive broad forms and gloss over discontinuities, we round out and round off. And at the other extreme, in the high flights of theoretical physics, man’s creativity is overwhelming. Physical theory is indeed uncannily successful in the corroborations that it predicts and in the power over nature that it confers, but even so it is ninety-nine parts conceptualization to one part observation. May there not be some radically alternative conceptual structure, undreamt of, that would fit all the past observations and all the predicted ones equally well, and yet be untranslatable into our scheme? Our own physical theory and that one would be two world versions, equally sound. Two versions of the world? But what world is that? To describe it we must retreat into one or the other version; they share no neutral description. Recognize the two versions, Goodman says, and leave it at that.

This much will already estrange many of Goodman’s readers. Not me. But then he presses on where I falter. Another world version that he treats with respect is the commonsense one, which depicts a world not of atoms and electrons and nuclear particles but of sticks, stones, people, and other coarse objects. He sees further world versions, more fragmentary, in the styles of various painters. Thus he contrasts the world of Rembrandt with the world of Rouault and the world of Picasso. Shunning even the restraints of representational art, he forges on to abstract painting and to music: here again are world versions in their lesser ways. How, when they depict nothing? Well, they refer in another way: they stand as samples of interesting strains or qualities. There is a significant continuity, Goodman argues, between exemplification and depiction as well as between depiction and description.

One feels that this sequence of worlds or versions founders in absurdity. I take Goodman’s defense of it to be that there is no reasonable intermediate point at which to end it. I would end it after the first step: physical theory. I grant the possibility of alternative physical theories, insusceptible to adjudication; but I see the rest of his sequence of worlds or world versions only as a rather tenuous metaphor.

Why, Goodman asks, this special deference to physical theory? This is a good question, and part of its merit is that it admits of a good answer. The answer is not that everything worth saying can be translated into the technical vocabulary of physics; not even that all good science can be translated into that vocabulary. The answer is rather this: nothing happens in the world, not the flutter of an eyelid, not the flicker of a thought, without some redistribution of microphysical states. It is usually hopeless and pointless to determine just what microphysical states lapsed and what ones supervened in the event, but some reshuffling at that level there had to be; physics can settle for no less. If the physicist suspected there was any event that did not consist in a redistribution of the elementary states allowed for by his physical theory, he would seek a way of supplementing his theory. Full coverage in this sense is the very business of physics, and only of physics.


Anyone who will say, “Physics is all very well in its place”—and who will not?—is then already committed to a physicalism of at least the nonreductive, nontranslational sort stated above. Hence my special deference to physical theory as a world version, and to the physical world as the world.

Component essays in Goodman’s congeries are rewarding quite apart from the polycosmic motif that strings them together. There is a bright one on samples, which dramatizes in deft parables the relation of sample to purpose. A swatch represents its bolt in point of texture, pattern, and color, but, unlike a sample cupcake, it deviates in size and shape. But a swatch can also be used to exemplify swatches; and then, he observes, its size and shape count for much and its particular texture, pattern, and color for little.

The chapter on quotation explores the possibility of analogues of quotation, direct and indirect, in painting and music. Goodman finds that depictions of picture frames will not quite do as analogues of quotation marks, and that indirect quotation has better affinities in painting than direct. Both sorts of quotation come off badly in music. The reasons he gives for these similarities and contrasts afford some worthwhile semantical insights.

The chapter on perception begins with the familiar illusion induced by closely paired flashes, which the eye perceives as a single moving light. Remarkable variations of this phenomenon are then reported, stemming from experiments by Paul Kolers. Illusions can be created of elaborate permutations of position along quite unexpected lines, accompanied by changes of shape. Analogous affects are not obtainable in the case of changes of color, and Goodman offers an ingenious and convincing reason why they should not be.

In his chapter on art he dismisses the traditional problem of defining a work of art. He shows that the notion is hopelessly entangled with the fancied distinction in metaphysics between internal and external relations, between intrinsic and extrinsic, between essence and accident, which he laudably rejects. He finds more significance in describing the circumstances in which a thing functions as art; and it is as a means to this venture that his discussion of swatches and other samples fits in.

There are engaging passages. The scientist

seeks systems, simplicity, scope; and when satisfied on these scores he tailors the truth to fit…. He as much decrees as discovers the laws he sets forth, as much designs as discerns the patterns he delineates.

…we must distinguish falsehood and fiction from truth and fact; but we cannot, I am sure, do it on the ground that fiction is fabricated and fact found.

…the philosopher like the philanderer is always finding himself stuck with one or two many.

How can anyone thus sensitive to words have coined “acquacentric”? Why the Italian, or indeed anything to do with Latin, when he could have played “hydrocentric” straight?

This Issue

November 23, 1978