WHAT ART IS: THE ESTHETIC THEORY OF AYN RAND. By Louis Torres and Michelle Kamhi. Chicago and La Salle, Illinois: Open Court, 2000. Pp. xvi, 523.

What Art Is grew out of a rather brief series of essays in which Louis Torres and Michelle Kamhi attempted some years ago to clarify and sort out the content of several of Ayn Rand's essays on esthetics.(1) In the process, their project changed considerably. The book as we now have it is a big one, and its subject extends well beyond the interpretation of Rand. It not only establishes serious philosophical discussion of Rand's esthetics (a thing previously almost unheard of) but constitutes a major contribution to objectivist thinking on the arts. The breadth and depth of the learning the authors display is strikingly evident, even from a cursory look at the dense and fascinating (and very plentiful) footnotes. I am not sure that I have ever reviewed a book from which I have learned so much.

This book actually has two quite different though closely related purposes, one more ambitious than the other. The first is to present Rand's esthetic theory in a way that sorts out the basic elements in it that the authors believe have great merit and deserve our most thoughtful attention from those aspects of it that, in their view, are less essential to it and have considerably less value. The second purpose is to use the core of her theory, thus isolated and cleaned up, in order to directly engage in one of the great issues of twentieth century esthetic theory.

This issue is not merely about the theory of the last hundred years or so but, in addition, touches on a real world phenomenon that has been frequently described and amply documented. As everyone knows, many movements in twentieth century art have aimed at "breaking down the barriers" between art as traditionally understood on the one hand and things that lie outside the traditional boundaries of art on the other. Further, as is also common knowledge, these movements have done so by producing works that (as their creators like to say) "challenge" these traditional notions. They have produced paintings that depict nothing; music without melody, rhythm, or harmony; poetry that lacks every attribute of poetry except for being cut up into lines on the page on the page -- and so forth. And yet, as many people have observed, average people have shown a really remarkable resistance to having their barriers blurred, despite the fact that these sorts of "challenges" have repeatedly been inflicted on them for almost a hundred years. To this day, most people react to such works -- generally characterized as avant garde -- with a mixture of boredom and hostility.

The second of Torres and Kamhi's main purposes is to use Rand's basic theory to attack the claims of the avant guard to be producing art. Most of the eight chapters of Part II or their book are attacks on the claims that modern esthetic theory has made over the years that one form or another of (in their view) non-art is actually art. The list of "pseudo-art" genres that are discussed in them is very long, and includes (for instance) abstract painting, aleatory music, photography, quilt-making, some (though not all) primitive art, avant-garde (but not modern) dance, conceptual art, video art, and installation art.

Objectivists will find a great deal in Part II to applaud. Their reaction to Part I, though, might be more mixed. This is the part in which the first of the authors' two main purposes -- that of carrying out the sorting-out function -- is principally realized. It consists, to a large extent, of a detailed discussion of each of the four principal essays printed in Rand's volume, The Romantic Manifesto: "The Psycho-Epistemology of Art," "Philosophy and Sense of Life," "Art and Sense of Life," and "Art and Cognition." One of the authors' concerns here is the quite laudable one of pointing out Rand's rather cranky shortcomings -- her tendencies to overstatement and oversimplification, her giving short shrift to other thinkers and other schools of thought, and her excessively moralistic tone -- and crafting formulations of her basic theory that are free of these defects. What I eventually found a little depressing, however, was the fact that they seem to miss no opportunity to point out these foibles of hers, or to point out her lapses from rigor. Though I appreciated being told, for instance, that Rand's beloved Aristotle quotation, which says that literature presents "things as they might be and ought to be," is actually a misquotation, I also thought that they made the point with a little too much emphasis (pp. 63-64, 355 nn. 13-19), especially given the fact that (as they point out) the misquotation does represent a defensible interpretation of what Aristotle's position actually is. The sheer accumulation of these comments might give some readers the vague impression of a slight hostility toward their subject, as if they admire her ideas but find her annoying as a person. I incline to think, though, that this is a sort of optical illusion produced by the fact that, as I say, these comments do form a rather obtrusive pattern. It could be that they do share one of Rand's frailties: the inability to trust that the reader has gotten an important point that has been made just once.

Of much greater importance, of course, is the treatment by Torres and Kamhi of Rand's ideas themselves. Here, again, it is likely that they will ruffle a few objectivist feathers, since they disagree with Rand surprisingly often. On the whole, they tend to see her observations on particular genres as less valuable than her identification of the general principles of art. As I will point out later, they have rather important differences with Rand regarding her observations on music and architecture.

However, their differences with Rand go rather deeper than this. Before I say any more, however, about where they agree or disagree and why, I think I should make a brief comment about how Rand's ideas about art are structured. These ideas can be thought of as occupying four different theoretical levels. At one level, the most concrete, we have (1) Rand's many comments, generally fascinating but often quirky, on specific works of art that she loved or hated. Then there are the ideas she present as a more or less direct justification of these comments, which consist of (2) her account of the difference between Romanticism and Naturalism. This includes both a theory about how these two sorts of art differ and a defense of Romanticism as altogether superior to Naturalism on the ground of its different treatment of the human power of volition. Undergirding (2), or at least purporting to, is (3): the account of the nature of art, in which it is defined as "a selective re-creation of reality according the an artist's metaphysical value-judgements." This in turn is presented by Rand as logically based on (4) a discussion of the nature of "metaphysical value-judgements" and of the related notion of a sense of life, as well as of the relation between the abstract and the concrete in human psychology. It is (4) that gives the vital function of art, as defined in (3).

On the most concrete of the levels in which Rand works (1), Torres and Kamhi spend virtually no time. Their few remarks on this material are to the general effect that these are Rand's "personal" reactions to various works and do not follow from her basic esthetic theory. A good part of the reason they are so dismissive of this aspect of Rand's writings on esthetics is that they also take a rather dim view of the ideas in (2). Her treatment of Romanticism and Naturalism, they claim, should be seen as part of her "personal esthetic of literature": it tends to be "misleading" if we view it in the context of her theory of art as such (p. 31). Considering the importance of these ideas (at least in Rand's own opinion) it is disappointing that their comments on them are as brief as they are (pp. 31-33). They advance two main contentions concerning them. First, they point out that her definition of Romanticism, as "a category of art based on the recognition of the principle that man possesses the faculty of volition," seems to be applicable only to arts that tell a story. As they suggest, it is indeed hard to see how an art can depict human beings as having "the faculty of volition" unless it shows them choosing and acting, and it seems foolish to expect such non-narrative (or not necessarily narrative) arts as painting and music to do this. Second, they contend that on those few occasions when Rand attempts to show how such arts can be Romantic, she gives an account which is seems plainly a-historical, as when she speaks of pictorial artists that depict human beings as heroic, or that choose to depict subjects that are beautiful. As Torres and Kamhi point out, art that has either or both of these characteristics can be found in periods before the advent of Romanticism and, as Rand rightly says in other contexts, this crucial event did not occur until the nineteenth century.

Though their comments on this particular nest of issues are, as I have said, regrettably sparse, I believe they are also incisive and worthy of close attention. Clearly, they do have a valid point here: though Rand's claim that literature that is Romantic in her sense could not exist until the relevant historical forces (roughly, the high tide of the influence of reason coupled with the birth of modern freedom in the political realm) had made it possible, it would be obviously false to say that no artist before that time had consistently chosen to represent beautiful subjects. A good deal more work needs to be done on an Objectivist definition of Romantic art, at least if one wishes to insist (as Rand of course did) that it is a concept that applies to art in general and not merely to literature. However, I think Torres and Kamhi are going too far when they suggest that, even if we regard Rand's comments on Romanticism as applying to literature alone, they constitute mere personal statements that need not be taken seriously as theory. Her account of Romantic literature is an application to literature of her account of art in general, which holds that it embodies the artist's "metaphysical value-judgements." These are beliefs about the world and human beings that are most fundamentally relevant to human well-being. If one is dealing with the subject matter of the narrative arts -- that is, if one is dealing with human action -- then beliefs about whether the lives of human beings are directed by their own choices will clearly belong in this category. Such judgments will profoundly affect the sort of meaning and value action can have. This of course means that Rand's conception of Romanticism, and the case she makes for its superiority over Naturalism, are based not only on her esthetic theory but on her theory of the freedom of the will and, to some extent, on her ethics as well. That would mean that to understand and appreciate it properly we must view it as a theoretical matter and not merely as a personal preference of hers.

Torres and Kamhi give considerably fuller treatment to the issues raised by (4) than they do to those in (1) and (2), but concerning the one issue which is most crucially at stake -- namely, the function of art in human life -- the space they devote to it (two paragraphs on p. 27, one paragraph on pp. 33-34, and one paragraph on p. 57) is once again more sparse than one might have expected. Again, at least part of the reason for this seems to be a disagreement with Rand that prevents them from taking what she has to say on this subject very seriously. At the conclusion of the essay "The Psycho-Epistemology of Art," Rand says that the reason art is so personally important to people is that it "confirms or denies the efficacy of a man's consciousness, according to whether an art work supports or negates his own fundamental view of reality." Their brief comment on this point is that "such claims are simply not true" (p. 57). In their own experience, they point out, their reaction to works that contradict their view of reality tends to be one of distaste, rather than doubting their own mental efficacy.

As before, I think they are dismissing what Rand is saying rather too quickly. They remark at one point that, when Rand mentions this notion about confirming or denying the efficacy of one's mind at the end of this essay, she is, "surprisingly," raising "an entirely new point," one that she has not broached earlier in it (p. 33). They are quite right to find it surprising that she would do such a thing (if that is indeed what she is doing) but I actually doubt that the point she is raising here really is a new one. After all, she had already mentioned the fact that art is "of profoundly personalconcern to most men" (the emphasis is Rand's) at the very beginning (second paragraph) of the essay, and her tone of voice there struck me as one that would be appropriate to stating the basic theme of the whole piece. The fact that she comes around to this fact again at the end of the essay, in a style that would be appropriate to a summary, only reinforces this impression: she is not abruptly introducing a new point but, on the contrary, is sounding a theme that she has really been talking about all along. I would read the concluding comment about "confirming or denying" in terms of various other things she has said since the beginning, including a remark that comes just one page before the summary-like ending. There she says that "To acquire the full, persuasive, irresistible power of reality, man's metaphysical abstractions have to confront him in the form of concretes -- ie., in the form of art." Art, she says, serves to counteract a natural feature of human consciousness: in the "day-to-day existence" of the individual human being "he is often in danger of losing his perspective and the reality of his convictions."(2) I interpret this language, which speaks of "perspective," "reality," and "power," as addressing a problem that is rather different from that of mere certainty that one's convictions are true. It is of course true enough that one's sense of the efficacy of one's consciousness can be enhanced by a heightened sense of certitude, but it is also true that it can also be enhanced by a renewed grasp of what one's fundamental convictions really mean. On the other hand, inefficacy can be experienced as uncertainty and it can also be experience as confusion.

Perhaps the function of art for Rand is to combat a certain sort of confusion by enhancing, over the long run, one's grasp of one's most fundamental principles. At present, I do not have a well-developed interpretation of Rand's views on these matters but (and this is really my only point for now) these views do seem worthy of a more extended examination than Torres and Kamhi have given them so far.

As I have suggested, by far the greatest part of What Art Is is devoted to detailed examination of the many issues raised by (3): namely, the definition of art in general. Their discussion of what art is and is not takes up, in one way or another, most of Part II as well as most of Part I. It is here that the book's considerable virtues can be fully appreciated. The number of interesting questions they address in this part of their work is vast, and there is no way I could hope here to do them justice here. I will have to content myself with a few very brief remarks.

Rand's account of what art is that has unorthodox implications concerning the issue of which genres fall inside and which fall outside the borders of art. There are really two features of this account that are relevant to these implications. One is the idea that art is a "recreation" of reality. As Torres and Kamhi illuminatingly suggest, Rand's theory is part of the grand tradition, which begins with Plato and Aristotle, that holds art to be essentially mimesis -- a word that has at times been mistranslated as "imitation" but really means something much more like "representation." It is of course because art is in this sense a "recreation" that non-representational painting cannot be art. In addition to this, however, art must also on Rand's account be a particular sort of reproduction, one that the artist builds detail by detail, selecting them in accordance with his or her fundamental values. This is why photography, in which the representation is built all at once by a mechanical device, cannot be art.

However, those of us who accept Rand's conception of art cannot content ourselves with magisterially declaring that this or that thing is not really art. This conception, and indeed any mimetic conception, has implications that are rather more far-reaching than is entirely consistent with such magisterial ease. It also seems to imply that one or two things that we ourselves wish to keep inside the realm of art actually fall outside its borders. Rather obviously, music appears to be generally non-mimetic. It need not, and usually does not, depict anything beyond itself. Similar statements can be made about architecture: unlike paintings, novels, and statues, buildings are not representations of things. Both these genres raise problems for a consistently mimetic theory.

Torres and Kamhi claim (very plausibly, I think) that Rand's discussions of music and architecture in "Art and Cognition" fail to solve these problems. Her account of music successfully avoids being implausibly mimetic, but it does so at the price of regarding music as entirely non-representational. They point out that this seems to conflict with her general account of art, and that she does nothing to resolve the discrepancy (pp. 85-86). They briefly sketch out a possible account of music which, they maintain, would be consistent with her general theory. According to this account, musical tones are stylized versions of natural human sounds, including tones of voice, for instance, and sequences of such tones. Music is a selective recreation of this aspect of reality.

Regarding architecture they see the same fundamental problem in Rand's account that they saw in her discussion of music: that she came up with an account that characterized architecture as non-representational, but that thereby contradicted her theory of art. Their own solution to this problem, as in the case of music, is a significant departure from Rand, but the departure in this case is in a different direction. The solution they propose is to do to architecture what they do to clothing design and quilting: they deny that it is an art at all. Thus there is no need to integrate it into a mimetic general theory of art, except by showing that the false view that it is an art causes certain sorts of mischief, as similar claims about other pseudo-arts do.(3)

As in so many other cases, I find what Torres and Kamhi have to say both interesting and plausible. So far, though, they have not managed to convince me. Among the reasons why I balk, there is one that might be worth articulating here. It is that I am not sure that the account of music that they suggest really does fit into the particular sort of mimetic theory that we find in Rand.

What counts as a "recreation of reality" for purposes of this theory? As I have suggested above, painting, sculpture, and the novel recreate things, according to this theory, in the sense that they depict or represent them: they are representations of a vase of flowers, a naked man, the life of brilliant architect. In that case, though, the account of music that Torres and Kamhi have recommended will apparently not fit into this theory, because this account does not seem to assert that, for instance, symphonic music represents or depicts human speech patterns, tones of voice, moans and shrieks. Indeed, the account should not say such a thing: music is not about tones of voice in the way that a novel is about an architect.

This might sound like a quibble, and in a way perhaps it is. If this is to be more that an issue about the meaning of a piece of technical terminology, we it will need to know more about what the point or value of the "recreation of reality" is, according to Rand's theory. Perhaps the reference the novel makes to the architect is crucial to the value of the novel has for us, and perhaps the same thing can be said about the value of a symphony and the reference it makes to natural human sounds. In that case, the fact that the symphony does selectively recreate reality (supposing that is what it does) will be a very important fact about it; it will give us a powerful reason to classify these two sorts of works together, as works of art. The reason would be that the reality-recreating features of each work have the same sort of value, and they have it for the same reason.

At first glance, though, this does not seem to be a very plausible. As we read the novel, we are trying to find out about the architect; as we look at the painting, we are peering at those flowers. Our interest in these objects (which of course may be entirely imaginary) is a crucial part of the value the these works have for us. However, we do not listen to a piece of music because we want to find out about the natural sounds that are referred to and stylized in them..

The elements of reality that are selected and stylized in the novel and in music are not only very different in kind, their presence in each case plays a radically different role. Yet they have one very important characteristic in common: though they do it in very different ways, they do serve the very same function. They both enable the work to concretize very basic values in such a way as to enable us to perceive these values directly, as if they were percepts. They serve to embody a particular sense of life. This suggests a possible alternative way to integrate music into an objectivist account of art.

As at least one commentator has already pointed out, Rand does not define art in terms of its function.(4) What her definition does mention is the process by which art is created. Psychologically, this is understandable, given that Rand was herself an artist. She defines art from the artist's point of view, as the result of a certain sort of process.. Understandable though it is, it is probably not the best way to define art. The reason is that it lays great emphasis on an aspect of art which, whether or not we accept the Objectivist theory that it is shared by all works of art, is clearly not equally important in all the arts. The fact that the novel depicts putative facts is very crucial to the way in which it embodies values and, consequently, it is also very crucial to the value that the work has for us. The fact that music also in some way depicts something is far less important, and such importance as it does have is probably very different in kind. Despite this fact, the Objectivist theory that both arts are valuable to us because they do embody values is eminently defensible, and does not select an aspect of art that obviously varies enormously from one genre to another.

What I suggest is that the definition of art ought to focus on this aspect. In other words, we should define art in terms of its function. This function is, to put it somewhat differently than previously, to create a certain sort of meaning, in which values are embodied in concrete form. In order to create this sort of meaning, the representational aspect of art is, according to Objectivist theory, an indispensable means. But it is only a necessary condition of something's being a work of art. It is not what makes it art. If we were to construct a definition of art more or less along the lines I am suggesting (and I must admit that at present I do not have such a definition to offer) the obvious differences between the novel and music with respect to their representational features might no longer be an embarrassment.

I make this suggestion with some hesitation, because it does seem to involve a substantive change in the Objectivist account of art. Fundamentally, my suggestion would promote the broadly cognitive function of art to a status that is more central to the account, and it would demote the idea that it is a selective re-creation of reality to a more peripheral sort of status. An Objectivist account of art that has undergone this sort of reconstruction might well be unable to apply the notion of recreating reality, as Torres and Kamhi and indeed Rand herself often do, as a mechanical litmus test that shows, without any other considerations involved, that this or that genre cannot be art. It may not seem so obvious, for instance, that photography cannot be art, simply because the details that the photographer built a picture of reality by selecting a detail at a time. The issue raised for a reconstructed Objectivist esthetic by a genre like photography would be this: Given that it does contain representational elements, and given that the representation is achieved by acts of selection on the part of its creator, do the works in this genre have enough of these features to enable it (given in addition the resources of its unique media of expression) to do what art does? For a functional account, what art is follows from what art does.

Many of the issues broached in this book cry out for exploration. I should emphasize, before I take leave of the subject for the present, tha What Art Is takes enormous first strides in exploring all the questions raised in its pages. My complaints about the book, if they can be called that, tend to concern things it fails to explore or to explore more fully, rather than what its authors actually do say. I hope that in future publications they will make up for these omissions. In particular, it would be wonderful to see them discuss in greater detail the issue of the function of art and the less fundamental but equally interesting matter of Rand's defense of Romanticism. Better yet, I hope many others with do so, making objectivist esthetics the lively arena of intellectual give and take I have long thought it can be and ought to be.

1. Torres and Kamhi tell us that Rand preferred the Anglo-Saxon to the latin spelling of this word.

2. In The Romantic Manifesto: A Philosophy of Literature, second revised edition (New York: Signet, 1975), p. 23.

3. They illustrate this point with some deflating comments on the work of Frank Lloyd Wright. His buildings, they say, are often "maintenance horrors," the Johnson Wax offices "a haven for mice," and are characterized by other design flaws. Their point is that the Wrightian mind set that regards architecture an art tends to interfere with sound, practical design, which is essential for a building to be a good one.

4. See p. 105 and p. 378 n. 52, as well as Torres and Kamhi's "Critical Neglect of Ayn Rand's Esthetics,"The Journal of Ayn Rand Studies, vol. 2, no. 1 (Fall 2000), pp. 15-17 and n. 33.