Production Theories and Artistic Value

1. Housekeeping Points[1]

It needs to be made perfectly clear that the value at issue here is not aesthetic value but artistic value, the value that works of art have in respect of being works of art. Although the distinction between artistic value and aesthetic value breaks down in Beardsley's account, accounts which are affective (and so, presumably, expressionist) or cognitivist are solely about works of art. Experiences of many natural objects (and events), and experiences of many non-art artifactual objects, can be aesthetic or have a strong aesthetic component. Indeed, I would wager that most aestheticians today believe that an aesthetic perspective can be taken to any object, so long as that object is phenomenal or, in principle, sensory. So aesthetic value as a category is much wider than artistic value, since only a small percentage of artifactual objects are works of art. I will argue, though, that artistic value is not a species or a subset of aesthetic value. Some art objects are not commonly viewed from an aesthetic perspective; I believe that taking an aesthetic perspective to some artworks is to miss what is most important about those works as art.

Each of the production theories under discussion here can be understood as having two distinct functions. One function is evaluative. One can assess the merits of a work of art through consideration of the presence and strength of the features on which these accounts focus. This is something that Dickie makes evident in his Evaluating Art,[2] at least with regard to the aesthetic and cognitivist models (chapters four and six, respectively). If we are using Tolstoy's work as our affective model, then one need only recall that Tolstoy himself used his model for evaluation: "And not only is infection [of expressed feeling] a sure sign of art, but the degree of infectiousness is also the sole measure of excellence in art."[3] Frankly, the value-focused nature of each theory is the very point of this paper, so this is a matter on which no more time need be spent just now.

The other function is definitional. Each theory defines what characteristics must be present for an object to be properly considered a work of art. In the cases of Tolstoy and Goodman, this is straightforwardly the case. It is less obvious in Beardsley's, though. However, in a late work (1979), "Redefining Art," Beardsley writes:

"I say that an artwork is either an arrangement of conditions intended to be capable of affording an experience with marked aesthetic character or (incidentally) an arrangement belonging to a class or type of arrangements that is typically intended to have this capacity.... I can hasten to add at the moment is that it is to be understood from the start that the arrangements I speak of often are created with more than one intention, but what makes them art, on this definition, is that the aesthetic intention described above is present and operative."[4]

This is enough to be able to say that Beardsley's account had a definitional component.

Would it make sense, given the definitional facet of each of these views, to look for other production theories that are grounded in definition? I would consider the "artworld" views of Arthur Danto here,[5] but they are a bit abstract for the modesty of the point I want to make, so it would be better to consider the more concrete Institutional Theory of Art advanced by Dickie. Dickie writes: "A work of art is an artifact of a kind created to be presented to an artworld public."[6] This appears to be a revision to his earlier and perhaps more famous definition: "A work of art in the classificatory sense is (1) an artifact (2) a set of the aspects of which has had conferred upon it the status of a candidate for appreciation by some person or persons acting on behalf of a certain social institution (the artworld)."[7]

Although there may not be an apparent value component in this account to exploit in recasting this view as a value-production theory, we might import a means by which to consider it as a value-production or value-teleological theory, namely one from Aristotle. Aristotle held that the goodness of an object can be judged on how well it performs its function, in respect of the sort of thing it is; something is good if it is a highly functioning one of its kind. If it is legitimate to bring this to bear on art theories, then perhaps a good work of art, following Dickie and Aristotle, is a work that functions highly as "an artifact of a kind created to be presented to an artworld public."

Out of this account, we might be able to wrestle the notion that if a work of art is wellaccepted by the artworld - with "accepted" being the productive end of being presented - then it is a good work. But then the question has to be: Accepted how? What is it that constitutes the nature of the acceptance? And with that question, we move back down to the level of detail offered in the three theories with which we started. Even if I take massive liberties with Dickie's account, it seems that coherent value accounts cannot come from definitional accounts wholesale. This gives me confidence that the scope of my argument is about right.

If my focus is on discussing what makes works of art valuable, I do need a theory of what counts as a work of art. I need a means of demarcating the range of my claims, and so here I will turn to Danto's artworld theory. I reject artworld theories as sufficient for providing accounts of the value of art, but since some non-question-begging means of accounting for my subject matter is necessary, I accept the "classification function" of the artworld as expressed in Danto's work.

Using an artworld theoretic approach to classifying what counts as art is to limit the range of my discussion and the range of my claims to the range of the artworld. "Artworlds," if a plural use of that term is appropriate, have boundaries that most likely are coextensive with boundaries of culture. A European-American artworld may at times overlap with an East Asian artworld or with a Middle Eastern artworld, but judging from what may count as a member of the canon of the European-American artworld and what may count as a member of the canon of the East Asian artworld, and judging from the sorts of aesthetic and artistic sensibilities that seem prevalent in (e.g.) these two spheres, it seems the most honest position is to allow the pluralization, to speak of "artworlds."

To take a step further, it may not even make sense, relative to some cultures or societies, to speak of an "artworld" in those cases where a particular culture or society has no objects that are relevantly like the objects European-Americans take to be works of art. That is, there may be cultures or societies which do not have "art" and "non-art" as part of their ontologies, and so all theory about art and artworlds would be meaningless as applied to them.

If this is the case, then my use of Danto's artworld theory for circumscribing the range of my discussion may necessarily limit what I have to say to just the European-American artworld. I do not take this as a grave limitation; we all have our contexts and our perspectives. To locate a theory in a context and thereby limit it to that context may be less ambitious than to create a theory that transcends cultural context, but to do the latter involves inherent dangers that make that level of ambition potentially unwise.

Each of the production theories under discussion here purports to be essentialist. Although I would be surprised to learn that any one of these theorists made an explicit issue out of the completeness of his theory, I take it that these theorists believed that their accounts captured what is fundamentally the case about artistic value. This assumption on my part is necessary if I am to argue against their completeness. If I am wrong in this, it needs to be shown.

I will not discuss my rejection of object-intrinsic value accounts here. I take that up a bit in the "Artistic Value" paper, and the arguments that focus on epistemological access problems to intrinsic value properties as well as the metaphysical problems in trying to come to grips with the nature of these properties are well known. At the very least, if not reason for outright rejection of such theories, the problems warrant avoidance of inclusion of them here. Whether the instrumental accounts discussed in this paper produce values that are intrinsic or instrumental for the production of still other values is a secondary matter and one beyond the present scope.[8]

Given that the accounts I am considering are instrumentalist in nature, the obvious first question to be asked is: Do real world experiences of artworks always produce the "deliverable" that such instrumental accounts promise? My strategy in criticizing the individual theories will follow this question.

2. Beardsley's Aesthetic Account

Beardsley's original account of artistic value comes from his 1958 book, Aesthetics: Problems in the Philosophy of Criticism[9], in the chapters entitled "Critical Evaluation" and "Aesthetic Value." There he described the three General Canons of aesthetic merit in works of art: the degree of unity or disunity in a work, the degree of complexity or simplicity, and the degree of intensity or lack of intensity.[10] Each of these canons represents some quality of an aesthetic experience rather than of an aesthetic object per se, though these qualities are objectively focused, that is, focused on the formal qualities of the object under aesthetic consideration. He writes:

"First, an aesthetic experience is one in which attention is firmly fixed upon heterogeneous but interrelated components of a phenomenally objective field - visual or auditory patterns, or the characters and events in literature... Second, it is an experience of some intensity... But this discussion already anticipates the two other features of aesthetic experience, which may both be subsumed under unity. For, third, it is an experience that hangs together, or is coherent, to an unusually high degree. Fourth, it is an experience that is unusually complete in itself... [B]ecause of the highly concentrated, or localized, attention characteristic of aesthetic experience, it tends to mark itself out from the general stream of experience, and stand in memory as a single experience.... One aesthetic experience may differ from another in any or all of three connected but independent respects... I propose to say that one aesthetic experience has a greater magnitude - that is, it is more of an aesthetic experience - than another; and that its magnitude is a function of at least these three variables."[11]

The subjective focus of Beardsley's criteria come out more strongly in his 1979 revised list,[12] and the instrumental character comes out strongly when he writes:

" 'X has greater aesthetic value than Y' means 'X has the capacity to produce an aesthetic experience of greater magnitude (such an experience having more value) than that produced by Y.' Since this definition defines '"aesthetic value' in terms of consequences, an objects's utility or instrumentality to a certain sort of experience, I shall call it an Instrumentalist definition of 'aesthetic value.' "[13]

It is the marriage, so to speak, of his chapter on "Aesthetic Value" with his chapter on "Critical Evaluation" that establishes the point that Beardsley understands artistic value in terms of aesthetic value. The union is strengthened when he writes: "[A]n artwork can be usefully defined as an intentional arrangement of conditions for affording experiences with a marked aesthetic character."[14]

While it may well be true that one can take an aesthetic perspective, in line with Beardsley's description of it, to any object (so long as that object is phenomenal or in principle sensory), it would be odd indeed if that perspective turned out to be appropriate when it comes to many of the objects created within the past century. Aesthetic accounts of artistic value, when they are presented as complete analyses of artistic value, suffer from the presence of too many available counterexamples. Consider Marcel Duchamp's In Advance of a Broken Arm. It is physically a snow shovel; it is green and red and was purchased by Duchamp right off the rack. In Advance of a Broken Arm, like any other of Duchamp's so-called Readymades, was not originally a work of art. However, the adoption of it by Duchamp as art rendered the object art, or at least his act introduced the candidacy of the object to be recognized as art.

What makes the snow shovel with which Duchamp left the hardware store different from the ones he left behind? Physically the set of snow shovels is identical. We know this, because were they different from one another, Duchamp's statement in choosing the shovel to be "elevated" above the rest would be lost. Readymades are all essentially not physically distinctive. This being the case, we would well say of someone in a gallery setting who was concentrating on the phenomenal features of the shovel that he "just didn't get it," that given the historical context of the object, an aesthetic perspective is not only inappropriate for reaching the true value of In Advance of a Broken Arm as a work of art, but it will prove rather, perhaps distinctly, unrewarding. In Advance of a Broken Arm is but one of a long list of similar works.

Recent art exhibitions have only added fuel to this. Damien Hirst, one of the Young British Artists, recently found himself, with artist Christopher Ofili, at the center of a major controversy surrounding the Brooklyn Art Museum's exhibition of a show entitled Sensation. If Hirst's work, or at least these pieces, has artistic value, surely it does not lie in its potential to create in viewers experiences that are aesthetic. To add yet a bit more to this point, the famed Sister Wendy Becket says, in allusion to the work of Jasper Johns, as she is discussing the conceptual nature of Modern Art:

"[W]hat he really want to communicate is an idea. Is this a flag or is it a work of art? A concept? Now this conceptual art is very popular at the moment - popular with the artworld, not with the rest of us. And often you see the stuff; you get the concept, and then you move on. You've lost interest. So here's another question: When is conceptual art great art? And the answer is: when it gives deep visual satisfaction, like Jasper John's flag.[15]

Sister Wendy draws a distinction between conceptual art which gives deep visual satisfaction and conceptual art which does not. I take "deep visual satisfaction" to constitute a rewarding aesthetic engagement, using the word "aesthetic" in line with Beardsley's views. She suggests that a good deal of conceptual art does not provide any satisfaction except a cognitive one, and there is a slight suggestion that this cognitive engagement is at times fleeting, perhaps even unrewarding. It is just the distinction that Sister Wendy points out that is at issue here.

The conclusion has to be that aesthetic experiences and art experiences - if it makes sense to use that second expression - are essentially different things. There are many aesthetic experiences that are not experiences focused on art objects. Certainly that's uncontroversial. But there are a good many art experiences, if indeed we want to follow the artworld's lead on what counts as art, which are not best viewed aesthetically -- which when viewed aesthetically actually lose value. This problem makes it appear that aesthetic-experience production theories cannot be the whole story.[16]

3. Goodman's Cognitivist Account

I take Goodman's account from his book, Languages of Art,[17] in which he theorizes that art is essentially symbolic. A given work of art functions as a symbol (or sign), or a set of symbols. Goodman differentiates between art symbol systems and non-art symbol systems through a series of distinctions.[18] This is of course important because there are many symbol systems that have nothing to do with art. One of the first things to recognize about Goodman's theory is that it is, at heart, a representational theory. If works of art are symbols, they must refer. To what they refer is not really the point, but reference is essential. "The plain fact is that a picture, to represent an object, must be a symbol for it, stand for it, refer to it."[19]

The second thing to recognize is that symbol systems are human creations, human developments, and must be learned in order to be applied and to be understood, or, in Goodman's terms, "read." "Pictures in perspective have to be read; and the ability to read has to be acquired."[20]

Perhaps the most important thing about Goodman's theory for present purposes is its focus not on the sensory or phenomenal, but on the cognitive. Goodman writes:

"Use of symbols beyond immediate need is for the sake of understanding, not practice; what compels is the urge to know, what delights is discovery, and communication is secondary to the apprehension and formulation of what is to be communicated. The primary purpose is cognition in and for itself; the practical, pleasure, compulsion, and communicative utility all depend upon this. Symbolization, then, is to be judged fundamentally by how well it serves the cognitive purpose: by the delicacy of its discriminations and the aptness of its allusions; by the way it works in grasping, exploring, and informing the world; by how it analyses, sorts, orders, and organizes; by how it participates in the making, manipulation, retention, and transformation of knowledge."[21]

This clearly places the cognitive function in the center as regards both the understanding of what art is (the definitional component) and assessing whether a given work of art succeeds and to what degree it succeeds (the evaluative component).

Does Goodman's account succeed? I would reject Goodman's theory as a complete account of artistic value because, while we all no doubt have had the experience of attending to a work of art in a problem-solving or puzzle-solving frame of mind, the very fact that we can identify when our experience of art is cognitive suggests a distinction from those art experiences that are not. I can take a puzzle-solving attitude toward an art object, working out for myself the internal logic of the piece, the rules that this particular artwork instantiates and follows, and even, for good measure, understanding how the object refers to and represents other things.

But I can just as easily, even through a conscious and volitional choice, adopt an attitude of passive acceptance of what is being offered (to my senses), taking delight merely in the sensory stimulations themselves or in how they make me feel. I am not very familiar with the technical aspects of music, so my entree into appreciating it is generally through the affective. I am more familiar with the technical aspects of dance, but I find I can move between two attitudes, one cognitive and the other not, easily and fluidly.

While a critical appraisal of the object under consideration may issue forth from the cognitive-engagement frame of mind more readily or easily than from a different frame - keeping in mind that we are considering value here - I submit that this is in large part because linguistic articulation of the value that I am experiencing flows more easily or readily when I am already cognitively engaged. Words flow a bit less fluidly when my mind-set is emotive or purely focused on the phenomenal. But I would suggest that on many occasions, the value of the work under review is heightened when considered from a non-cognitive vantage point. While Goodman does not discount the emotive or the sensory - as Beardsley does not discount the cognitive - his primary focus in accounting for artistic value is centrally lodged in the cognitive, and this seems too narrow.

4. Tolstoy's Affective Account

Leo Tolstoy envisioned art as essentially a form of communication.[22] Art is meant to communicate universal emotion, which is felt by the artist and is the subject of her work, and is then communicated to her audience. It is not enough for the artist to have felt something and produce some artifact resulting from that feeling. What has to take place, for the work of art to be successful, is for us to feel what the artist felt, or at least for us to feel what the artist's work can make us feel.

Every true work of art causes the viewer to enter into a special relationship with the artist, and not only with the artist, but with everyone else who has at one time or other entered into that same relationship. The "artistic relationship" between artist and audience builds a community, a community of creator, of object, and of all those who experience the object. The artist's job is to evoke in herself some feeling once experienced, and then, once having evoked it in herself, to communicate it to her audience through some sensual medium, through colors, shapes, melodies, harmonies, figures, movements, and so on. The artist seeks to infect her audience with these feelings: "the degree of infectiousness is the sole measure of excellence in art." "Infectiousness" translates into how intensely the viewer experiences the artist's emotion, how clearly she feels it, and how sincere it is.

Problems with affective theories were known very early. Tolstoy's contemporary expressionist theorists, Benedetto Croce and R. G. Collingwood, had already created theories that jettisoned the focus on pure feeling and the infectious communication of that feeling for expression of, in Croce's case, intuitions, and in Collingwood's case, individualized emotion properly explored, contextualized, and demonstrated. However, interpretation of art that focuses on the affective is still alive and well. Consider the following from Sister Wendy:

"I'm not afraid you won't think this Mark Rothko beautiful, but what I am afraid, a little, somebody might think it's just beautiful. Lovely colors. No meaning. But meaning is what he was all about, and he would have been furiously angry if anyone thought that, and told you so in suitably salty language. It was subject matter that mattered most to him. And the subject matter was the emotions. Not small, personal emotions - up today, down tomorrow - but the great timeless emotions. How we feel about death, and courage, and ecstacy. He was convinced that if you would just encounter his paintings, that emotion would be communicated to you with absolute clarity. So to achieve this he painted very large. Because in a small painting - big you, little painting - you can control it. But with a large painting, it controls you. You're taken into it. Unless of course you look at it from a distance, that killing, assessing look. So to combat that, he insisted that always the light be very dim, so you couldn't actually see the thing until you were right up against it. And then something does begin to happen. He painted with very thin mists of paint, feathering it on, breathing it on. And you are taken up, out of yourself, into something greater, something transcendent and majestic. If you can think of a religious painting without religion, this is what you experience here. It's so timeless, that when I've had this encounter, I feel to return to the world of time, I have to shake my head and bring myself down to earth again.[23]

Sister Wendy's work is as good a popular sort of art interpretation as any, I imagine, and if affective treatments like the one she offers of, and actually ascribes to, Mark Rothko, are still effective in communicating the value of a work, then affective production theories belong in this paper (although I will admit that I went for the easiest and most straightforward one with Tolstoy).

One problem for the production-of-emotion model is that degree to which the feeling must be, to use Tolstoy's work, infected. Clearly Tolstoy's artistic aim is not simply cognitive appreciation of the expression of emotion. The emotion has to be felt by the audience. But then the questions are: How and How Much? The edict to infect the audience with a sincere level of the feeling being expressed leaves one with these sorts of question. Another problem is understanding or accessing artist intentions as they constitute the source of the artist's emotions. Though we commonly expect that the artist can and probably does know her intention regarding a given art creation, and though she can communicate this to others who wish to criticize, interpret or merely appreciate her work, the intention of the artist is often something that cannot be readily discovered.

Although one can be reasonably certain of, say, Michelangelo's intention regarding the creation of the Pieta, a casual viewer at New York's Museum of Modern Art may be hard pressed to explain the intention behind any one of the mature, untitled works of Jackson Pollock. The difficulty here is a simple one. How is it that we can know that a work contains or is an expression of emotion? This may be obvious in many works, but this is a more difficult task when it comes to formalized or highly abstract works. If one is relegated to having to fathom the intention of the artist in order to determine whether the work is an expression of emotion, one may find oneself silent. This problem, coupled with the earlier one, renders expressive-affective theories of artistic value lacking.

5. Goldman's Alternative World Account

Alan Goldman, in his Aesthetic Value, offers a production account of value. His view is broad, incorporating all perspectives:

"The value of such works lies first in the challenge and richness of the perceptual, affective, and cognitive experience they afford. Symbolic and expressive density combines here with sensuous feel. From the subjective side, all one's perceptual, cognitive, and affective capacities can be engaged in apprehending these relations, even if one's grasp of them is imperfect or only implicit. These different facets of appreciation are not only engaged simultaneously but are also often indissolubly united, as when formal relations amount musical tones or painted shapes are experienced as felt tensions and resolutions and perhaps as higher-order or some ordinary emotions as well."[24]

His account is this:

"When we are so fully and satisfyingly involved in appreciating an artwork, we can be said to lose our ordinary, practically oriented selves in a world of the work.... [It] can engage us so fully as to constitute another world for us, at least temporarily."[25]

Goldman evades complaints about narrowness of scope in constructing a theory that is very broad indeed. In the production of experiences of alternate worlds, the artwork can trigger a huge range of different sorts of experiences that will be subjectively efficacious. For me, a single combination of smells can invoke another whole world (in my case, cigarettes, perfume, and diesel exhaust put me in London instantly and thoroughly). If one has a whole book or an entire film in which to develop alternate world cues and contexts, the effect - if the book or film is good - will surely be pronounced. Just think of all the people influenced by Tolkien's trilogy and who, even these decades later, have never really gotten back out of Middle Earth. If Goldman's account is so broad, and if we take it to constitute a theory of artistic value, then is there any criticism left to make of it? If the "deliverable" that Goldman promises as the instrumental product of art is so broad, and can be produced in such an incredible variety and number of ways, does his account fully succeed?

6. The Modification Problem

I have only one criticism left, but it is a criticism of all production theories of artistic value. One difficulty for all of the accounts we surveyed, Goldman's included, concerns modification of the object under consideration. If the worth of an art object is grounded in its potential for producing aesthetic experiences of a decently high magnitude, and better works of art are those that produce experiences of higher magnitudes (to use Beardsley's word, but to think of this in terms of each of the accounts) than lesser works of art, then it would seem that we could do artworks and art audiences a service if we modify works of art of lesser artistic quality in ways that enhance their artistic value.

In 1919, Duchamp drew a moustache on the Mona Lisa and named it L.H.O.O.Q. Duchamp did not draw a moustache on the actual Mona Lisa, of course, but on a copy. If Duchamp had drawn a moustache on the original Mona Lisa, the one painted by Leonardo's own hand, then I would wager that very few people would have been okay with that. In 1959, Robert Rauschenberg asked Willem de Kooning for permission to erase one of his drawings. De Kooning gave his permission, the erasure was made, and the erased drawing was displayed under Rauschenberg's name with the title Erased De Kooning Drawing.

While Rauschenberg was able to get away with this, he could do it only once - Rauschenberg was the right artist in the right context at the right time - and he did it only after securing de Kooning's permission. No doubt there are other cases of artistic modification, but the number is extremely small. In general, art audiences believe there is something seriously wrong about the modification of a work of art. Yet, if the value rests exclusively in the productive value of the work, and the "deliverable" can be increased through modification - either by the artist herself at some point after the work has been presented for viewing or by another, perhaps more gifted, artist - then artistic modification should not affect us so. Indeed, in many situations, we should welcome it.

I want to be clear here in saying that in rejecting wide-spread modification of works of art, I am trying to account for what I take to be a very strong intuition about art. I believe there is a strong intuition that once a work of art is complete -- once the artist has set aside her brushes or chisels or pen -- the work has a certain value in terms of its being a work of art. To modify a completed work is to jeopardize that value, to put it at risk or even to destroy it. Even in those cases where one owns a work of art, modification of that work - say, cutting it to fit a particular frame - seems a cause of distress to art lovers, regardless of considerations of property rights.

The value of art transcends property ownership, or at least the intuition of many art lovers is that it does. What value completed works of art have is what is at issue in this paper, but one thing seems very clear: Production accounts of the value of art do not and cannot take seriously the intuition that most lovers of art feel about the prohibition against modifying works of art. Production theories cannot account for this.

The value at issue here is the value the work has in virtue of its being a work of art. Surely there are many contextual or historical features of works of art that add to, or perhaps even primarily account for, the value of certain works - provenance, the ability of a work of art to communicate great religious meaning, teach a valuable moral lesson, or serve as a source of social or political unification - but I mean to focus narrowly on the artistic value of these works.

One may argue that there are at least three reasons why permission to modify artworks does not work as a counter-argument to production theories. First, modification does not, as a matter of fact, increase the artistic value of modified works of art. Second, the modified work of art is no longer the same work of art as the original; it is a separate and distinct second work. And, third, the reasons that we find modification of works of art objectionable is not because of considerations of artistic value, but because modification actually diminishes the value of the work in other ways.

In response to the first point, that modification does not, as a matter of fact, increase the artistic value of modified works, I would make two points. First, my argument concerning modification is a logical one, not an empirical or contingent one. Production theories allow the possibility of value-enhancement through modification, and if we find modification objectionable, then we should reject production theories of artistic value. To say that modified works are never artistically better than the original works, in a way where this claim is not empirical and contingent, requires a theory of artistic value, and this is precisely what is at issue. To use as evidence the facts that we do not find modified works better is to leave open the possibility that we may in some future cases.

The second point I would make in response to this first objection is that there may be, right now, cases where we think modified works are superior. I think few people would believe that Duchamp's L.H.O.O.Q. is a superior work to Leonardo's Mona Lisa, but I am not as secure in this same intuition when it comes to Rauschenberg's Erased De Kooning Drawing. I do not know what the original De Kooning drawing looked like, but I know that Rauschenberg's work is, to the extent that there is any agreement about such things, a work of art of some value.

Throughout this paper I tend to focus on paintings, but we may consider other art forms. It is certainly within the realm of possibilities that a majority of people find colorized versions of certain films better than the originals. Ted Turner may be a member of such a majority. Younger viewers or simply those without much experience of black-and-white film or television may find colorized versions of films more aesthetically accessible and so perhaps more engaging. Better examples may come from music, where variations by composers of the works of earlier composers are standard. The chances are great, I would wager, that a majority of listeners actually find Mozart's variations better than their originals, for instance, Mozart's variation on Salieri's Mio Caro Adone. But one may argue that Mozart's Mio Caro Adone and Salieri's Mio Caro Adone are different works, as Frank Capra's original It's a Wonderful Life is a different film from Ted Turner's colorized version.

This then leads us to the second objection, that in the examples I offer we are not talking straightforwardly about one work that undergoes modification; we are rather talking about two separate works: the original and the modified version. This is, one may argue, why some (although not Woody Allen) find colorization of films acceptable. One is not damaging the original film in making a colorized copy;[26] one is creating a separate film. This is perhaps even easier to say in the case of musical variations.

This sort of position is consonant with the view of Mark Sagoff[27] concerning artwork restoration and copying. Sagoff argues that works of art are highly individual because they are the products of a particular artistic process. If that work undergoes a secondary process, not part of the original process of the original artist, the resulting work is no longer the original work but a second new work. Sagoff talks about the restoration of Michelangelo's Pietà  after it was attacked with a hammer in 1972. He praises the restorer, Redig de Campos, for taking pains to ensure that what changes he made to restore the Pietà  to a condition that is visually undetectable from its pre-1972 state could be easily detected and easily reversed by future restorers or caretakers of the work. The 1972 lunatic changed Michelangelo's Pietà ; to change it further, even with the intent of (phenomenally) restoring it to its original state would not be to reverse the imposition of the lunatic's "new process" but actually to add a third "process" to the history of this work.

I should point out that I am in great sympathy with Sagoff's ultimate point, which I take to be a rejection of Beardsley-style arguments for artistic value being a matter of production of aesthetic experiences, and with his ancillary point concerning the impermissibility of modification. The only place I part company with him is over the ontological status of the modified work. I differ for two reasons. First, while it may be readily acceptable that Mozart's Mio Caro Adone and Salieri's Mio Caro Adone are different works, this is less clear in the Rauschenberg/de Kooning case. Should we rather say that de Kooning's drawing has ceased to exist or say that Rauschenberg's work is an evolution of de Kooning's? Were I de Kooning, I would certainly prefer the latter, and it would be on that basis that I would be motivated to grant permission for the erasure to be done.

My intuition is that de Kooning is every bit as important as a part of the artistic process which resulted in Erased De Kooning Drawing as is Rauschenberg. Rauschenberg did not erase any old thing. He erased a de Kooning drawing. The de Kooning drawing was not destroyed or made to cease to exist; it was one step of a single artistic process that resulted in the work Erased De Kooning Drawing. But more to the point, the "provenancial process" that led to the creation of this work did indeed include a modification of one artist's artwork (a bona fide artwork in its own right) by another artist, who, I have to imagine, was motivated to create a work of greater artistic value, greater artistic significance, than the original drawing. Based on the attention that Erased De Kooning Drawing has received in the artworld, my intuition is that they succeeded.

My second reason for differing with Sagoff is this. When an artist creates a work of art, it is the artist herself who, at some point of her own choosing, pronounces a work complete. From a non-practitioner's point of view, I may be tempted to call this point arbitrary. The difference between a painting with one less (minute) brush stroke and one more (minute) one is, in the vast majority of paintings throughout the history of art, not significant. Certainly some paintings could not bear one stroke more or less, but this is, in art history, a small minority of works.

Whether my intuition about this is shared or not does not matter; the point is this: If an artist pronounced a work complete but two days later reconsidered and put in a few more brush strokes, it is difficult to see how this meaningfully constitutes the imposition of a second artistic process. (I wager that most writers of philosophy papers have had the experience of believing a paper to be complete, but then having another thought and returning to the computer.) I contend that what this returning artist does to her work is on a continuum with what Rauschenberg did to the de Kooning drawing. The difference is one of degree but not of kind. If this is the case, the modification of one work which results not in a second work but rather merely an evolution of the first original work is a possibility. If such modification is a possibility, production theories of artistic value allow for it.

The last objection I want to take up concerns whether the modification of works of art is objectionable because of artistic value considerations or because modification diminishes the value of the work in other ways. A good example of this comes from a world related to the artworld, the world of antiques, specifically antique furniture.[28] A chair, say, that is quite old will have become dark and dull with age. The novice collector of such an antique chair may think that he can restore the chair to its original brilliant condition by stripping off the finish and putting a new, probably more protective, finish in its place. This will bring out the woodgrain, brighten up the piece, and, generally, make the piece more directly aesthetically pleasing. The problem with this, as all viewers of the Antiques Roadshow will know, is that such a restorative act will actually diminish the worth, i.e., the value, of the chair immensely. The value, it may be argued, that needs to be protected is not the aesthetic value per se but rather the value of the chair as something that has been around a very long time, a value of longevity that is indicated in its economic value.

I have two answers to this. First, modification of the chair in this example is motivated precisely by a theory of (artistic) value that I reject. In this example, one modifies the chair to enhance one's (direct, sensory) aesthetic experience of the chair. The chair unmodified, on the other hand, has a much closer connection to the chair that was, to return to Sagoff's view, the product of the process of a particular artist or, in this case, the furniture maker. But this answer does not perhaps get at the root of the objection. So, second, I would answer that my goal in this paper is not to put forward a theory of artistic value. I do that elsewhere. My goal is simply to show that production theories of artistic value are insufficient. This being the case, it may be that considerations such as longevity do indeed play a role to a sufficient degree (or sufficient degrees) of artistic value, perhaps even along the lines hinted at above. I am not obliged to say at this point how this would be.

Production theories of artistic value inherently and logically allow the modification of art works where that modification will enhance the instrumental "deliverable" effectiveness. If we believe that art work modification, on the whole, is not appropriate, then production theories suffer. The embracing of modification by production theories is necessary because their very logic is predicated not on the value of the object per se but on the experience of the viewer. The better the experience, the higher the artistic worth of the object that produces it. The better the object, the better the experience. Modify away!

7. A Non-Instrumental, Extrinsic Value Account

I have not in this paper offered a strong argument against modification, and to some degree I have celebrated the modification by Rauschenberg of de Kooning's drawing. The case against the general modification of works of art is made to some degree in the paper where I advance a theory of artistic value, but essentially this case rests on very widely shared intuitions. It is indeed the rare individual, even the rare libertarian capitalist, who would believe that once he owns a work of art, it is really his to do with as he wishes. Owners of artworks are caretakers of something whose full value cannot be measured on the same scale or in the same terms as the scope of their ownership.

The modification problem is a species of a larger problem, and that concerns the way in which we actually do value art. We build museums and galleries for art works, and industrialists pay millions of dollars for a single piece. But we do not do this for any other objects or events that possess or deliver the value(s) that production theories claim for art. We do not build such houses or pay such prices for purely aesthetic objects, or purely cognitive objects (the closest would be an arena for chess matches; puzzle museums don't count), or purely emotive objects (the closest is a movie theatre, but that's only if the film, either the particular film or film in general, does not count as art).

If we understand the "symptoms" of real-world valuing in terms of money and care, works of art have a value that far surpasses the price of pigment and canvas, etc., in the actual world. For us to say that their value lies in producing certain experiential states (or, really, producing anything) should be to say that we would pay and care equally for non-art objects that produce those same sorts of states, but in reality we do not. To chalk this up to our being acculturated or socialized to take care of art, without regard for these philosophical considerations of its value, is not to do service to the fields of everyone reading this journal.

In my paper "Artistic Value," I argue for an alternative to production theories. It seems to me that even if we reject object intrinsic artistic value accounts, this leaves us not just with instrumental accounts but with extrinsic accounts, of which instrumental accounts are a species. I suggest that we consider, as an extrinsic account of artistic value, a focus not on the audience but on the artist herself. The value of a work of art is located subjectively in individuals who respect the art object as the product of the artist, her time, talent, skills, labor, concentration, and perhaps above all as the instantiation of her valuable and irreducible expression. The respect we accord a given art object is borne on a respect for the artist's efforts.

This jibes well I think with the actual way we - perhaps "we" as Westerners - value art. We tend to understand and appreciate art in terms of who it came from, who the artist was, what her influence was, and the rest. This may not be a good thing, it may be snobbish and elitist and impure and all the rest, but it explains our ordinary experiences with art. It explains why an industrialist will pay millions for a Monet. It explains what we choose to house in the Louvre, the National Gallery, the Tate, and MOMA.

This view is argued for in that previous paper, of course, so it does not make sense to repeat the argument here. I simply wanted to close with this suggestion so that my rejection of production theories of artistic value could end on a positive note.