Excerpted from Art and Its Objects by Richard Wollheim. Copyright © 1980 by Cambridge University Press. Reprinted by permission of Cambridge University Press.
It is a deficiency of at least the English language that there is no single word, applicable over all the arts, for the process of coming to understand a particular work of art. To make good this deficiency I shall appropriate the word ‘criticism’, but in doing so I know that, though this concurs with the way the word is normally used in connection with, say, literature, it violates usage in, at any rate, the domain of the visual arts, where ‘criticism’ is the name of a purely evaluative activity.
The central question to be asked of criticism is, What does it do? How is a piece of criticism to be assessed, and what determines whether it is adequate? To my mind the best brief answer, of which this essay will offer an exposition and a limited defence, is, Criticism is retrieval. The task of criticism is the reconstruction of the creative process, where the creative process must in turn be thought of as something not stopping short of, but terminating on, the work of art itself. The creative process reconstructed, or retrieval complete, the work is then open to understanding. To the view advanced, that criticism is retrieval, several objections are raised.
The first objection is that, by and large, this view makes criticism impossible: and this is so because, except in exceptional circumstances, it is beyond the bounds of practical possibility to reconstruct the creative process.
Any argument to any such conclusion makes use of further premisses—either about the nature of knowledge and its limits, or about the nature of the mind and its inaccessibility—and the character of these further premisses comes out in the precise way the conclusion is formulated or how it is qualified. For, though an extreme form of the objection would be that the creative process can never be reconstructed, the conclusion is likelier to take some such form as that criticism is impossible unless the critic and the artist are one and the same person, or the work was created in the ambience of the critic, or the creative process was fully, unambiguously, and contemporaneously documented by the artist. This is not the place to assess the general philosophical theses of scepticism or solipsism, or their variants, but it is worth observing that these theses ought not to be credited with greater force outside general philosophy than they are inside it. The observation is called for, because traditionally philosophers of art permit the creative process, or, more broadly, the mental life of artists, to give rise to epistemological problems of an order that they would not sanction in inquiry generally.
These difficulties apart, the objection in its present form offers a persuasive rather than a conclusive argument against the retrieval view. For maybe the truth is that criticism is a practical impossibility, or is so outside very favoured circumstances. But sometimes the objection is stated to stronger effect, and then an incompatibility is asserted between the sceptical or solipsistic premisses, however framed, and not just the practice of criticism as retrieval but the view that criticism is retrieval.
A step further, and it is asserted that from these same premisses an alternative view of criticism follows. This alternative view may be expressed as, Criticism is revision, and it holds that the task of criticism is so to interpret the work that it says most to the critic there and then. Assuming the critical role, we must make the work of art speak ‘to us, today’.
It is clear that this derivation too must require further premisses, though less clear what they would be. One thing seems certain, though it is often ignored by adherents of the revisionary view, and that is this: If criticism is justifiably revision when we lack the necessary evidence for reconstructing the creative process, then it must also be revision when we have, if we ever do, adequate evidence for retrieval. We cannot as critics be entitled to make the work of art relate to us when we are in a state of ignorance about its history without our having an obligation to do so, and this obligation must continue to hold in the face of knowledge. Otherwise revision is never a critical undertaking: it is only, sometimes, a pis-aller, or a second best to criticism. Indeed, the strongest case for the revisionary view of criticism draws support from a thesis which appears to dispense with scepticism or, at any rate, cuts across it.
The thesis I have in mind, which is generally called ‘radical historicism’ and is best known through the advocacy of [T. S.] Eliot, holds that works of art actually change their meaning over history. On this thesis the task of the critic at any given historical moment is not so much to impose a new meaning upon, as to extract the new meaning from, the work of art. That works of art are semantically mobile in this way is to be explained not simply—to take the case of a literary work—by reference to linguistic change or to shifts in the meaning of words and idioms, but, more fundamentally, more radically, by appeal to the way in which every new work of art rewrites to some degree or other every related, or maybe every known, work of art in the same tradition. To this central contention the thesis adds the corollary that, as some particular meaning of a work of art becomes invalid or obsolete, it also becomes inaccessible: It ceases to be a possible object of knowledge.
Radical historicism is a doctrine, like the [Benjamin] Whorfian thesis about the non-intertranslatability of natural languages, with which indeed it has much in common, that has its greatest appeal when it gets us to imagine something which on reflection turns out to be just what it asserts is unimaginable. So, for instance, under the influence of radical historicism (or so it seems) we start to imagine how a contemporary of Shakespeare’s would find the inherited reading of [Geoffrey] Chaucer’s Troilus dull or dead, and we find ourselves readily sympathizing with his preference for a new revitalized reading inspired by Troilus and Cressida. And then we reflect that, if radical historicism is indeed true, just such a comparison was not open to one of Shakespeare’s contemporaries, and is even less so to us. To him only one term to the comparison was accessible: to us neither is.
A second objection to the retrieval view of criticism goes deeper in that it concentrates upon the view itself and not merely upon its consequences. According to this objection, retrieval is, from the critical point of view, on any given occasion either misleading or otiose. From the outset the objection contrasts retrieval with its own favoured view of criticism, which may be expressed as, Criticism is scrutiny—scrutiny of the literary text, of the musical score, of the painted surface—and it holds that retrieval is misleading when its results deviate from the findings of scrutiny and it is otiose when its results concur with the findings of scrutiny. In this latter case it is (note) retrieval that is reckoned otiose, not scrutiny, and the reason given is that reliance upon retrieval presupposes scrutiny but not vice versa. Scrutiny is presupposed because it is only with the findings of scrutiny also before us that we can be certain that we are dealing with a case where the results of retrieval merely reduplicate those of scrutiny, and hence that retrieval is not misleading. So, overall, retrieval can never do better than scrutiny, sometimes it can do worse, and which is the case cannot be determined without the benefit of scrutiny.
But how does this objection characterize the difference between the cases where retrieval does no worse than, and those where it does worse than, scrutiny? The cases are distinguished in that, given a work of art and the creative process that terminates on it, there are two possibilities. One is that the creative process realizes itself in the work of art: the other is that it fails to. Now it is in the latter case that retrieval is misleading, whereas in the former case it is merely otiose. In the former case, scrutiny will show the critic that the work is as retrieval laboriously allows him to infer that it is: in the latter case, retrieval will lead him to infer that the work is as scrutiny will soon reveal it not to be. This objection to the retrieval view shows itself vulnerable on a number of counts.
In the first place, though it is indubitably true that the creative process either is or is not realized in the work of art, nevertheless, if ‘realized’ means (as it presumably does) ‘fully realized’, this is not, from the point of view of criticism, the best way of setting out the alternatives. For critically it is a highly relevant fact that the creative process may be realized in the work of art to varying degrees. (There are, indeed, theoretical reasons of some strength, which I shall not assess, for thinking that the creative process is never realized in a work of art either to degree 1 or to degree 0: realization must always be to some intermediate degree.) But, it might be thought, this presents no real problem. For the objection can surely concede that the creative process may be realized to varying degrees, and can then further concede that sometimes, even when the creative process has not been fully realized, retrieval may not be misleading. All that it has to insist upon, surely, is that, if the creative process may be harmlessly, though otiosely, reconstructed up to the point to which it was realized in the work of art, retrieval is misleading if, and as soon as, it is carried beyond this point. However, as we shall see, this concession brings its difficulties in train.
Secondly: Suppose we confine ourselves (as the objection says) to that part of the creative process which is realized in the work of art. It becomes clear that there is something that reconstruction of this part of the process can bring to light which scrutiny of the corresponding part of the work cannot. It can show that that part of the work which came about through design did indeed come about through design and not through accident or error. Scrutiny, which ex hypothesi limits itself to the outcome, cannot show this. (A parallel in the philosophy of action: if an action is intentional, then, it might be thought, reconstruction of the agent’s mental process will not tell us more about it than we could learn from observation of the action: but we can learn this from observation of the action only if we already or independently know that the action is intentional.) Accordingly—and as yet the point can be made only hypothetically—if criticism is concerned to find out not just what the work of art is like but what the work is like by design, then, contrary to what the objection asserts, scrutiny, to be a source of knowledge, must presuppose retrieval.
Thirdly: The objection, as emended, states that that part of the creative process which is not realized in the work of art is not to be reconstructed. But how is this part of the process to be identified? There are two distinct grounds on which the distinction could be effected, and they give different results. We could exclude from critical consideration any part of the creative process in which the work of art is not … more or less directly prefigured: Alternatively, we might exclude only that part of the creative process which has no bearing at all upon the character of the work. Two kinds of case show how crucial it is which way the distinction is effected. The first case is where the artist changes his mind. Rodin’s Monument to Balzac started off as a nude sculpture. Is the critically relevant part of the creative process only that which includes Rodin’s change of mind to, and his subsequent concentration upon, the draped Balzac: or should it also embrace his concentration upon, and his subsequent change of mind from, the naked Balzac? The second case is where an artist sticks to his intention but fails in it. In writing The Idiot [Fyodor] Dostoievsky set out to portray a totally good man. Prince Myshkin is not a totally good man, but Dostoievsky’s depiction of him is clearly not unaffected by the original aim: it is the failed depiction of a totally good man. Should we, or should we not, regard Dostoievsky’s original aim, unsuccessfully realized though it is in the work of art, as a critically relevant part of the creative process?
In the light of the next, or fourth, point, the previous two points can be sharpened. For the objection, in claiming that scrutiny can establish everything that at one and the same time is critically relevant and can be established by retrieval, totally misconceives the nature of the interest that criticism might take in the creative process and, therefore, what it stands to gain from reconstructing it. For the objection appears to assume that, if the critic is interested in the creative process, this is because, or is to be accounted for by the degree to which, it provides him with good evidence for the character of the work. The critic seeks to infer from how the work was brought about how it is. Now, of course, if this were so, then there would, on the face of it at any rate, be reason to think that retrieval was at best a detour to a destination to which scrutiny could be a short cut. But that this is a misconception is revealed by the fact that the critic committed to retrieval is not committed to any assumptions about the likely degree of match between the creative process and the resultant work and he will continue to be interested in the creative process even in the case when he knows that there is a mismatch between the two. The critic who tries to reconstruct the creative process has a quite different aim from that which the objection to the retrieval view assumes. He does so in order to understand the work of art—though it would be wrong to say, as some philosophers of art tend to, that he seeks understanding rather than description. Understanding is reached through description, but through profound description, or description profounder than scrutiny can provide, and such description may be expected to include such issues as how much of the character of the work is by design, how much has come about through changes of intention, and what were the ambitions that went to its making but were not realized in the final product.
But, fifthly, and finally, the objection, in opposing scrutiny to retrieval presents scrutiny as though it were itself quite unproblematic: or as though, given a work of art, there would be no difficulty, or at any rate no theoretical difficulty, in dividing its properties into those which are accessible and those which are inaccessible to scrutiny. In considering the objection I have gone along with this, particularly in the second point I raise. However, in the main body of Art and its Objects I rejected this traditional assumption … though I preferred to make my point by considering specific properties that resisted the dichotomy. … Here I shall consider the matter more directly.
Crucially the view that criticism is scrutiny is seriously under-defined until an answer is given to the question, Scrutiny by whom? The following cases illustrate the problem: The listener who is ignorant of the mission of Christ will miss much of the pathos in the St Matthew Passion; a viewer who has not gathered that [Giovanni Lorenzo] Bernini’s mature sculpture requires a frontal point of view, as opposed to the multiple viewpoint against which it reacted, will fail to discern the emotional immediacy it aims at; a reader’s response to [Thomas] Hardy’s ‘At Castle Boterel’ will be modified when he learns that the poet’s wife had just died, and then it will be modified again as he learns how unhappy the marriage had been; the spectator who is made aware that in the relevant panel of the S. Francesco altarpiece [Stefano di Giovanni] Sassetta uses to paint the cloak that the Saint discards, thereby renouncing his inheritance, the most expensive and most difficult pigment available will come to recognize a drama first in the gesture, then in the picture as a whole, of which he had been previously ignorant. With any form of perception—and scrutiny is a form of perception—what is perceptible is always dependent not only upon such physical factors as the nature of the stimulus, the state of the organism, and the prevailing local conditions, but also upon cognitive factors. Accordingly, the scrutiny view needs to be filled out by a definition of the person whose scrutiny is authoritative, or ‘the ideal critic’, and any such definition must be partly in terms of the cognitive stock upon which the critic can draw. There are a number of possible definitions, for each of which the appeal of the scrutiny view, as well as its right to go by that name, will vary.
A heroic proposal, deriving from Kant, the aim of which is to ensure the democracy of art, is to define the ideal critic as one whose cognitive stock is empty, or who brings to bear upon the work of art zero knowledge, beliefs, and concepts. The proposal has, however, little to recommend it except its aim. It is all but impossible to put into practice, and, if it could be, it would lead to critical judgments that would be universally unacceptable.
Another proposal is to define the cognitive stock on which scrutiny is based as consisting solely of beliefs that could themselves have been derived—though in practice they may not have been derived—from scrutiny of the work of art concerned. But this takes us round in a circle: for what requirement is placed upon the cognitive stock on which the scrutiny that gives rise to these beliefs itself depends?