Jump to ContentJump to Main Navigation
The Critical Imagination$

James Grant

Print publication date: 2013

Print ISBN-13: 9780199661794

Published to Oxford Scholarship Online: May 2013

DOI: 10.1093/acprof:oso/9780199661794.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2019. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see www.oxfordscholarship.com/page/privacy-policy). Subscriber: null; date: 15 February 2019

The Aims of Criticism

The Aims of Criticism

(p.5) 1 The Aims of Criticism
The Critical Imagination

James Grant

Oxford University Press

Abstract and Keywords

Criticism of the arts is a prominent part of our cultural life. Philosophers and other theorists of the arts have long disagreed, however, about what the aims of art criticism are. Is the point of criticizing an artwork to evaluate it, to explain something about it, to help readers choose what artworks to experience, to modify their responses to the work, or to help them appreciate the work better? Each of these views on the aims of criticism is assessed in this chapter. Monroe Beardsley, Arnold Isenberg, Noël Carroll, Frank Sibley, and Arthur Danto, among others, have endorsed these claims. None of them, it is argued, identifies an aim shared by all criticism. A good account of the aims of criticism, however, ought to accommodate the truth in each of these views.

Keywords:   art criticism, literary criticism, aims of criticism, evaluation, appreciation, explanation, isenberg, baxandall, beardsley, sibley, carroll

To characterize the role imaginativeness plays in criticism of the arts, we need to know what the aims of criticism are. If we know what criticism is an attempt to achieve, we shall be better placed to say how imaginativeness enables critics to achieve it.

Philosophers and other theorists of the arts have often made claims about what the aims of criticism are. There are at least two things we might be interested in when we ask this question. First, we might want to know what the constitutive aim or aims of criticism are. Being written with this aim or these aims would be part of what makes a piece of writing an instance of art criticism. If criticism has a constitutive aim, it would be both interesting in itself to know what it is, and useful for saying what role endowments such as imaginativeness play in criticism.

Second, we might want to know what makes a piece of criticism a piece of good criticism. Perhaps there are aims such that achieving them (or being such as to achieve them) makes a piece of criticism that has them good criticism. That is to say, it makes it good as criticism: a passage can be an example of good writing but bad or indifferent criticism. Aims of this kind may or may not also turn out to be constitutive aims, as we shall see. If there are such aims, knowing what they are, too, would enable us to say what the role of imaginativeness in criticism is.

In what follows, I will identify both a constitutive and a non-constitutive aim of criticism. Saying how imaginativeness enables a critic to achieve these aims will show what its role in criticism is. In this chapter, I will consider five answers to the question of what the aims of criticism are. I will argue that none of them is an aim of all criticism. However, each proposal is instructive for what it reveals about the variety of things critics do and attempt to achieve. An account of the aims of criticism should make sense of the fact that critics do and attempt to achieve these things. In Chapter 2, I will provide such an account.

(p.6) 1. Helping readers choose

Monroe Beardsley argues that the primary aim of criticism is to help the critic’s readership decide which artworks to choose to experience. He says that critics, in the strict sense, are ‘those who set themselves up, or are set up by others, to make public judgments for the purpose of guiding the choices of others who are less qualified than they, perhaps by the lack of talent or time’.1 He calls this the ‘Consumers’ union model of (professional) criticism’, and argues that it is ‘essentially correct as an account of art criticism, capturing its primary character, on which its other features depend’.2 Critical activities such as explaining and interpreting are undertaken for this purpose.3

Beardsley’s model is best suited to much journalistic criticism, particularly reviews. Many reviews by theatre critics, film critics, music critics, etc., are written with the intention of helping readers to decide whether to see, watch, listen to (etc.) the work. Virginia Woolf makes this aim explicit in her essay on The Faerie Queene, introducing her piece as ‘some general observations made by one who has gone through the experience, and wishes to urge others, who may be hiding their yawns and their polite boredom, to the same experience’.4

However, Beardsley’s account is a poor model of much academic criticism. It is implausible that guiding the choices of others is the aim of much academic criticism. Such criticism is normally written for readers who have already chosen to experience the work. It is usually presumed that they are reading the criticism because they are studying the work. Good academic criticism is certainly capable of providing information that would be useful to audiences trying to decide whether to experience the work. But it is implausible that being useful in this way is what makes it good criticism, or that it is defective if it is not useful for this purpose.

(p.7) Furthermore, not even all journalistic criticism has this aim. Some reviews (for example, of a theatre production that has already finished) may simply be written to inform the reader of what was of interest in the work and why it was of interest. It may be of great interest to know how a certain director adapted a certain play, in what respects it was successful or unsuccessful, etc. For works that the reader can no longer experience, the aim of criticizing them cannot be to help the reader to decide whether to choose to experience them. But it is clearly not pointless to write criticism of such works.

2. Perception

Many hold that critics describe works in order to cause their readers to perceive features of the work. Stuart Hampshire, Frank Sibley, Michael Baxandall, and many others have held this view.5 The main argument offered in support of it is presented by Arnold Isenberg. Isenberg’s paper has exercised great influence.6 It is widely regarded as the classic defence of particularism in aesthetics. And his view that the aim of critical description is to cause perception has been not only supported but taken for granted by many aestheticians.

(p.8) Isenberg begins his argument by examining

a theory of criticism…which divides the critical process into three parts. There is the value judgment or verdict (V): ‘This picture or poem is good—.’ There is a particular statement or reason (R): ‘—because it has such-and-such a quality—.’ And there is a general statement or norm (N): ‘—and any work which has that quality is pro tanto good.’7

Isenberg agrees that critics make value judgements. He also agrees that ‘when we speak of “justifying” or “giving reasons” for our critical judgments, we refer to something which…does go on in the world’.8 And he agrees that the value judgement is ‘in some sense conditional upon R’.9

His objection to the theory of criticism he describes is that, though V is in some sense conditional upon R, ‘the truth of R never adds the slightest weight to V’.10 What we describe as ‘giving reasons’ in support of our value judgements is not a case of making statements whose truth supports our judgements.

He uses this claim to argue that getting the reader to perceive is the aim of critical description. The overarching structure of his argument is as follows:

  1. (A) The truth of R offers no support for the value judgement.

  2. (B) If the truth of R offers no support for the value judgement, the best explanation of R’s function is that using it is a way of getting the reader to perceive.


  1. (C) The best explanation of R’s function is that using it is a way of getting the reader to perceive.

Isenberg devotes most effort to defending (A). This defence involves denying that critics do rely on norms of the kind on which, according to the simple theory he rejects, they rely.11

(p.9) His argument for (A) appeals to the following passage by the art critic Ludwig Goldscheider. Discussing El Greco’s The Burial of Count Orgaz,12 Goldscheider writes:

Like the contour of a violently rising and falling wave is the outline of the four illuminated figures in the foreground: steeply upwards and downwards about the grey monk on the left, in mutually inclined curves about the yellow of the two saints, and again steeply upwards and downwards about…the priest on the right. The depth of the wave indicates the optical center; the double curve of the saints’ yellow garments is carried by the greyish white of the shroud down still farther; in this lowest depth rests the bluish-grey armor of the knight.13

Isenberg comments:

This passage—which, we may suppose, was written to justify a favorable judgment on the painting—conveys to us the idea of a certain quality which, if we believe the critic, we should expect to find in a certain painting by El Greco. And we do find it: we can verify its presence by perception.… But the same quality (‘a steeply rising and falling curve,’ etc.) would be found in any of a hundred lines one could draw on the board in three minutes. It could not be the critic’s purpose to inform us of the presence of a quality as banal and obvious as this.14

The point, it seems, is that there is no true norm to the effect that any work which has a steeply rising and falling curve, etc., is pro tanto good. So the argument appears to be:

  1. (A1) If the truth of R supports the value judgement, then there are true norms to the effect that any work with the property attributed by R is pro tanto good.

  2. (A2) But there are no true norms to this effect.


  3. (A) The truth of R offers no support for the value judgement.

That result raises a problem: ‘as long as we have no alternative interpretation of the import and function of R, we must assume either that R is perfectly arbitrary or that it presupposes and depends on some general claim.’15 The above argument seems to show that R does not presuppose (p.10) and depend on some general claim, so it seems that R is perfectly arbitrary. This leads Isenberg to his alternative interpretation of the function of R.

He continues the discussion of Goldscheider in this way:

It seems reasonable to suppose that the critic is thinking of another quality, no idea of which is transmitted to us by his language, which he sees and which by his use of language he gets us to see. This quality is, of course, a wavelike contour; but it is not the quality designated by the expression ‘wavelike contour’. Any object which has this quality will have a wavelike contour; but it is not true that any object which has a wavelike contour will have this quality.…Now the critic…gives us directions for perceiving, and does this by means of the idea he imparts to us, which narrows down the field of possible visual orientations and guides us in the discrimination of details, the organization of parts, the grouping of discrete objects into patterns.…It is a function of criticism to bring about communication at the level of the senses, that is, to induce a sameness of vision, of experienced content.16

Isenberg’s positive account thus has two components. There is the inference to the best explanation of the aim of criticism in providing R:

The aim of criticism in providing R is to get the reader to perceive certain properties.

Isenberg defends the view that this is the best alternative explanation of R’s function by claiming that ‘reading criticism, otherwise than in the presence, or with direct recollection, of the objects discussed is a blank and senseless employment.’17

However, his view has a second component independent of this one. Isenberg also claims that

The properties criticism aims to get the reader to perceive are not the properties attributed by the critic in providing R.

One might have thought the aim is to cause the reader to see the property the critic attributes (even though the truth of the claim that the object has that property does not support the value judgement). But this is not Isenberg’s view. He is explicit that ‘criticism does not actually designate (p.11) the qualities to which it somehow directs our attention’.18 Presumably, the argument for this additional claim is that, as he says, the property attributed by the critic in providing R is banal and obvious. Not only can it not be the critic’s purpose to ‘inform us of the presence’ of such a banal and obvious property: it also cannot be her purpose to make us see it.

I will take issue with both components of Isenberg’s view. However, there is some truth in what Isenberg says. In fact, it is obvious that critics sometimes try to draw our attention to, or get us to perceive, various properties of a work. Critics include reproductions of paintings in their work to get us to look them, and quote lines of poetry to get us to read them. Sometimes critics explicitly instruct us to look at certain features. Nevertheless, Isenberg’s account is flawed.

The first problem is that, when critics try to get us to see a certain quality, they do normally designate this quality. It is worth examining how Isenberg is led to deny this. Other philosophers have also held this. Mary Mothersill, for instance, says that Isenberg has shown that, in a certain sense, ‘critics do not…“mean” what they “say”’.19 But even in the passage of criticism Isenberg uses to support his point, the critic designates the quality he gets us to see.

The argument for this is simple. The quality Goldscheider gets us to see is the outline of the four illuminated figures in the foreground. He designates this quality with the definite description ‘the outline of the four illuminated figures in the foreground’. Therefore, he designates the quality he gets us to see.

Isenberg claims that the quality Goldscheider gets us to see ‘is not the quality designated by the expression “wavelike contour”’. This may well be true. But even if it is true, it does not establish the conclusion that ‘criticism does not actually designate the qualities to which it somehow directs our attention’. Goldscheider designates the quality to which he directs our attention. It is just that he does not designate it with the expression ‘wavelike contour’. He designates it with a different expression—namely, ‘the outline of the four illuminated figures in the foreground’.

(p.12) Goldscheider does not use the expression ‘wavelike contour’ in the passage Isenberg quotes: he uses the predicate ‘x is like the contour of a violently rising and falling wave’. I assume Isenberg’s claim is a concise way of making the point that the quality he gets us to see is not designated by that predicate. And that claim is plausible. I assume that, in Isenberg’s usage, what is designated by an expression is what is denoted or referred to with (or by) that expression. If anything is designated, in this sense, by the predicate ‘x is like the contour of a violently rising and falling wave’, then it is the quality being like the contour of a violently rising and falling wave. So Isenberg is clearly right that the quality Goldscheider gets us to see is not designated by this predicate. The outline of the figures in the foreground (the quality he gets us to see) is not identical to the quality, being like the contour of a violently rising and falling wave. But this is an uninteresting result. For Goldscheider does designate the quality he gets us to see—with a different expression.

What is certainly true is that Goldscheider characterizes the quality he gets us to see with the predicate ‘x is like the contour of a violently rising and falling wave’. He characterizes, but does not designate, the outline of the figures with this predicate.20 What is also true is that Goldscheider gets us to see the outline, and gets us to see that the outline has the quality of being like the contour of a violently rising and falling wave. He gets us to see one quality (the outline), and he gets us to see that it has a further quality (being like the contour of a violent wave). Isenberg does not make the distinction between designation and characterization, or the distinction between seeing and seeing-that. But if we make these distinctions, then another thesis suggests itself.

Isenberg might mean that Goldscheider does not characterize the outline as having the quality he gets us to see that it has. Isenberg believes that ‘it could not be the critic’s purpose to inform us of the presence of a quality as banal and obvious as’ being like the contour of a violently rising and falling wave. This is the quality he thinks could be possessed by ‘any of a hundred lines one could draw on the board in three minutes’. And this is the quality Goldscheider characterizes the outline as having. So perhaps Isenberg (p.13) means that Goldscheider’s purpose is not to get us to see that the outline has this quality. Rather, his aim is to get us to see that the outline has some other quality, which he does not characterize it as having.

This claim is implausible for different reasons. First, even if it is not the critic’s purpose merely to inform us that the outline is wavelike, it does not follow that it is not his purpose to get us to see that the outline is wavelike. It may be interesting for the critic’s readers to see that this is the case, but not especially interesting merely to be informed that it is the case.

Second, even if one can easily draw hundreds of wavelike lines on a blackboard, it does not follow that the fact that the outline of the figures in El Greco’s painting is like a violent wave is banal or obvious. How easily we can draw a wavelike line has nothing to do with whether this fact about the El Greco is interesting or banal. If El Greco had given one of the figures a square head, that certainly would be an interesting fact, even though squares are easily reproduced on the blackboard, too.

So we should not accept Isenberg’s claim that critical communication differs from ordinary communication in the ways he says it does—even in those cases in which the critic is trying to get her readership to see. Goldscheider gets us to see the outline of the figures, and he gets us to see that this outline is like a violent wave. He designates the quality he gets us to see, and he characterizes it as having a quality he gets us to see that it has.

Further problems attach to the other component of Isenberg’s conclusion: the view that getting us to perceive properties is the critic’s aim. The first point to highlight is that Isenberg’s argument only supports the conclusion that this is the point of making ‘R’-type statements. As I have noted, his argument is presented as an examination of the function of R (rather than that of V or N), and it only supports such a conclusion as regards R-type statements.

This point is worth emphasizing. The upshot of it is that, for all Isenberg has shown, critics do support their value judgements—only not with R-type statements. They may, for example, support some of their evaluations with other evaluations. As I will explain further below, Isenberg regards R-type statements as descriptive, non-evaluative claims. So even if Isenberg’s argument about R-type claims were flawless, it would not show that one value judgement cannot support another.

Discussions of Isenberg tend to overlook the fact that the scope of his thesis is limited in this way. For example, Daniel Kaufman and Noël (p.14) Carroll have recently objected to Isenberg on the grounds that such claims as ‘Roger van der Weyden’s Descent from the Cross displays Christ’s humanity well’ or ‘Harold Lloyd’s Safety Last contains many successful pratfalls’ support, respectively, such value judgements as ‘Roger’s Descent from the Cross is good’21 and ‘Safety Last is good (pro tanto)’.22

Kaufman and Carroll are wrong to claim that these are counterexamples to the claim Isenberg makes. These are cases of supporting one evaluation with another: to say that Christ’s humanity is displayed well and that the pratfalls are successful is to evaluate. Isenberg’s problem, however, is whether evaluations can be supported by descriptive, non-evaluative claims.

It is true, however, that it is possible to support one value judgement with another. It would be absurd to hold that the truth of the claim, ‘Safety Last contains many successful pratfalls’, cannot add the slightest weight to the claim that it is a good slapstick comedy. So not only does Isenberg’s thesis not imply that value judgements cannot be supported by other value judgements: it is clear that they can indeed be so supported, and that they often are in criticism.

There is another little-noted consequence of the fact that Isenberg’s thesis is restricted to R-type statements: Isenberg needs to tell us what an R-type statement is. If his thesis about their function is true, they cannot really be those statements whose function is to support value judgements, even though this may appear to be their function. What, then, are they, according to Isenberg?

Isenberg says that ‘R is a statement describing the content of an art work’.23 He contrasts descriptive statements with evaluative statements, as many philosophers have done. He acknowledges that this is an idealization:

V and R, it should be said, are often combined in sentences which are at once normative and descriptive. If we have been told that the colors of a certain painting are garish, it would be astonishing to find that they were all very pale and unsaturated; and to this extent the critical comment conveys (p.15) information.…We shall be concerned exclusively with the descriptive function of R.24

As I mentioned above, Isenberg also says that the critic’s value judgement is ‘in some sense conditional upon R’.

This is all he tells us by way of characterizing R. And this creates a number of problems. Suppose Isenberg means that an R-type claim is any descriptive, non-evaluative critical claim made about a work (including the descriptive, non-evaluative claim made with ‘sentences which are at once normative and descriptive’) upon which the value judgement is in some sense conditional. If that is so, then it is not at all plausible that the function of every such claim is to get the reader to perceive properties.

First, various kinds of review provide counterexamples. Reviews of works the critic’s readers cannot perceive, such as a theatrical production that has completed its run, are counterexamples. So are negative reviews discouraging the reader from perceiving the work. The aim of making the descriptive claims in such reviews cannot be to cause the reader to perceive features of the work. Reviews like these sometimes do reproduce the work or parts of it, to enable the reader to perceive certain properties. But we do not necessarily fault the review if it does not do this, even if it contains many non-evaluative descriptions on which a value judgement is somehow conditional. Indeed, it is not always possible to enable the reader to perceive every single property ascribed with such descriptions in a review. Reproducing enough of a novel, film, or live performance to allow the reader to perceive every property ascribed in the review is often impractical or impossible. But the review may be none the worse for that.

Second, there are several kinds of descriptive critical statement on which a value judgement is sometimes conditional, but which do not always have causing perception or directing perceptual attention as their aim. For example, critics sometimes try to persuade a reader that something is true in the world of the work: that Hamlet only feigns madness; that the governess in The Turn of the Screw did see ghosts; or that Ugolino in the Inferno ate his children. Beardsley has called this the ‘elucidation’ of a representational artwork.25 A value judgement may well be conditional (p.16) on the truth of such a claim. A critic of The Turn of the Screw may regard the story as better if the governess is suffering from the effects of repression than it would be if she really saw ghosts. But the aim of an elucidation is often to cause the reader to believe that something is true in the world of the work, not to guide perception.

Other examples are easily found. The aim of some description may be to get the readership to interpret the work in a certain way, to get them to believe that the work belongs to a certain artistic category or genre, or to inform them of something about the work’s historical or cultural context (e.g., that it had a certain religious or political function). A value judgement can be conditional on the truth of any such claim, even though the aim is not to guide perception.

So if an R-type claim is any descriptive, non-evaluative claim on which the value judgement is somehow conditional, Isenberg’s thesis about the function of such claims is false. But if an R-type claim is not just any claim of this kind, then it is not clear which claims are R-type claims. Isenberg offers no more clarification of this than what I have quoted. So Isenberg’s position is problematic. What he writes is unclear, and there do not appear to be clarifications of what he means that render his claims plausible.

It is certainly true that the point of some art-critical statements is to get the critic’s readership to perceive properties, or to perceive that the work has certain properties. In fact, this is not only true, but obvious. Isenberg’s thesis is the more ambitious claim that every critical statement of a certain very general kind has this function. Without a characterization of the kind of statement he is talking about that makes his thesis plausible, we are left with the more modest claim that some critical statements have this function.

3. Evaluation

Carroll argues that the aim of criticism is to provide a sound justification for an evaluation of a work. As he puts it, ‘criticism, properly so-called, is not merely a matter of evaluating an artwork—of giving it a thumbs-up or thumbs-down. Critics are expected to supply reasons—indeed, good reasons—in support of their evaluations.’26 Other critical operations, (p.17) such as ‘description, classification, contextualization, elucidation, interpretation, analysis…typically function as grounds for evaluation’.27

His argument for this is that evaluation is ‘the feature that sets criticism off from comparable discourses’28 about art:

For example, certain forms of historical discourse about art will mobilize description, elucidation, contextualization, classification, interpretation, and/or analysis. For example, an economic historian of art might describe and analyze Rembrandt’s tendency to have large swaths of black in his pictures in order to explain that in this way Rembrandt was able to undertake, for the purpose of maximizing his profit margin, a very large number of commissions, since those empty, unarticulated, black spaces of canvas could be painted very quickly.…The notion that, additionally, criticism engages in evaluation provides us with a ready differentia or rationale, which suggestion is also amply supported in everyday speech.29

Thus, ‘evaluation is an essential feature of criticism such that if a piece of discourse lacks explicit or implicit evaluation, it would not qualify as criticism’.30

Carroll holds that reasoned evaluation is possible because, contra Isenberg, there are indeed principles of critical evaluation. But they do not take the form: ‘Every artwork with such-and-such a characteristic is good (pro tanto)’. Genuine critical principles are not claims about all artworks whatsoever. Rather, they are claims about artworks of a given category.31 According to Carroll, an artwork of a given category that fulfils the artistic purposes or function of artworks of that category is, normally, good (pro tanto). If this is so, he argues, then there are many true critical principles. For example, the principle he uses to support his judgement of Lloyd’s Safety Last is: ‘Given the purpose or function of slapstick comedy, slapstick comedies that contain many successful pratfalls, all other things being equal, are good (pro tanto).’32

(p.18) To illustrate such category-relative evaluation, Carroll uses a review by the dance critic Joan Acocella. Acocella is reviewing a production of Mark Morris’s Mozart Dances, a work of modern abstract choreography. She writes: ‘Why is he so popular? One reason, I think, is that he gives people the modern pleasure of seeing abstract work without leaving them scratching their heads over what it was about. Though he may not have a story on the surface, he always has one underneath, in the form of movement motifs.’33 She then describes these movement motifs and the story they suggest. Carroll says of her review:

Acocella enables her readers to understand her grounds for maintaining that Morris has subtly articulated the outline of a story. This, in turn, she maintains, gives the viewer a way into a dance of the sort that is often confusing to audiences, presupposing, as she does, that a narrative, typically, enhances accessibility.…She is not supposing that a suggested narrative is a good-making feature of every artwork. Rather, she is restricting her claim to works of modern abstract choreography and saying that, all things being equal, it is a good-making feature in such works.34

As I have said, Carroll is right to hold both that it is possible to provide support for a critical evaluation and that critics do this. However, like Beardsley and Isenberg, he overstates the prevalence of the aim he identifies.

First, some good criticism merely provides evaluations of artworks without supporting them. For example, many entries in The Penguin Guide to Recorded Classical Music simply rate recordings on a scale of one star to five stars without further comment.35 This is bad criticism if the evaluation is wrong, or if the critic herself is not justified in giving it the rating she gives it. But it is not necessarily bad criticism if she does not provide a justification for the verdict. The readership being addressed may be one that is only interested in the verdict of a good critic, rather than a justification for it. They may look to the critic, as Beardsley held, to guide (p.19) their choices, and for this purpose they may only be interested in her verdict.

Second, Carroll gives us no good reason to believe that the point of other critical operations, such as interpretation and classification, is always to support an evaluation. Some of the counterexamples to Isenberg’s view also serve as counterexamples to Carroll’s view. For example, it seems that a convincing or plausible elucidation, in Beardsley’s sense,36 of a controversial question about the world of a work could be good criticism even if it is not used to support an evaluation. If a critic came up with a convincing answer to the question of why Hamlet procrastinated, that in itself would be excellent criticism. It would not be flawed if the elucidation was not presented in support of an evaluation.

Carroll responds to the objection that much good criticism (e.g., of canonical works) appears not to include evaluation:

Since it is through operations like interpretation, description, analysis, classification, contextualization, etc. that one grounds one’s evaluations, when it comes to criticizing canonical works, if, for example, these routines bring to readers a view of the unity, complexity, sophistication, and wisdom of the work, then it may not be necessary to round off one’s critical remarks with overt commendation. The recommendation may be implicit. But this only shows that the evaluative moment in criticism need not be explicit.37

This reply is weak because at most it establishes that, when there is no explicit evaluation, there may be an implicit evaluation. This is true. But that does not answer the objection that it seems possible to produce good criticism that contains neither an explicit nor an implicit evaluation.

It seems, therefore, that Carroll must rely on his argument that a piece of criticism must contain evaluation, because otherwise criticism would be indistinguishable from similar discourses about art, which can also contain the other operations he mentions. But this argument is also unsuccessful. If the claim is that no piece of writing about an artwork lacking evaluation of the work could be criticism, then the elucidation of Hamlet I suggested seems simply to be a counterexample. Such an elucidation would generally be acknowledged to be criticism. The fact that such a piece of writing (p.20) appears to be criticism is a datum that any account of criticism must respect or explain away. Carroll’s theory does not do this. Carroll is right that ‘the challenge that confronts the skeptic regarding the claim that criticism is essentially concerned with evaluation is to propose another distinction—a more effective and more persuasive dividing line between criticism and comparable modes of discourse—than the one I am advancing.’38 I will answer this challenge in Chapter 2. But we do not have to answer it to show that the claim that all criticism contains implicit or explicit evaluation is mistaken.

So it seems there can be good criticism containing evaluation but no support for it, and good criticism containing no evaluation. There is a third reason to doubt that providing support for an evaluation is always the aim of criticism. Even when criticism contains evaluations, and even when what the critic says could be used to support those evaluations, it is not always plausible that the point of the criticism is to support those evaluations.

For example, often the evaluations implicit or explicit in criticism of well-known works are already very well supported. This makes it implausible that the point of criticizing those works again is to provide further support for those evaluations. The readership being addressed may not need to be persuaded, and may also be justified in believing, that King Lear is a masterpiece. Their experience of the work may justify them in believing this, and past criticism may have provided ample support for this evaluation. It may therefore be unnecessary to provide further support for it. But it is not pointless to criticize such works further.

Carroll might reply that, although this may be true of judgements of the overall value of a well-known work (e.g., the claim that the work is a masterpiece), those are often not the kinds of evaluation critics argue for. He writes that ‘the category-relative evaluation of an artwork is a pro tanto evaluation’.39 For instance, the evaluation in his Harold Lloyd example is not a judgement of the comedy’s overall value. It takes the form: ‘This work is good (pro tanto)’.40 So Carroll might reply that, although it is often unnecessary to provide support for judgements of the overall value of well-known works, it is often necessary to provide support for various pro tanto evaluations of such works.

(p.21) What, however, does ‘good pro tanto’ mean? This expression is frequently used in philosophy without explanation. Some writers say they use ‘“good pro tanto” to refer to something’s being good in some respect’ or ‘good in a way’.41 This clearly will not help Carroll. A critic’s readers also frequently have plenty of evidence that King Lear is good in a way—indeed, that it is good in many ways. The point of criticizing such works cannot be to provide evidence for this extremely modest claim.

The obvious alternative is to construe ‘good pro tanto’ as meaning ‘good in that way’ or ‘good in that respect’. The dictionary meaning of ‘pro tanto’ is: ‘to such an extent, to that extent’. ‘Good to that extent’ or ‘good to such an extent’ seem to be equivalent to ‘good in that respect’ or ‘good in that way’. If this is the form that a pro tanto evaluation takes, then the claim that something is pro tanto good, unlike the claim that there is some respect or way in which a certain thing is good, can only be understood in a context that makes it clear what way or respect is being referred to. ‘Pro tanto good’ is comparable to a phrase containing a demonstrative referring to a respect or way in which the work is being said to be good. For instance, the claim that Safety Last is pro tanto good, as it occurs in Carroll’s example, amounts to the claim that its many successful pratfalls are a good thing about it. So perhaps the claim Carroll would make is this: when the overall value of the work is already well established, the aim of criticism is to support the claim that the work is good (or bad) in such-and-such a respect or way.

It is certainly true that, even when we have ample justification for the belief that a work is a masterpiece, and for the belief that there are many ways in which it is good, it remains of great interest to learn what is good or bad about it. We can continue to learn such things long after we have established a work’s overall value. It is also true that much criticism involves telling us what is good or bad about a work.

What is doubtful is that the aim of criticism is always to provide support for the claim that such-and-such is a good or bad thing about a work. Much good criticism simply asserts or implies that this or that is a good thing about the work. This can be of great interest. But the criticism often does not supply reasons to back up the claim that this or that is a good thing about the work.

(p.22) For example, Samuel Johnson, discussing Othello, writes: ‘The scenes from the beginning to the end are busy, varied by happy interchanges and regularly promoting the progression of the story; and the narrative in the end, though it tells but what is known already, yet is necessary to produce the death of Othello.’42 William Hazlitt writes: ‘Macbeth and Lear, Othello and Hamlet, are usually reckoned Shakespeare’s four principal tragedies. Lear stands first for the profound intensity of the passion; Macbeth for the wildness of the imagination and the rapidity of the action; Othello for the progressive interest and powerful alternations of feeling; Hamlet for the refined development of thought and sentiment.’43 Here we have several claims about what is good about some acknowledged masterpieces. Johnson claims that it is a good thing about Othello that the scenes are busy and varied by happy interchanges, and that they regularly promote the progression of the story. Hazlitt claims that the profound intensity of the passion is a good thing about Lear, and that its refined development of thought and sentiment is a good thing about Hamlet.

But Johnson and Hazlitt do not supply us with any reasons in support of their claims that these are good things about these plays. Moreover, it is clear why they do not: it is unnecessary to provide support for the claim that these are good things about the plays. It would certainly be necessary to provide support for the claim that it is a bad thing about Othello that its scenes are varied by happy interchanges and regularly promote the progression of the story. But it seems no more necessary to support the pro tanto evaluation that this is a good thing about Othello than it would be to support the all-things-considered evaluation that Othello is a great tragedy. An important difference between these evaluations is that a reader may not have entertained the thought that this is one good thing about Othello, whereas she is almost certain to have entertained the thought that it is a great tragedy. But it is no more necessary to support the one claim than it is to support the other.

So if this is what a pro tanto evaluation is, then we should agree with Carroll that much good criticism contains many such evaluations. But even in such criticism, it is implausible that the aim is always to provide (p.23) reasoned pro tanto evaluations. Often, it will indeed be necessary to provide justifications of one’s pro tanto evaluations. But it may be enough that one’s criticism provide numerous correct and insightful pro tanto evaluations of the work. Johnson and Hazlitt’s criticism is flawed if what they say are good things about the plays are not good things about them. But their criticism does not seem to be flawed on account of not including justifications of their pro tanto evaluations.

4. Explanation

It is often said that critics explain why a work has the value or the aesthetic properties it has. For example, Arthur Danto holds that criticism is a kind of education, and that the aim of criticism is to provide explanations of a certain kind: ‘Education is not training people to say, Mitchell’s Hemlock is better than Marden’s Cold Mountain. It is rather explaining how and why each of them is good in its own way.’44 Sibley discusses explanation in criticism at greater length than many. He writes:

A critic frequently tries, as one of his central occupations, to say why a picture is unbalanced, or what gives a complex work its grace, unity, or serenity.… He may mention a concentration of blues and greys as responsible for the unity of tone, certain wavy lines as giving a restless quality, a change in key as giving a sombre or indecisive character.45

Sibley describes explanation as ‘one of the central activities of critics’,46 though he does not claim that providing explanations is the aim of all criticism.47

There are, indeed, several kinds of fact that critics explain, apart from those mentioned by Danto and Sibley. In addition to facts about a work’s value and aesthetic properties, critics also explain facts about the world of a work. Sometimes these facts are explained by appealing to or postulating other facts about the world of the work. I have already mentioned the (p.24) example of attempting to explain why Hamlet procrastinated, which might proceed by postulating facts about Hamlet’s psychology. Sometimes, by contrast, facts about the world of the work are explained by citing other kinds of fact. Peter Lamarque holds that an incident in a fiction can sometimes be explained by citing a fact about ‘the contribution the episode makes to the completed artistic structure’ of the work, as when a critic explains why Tess killed Alec d’Urberville by saying that it signifies an end to her journey.48 Similarly, a commonplace about Henry IV, Part One is that Hotspur is impetuous and hot-headed because this makes him a foil to the calculating Prince Hal.

Critics also sometimes explain why a work leaves a certain impression or elicits a certain response. Thomas De Quincey does this in his essay, ‘On the Knocking at the Gate in Macbeth’. De Quincey writes that, ever since his childhood,

the knocking at the gate, which succeeds to the murder of Duncan, produced to my feelings an effect for which I never could account: the effect was—that it reflected back upon the murder a peculiar awfulness and a depth of solemnity: yet, however obstinately I endeavoured with my understanding to comprehend this, for many years I never could see why it should produce such an effect.49

His answer is that the knocking at the gate marks ‘the re-establishment of the goings-on of the world in which we live’ after the murder, and this ‘makes us profoundly sensible of the awful parenthesis that had suspended them’.50 De Quincey argues that if the reader reflects on cases in which ordinary goings-on are dramatically interrupted and then resumed, ‘he will be aware that at no moment was his sense of the complete suspension and pause in ordinary human concerns so full and affecting as at that moment when the suspension ceases, and the goings-on of human life are suddenly resumed’.51 This is so, he suggests, when one sees a woman revive after fainting, or hears the rattling wheels of the carriage break the silence of the funeral procession of a great national hero; and the same kind of effect, he argues, occurs in Macbeth.52

(p.25) Not only are there several kinds of fact that critics explain. There are also several kinds of explanation that critics provide. Bas van Fraassen claimed that ‘an explanation is an answer to a why-question.’53 Explanations of why something is the case loom large in aestheticians’ examples of the kinds of explanation critics provide. But as several philosophers have pointed out, not every explanation is an explanation of why something is the case.54 And the most important explanations in a piece of criticism are not always explanations of why something is the case.

Some explanations provide answers to how-questions. For example, John Summerson explains how the Romans solved the problem of combining the Ionic order of Greek temple architecture with the arch-and-vault system needed for major Roman buildings like the Colosseum (see Figure 1.1):

Here you have a grammatical construction which is a pretty complete thing. It is controlled by an Ionic order which obeys nothing but its own traditional aesthetic rules. The shape and size of the piers behind the columns and of the arch, on the other hand, have come about through the exigencies of convenience and construction. The two disciplines have got to meet each other harmoniously and I think we may agree that they do. The pedestal moulding of the order ranges with the sill height of the arched gallery. The impost of the arch strikes the columns a little above half their height and the arch sits comfortably between the columns and the architrave above. If this arrangement is satisfactory it has been achieved by a very careful balancing of needs, the aesthetic dictatorship of the Ionic order and the practical needs of the building as a thing of use.55

So too, some explanations provide answers to what-questions. You can explain what a word means, what a gesture signifies, or what an image symbolizes. Such explanations, unlike the previous ones we have considered, do not take facts as their explananda. For example, when you explain what a word means, what you explain is its meaning, not the fact that it has a meaning, or any other fact. Clearly, explanations of this kind, too, are prevalent in criticism.


                     Figure 1.1. C. Elizabeth Grant. Drawing after Antoine Babuty Desgodets, Second Ordre du Colisée, à Rome. Pen on paper. After Antoine Babuty Desgodets, Les édifices antiques de Rome: dessinés et mesurés très exactement (Paris: Jean Baptiste Coignard, 1682), p. 263. © 2012 C. Elizabeth Grant. With permission.

Figure 1.1.Drawing after Antoine Babuty DesgodetsAntoine Babuty Desgodets, Les édifices antiques de Rome: dessinés et mesurés très exactement (Paris: Jean Baptiste Coignard, 1682), p. 263.

It would appear, then, that explanation plays a significant role in criticism. This is all the more obvious when we remind ourselves of the various kinds of explanation that there are and the various explananda we want critics to explain. And it is unsurprising that explanation has such a significant role in criticism. Good explanations provide us with understanding, and this is one of the benefits of good criticism.

Nevertheless, it is not the aim of all criticism to provide explanations. First, several of the examples mentioned in section 3 tell against this claim, such as criticism that merely provides unsupported verdicts. Second, although some criticism explains why a work elicits certain responses, some criticism tells us what responses the work elicits, or can elicit, from (p.27) someone who appreciates it, without providing an explanation of why it does so. Third, I noted in section 3 that critics often tell us what is good or bad about a work. Telling us what is good or bad about a work may or may not be part of an explanation. You can certainly explain why a work is good by telling us what is good about it. But you can also tell us what is good about it without explaining why it is good: you can tell us what is good about it without believing, claiming, or implying that it is good, and therefore without explaining why it is.

5. Aiding appreciation

As I have stressed, critics sometimes do guide perception, provide evaluations, support them, elucidate, interpret, explain, and describe appreciative responses. This is not meant to be an exhaustive list. It might be that all of these, and perhaps more, are the aims of criticism, and that there is no further aim. Clearly, however, it would be desirable to find an aim for the sake of which critics do all of these things.

An attractive suggestion is that the aim of criticism is to enable the critic’s readers to appreciate the work better than they would be likely to if they experienced the work (or a suitable reproduction, performance, token, etc.) without having read the criticism. This is meant to include the case in which the readers would probably not appreciate the work at all without the criticism. Plausibly, all of the above critical activities are ways of enabling a person to appreciate a work better.

This, I shall suggest, is close to the truth of the matter. In Chapter 2, I will show how all of the critical activities we have identified can aid appreciation. However, there are also counterexamples to the claim that aiding appreciation is the aim of all criticism.

Criticism of works that the critic knows cannot be experienced anymore, and of which she knows there are no appropriate reproductions, performances, etc., provide counter examples. Such criticism can be addressed to readers who never have been and never will be in a position to appreciate the work. The aim of criticizing them cannot be to aid appreciation of such works.

Furthermore, when a work can still be appreciated, the information provided by good criticism may not enable the readership to appreciate it better than they would be likely to without the criticism. Reviews that (p.28) describe the work in such a way as to provide information that is useful to someone trying to decide whether to experience the work provide examples. It might be very useful to know whether a film is terrifying, funny, challenging, clichéd, thrilling, etc., when trying to decide whether to watch it. However, such information might be useless for the purpose of gaining a better appreciation of the work than one would be likely to acquire by experiencing it without the criticism. For those who have seen or are watching the film, being told that it received four stars will probably not enable them to appreciate it any better. Similarly, anyone who sees the film might be unlikely to fail to see that it is terrifying, clichéd, disturbing, etc., and so this information will not enable them to appreciate it any better than they would be likely to without the criticism.

Nevertheless, there clearly are close links between criticism and appreciation. And we ought not to abandon the attempt to identify an aim that unifies the various critical activities. In Chapter 2, I shall suggest that such an aim can be found, and that it does have something to do with appreciation. To say what it is, however, we shall first have to say more about appreciation.


(1) Monroe C. Beardsley, ‘What Are Critics For?’, in The Aesthetic Point of View: Selected Essays, ed. Michael J. Wreen and Donald M. Callen (Ithaca: Cornell University Press, 1982), p. 149.

(2) Ibid.

(3) Ibid., p. 160. Stein Haugom Olsen expresses a similar view of the aim of what he calls criticism in ‘the judgmental sense’. Criticizing, in this sense, is ‘pointing out good or bad qualities in a literary work in order to make a recommendation of some sort either to the artist or to the reading public’ (Stein Haugom Olsen, ‘Criticism of Literature and Criticism of Culture’, Ratio 22/4 (2009): 439–440).

(4) Virginia Woolf, ‘The Faery Queen’, in Hugh Maclean and Anne Lake Prescott, eds, Edmund Spenser’s Poetry, 3rd edn (New York: Norton, 1993), p. 672.

(5) See Stuart Hampshire, ‘Logic and Appreciation’, in William Elton, ed., Aesthetics and Language (Oxford: Blackwell, 1959), p. 165; Frank Sibley, ‘Aesthetic Concepts’, in Approach to Aesthetics: Collected Essays on Philosophical Aesthetics, ed. John Benson, Betty Redfern, and Jeremy Roxbee Cox (Oxford: Oxford University Press, 2001), pp. 15–20; Frank Sibley, ‘Aesthetic and Non-Aesthetic’, in Approach to Aesthetics, p. 38; Michael Baxandall, ‘The Language of Art History’, New Literary History 10/3 (1979): 455; James Shelley, ‘The Character and Role of Principles in the Evaluation of Art’, British Journal of Aesthetics 42/1 (2002): 51.

(6) See, e.g., Mary Mothersill, ‘Critical Reasons’, The Philosophical Quarterly 11/42 (1961): 74–78; John Casey, The Language of Criticism (London: Methuen, 1966), pp. 172–173; Mary Mothersill, Beauty Restored (Oxford: Oxford University Press, 1984), p. 354; Ted Cohen, ‘On Consistency in One’s Personal Aesthetics’, in Jerrold Levinson, ed., Aesthetics and Ethics: Essays at the Intersection (Cambridge: Cambridge University Press, 1998), pp. 112–113; Joel J. Kupperman, Value…and What Follows (Oxford: Oxford University Press, 1999), p. 53; Dominic McIver Lopes, Sight and Sensibility: Evaluating Pictures (Oxford: Oxford University Press, 2005), pp. 105–106; Robert Hopkins, ‘Critical Reasoning and Critical Perception’, in Matthew Kieran and Dominic McIver Lopes, eds, Knowing Art: Essays in Aesthetics and Epistemology (Dordrecht: Springer, 2007), pp. 137–153; James Shelley, ‘Critical Compatibilism’, in Matthew Kieran and Dominic McIver Lopes, eds, Knowing Art: Essays in Aesthetics and Epistemology (Dordrecht: Springer, 2007), pp. 125–136; Keith Lehrer, Art, Self and Knowledge (Oxford: Oxford University Press, 2012), pp. 21, 29.

(7) Arnold Isenberg, ‘Critical Communication’, Philosophical Review 58/4 (1949), p. 330.

(8) Ibid.n. 3

(9) Ibid.

(10) Ibid.

(11) Shelley, ‘Critical Compatibilism’, pp. 128–129

(12) For an image of this work, see 〈http://en.wikipedia.org/wiki/The_Burial_of_the_Count_of_Orgaz〉 accessed 14 June 2012.

(13) Isenberg, ‘Critical Communication’, p. 335

(14) Ibid.

(15) Ibid.

(16) Ibid.

(17) Ibid.

(18) Isenberg, ‘Critical Communication’, p. 339. Elsewhere he says that ‘the “grounds” to which [the critic] is really appealing are not the same as those which he explicitly states or designates’ (ibid., pp. 343–344).

(19) Mothersill, ‘Critical Reasons’, p. 77. Compare Baxandall, ‘The Language of Art History’, pp. 455–456, and Lehrer, Art, Self and Knowledge, p. 21.

(20) Again, assuming that what is designated by an expression is what is denoted or referred to with (or by) that expression. If Isenberg thinks what is designated by an expression is what is characterized with or by that expression, then (contra Isenberg) Goldscheider does designate the outline with ‘x is like the contour of a violently rising and falling wave’.

(21) Daniel A. Kaufman, ‘Critical Justification and Critical Laws’, British Journal of Aesthetics 43/4 (2003): 399.

(22) Noël Carroll, On Criticism (New York: Routledge, 2009), p. 167.

(23) Isenberg, ‘Critical Communication’, p. 331

(24) Ibid.contentrepresentational

(25) Monroe C. Beardsley, Aesthetics: Problems in the Philosophy of Criticism, 2nd edn (Indianapolis: Hackett, 1981), pp. 242–247

(26) Carroll, On Criticism, p. 13

(27) Ibid.

(28) Ibid.

(29) Ibid.

(30) Ibid.

(31) Compare Kaufman, ‘Critical Justification and Critical Laws’, and Daniel A. Kaufman, ‘Normative Criticism and the Objective Value of Artworks’, The Journal of Aesthetics and Art Criticism 60/2 (2002): 151–166. See Richard Shusterman, ‘Wittgenstein and Critical Reasoning’, Philosophy and Phenomenological Research 47/1 (1986): 94–95 for a brief discussion of philosophers who have supported this idea (e.g., Aristotle, Collingwood, and Weitz) and of critics who have applied it (e.g., Addison, Johnson, and Coleridge).

(32) Carroll, On Criticism, p. 167. I take it that, by ‘other things being equal’, Carroll means ‘normally’. He writes: ‘The Isenbergian may grumble about my Harold Lloyd example, suggesting that it can be the case that under some strange conditions a particular pratfall might not contribute to the goodness of a slapstick comedy. But that is why the ceteris paribus clause has been added to our formulation’ (ibid., p. 168). The mention of ‘strange’ conditions suggests that, in Carroll’s usage, ‘all other things being equal’ means ‘normally’.

(34) Carroll, On Criticism, pp. 154–155

(35) Ivan March et al., The Penguin Guide to Recorded Classical Music 2010 (London: Penguin, 2009).

(36) Carroll, On Criticism, pp. 108–110

(37) Ibid.

(38) Ibid.

(39) Ibid.

(40) Ibid.

(41) Christine Tappolet, ‘Through Thick and Thin: Good and Its Determinates’, Dialectica 58/2 (2004): 210.

(42) Samuel Johnson, ‘Selections from the Notes to the Edition of Shakespeare’s Plays’, in Samuel Johnson on Shakespeare, ed. H. R. Woudhuysen (London: Penguin, 1989), p. 247.

(43) William Hazlitt, ‘Macbeth’, in Selected Writings, ed. Jon Cook (Oxford: Oxford University Press, 1991), p. 336.

(44) Arthur C. Danto, ‘The Fly in the Fly Bottle: The Explanation and Critical Judgment of Works of Art’, in Unnatural Wonders: Essays from the Gap Between Art and Life (New York: Farrar, Straus and Giroux, 2005), p. 361. See also Monroe C. Beardsley, ‘Critical Evaluation’, in The Aesthetic Point of View, p. 321.

(45) Sibley, ‘Aesthetic and Non-Aesthetic’, p. 36

(46) Ibid.

(47) ibid.

(48) Peter Lamarque, The Philosophy of Literature (Oxford: Blackwell, 2009), p. 207.

(49) Thomas De Quincey, ‘On the Knocking at the Gate in Macbeth’, in Confessions of an English Opium-Eater, and Other Writings, ed. Grevel Lindop (Oxford: Oxford University Press, 1985), p. 81.

(50) Ibid.

(51) Ibid.

(52) Ibid.

(53) Bas C. van Fraassen, The Scientific Image (Oxford: Oxford University Press, 1980), p. 134.

(54) Sylvain Bromberger, ‘Why-Questions’, in R. Colodny, ed., Mind and Cosmos: Essays in Contemporary Science and Philosophy (Pittsburgh: University of Pittsburgh Press, 1966), pp. 89–90.

(55) John Summerson, The Classical Language of Architecture (London: Thames and Hudson, 1980), pp. 21–22.