Values of Art and the Ethical Question

I. THE ETHICAL QUESTION

DOES THE ethical value of a work of art ever contribute to its aesthetic value? In what follows I shall refer to this as ‘the ethical question’. Philosophical tradition offers at least three different kinds of answer to the ethical question.1

Ethicists give an affirmative answer to the ethical question.2 According to the more plausible forms of ethicism, the ethical value of a work of art is aesthetically relevant insofar as it is suitably manifested by the intrinsic properties of that work. One canonical form of suitable manifestation is expression, as instantiated in the moral and religious vision propounded in a poetic work such as Dante's The Divine Comedy. According to the ethicist, the expression of an ethical attitude by a work is aesthetically relevant, not only insofar as recognizing its expression is a condition of understanding the work, but also insofar as ethically evaluating what is expressed is a condition of engaging with the work as the intentional product that it is. Ethicists differ over the exact form that competent aesthetic engagement may take.3 In one of its forms, the case for ethicism is based on the idea that some ethical aspects of artworks are genuinely beautiful, beauty being a paradigm aesthetic quality.4 In a second form, the case for ethicism is based on the cognitive or affective ethical rewards (or lack thereof) offered by some works of art.5 Yet another form of ethicism is based on the idea that some art invites or prescribes an ethically relevant response on the part of the audience, a response that may or may not on reflection be regarded as merited.6 Ethicism in all its manifestations entails that the aesthetic value of artworks varies positively with their ethical value. To this extent, the ethicist position might be thought to capture one distinctive sense in which art matters. Good art can have an ethically reinforcing impact on its audience insofar as the audience is able to engage sympathetically with the ethical values embodied within it.

So-called aesthetic ‘contextualists’ also give an affirmative answer to the ethical question. Yet contextualists deny that aesthetic value always varies positively with ethical value.7 According to the more plausible forms of contextualism, while the aesthetic value of a work of art can be enhanced by its ethical value, in some cases it can equally be enhanced by its ethical disvalue. Contextualism is sometimes motivated by appeal to deliberately amoral, immoral, or transgressive art that is designed to challenge existing norms, practices, or expectations. One kind of example that is sometimes used to illustrate the contextualist view is that of extremely violent or ethically subversive movies of the kind associated with film directors such as Martin Scorsese or Quentin Tarantino. When considered on its own terms, contextualism could also be thought to capture a distinctive sense in which art matters. Good art can have an intellectually stimulating effect on its audience insofar as art can engage or awaken ideas traditionally suppressed, negated, or disapproved of by the prevailing system of conventional morality.

So-called aesthetic ‘autonomists’ give a negative answer to the ethical question.8 According to the more plausible forms of autonomism, ethical values are irrelevant to genuinely aesthetic appreciation of artworks. So understood, the autonomist does not deny that artworks can manifest ethically relevant features that substantially affect our experience of these works. Nor does the autonomist necessarily deny that the ethically relevant features of an artwork can be aesthetically relevant. What the autonomist denies is that the ethical significance of these features is aesthetically relevant. Thus, the manner in which a face has been drawn could be funny on the one hand and an immoral insult to its subject on the other. According to the autonomist, the former fact is of aesthetic relevance while the latter is not. Autonomism could be thought to capture a third distinctive sense in which art matters. Art is special insofar as it allows the audience to engage in an intrinsically rewarding form of practically disinterested contemplation of the manifest features of artistic objects with the sole aim of experiencing the intrinsic reward that this engagement offers.9

One natural way to approach the ethical question is to make use of the standard tools of conceptual analysis.10 Such attempts would normally focus on identifying the necessary or essential properties of art, the aesthetic, and the ethical by means of an a priori analysis of the associated linguistic terms or underlying concepts. In this way, one might hope to understand the concepts of goodness and beauty, for example, in order to determine whether there are any interesting conceptual connections between the two. Or one might analyse the concept of the aesthetic to determine whether it is necessarily coextensive with the presence of a distinctive intellectual attitude. And so on. This analytical approach to the ethical question has been dominant in philosophical aesthetics for some time.11 It has not, however, succeeded in generating universal convergence on an answer to the ethical question. In this paper, I shall consider one possible explanation why this has been so.

II. AN ARGUMENT FOR ETHICISM

The case for ethicism is perhaps most plausible in the case of non-controversially representational forms of art such as literature. One of the more compelling arguments for ethicism about literary art in particular maintains that the direct application of ethical concepts is a necessary condition of competent engagement with at least some artworks of this kind.12 Thus, many works of fiction, such as Tolstoy's War and Peace, are not only characterized by a liberal use of explicitly ethical language, but also express ethical attitudes by means of this language, and either invite or prescribe that the reader share in those attitudes. It is therefore natural to think that it is a necessary condition of understanding such works both to grasp the ethical content, and to judge the ethical merit, of the attitudes expressed.

Perhaps the most plausible version of this argument takes it to apply to literary descriptions of individuals and situations in terms of so-called ‘thick’ ethical concepts, such as cruel, brutal, sentimental, decadent, and so on.13 Thick concepts, on at least one widely accepted account, combine evaluative content with descriptive content, the former providing their evaluative direction and the latter their world-guidedness and candidacy for objective truth and falsity.14 It might seem to follow that to endorse the application of a thick concept entails endorsing its evaluative content. When applied to the literary arts, for example, one consequence would seem to be that in order to grasp a literary description using a thick ethical concept it is necessary to grasp its evaluative point, and thereby to sympathetically enter into the ethical perspective expressed by that concept. A fictional work presenting a thick ethical description with which the reader cannot ethically sympathize will therefore be aesthetically problematic at the basic level of grasping the work's content. For without engaging fully and sympathetically with the work evaluatively, the reader will be unable to engage fully and sympathetically with the work descriptively.

In contrast to some of its more superficial cousins, this argument is arguably both philosophically deep and genuinely challenging. It is not, however, without its limitations. True, in order to grasp the application of a thick ethical concept it is necessary to find this application minimally intelligible. It is not, however, necessary to agree with it. A reader might disagree with the description of a fictional character as ‘a slut’, for example, at two different levels. First, the reader might disagree with the application of the concept because the (real or fictional) situation to which it is applied contingently fails to merit it. Thus, on the basis of what I take to be the author's unconvincing characterization I might disagree with the description of a fictional character as a slut even though I would be willing to describe some other character (whether real or fictional) as genuinely slutty. Second, the reader might disagree with the application of a thick concept because of her rejection of the central evaluative connotations of that concept. Thus, I might disagree with the unsympathetic description of a fictional character as ‘a slut’ on grounds that entail my refusal to apply this term to any character (whether real or fictional) whatsoever. In either case, however, it is possible for me to find the application of this term by another person intelligible in the context of his or her ethical outlook, even if I fail to agree with it, either in general or in this particular case. Given this fact, it will in a wide range of cases be possible for me to bracket my ethical disagreement (or in the complementary case, agreement) with the application of a thick concept as part of an episode of engaging with the wider description of which it is a part, this with a view to discovering what the description thus interpreted has to offer by way of aesthetically valuable experience. It therefore does not follow from the fact that a literary work includes a set of thick ethical concepts as part of its content that it is impossible to evaluate that work aesthetically in abstraction from the question of the ethical merits of the attitudes expressed by the thick concepts included in that work. Nor does it follow that it is impossible for someone to evaluate a literary work of art more rather than less by virtue of the way it makes vivid or interesting an ethical viewpoint with which the reader cannot, on reflection, agree. The basic entry points of both autonomism and contextualism are therefore left untouched by what is arguably one of the strongest arguments available for an ethicist answer to the ethical question in the case of the literary arts.

This conclusion obviously leaves untouched the other arguments for ethicism I mentioned in Section I above. I do not, however, propose to discuss the merits or demerits of these arguments here. Instead, I shall discuss two alternative theses that jointly suggest a different way to understand the debate about the ethical question, namely one according to which both ethicists and their opponents can be seen to put forward an approach to aesthetic evaluation that is, at least in principle, perfectly intelligible on its own terms.

III. THE INDETERMINACY THESIS

According to the first thesis I shall consider, the ethical question has no determinate answer, at least when interpreted as an a priori question of conceptual analysis. While potentially counterintuitive, this claim gains a modicum of initial plausibility from reflection on the history of some of its constituent ideas. The term ‘aesthetic’, for example, is a term of art, apparently first introduced into philosophical discourse in the early part of the eighteenth century to characterize what at the time was considered to be a canonical form of engagement with some works of art, including, but not exclusively so, the art of poetry.15 What we now think of as aesthetic interest obviously pre-dates the introduction of this term, as does the activity of what we now call aesthetic criticism and philosophical theorizing about its nature, scope, and limits. The history of these different activities is characterized by changes of emphasis along a number of dimensions. These include the many and various possible connections between art and nature, art and craft, art and science, art and religion, art and introspection, art and education, as well as art, ethics, and politics. The connotations associated with the term ‘aesthetic value’ or any of its rough equivalents at any given time and place have obviously varied in response to how important the various connections listed above have been for the people engaging critically with artworks there and then. The connotations associated with the term ‘aesthetic value’ at the present time in the parts of the world inhabited by contemporary analytical aestheticians are equally a product of how important the various connections just listed are for people engaging critically and competently with artworks here and now. As the present state of philosophical discussion arguably testifies, there is disagreement among competent speakers as to what the defining connotations of the term ‘aesthetic value’ currently are. On a charitable reading, this disagreement is not a superficial terminological dispute about the right to use an intrinsically arbitrary label, but rather a deeper dispute about the boundaries of the concept (or concepts) the term is used by competent speakers to express. Any resolution of this disagreement presupposes the existence of an account that adequately characterizes the underlying concept (or concepts) in question. It is a surprising, if notable, feature of much philosophical writing on the ethical question that the explicit articulation of such a theory is a task that is rarely, if ever, attempted.

There are at least two standard approaches that an analysis of the central notions involved in the ethical question could adopt. The first is to understand the relevant terms by analogy with natural kind terms. On one common version of this view, aesthetic value is, roughly, that phenomenon in the world, whatever it may be, that is the causal origin of our thought and talk when we employ the term ‘aesthetic value’.16 This natural kind approach is prima facie unpromising when applied to the concept of aesthetic value, for at least two reasons. First, arguments about the nature of aesthetic value are not primarily (if at all) driven by the project of discovering the nature of whatever contingent phenomenon was initially labelled as aesthetically valuable by a particular group of European intellectuals at some point in the eighteenth century. No doubt such a phenomenon exists. Thus there are aspects of Kant's third Critique that strongly suggest an interest in discovering the essence of a particular psychological faculty that he called ‘judgement’ (or ‘Urteilskraft’). Yet this phenomenon is only one contingent take by some people on what we have come to recognize as the locus of aesthetic value, as opposed to a uniquely salient phenomenon that serves to fix the meaning and reference of ‘aesthetic value’ once and for all. Second, aesthetic value is not prima facie a strong candidate for classification as a ‘natural kind’ on any of the currently plausible conceptions thereof. On the contrary, aesthetic value, such as we are able to identify it in extension, is a historically contingent and dynamic phenomenon, subject to development and change in response to artistic and critical developments in historical time. In this way, aesthetic value arguably differs markedly from paradigm examples of natural kinds, including the chemical elements and natural properties targeted by the natural sciences, and appealed to as paradigmatic in philosophical analyses of natural kind terms.

A second approach to the conceptual analysis of aesthetic value is to understand the term ‘aesthetic value’ in terms of its constitutive connotations. On this view, aesthetic value is that phenomenon in the world, whatever it may be, that is picked out by the instantiation of all, most, or enough of the connotations associated with the term ‘aesthetic value’ by speakers who count as having competently mastered the term at any given time or place.17 This approach is prima facie more promising, for at least two reasons. First, it avoids the difficulties confronted by the construal of ‘aesthetic value’ as a natural kind term. For on this second approach, the extension of ‘aesthetic value’ is implicitly sensitive to the changes and critical development that both the term and the underlying phenomenon it picks out have undergone over time. Second, this second approach connects more clearly with what arguments about the ethical question have traditionally been about—that is, not the empirical constitution of an underlying natural phenomenon causally responsible for our use of the term ‘aesthetic value’, but rather the intelligible boundaries of a concept that defines a distinctive, but historically variable, kind of intellectual interest. I shall describe this approach as ‘the descriptive approach’ to the analysis of aesthetic value, and I shall henceforth assume it in the discussion that follows.

The different connotations associated with the notion of aesthetic value by competent speakers are variously spread out across a wide spectrum. Some connotations arguably make up a relatively uncontested conceptual core. This core includes among its elements such comparatively formal values as derive from intrinsic aspects of artworks like their structure, universally (or near enough) agreed among competent speakers to ground genuine attributions of aesthetic value. According to the indeterminacy thesis, the answer to the question whether these comparatively formal values are indeed aesthetic is affirmative, and determinately so. Thus, no party to the debate seriously denies that the values trumpeted as aesthetic by autonomist answers to the ethical question are genuinely aesthetic, on any understanding of that notion that we are likely to recognize.

Other connotations of aesthetic value, while common or widespread among competent speakers, are not universally shared. Among these connotations, some are seriously contested. This domain of connotations arguably includes among its elements such comparatively substantial values as derive from extrinsic features of artworks such as their originality, as well as certain intrinsic features of artworks such as some aspects of their representational content. While not universally agreed among competent speakers to ground attributions of aesthetic value, these features can be agreed to ground some kind of value, the categorization of such value being potentially subject to reasonable disagreement. According to the indeterminacy thesis, the answer to the question whether these comparatively substantial values are indeed aesthetic is indeterminate, thus forming an area of vagueness with respect to the categorization of values as aesthetic or non-aesthetic. In this way, there can be intelligible controversy between ethicists on the one hand and autonomists on the other about which aspects of the representational content of literary artworks are genuinely relevant to their aesthetic value and which are not.

Beyond this area of vagueness there is an indefinitely large domain of features that are rarely, if ever, associated with the notion of aesthetic value by any reasonable and competent speaker. Thus, it is rarely (if ever) considered aesthetically relevant that an artwork is made of a certain number of atoms (although this fact might conceivably be considered relevant for some different intelligible purpose—a particularly risky bet, say). Such features are (near enough) universally agreed by competent speakers not to ground genuine attributions of aesthetic value. According to the indeterminacy thesis, the answer to the question whether any values grounded in such features are aesthetic is negative, and determinately so. Thus, no party to the debate seriously denies that there are some values attributable to artworks that are not genuinely aesthetic, on any recognizable understanding of that notion.18

The boundary between the outer domain of determinate aesthetic irrelevance and the range of indeterminacy is arguably itself vague. The same applies to the boundary between the range of indeterminacy and the conceptual core of aesthetic relevance. There is consequently no determinate answer to exactly where one domain ends and another begins. It does not follow that there is no difference at all between the three domains, on pain of paradox.

According to the indeterminacy thesis, answers to the ethical question fall into the second, indeterminate, domain of non-universal and reasonably contestable connotations of aesthetic value. We can initially motivate this claim by appealing to the following three considerations. First, the association of ethical with aesthetic value is currently widespread but not universally accepted among competent speakers, including competent practitioners of aesthetic criticism. Second, the association of ethical with aesthetic value has been subject to significant variation among basically competent speakers, including practitioners of aesthetic criticism, at different points in the history of aesthetic thought. Both of these considerations can be verified by a cursory glance at the existing literature. I shall therefore simply take them as given here. Third, the association of ethical with aesthetic value can coherently be subjected to reasonable contest within the practice of aesthetic criticism itself, with intelligible, interesting, and innovative arguments existing on both sides. I shall address the significance of this third claim shortly.

IV. THE NORMATIVE THESIS

While some attempts to engage with the ethical question are undermined by the indeterminacy thesis, other attempts may not be so vulnerable. An answer to the ethical question does not have to take the form of an a priori descriptive claim about the correct analysis of the concept of aesthetic value. An attempt to answer the ethical question could equally be a substantial and normative proposal about some critically appropriate focus of aesthetic engagement and reward. This kind of answer to the ethical question would arise from substantial, or first-order, engagement with artworks themselves, and would connect with conceptual questions of analysis and essence only at the margins. In the terms of such practical, or first-order, involvement, the ethical question is one about how to critically approach artworks in an informed, intelligent, and potentially rewarding way.19 Thus understood, the ethical question is a normative question, itself an internal part of aesthetic criticism. To think of the ethical question in this way is a source of one of the more charitable interpretations of the wider historical debate about the nature of aesthetic value and its various relations to our ethical sensibilities.

According to the normative thesis, then, some answers to the ethical question are best understood as attempts to promote some connotations of ‘aesthetic value’ over others in the course of aesthetic engagement. At the second-order level of conceptual analysis, such attempts can be thought of as negotiating a vague conceptual boundary by influencing how the line between conceptual core and indeterminacy is drawn. At the first-order level of aesthetic engagement, such attempts can be thought of as promoting the value of approaching a given set of artworks in terms of some connotations rather than others within the domain of conceptual core and indeterminacy. The guarantee of reasoned convergence on one answer to the ethical question in the course of aesthetic engagement would obviously reintroduce at a first-order level a form of determinacy assumed to be lacking at the second-order level. Yet it would be just as foolish to assume that there is necessarily only one reasonable answer to the ethical question so interpreted as it would be to think that there is necessarily only one reasonable answer to the ethical question for all artworks. Given that there is a plurality of different ways of engaging rewardingly with artworks in an informed and intelligent way, there will be a plurality of different ways of reasonably combining the different connotations of ‘aesthetic value’ so as to realize that reward in aesthetic experience. This is not to say that any package of connotations would be as reasonable as any other. Given that any activity we could recognize as a form of aesthetic engagement must be guided by the features of its object, there will be elements of the conceptual core that no reasonable form of aesthetic engagement would ignore. Thus, it is impossible to competently judge a painting by Rothko, for example, without considering his use of colour. Likewise, there will be elements at or beyond the periphery of connotations of aesthetic value that only an impossibly artificial conception of the notion of aesthetic would tolerate. Thus, no serious critic would be likely to hold it against a piece by Rothko that it has been reproduced in poster form, one copy of which was once involved by mere accident in setting fire to someone's curtains. At this point, we reach the realm of the unrecognizable, where what started out as an interesting and reasonable debate has indeed been transformed into what is at best an uninteresting terminological dispute about the right to use an intrinsically arbitrary label.

The explanatory connection between the normative thesis and the indeterminacy thesis is as follows. On the one hand, the normative thesis explains why the conceptual boundaries of aesthetic value are contested in the course of aesthetic criticism. On the other hand, the indeterminacy thesis locates the domain of reasonable contest within the domain of indeterminacy. In this way, the two theses cohere to make up an explanatorily potent analytical tool, provided they can be shown to be plausible when actually applied in the course of philosophical and aesthetic criticism. I have obviously made no serious attempt to apply this tool as part of aesthetic criticism in this paper.

V. INTERPRETING THE TWO THESES

The normative thesis and the indeterminacy thesis can each be interpreted in a number of different ways. Some of these interpretations are potential sources of misunderstanding. Here I shall mention two. First, it might be thought that the normative thesis and the indeterminacy thesis jointly entail a form of contextualism. This would be a mistake. The normative thesis and the indeterminacy thesis are inconsistent with contextualism as defined in Section I. True, the contextualist denies the a priori conceptual claim that aesthetic value varies positively with ethical value. To this extent, contextualism shares a negative commitment with the two theses put forward in this paper. Yet contextualism also entails that it is determinately the case that in some contexts the aesthetic value of an artwork is enhanced by its ethical value, and that in other contexts the aesthetic value of an artwork is enhanced by its ethical disvalue. This claim is inconsistent with the indeterminacy thesis, and therefore incompatible with the normative thesis and the indeterminacy theses combined. According to the indeterminacy thesis, it is not determinately the case that the ethical value or disvalue of an artwork either enhances or detracts from its aesthetic value, even when the contextual facts are known. What is consistent with the indeterminacy thesis is the claim that on some possible ways of removing indeterminacy (by stipulation or otherwise) the ethical value of artworks can affect the aesthetic value of those artworks. This claim is further consistent with the claim that the way in which ethical value affects aesthetic value so understood is correctly describable along contextualist (as opposed to ethicist) lines. Yet the indeterminacy thesis does not entail this claim, either alone or in conjunction with the normative thesis. Accepting the two theses is therefore not to accept a form of contextualism.

Second, there is more than one way to understand the indeterminacy thesis. On one interpretation, indeterminacy is a standard case of ambiguity. The idea that the indeterminacy embodied in the indeterminacy thesis can be analysed as a standard case of ambiguity is supported by the fact that ‘aesthetic value’ is a part-colloquial, part-technical, term with a number of different contexts of application. In standard cases of ambiguity, initial indeterminacy can be removed by specifying which among a number of intelligible meanings a speaker has in mind, depending on such different factors as communicative context, speaker intention, assumed reference-class, and so on. In the case of an ambiguous term such as ‘house’, for example, initial indeterminacy can be resolved by explicitly specifying whether in a given context the term ‘house’ is to be understood as being coextensive with the general ‘dwelling’ or the more particular ‘free-standing dwelling’, say. Successful removal of indeterminacy by disambiguation along these lines would normally remove semantic confusion and any associated grounds for disagreement. If the indeterminacy attributed to the term ‘aesthetic value’ by the indeterminacy thesis were a standard case of ambiguity, it should therefore be possible to remove semantic confusion and any associated grounds for disagreement in a similar way by specifying whether in a given context the term ‘aesthetic value’ is meant to be understood in a narrow and formalistic sense, or alternatively as including in its extension a wider range of expressive, cognitive, and ethical values.

It is not plausible to think that this possibility obtains for ‘aesthetic value’. Some acts of disambiguation are compatible with the indeterminacy thesis and could in principle remove some sources of semantic confusion. Yet it is not clear that any such act would remove all associated grounds for disagreement. Suppose we grant the autonomist her title to the disambiguated term ‘aesthetic value (narrow)’ and the ethicist (or contextualist) their title to the disambiguated term ‘aesthetic value (wide)’. Given that the functional role played by the term ‘aesthetic value’ in our thinking is to licence evaluative inferences and associated actions in our practical engagement with artworks, the question whether to base these inferences and actions on ‘aesthetic value (narrow)’ or ‘aesthetic value (wide)’ will remain even after disambiguation. This is where the normative thesis comes in, stating that this question is a matter of substantial, first-order, evaluative engagement with artworks themselves. No mere act of disambiguation can answer this normative question. The indeterminacy embodied in the indeterminacy thesis is therefore not best understood as a standard case of ambiguity.

On a second interpretation, indeterminacy is a standard case of vagueness. This interpretation is suggested by the fact that vagueness can survive disambiguation by leaving borderline cases. Thus, whether we understand by ‘house’ either ‘dwelling’ or ‘free-standing dwelling’, it remains vague in some cases whether a given location of occupancy qualifies as a ‘house’ on either disambiguation. The most obvious cases of vagueness are ones where a line is drawn along a continuum, such as those of height, weight, or size. Yet as the case of ‘house’ illustrates, this is not an essential feature of vagueness, in which case the absence of a relevant continuum in the case of ‘aesthetic value’ is not an obstacle to describing the indeterminacy in its extension as a standard case of vagueness.

Nevertheless, there are several obstacles to analysing the indeterminacy embodied in the indeterminacy thesis as a standard case of vagueness. Here I shall mention two. First, the potential for disagreement left over from a disambiguation of ‘aesthetic value’ does not arise from a residue of borderline cases in the extension of either ‘aesthetic value (narrow)’ or ‘aesthetic value (broad)’, but rather from the question of whether to treat ‘aesthetic value (narrow)’ or ‘aesthetic value (broad)’ as the basis for paradigmatic inferences and actions associated with the activity of engaging critically with artworks. Second, and according to the normative thesis, the choice of a preferred interpretation for the term ‘aesthetic value’ is a decision the guiding values of which are not conceptually separable from the substantial, first-order, activity of engaging intelligently with artworks themselves. In this respect, the case of ‘aesthetic value’ arguably differs from a wide range of vague terms (such as ‘tall’ or ‘chair’) for which different sharpenings within a domain of indeterminacy could reasonably be proposed in various contexts on purely pragmatic grounds (such as convenience or simplicity) that bear no interesting conceptual relation to the question of what, if anything, qualifies an object as falling within the extension of the relevant term.

Between them, the normative thesis and the indeterminacy thesis describe a form of indeterminacy that, even though it shares important features with standard ambiguity and vagueness, is not straightforwardly reducible to either. It does not follow that the two theses fail to describe a genuine semantic phenomenon. The challenge of understanding this phenomenon as a semantic category that combines controversy over conceptual boundaries with a shared and incontestable conceptual core goes straight to the heart of our understanding of value-laden terms in general. This challenge is arguably related to the reason why W. B. Gallie has described some evaluative concepts as ‘essentially contestable’. Essentially contestable concepts are concepts for which all parties can in principle agree that (a) some paradigm cases fall non-controversially within their extension, (b) the criteria applied by the opposition are intelligibly applicable on their own terms, and (c) the choice of which criteria to use can be a matter of genuine, substantial, and persistent disagreement among reasonable people. The idea of essential contestability and the case for thinking about aesthetic value along these lines is one about which much more could usefully be said than I have space to do here.20

VI. APPLYING THE TWO THESES

The normative thesis and the indeterminacy thesis jointly account for a number of salient facts about the existing literature on the ethical question. Here I shall mention three. First, the two theses jointly explain the apparent interminability of this debate. This appearance of interminability should come as no surprise once the ethical question is understood as a substantial and evaluative question about informed, intelligent, and potentially rewarding ways of approaching artworks with understanding. Given a reasonable pluralism about valuable ways of approaching artworks with understanding, it is hardly surprising that philosophers, critics, and laypersons have come to different conclusions in different circumstances about the experiential reward offered by different degrees of ethical engagement when approaching different artworks. Thus, a contextually informed experience of watching a film such as Triumph of the Will, for example, could undeniably be affected by whether it takes place before or after the events of the Second World War. When interpreted in terms of the two theses, the apparent interminability of the debate about the ethical question is a predictable symptom of the plurality of aesthetic value broadly understood, and not necessarily a symptom of philosophical misconceptions of the a priori necessary conditions of competent aesthetic judgement.

Second, the two theses combine to account for why some judgments of aesthetic relevance are comparably uncontested and stable. The classification of connotations into the three domains of conceptual core, indeterminacy, and determinate periphery captures this division as it has been made by different speakers at different times and places. At least with respect to the conceptual core, it is a plausible conjecture that some of its elements have remained comparatively stable across space and time. The same arguably holds for an indefinitely wide region beyond the periphery of indeterminate aesthetic relevance. The domain of indeterminacy, however, including its vague boundaries between core and periphery, is predicted by the two theses to be subject to contest and change. This prediction is arguably borne out by the variable defeasibility with which connotations involving ethics, politics, religion, science, craft, and other practices have been associated with the notion of aesthetic value in writings on art and aesthetics at different times and places.

Third, the two theses jointly accord with the holistic and indeterminate character of first-order aesthetic criticism. Different competent descriptions of an artwork may involve different emphases on the various connotations associated with the term ‘aesthetic value’ and may relate these different connotations to each other in indeterminately different ways. Thus, the relative importance attached to intrinsic and extrinsic features of a work may vary from one competent description to another, depending on the focus and interest that drives an episode of aesthetic engagement. One competent description of Rubens's painting The Adoration of the Magi, for example, may emphasize its colouring and structural features over its representational properties. A second competent description may focus more heavily on the painting's representational properties, such as its religious content. A third description may focus on such extrinsic features of the painting as its manner of conception, its originality, the cultural significance of its existence, and the context of its current presentation as hung above the altar in King's College Chapel. Each kind of description can obviously be combined with aspects of the others in an indefinite number of ways, while engaging with the painting as the object that it is. Each kind of description is in principle capable of combining with a valuable experience of the work, the value of this experience being such as to merit the label ‘aesthetic’ with respect to each of its relevant aspects. To deny this degree of holism and indeterminacy in the first-order practice of aesthetic criticism would be to arbitrarily disconnect the project of accounting for the nature of aesthetic value from the actual practice of critical engagement that the project of philosophical aesthetics is trying to understand.

VII DEFENDING THE TWO THESES

It might be objected that the two theses jointly imply an excessively liberal view of competent aesthetic judgement. One thought behind this objection might be that the admission of a wide plurality of connotations as aesthetically relevant embodied in the indeterminacy thesis fails to respect the obvious fact that sound aesthetic engagement is constitutively bound up with the nature of its object, and that all considerations of genuine aesthetic relevance must therefore be rooted in the intrinsic properties of the artwork itself.21

This objection fails, for two reasons. First, the indeterminacy thesis is consistent with the claim that sound aesthetic engagement is grounded in the nature of its object. Indeed, many of the narrowly intrinsic features of an artwork are among the features associated with the conceptual core of aesthetic value, including the values grounded in the use of colour and light in The Adoration. Second, the claim that sound aesthetic engagement is necessarily grounded purely in the intrinsic features of its object is arguably too strong. A wide range of extrinsic features of artworks have historically been considered as being of paradigmatic aesthetic significance among competent art critics, including such features as originality, rarity, and historical origin. For the vast majority of interested parties there is a world of difference between the aesthetic value of The Adoration itself and any digitized reproduction thereof downloaded from Google Images, for example.

A second objection would claim that the two theses jointly fail to account for the fact that some forms of aesthetic engagement count as ‘canonical’ forms of engagement with works of art as works of art. Such canonical forms of aesthetic engagement are arguably definitional of the concept of aesthetic value in its historically paradigmatic and incontestable applications.22 Thus, a focus on the intrinsic surface features and compositional structure of The Adoration, for example, would form an incontestable part of any canonical aesthetic engagement with that painting as a work of art. The two theses, however, might seem to have the implausible implication that no subset of possible approaches to an artwork can count as aesthetically canonical in the required sense.

In fact, the two theses have quite the opposite implication. First, the idea of canonical aesthetic engagement can be consistently grounded in the conceptual core of connotations associated with the notion of aesthetic value by competent speakers. Second, the domain of indeterminacy and the vague boundary between this domain and the conceptual core accounts for how modes of aesthetic engagement considered as canonical might differ at different times and places (something that has undoubtedly been the case). Third, in at least some of its critical applications the notion of a canonical mode of aesthetic engagement is most charitably interpreted as a substantial normative idea, primarily serving to recommend the focus of aesthetic engagement on some connotations of aesthetic value rather than others. Thus, a number of historically influential formulations of ethicism as applied to the art of literature, for example, could reasonably be interpreted along these lines, as could some classical formulations of autonomism as applied to the visual arts.23

A third objection is focused on the methodology behind the argument for the indeterminacy thesis. By focusing on the connotations associated with a linguistic term such as ‘aesthetic value’ it might be thought that this argument fails to preserve the connection between the materials of philosophical analysis on the one hand and the underlying objective phenomenon the understanding of which is the ultimate target of the conceptual inquiry on the other. I have already argued that this objection is implausible if it is taken to suggest that aesthetic value is a natural kind of which the term ‘aesthetic value’ serves as a so-called ‘rigid designator’.24 Quite apart from this point, however, the objection is misguided if it assumes that the truth of the two theses is incompatible with the existence of a genuine phenomenon of aesthetic value that exists independently of our application to it of some arbitrary linguistic label. On the contrary, the two theses jointly constitute an account of how this independently existing phenomenon can be characterized. What the two theses do imply is that this phenomenon is indeterminate in its boundaries and potentially plural. The conclusion that the phenomenon does not objectively exist does not therefore follow.25

A more serious objection to the two theses is based on the claim that the nature of artworks as intentionally produced objects entails substantial and determinate constraints on sound aesthetic engagement in such a way as to rule out the promiscuous degree of vagueness at the boundaries of aesthetic value entailed by the indeterminacy thesis. As with the first objection considered above, this objection derives from the insight that sound aesthetic engagement with artworks is engagement with those artworks as works of art. Thus, if the intentional production of a given artwork is constitutively bound up with relational features of that artwork, these relational features of the work are aesthetically relevant. If not, they are not. The truth or falsehood of the religious viewpoint represented in a musical work, for example, can be aesthetically relevant in virtue of the artist's intention to represent the world as he sees it in that work. To this extent it might be considered correct for an atheist to judge Mozart's Requiem and Brahms's Requiem differently insofar as genuine religious conviction arguably played an essential part in the conception of the former that it may not have done in the case of the latter.26 If no such intention on part of the artist exists, either as manifested in the work itself, or in the context of the work's production, the question of truth is aesthetically irrelevant, and determinately so.27 It might be thought that the indeterminacy thesis fails to capture this distinction between aesthetic relevance and irrelevance, thereby characterizing as indeterminate a set of questions that in fact have perfectly determinate answers. If so, one might suspect that the putative domain of vagueness entailed by the indeterminacy thesis would simply evaporate in the face of sufficient knowledge of artworks themselves, considered as intentional products with objectively real features and an objectively given history of conception.

This objection is inconclusive. First, consider some contextual fact about a literary work such as War and Peace. Let this fact be one relating the novel to the political circumstances surrounding its manner of conception. An episode of competent engagement with War and Peace might take this fact into account as part of an aesthetic evaluation of that work or it may not. Either way, nothing can be inferred a priori about the spectator's understanding or misunderstanding of the work. Nor is it obvious that only one of these ways of approaching the work is capable of generating a valuable experience that is in some recognizable sense aesthetic. Second, consider the variety of different conditions in which an artwork may be approached for the intrinsic reward that it offers: to decide whether one wants to purchase or display it in a certain way, to consider its creator for a prize, to rank the work in comparison to others for the purpose of dividing one's attention or cutting one's art collection by half, to consider its artistic significance as a historical product, and so on. Each of these conditions is compatible with the engagement in question being recognizably aesthetic, even if its manner and focus may differ from case to case (and even if some of these ways of approaching an artwork may be impossible for a given spectator at a given time and place). It is arguably one of the strengths of the indeterminacy thesis and the normative thesis that they jointly account for the way in which the values attributed to artworks in the course of such different kinds of engagement can all merit the label ‘aesthetic’, without thereby entailing the much stronger (and implausible) claim that there is such a thing as the determinate essence of aesthetic value that they all share and that grounds a set of individually necessary and jointly sufficient conditions for meriting the use of the label ‘aesthetic value’. Indeed, by making room for a reasonable degree of pluralism about aesthetic value, the two theses arguably do better justice than their competitors to the way in which the activity of aesthetic engagement is both a self-consciously critical and a historically dynamic phenomenon.

At this point it is necessary to enter a caveat. As Hume points out, the ethicist is undoubtedly right in holding that the task of bracketing the question of the ethical merit of the attitudes expressed in a literary artwork can be more difficult the more widespread the ethically offending descriptions are in that work, and the less widespread the compensating presence of other, more rewarding, features of that work.28 Yet the indeterminacy thesis and the normative thesis do not together entail that an atheist who holds it against Dante's Divine Comedy, for example, that it expresses a system of what he regards as false ethical values is thereby necessarily making a mistake.29 Nor do they entail that imaginative resistance resulting from the attempt to enter into an ethical viewpoint with which one disagrees could not be a genuine aspect of an experience on the basis of which a competent judgement of aesthetic value is based. On the contrary, it is consistent with the two theses to think that the degree to which it is reasonable to bracket the ethical values associated with a work of art is intimately connected to the difficulty of doing so while engaging fully with it and extracting such aesthetic reward that it has to offer. The proof of the pudding lies in the eating. Over to you, art critics!30