1. The role of truth in art

Lady Gaga proclaims in her concerts ‘I hate the truth!’, adding in an interview that ‘in fact I hate the truth so much I prefer a giant dose of bullshit any day over the truth’.Footnote ¹ With less chutzpah but more philosophical sophistication, both Marxists and postmodernists debunk the suggestion that art aims at truth – on the grounds that art is ideological and, perhaps, that truth does not exist. Analytic philosophers are more cautious but equally sceptical, though artistic cognitivism has defenders among them. Gordon Graham, for instance, argues that we value art according to its ability to illuminate human experience, and that beauty and pleasure alone ‘cannot explain the value of art at its finest’.Footnote ² Even he does not defend artistic truth, however.

There is an obvious connection between cognitivism and artistic truth. However, except concerning the particular questions of the semantics of fictional discourse, and the impact of factual error on artistic value, truth has been little-debated in Anglo-American philosophical aesthetics. In asking whether the historical inaccuracy of an historical novel affects its aesthetic or artistic value, for instance, it is assumed too hastily that art is defined independent of truth. The focus is on how falsehood detracts from artistic value, not on how truth contributes to it.Footnote ³ Against dominant anti-definition or sociological accounts of art, I believe, one can talk at least of salient features, and among these is the property that high art aims at truth.

Artists often suggest that this is their intention. James Joyce said that in Dubliners his aim was to tell the truth; Picasso, it is said, commented that ‘Art is a lie that tells the truth’. As Danto argues, art – by which he means high or fine art, ‘Art’ with a capital ‘A’ – must be about something.Footnote ⁴ Presumably it says, suggests or hints things about that something – and therefore one can ask whether what it says or suggests is true.

A rare philosophical proponent of artistic truth is Iris Murdoch, who argues that art presents us with ‘a truthful image of the human condition in a form which can be steadily contemplated’, ‘the only context in which many of us are capable of contemplating it at all’.Footnote ⁵ However, her claim that practices of looking are educated by engaging with art rests on unacceptably Platonist assumptions. This article's more modest thesis carries no suggestion that art is privileged as a way of knowing; it implies no Romantic conception of poet as seer, and does not assume a metaphysical or absolute conception of truth, as those in the humanities who reject artistic truth often seem to assume. Rather, it advocates a deflationary, anti-metaphysical conception, and elucidates ‘Art aims at truth’ in terms of raising or addressing issues which an audience would discuss. Even this modest conception of artistic truth erodes the dominant philosophical paradigm of art, however.

That art aims at truth does not imply an intellectualist view of art; unconscious influences are vital in its creation, and art can aim at truths which the artist does not consciously intend. To say that it aims at truth is not to appeal only to the author's avowed or conscious intentions; intentions are manifested primarily through the work. But there is an essential reference to intentionality; mere illustration of truths is not sufficient. Artworks can illustrate a truth just as anything else can – a novel, a newspaper article and a radio interview can all show that racism was endemic in 1930s Britain, for instance – but that is not what is meant by art's aiming at truth.

Artworks can show how things might be, presenting possibilities for the audience's consideration. David Harrower's one-act play Blackbird, for instances, forces us to reconsider our definition of, and attitude, towards paedophilia – without, as didactic art would, prescribing some particular response.Footnote ⁶ It challenges us to consider the question whether a relationship between a man and an under-age girl could ever have value. In virtue of its form, the play slowly introduces its audience to the past relationship between the protagonists. There is a compelling dramatic rhythm to the gradual emergence of this information, that concludes with a revelation of the protagonists’ continuing feelings for each other. To fail to consider the issues concerning paedophilia that the play raises, is to reveal a very inadequate understanding of it.

To argue that art aims at truth, in the way that I propose, is to offer a humanistic treatment of art as continuous with other human activities. It implies a division between art that aims at entertainment, art that aims at usefulness, and art that aims at truth; and within the latter category, between art that aims at truth and fails, and art that aims at truth and succeeds. My claim is not that there is a special notion of ‘artistic truth’, but rather that high art aims at truth in the ordinary sense of that term. High art is not a purely social category, but it is an historically-conditioned one; it is a possibility fully manifested at a certain stage of art's historical development, though it may be present in earlier eras.Footnote ⁷ It is autonomous art, created not simply to satisfy a patron, but to capture authentic aims of the artist. The modernist picture is that such a possibility, though perhaps standing only a remote prospect of realisation, opens up when art enters the market-place; it becomes potentially autonomous at the same time as it becomes a commodity. Autonomous artworks have ceased to be products for an occasion, and are liberated from direct social function in service of court, aristocracy or church; a liberation that music, most socially retarded of the arts, finally gained only from the later 18^th century.

For both modernists and conservatives, autonomous art is consolation for the loss of common culture in an urban, industrialised age. Scruton for instance argues that high art, far from being an instrument of class oppression, emancipated people from the traditional common culture: ‘high culture…is the most reliable cure for the resulting loneliness. Bourgeois civilisation frees us from the bonds of common culture, and offers the consolation prize of art’.Footnote ⁸ However, although Scruton's treatment of art can hardly be described as ‘functionalist’, he neglects the defining characteristic of autonomous art, its purposiveness without a purpose – a concept that Adorno borrowed from Kant, transforming it to capture the unique, indirect social functions that autonomous art has in virtue of its lack of direct social function.Footnote ⁹ Autonomy, and classic status, involve separation of artworks from the original social circumstances of their production and consumption.Footnote ¹⁰

Art that aims only at gratification is art with a small ‘a’ – diversion or entertainment. Higher art is not merely pleasurable, though it can also entertain, as some works of Dickens and Shakespeare show; the contrast is with mere entertainment. The latter description is rather misleading when applied to an important borderline category, even so. To say that The Artist, Michel Hazanavicius’ recent acclaimed silent film, ‘aims at truth’, is to overstate its aims; it does not, for instance, show us much about the demise of the silent movie industry that we did not know already. The film is ‘knowing’ in a postmodern way, cleverly structured and with marvellous set-pieces, but essentially a wonderful entertainment. One could argue the same for Howard Hawks’ classic Chandler adaptation The Big Sleep, or his comedy Bringing Up Baby, though the boundary between high and popular arts is essentially vague. A diet of both high and popular art is ideal, and one that lacks higher art can be argued to be impoverished.

My central claim, therefore, is that high art aims at truth, while other kinds of art aim merely at pleasure, entertainment, fantasy or utility. The claim that high art aims at truth in some way parallels the familiar claim that belief aims at truth. Not all beliefs are true, and not all artworks have truth-content; the claim is normative. Truth is the fundamental dimension of assessment of beliefs, informing our epistemic norms and principles as a goal or theoretical value.Footnote ¹¹ For art, truth is one fundamental mode of assessingment; one can argue that an artwork fails to achieve truth despite the best efforts of the artist, for instance because their vision is distorted or incomplete.

In each case the claim can be contested; those who treat belief as an informational-state or disposition tend to deny that it aims at truth, while those who regard high art as an essentially self-reflexive or hermetic activity may deny the analogous claim.Footnote ¹² These alternative pictures should be resisted. There are duplicitous or lying assertions, but not lying beliefs; so belief cannot aim at falsehood. Art is more like the assertion than the belief; even so, if it fails to aim at truth, it is not high art. One could perhaps argue that art, in contrast, can aim at falsehood; perhaps inauthentic or insincere art does this. But I would prefer to argue that the main contrast is with art as entertainment, that aims at neither truth nor falsity.

An essential supplementary claim is that each artform aims at truth in its own way or ways. This claim is consistent with the possibility of syntheses between artforms, as it is with Greenbergian purism. Painting can aim at truth through lifelikeness or mimesis, while fiction can do so through didacticism; but in both cases, a higher and more sophisticated approach is possible. Art can be said to aim at truth in the following ways:

1. 1. Expression, presentation or consideration of truth or truths

2. 2. Truthful representation

3. 3. Truth to material or materials

4. 4. Authenticity, including authentic performance

5. 5. Historical truth

My focus here is on 1, and to a lesser extent 2 and 3.

The claim of artistic truth should be understood within the context of the historical development of the concept of art. Plato conceived of the aesthetic as sensual – and therefore objectionable – because he seemingly had no conception of art as other than the merely pleasurable, or the instrumentally didactic.Footnote ¹³ By instrumentally didactic, I mean art for religion's or politics’ sake, where the artwork is treated primarily as a vehicle for the transmission of truth. In such art, there is no internal relation to truth, that ‘aiming at truth’ which is consistent with art for art's sake in its broadest sense, the sense in which art becomes an end in itself.

In fact, in ancient Greek writings concerning activities now regarded as artistic, there is little alternative to the instrumental or diversion model, or the social function model, such as flute-playing at a symposium. I say ‘activities now regarded as artistic’, because the Greeks had no model of art as such, conceived of as manifested across different genres; that is, they had no system of the arts. The widely-accepted Kristeller thesis argues that the Western system of the five major arts – painting, sculpture, architecture, music and poetry – did not assume definite shape till the 18^th century, even though its ingredients went back to classical times.Footnote ¹⁴ It is generally agreed that the Greek term techne (Latin ars) does not distinguish between art and craft, in the modern senses of these terms, but embraced all kinds of human activities which would now be called arts, crafts or sciences.

The evolution of the modern system of the arts accompanied Kant's development of the concept of the aesthetic as a synthesis of the sensory and the intellectual. Other revolutionary developments in the world of the arts during the 18^th century included the separation of the value-spheres of ethical and aesthetic, art's growing commodification linked with a developing bourgeois public sphere of taste, and the rise of Romantic ideals of genius and self-expression that helps to constitute a post-Romantic conception of art. I would like to argue that another development of the time was that artistic truth no longer had to be treated in an instrumentally didactic way.

This latter change is important because the account of artistic truth offered here essentially requires freedom of interpretation; that is, that the highest art is non-didactic. Didactic refers, for instance, to the way that a poem might be ‘instructive’ in the moralising sense. As Dr. Johnson wrote of Milton: ‘Poetry is the art of uniting pleasure with truth, by calling imagination to the help of reason. Epic poetry undertakes to teach the most important truths by the most pleasing precepts, and therefore relates some great event in the most affecting manner.’Footnote ¹⁵ Johnson did recognise that poetry is an imaginative presentation of truth, differing from direct moral instruction such as a church sermon. Poetry of a lower status could be regarded merely as pleasurable diversion.Footnote ¹⁶

The Johnsonian model is applicable to wall-paintings in medieval churches, intended to instruct the unlettered peasantry in Gospel stories; and to socialist realist art or the agitprop films of Michael Moore. A more sophisticated cognitivism underlies the post-Romantic conception of art, with its requirement of free interpretation by the audience, involving critical debate. According to this conception, high or classic art is neither didactic nor pleasurable diversion; its truth is not reducible to anything as crude as a ‘message’, and artworks are concerned, rather, to raise possibilities for consideration. Art is autonomous, and so is its audience in their response.

‘Didactic’ suggests the settled world-view of Augustan writers such as Samuel Johnson; ‘message’ implies ideology and persuasion, a battle of ideas characteristic of less consensual times. There is a tendency for proponents of didactic art to deny that art has intrinsic value, and to treat it merely as a vehicle for truth – though Dr. Johnson would not have done so. Oscar Wilde's remark that ‘No work of art ever puts forward a view of any kind’ can be regarded either as narrow aestheticism, or as a salutary rejection of crudely didactic art in favour of a broader aestheticism that regards art as valuable in itself.Footnote ¹⁷ Art for art's sake can be interpreted as compatible with art's aiming at truth – if it is regarded as saying that art is valuable in itself, and not merely instrumentally. This is a broad aestheticism, as opposed to the narrow aestheticism of art for art as traditionally conceived. Indeed art for art in the broad sense, and art's aiming at truth, go together; art is not merely the means for projecting a truth, and is valuable in itself. Instrumentally didactic art is not high art, because it is merely a vehicle for transmission of truth; art that does not want to say anything is not high art either, but entertainment or some other variety of lower art.

We must now explore the assumptions of those who reject artistic truth.

2. The rejection of artistic truth

The contemporary separation of art and truth rests on several developments. Later 19^th century aestheticism, and the modernism that arose from it, scorned the concept of artistic truth. On this view, an artwork creates its own world, and refers to nothing beyond itself – suggested by Nietzsche's declaration that ‘We have art in order not to die of the truth’.Footnote ¹⁸ But while modernism's tendency was to abolish artistic truth, not all modernists wished to do so – we have already noted remarks by Joyce and Picasso, and Adorno is a notable exception among art theorists.

However, with postmodernism's incoherent attempt to abolish truth, influenced by Nietzsche and developed by Derrida, artistic truth came under further attack.Footnote ¹⁹ More interestingly, the Marxist ideological objection regards artists as subject to the dominant ideology, and so unable to see the truth about their own society. It undermines truth in all spheres of human activity, and we return to it later.

20^th-century Anglo-American philosophy, while rejecting the postmodern onslaught on truth, tends to deny artistic truth, for one or more of the following reasons:

1. (1) Truth is essentially propositional. Zuidervaart describes a ‘tunnel vision imposed by a propositionally inflected theory of truth’, but it is an understandable visual impairment.Footnote ²⁰ Horwich for instance, in considering the entities to which truth may be attributed, cites utterances, statements and beliefs, and propositions; but not novels, paintings or love.Footnote ²¹ Goodman hesitates to attribute truth to art because its ‘carriers…are literal and linguistic’, and talks instead of the ‘appropriateness’ or ‘propriety’ of a work.Footnote ²² Finally, Beardsley holds that, in order to have a truth-value – his interpretation of the claim of artistic truth – an artwork must either contain propositions or suggest and confirm hypotheses about reality, and it does neither.Footnote ²³

2. (2) Art is essentially ‘aesthetic’, emotive or formal in its appeal. I.A. Richards, for instance, considered art to be a non-propositional language of emotions, a view very commonly applied to music, most notoriously by Deryck Cooke.

3. (3) Artistic truth implies artistic knowledge, yet the latter is preponderantly banal; literature is replete with ‘epigrammatic observations hardly distinguishable from folk sayings’.Footnote ²⁴

This last formulation rests on a key error diagnosed by Adorno – that art says what its words say. It is true that when appreciating autonomous art, we do not simply regard it as a source of useful knowledge, nor as a way of stimulating the audience to a course of action, nor to assure them of the soundness of their own beliefs and actions.Footnote ²⁵ But this is not sufficient reason to reject artistic cognitivism or the concept of artistic truth; nor is the naive argument (2). Argument (1), as we will see, provides the most fundamental questioning of the concept of artistic truth.

Although the assumption of artistic truth is against the spirit of the age, it residually informs modern responses to literature, for instance; and so Wittgenstein exaggerates when he writes that ‘People nowadays think, scientists are there to instruct them, poets, musicians etc. to entertain them’.Footnote ²⁶ Toni Morrison's novel The Bluest Eye, which describes an incestuous relationship within an African-American family, was criticised for undermining communal solidarity – a response that treats her novel as didactic art with the wrong message, rather than art that yields imaginative or experiential truth. Similarly, J.M. Coetzee's Disgrace, which concerns an attack by a black criminal gang on a white father and daughter, has been criticised as racist – presumably for expressing the racist view that most violent crime in South Africa is by blacks on whites. Given that the mere fact of the assault rather than its treatment seems to be what is objected to, the criticism is justified only if black on white violence were a recurrent theme in Coetzee's novels, which it is not.

These examples are instructive, since the criticisms of Morrison or Coetzee assume that art aims at truth didactically, and that it must broadcast the right message – which, it is argued, Coetzee does not do. (I am not saying that a racist novel fails to aim at truth, though that might be argued; just that the objection assumes a didactic model.) That such objections are commonplace, shows that the postmodern questioning of truth has not been fully absorbed – unsurprisingly, given its incoherence - and the very same critics who allege racism would most likely deny truth. This objection is opposed by embattled proponents of artistic freedom, a value integral to the post-Romantic conception of art, that includes the freedom of the artist to shock and outrage.

3. A ‘grand narrative’ historicist conception of artistic truth

The view of art that I wish to defend, and which contrasts with accounts that reject artistic truth, may be labelled cognitivist. Cognitivism about art recognises that it is more than pleasurable diversion, and that its meaning transcends didactic utility; high art aims not just to give pleasure but to develop understanding – indeed to develop understanding through pleasure. To reiterate, cognitivism is represented in Anglo-American aesthetics; but the particular version of cognitivism advocated here, one that finds place for artistic truth, seems not to be.

Cognitivism involving a concept of artistic truth is most readily associated with German Idealist aesthetics. Despite the still common attribution to Kant of an exclusively ‘taste’ aesthetic, he did recognise broader functions of art, and his formalism does not extend to art itself.Footnote ²⁷ But Hegel is the most thoroughgoing artistic cognitivist, arguing emphatically that art's primary role is the disclosure of truth. He held that art is a way of discovering ourselves and the world, not merely a way of beautifying what has already been discovered: ‘In art we have to do not with any agreeable or useful child's play, but with an unfolding of the truth.’Footnote ²⁸ For Hegel, the content of art is not abstract, but is ‘the sensuous appearance [Scheinen] of the Idea’, that appeals through the senses to the mind or spirit.Footnote ²⁹ Hegel's pervasive philosophical impulse is to elevate purely conceptual modes of expression above sensory ones, and since art embodies metaphysical truth through a sensory medium, he is deeply ambivalent towards it. But he allows that as well as being understood conceptually, truth must be experienced sensuously through art, as well as felt and loved through religion.Footnote ³⁰

Hegel's discussion suggests that the concept of high art can be defined in terms other than those social ones – viz., the art of the socially dominant classes – assumed by Marxist critics; high art is autonomous art that aims at, or discloses, truth. However, while orthodox Marxism offers an ideological critique of artistic truth, members of the Frankfurt School, notably Adorno, were as much indebted to Hegel's cognitivist affirmation as to the Marxist critique of art. Adorno links truth-content with the ‘language-character’ that he attributes to all artworks, not just those whose medium is language. Even when its medium is linguistic, what the artwork says is not what its words say, and so music and literature are not so distinct: ‘No art can be pinned down as to what it says, and yet it speaks’.Footnote ³¹ Progressive art embodies an essential critical function within bourgeois culture, and ‘truth-content [is] the task of critique’.Footnote ³² For Adorno, artists control only the material, not the content of their work. Austen may have intended her novels as guidance to young women concerning matrimony, but for Adorno, the truth-content of Pride and Prejudice is a social truth of which its author may be unaware.

For him, the truth-content of artworks is neither factual nor propositional, but perceptible and structural. He assesses the integration and fracture of its form and content, considering the artwork in its own terms:

The ‘discursive form of answer’ is the conceptual or linguistic paraphrase that audiences are always tempted to derive from an artwork, and which Adorno rejects because what the artwork says is not what its words – if it has any – say.

Despite its insights concerning ‘what art says’, Adornian historicism is a contestable basis on which to defend artistic truth.Footnote ³⁴ The more modest, pluralistic conception that I wish to present is consistent with the modernist understanding of art's increasing social autonomy from the 18^th century onwards. Unlike Adorno's, however, it does not privilege socially critical art.Footnote ³⁵

4. A more modest conception of artistic truth

A pluralistic conception of artistic truth is found in Gadamer's critique of Hegel – a critique that I think applies also to Adorno's negative, pessimistic development of Hegelian dialectics. Though Hegel recognised ‘the truth that lies in every artistic experience [and] is…at the same time mediated with historical consciousness’, Gadamer writes, he denied art's historical multiplicity, mistakenly holding that it progresses towards the one, true art:

For Gadamer, there is knowledge in art, and experience of art contains a claim to truth that is not inferior, but simply different, to that of science: to do justice to ‘the truth of aesthetic experience [is to] overcome the radical subjectivisation of the aesthetic that began with Kant’.Footnote ³⁷ Gadamer rejects Adorno's view that the highest art is essentially critical – for him, it can be affirmative also – and thus the grand narrative historicism of which that view is a residue.

Like Gadamer, Theodore Meyer Greene regards artistic truth as non-conceptual, though confusingly he also seems to hold that one can regard a work of art as a judgment or assertion – an interpretation of reality – by the artist. He denies that truth is expressed adequately only through a conceptual medium: ‘certain aspects of reality can be apprehended and expressed…in and through the artistic media, and…what is thus apprehended and expressed cannot be translated into a conceptual medium without vital loss’.Footnote ³⁸ He continues:

Although artists can express propositions whose truth they neither deny nor assert, he holds, ‘serious’ art is the expression of the artist's sincere convictions.

Greene contrasts the enduring nature of artistic truth with that of science:

For Greene, ‘Artistic intuitions enjoy a degree of autonomy unrivalled in the scientific enterprise’. The artist's language can be ‘as perfectly adapted’ as the scientist's – artistic truth is not vague or imprecise by the standards of scientific truth, but aims at a perfect matching of precision and subject-matter.Footnote ³⁹

This last claim seems correct. The poet's or novelist's observation of nature or human psychology on which their work is based, and their search for the right phrase or word, is as exacting as the scientist's precise recording of observations and testing of hypotheses. However, Greene still does not sufficiently separate art from empirical inquiry. He is right that ‘a great work of art can be true for all time in a way in which no scientific theory can’. But it is misleading to say that ‘a given subject-matter is not as subject to correction by other individual approaches’ – for it is not subject at all. A later artwork cannot correct the artistic truths expressed by an earlier one, because – according to a post-Romantic conception - the pursuit of truth in art is an individual and not collective endeavour; indeed it is an endeavour bounded by the work in question. Art does not progress in the same way as science. One can talk of the development of a tradition, and of an artist achieving more fully what they were intending in an earlier work. But while Conrad's Heart of Darkness is a fully-realised artistic vision that transforms the stock devices of his earlier exotic romances, it is not a ‘correction’ of those earlier efforts, which still stand independently.

However, it is unusual for an artist to refine their a work through response to the judgments of critics – the collaboration of artist-photographer Jeff Wall and critic Michael Fried is a rare example. When an artwork is released into the world, in the form of publication or exhibition, it acquires a life of its own and cannot normally be retrieved by its creator – though obsessive revisers such as Bruckner have tried to do so. While a scientific theory can be discredited – indeed that is its normal fate – and becomes of purely historical interest, artworks are not discredited unless revealed as fakes; tastes change, and popularity is recognised as ephemeral, but classics endure. The artist has greater creative autonomy than the scientist, though to deny it to the latter entirely, is to subscribe to the unacceptably negative characterisation of scientific culture found in Kant.Footnote ⁴⁰

Reference has been made to Hegel's elevation of ‘conceptual truth’ over other modes, and to the suggestion, shared by Gadamer and Greene, that artistic truth is not ‘conceptual’. Because ‘conceptual’ has so many meanings, one could refer instead to ‘linguistic’ or ‘propositional’, which are slightly less broad in scope – while recognising that these terms are still badly in need of refinement. “Scientific truth” is often contrasted with artistic truth, but truth of inquiry is preferable, since it embraces the humanities as well as the sciences, and this description may also serve in place of ‘conceptual’. Critical discourse is ‘conceptual’, and the role of criticism includes interpreting and assessing the truth that a work aims at. But it is an aid to or expression of our understanding, not a paraphrase of what an artwork says. The truth expressed by an artwork is not a scientific, historical or philosophical truth, but rather, an experiential or imaginative truth – art makes truth real to the imagination. Imagination is essential to empirical inquiry too, and art is not the only activity that can express such truths; truths can be conveyed imaginatively through role-playing, psychotherapy and other means.

5. The most persuasive cases: literature, film and drama

We noted earlier Iris Murdoch's persuasive statement of artistic cognitivism:

Murdoch holds that we learn from the characters of Shakespeare or Tolstoy or the paintings of Velásquez or Titian

These claims are illustrated by David Harrower's one-act play Blackbird cited earlier. To reiterate, the play forces us to reconsider our attitude towards paedophilia, but it does not, as didactic art would, prescribe a particular response. It poses the question whether a relationship between a man and an under-age girl could ever have value. In virtue of its form, the play achieves a persuasive effect on its audience, who are slowly initiated into the existence of a past relationship between the protagonists, and into its social unacceptability. There is a compelling dramatic rhythm in the emergence of this information, involving an intense revelation of the protagonists’ continuing feelings for each other.

The highest art, such as Blackbird, that aims non-didactically at truth, exists at the opposite end of the continuum to didactic art, whether religious or political. One might attempt to elucidate the distinction in terms of Wayne Booth's contrast in The Rhetoric of Fiction between showing and telling. Booth argues that authors should not be forbidden from telling, and that it is easy to distinguish those who profess to be showing but are in fact telling.Footnote ⁴³ However, showing can be as didactic as telling; there is a continuum in art from didactic to non-didactic, from telling via showing to presenting possibilities and raising questions.Footnote ⁴⁴ It is the last of these that Harrower's play exemplifies.

The aesthete, narrowly defined, is someone who when asked ‘Did this play make you think about attitudes to paedophilia?’, might reply ‘I hadn't thought about that, I was just marvelling at its compelling formal construction’. Insofar as such an exclusive concern is possible, it amounts to a failure of artistic understanding. (To reiterate, aestheticism or art for art's sake, in the broader sense, regards the aesthetic as an end in itself, without assuming, with Wildean aesthetes, that it is the most valuable end.) Conversely, someone who concerned themselves only with what the play said about paedophilia, treating these as truths of inquiry, would also have an impoverished artistic understanding. To do so would not be a response to the play as art; but to ignore such issues is artistically inadequate also. It could be argued that I have shown not that art aims at truth, but that it aims at critical engagement. However, critical engagement involves a reaction to truth, and asks questions such as ‘Is it true that the relationship between an older man and an under-age girl could have value?’

Harrower does not need to have the expertise of the psychiatrist, psychologist or criminologist in this area. He may in fact belong to any one of those professions, but writing a play that focusses on the issue does not require access to any body of disciplinary knowledge. The playwright's knowledge of human motivation is not a disciplinary knowledge like the psychiatrist's or psychologist's (or historian's or philosopher's), but more like an ordinary person's, intensified or magnified.

These areas of expertise concern what I termed truths of inquiry. The playwright presents and explores human knowledge of the phenomenon – in Blackbird's case, paedophilia – in an imaginative way. The play makes vivid some of our misapprehensions, in a highly-charged way through its formal presentation. I may emerge from the very intense experience that it affords, wanting to read what the experts have to say, in that way deepening my understanding of paedophilia and social attitudes towards it. The play would not be cited as evidence in a government report, but members of a government commission might be recommended to see it as background to their official responsibility.

It is not even clear that the author requires direct acquaintance with their subject-matter. There is a tradition in literature of writing on subjects about which one has no direct experience. The Lime Twig, by American writer John Hawkes, a novel about gangsterism on the periphery of English horse-racing and betting, shows no acquaintance with English life, but is higher art than the novels of Dick Francis, who knew a great deal about these subjects. In The Remains of The Day, Kazuo Ishiguro recreates English country-house life of the 1930s; Conrad seems to have created the world of Nostromo mostly from his reading about Latin America.Footnote ⁴⁵

However, writers are often advised to write from what they know, and novels written from experience can be especially effective illustrations of Murdoch's claim that ‘[in art] we are presented with a truthful image of the human condition in a form which can be steadily contemplated’. The memoir-novel, or what is sometimes called ‘auto-fiction’, draws most directly on the author's experience. Solzhenitsyn's One Day In The Life of Ivan Denisovich had an explosive impact on Soviet society when it appeared in 1964, during the Khrushchev thaw, because in contrast to Stalin's favoured Socialist Realist morality tales, its genuine realism and truth relayed corroborated facts about the Gulag: ‘The sufferings of its heroes were pointless…The Party did not triumph in the end, and communism did not emerge the victor…[Its publisher Tvardovsky said] that the story had “not a drop of falsehood in it”’. Readers who were or had been in the Gulag ‘were overjoyed to read something which actually reflected their own feelings and experience. People afraid to breathe a word of their experiences to their closest friends suddenly felt a sense of release’.Footnote ⁴⁶

One reader wrote to Solzhenitsyn: ‘I wept as I read – they were all familiar characters, as if from my own brigade!’ The prisoners or ‘zeks’ in Dubravalag held a group reading, listening ‘without breathing’:

Many Gulag survivors, interviewed in the 1990s, insisted that they had witnessed scenes in books by Solzhenitsyn, Ginzburg or Shalamov, or recognised guards and NKVD interrogators represented there – though records show that this could not have happened:

Figes' final comment is important; art can supply a conceptual framework for their own experience that the individual lacks, something that helps them make sense of it in the way that Murdoch suggests.

There can be misidentification too, however. ‘Daddy’, the most notorious of Sylvia Plath's Ariel lyrics, is a shocking Grand Guignol outburst that portrays the writer's seemingly mild-mannered if foolishly obstinate father as a surreal monster. Anne Stevenson comments: ‘This distorting wilfulness…became for her poetic truth…The voice [in this poem] is finally that of a revengeful, bitterly hurt child storming against a beloved parent’, cursing him for dying.Footnote ⁴⁹ However, the result could be regarded as overwrought or out of control, self-absorbed, striking but not psychologically deep, with little reflection of the universal – even though some teenage readers may identify with it.

Literature's concern with truth therefore exists on a continuum, from the didacticism of Johnson's ‘The Vanity of Human Wishes’, through the residual moralising of Larkin's ‘An Arundel Tomb’ with its hint at an ‘almost-instinct almost true’, to Sylvia Plath's Ariel lyrics that seem to have nothing to do with truth. Revelation of the ‘child’ that Plath remained is not artistic truth in the relevant sense. As we saw, artworks do reveal truth in this way, and it is the task of what Danto called ‘deep interpretation’ to consider how they do. But this is not the ‘aiming at truth’ that this article is concerned with, since anything at all – commodified pop music, pornography, home movies – can be a source of truth in the sense that by reflecting on it, we arrive at truths otherwise inaccessible.

One could distinguish art concerned with a common reality – Johnson, Larkin, Levi, Conrad, Raymond Carver – from art that creates an alternative reality that beguiles us: Milton's ‘Lycidas’, Plath, Peake, Hesse, Woolf, Brian Marley. Larkin and Dickens created works that mirrored our social reality and understanding of it, in a highly personalised way that filters reality through their consciousness and heightens it by their craft. Theirs is not, strictly, an alternative to our social reality; although they present aspects of it that we may deny, or had hitherto failed to understand or recognise, nonetheless their works are strongly rooted in something we all share. Talk of a ‘common reality’ hopefully allows one to sidestep some of the pitfalls of the debate over literary realism.

In contrast, an alternative reality is presented in the Pastoral tradition in poetry that inspired Milton's ‘Lycidas’. This lyrical lament delineates the poet's grief, while portraying Milton and his late friend ‘untruthfully’ as shepherds. Dr. Johnson condemned its artificiality, complaining that ‘in this poem there is no nature, for there is no truth; there is no art, for there is nothing new’, its pastoral images ‘long ago exhausted’.Footnote ⁵⁰ Later audiences took a different view, holding that the poet's grief is persuasively expressed through pastoral convention. The reality in high art cannot be entirely alternative, however, as in science-fiction – though occasionally, as in Tarkovsky's Stalker, or Vonnegut's Slaughterhouse Five, science-fiction is high art.

The analogous issue in painting, to which artform we now turn, is illustrated by Watteau's more evanescent Arcadia. As John Golding writes, ‘Watteau lived in a world of his own, a dreamworld, one that could exist only in his eye and mind but one that he invites his audience to enjoy, to embrace. Et in arcadia ego’.Footnote ⁵¹ Watteau's rococo fêtes galantes express a human reality, otherwise we would have little interest in them; explorers of common and alternative realities can both aim at truth, whilst having different conceptions of its boundaries. Even so, Golding's suggestion that he was the greatest French artist of his century suggests that this was a silver age for French painting. We pursue the implications of this opposition between realism and idealism by considering how the visual arts aim at truth.

7. Truth in the visual arts, and the ideological objection to artistic truth

Religious art seems to be a paradigm of didactic art. However, in the Middle Ages, church paintings were not just art for religion's as opposed to art's sake. They were religious as much as artistic artefacts: ‘all images were…understood to be more than representations or commemorations, bringing the holder into spiritual contact and the semi-presence of the person depicted’.Footnote ⁵² Perhaps something like this is still believed in the modern era – for instance by Mondrian and metaphysical artists – and not just in more traditionalist cultures. Images of saints and the Virgin in particular became more common in medieval churches after the doctrine of Purgatory was confirmed in 1274; masses for the dead required saintly intercession.

The enduring sense that images bring us into contact with their subject meant that the didactic function of religious art always had an explicit personal dimension. This dimension became amplified, as Clifford Geertz explains:

Geertz argues that the relation between religious ideas and painted images was not simply expositive: ‘…they were not Sunday school illustrations…the religious painter was concerned with inviting his public to concern themselves with first things and last, not with providing them with a recipe or a surrogate for such concern…the relations of his painting to the wider culture was interactive’.Footnote ⁵³ Such a model construes artistic truth as inviting contemplation or understanding – as Iris Murdoch's discussion suggests – rather than as the presentation of a choice, or forcing the audience to take a view, as modernists often assume.

Religious art could be more primitively didactic, however, excluding a personal response. The high art of the fifteenth-century Italian Renaissance, where the viewer is already assumed to have received religious teaching or training, contrasts with the often rather crude representations that were once ubiquitous on English medieval church walls – didactic in a way that called for little individual contribution from the viewer. Minimally contributory didacticism is a minor genre in painting, illustrated by Jan Steen's ‘The Effects of Intemperance’ (1663–5, National Gallery, London) – portraying the ill-effects of alcoholic inebriation – or Holman Hunt's irretrievably moralistic ‘The Awakening Conscience’ (1853, Tate Britain, London).

Didactic aiming at truth contrasts with the representational truthfulness that art is often assumed to aim at. This is the sense in which paintings and drawings may be true to life; the subject may be more or less accurately represented. Although we admire ‘realistic’ pictures, we are now inclined to regard them as art with a small ‘a’, products of skill and craft, employing hard-earned techniques and materials. A true portrayal, in this sense, is no guarantee of artistic value and in some cases – Canaletto for instance – may substitute for insight and imagination. Conversely, portraits and landscapes of great merit may be anything but a true likeness. Among these, clearly, would be the portraits of Francis Bacon, yet he claimed that his paintings ‘tell us something true about the world we live in…art is recording… reporting’.Footnote ⁵⁴ Some artists even found a contradiction between didactic and representational truth, however. Burne-Jones commented that his paintings were ‘so different to landscape paintings. I don't want to copy objects; I want to tell people something’.Footnote ⁵⁵ With modernism's rejection of artistic truth, such didacticism appeared hopelessly conservative.

Although Geertz's richer notion of didactic art allows for a contribution from the audience, who are not merely passive recipients of the truths expressed, the post-Romantic conception of art has the stronger requirement of free interpretation by the audience. That development seems in tension with the patronage system's restriction of artistic autonomy. For instance, Titian in his portrait Pope Paul III and his Grandsons is not merely ‘reporting’ or ‘recording’, but presenting an artistic vision of the Pope's character. The audience are free, within limits, to interpret the portrait – to judge what kind of person or character the Pope is. But in an era of patronage, freedom of both artist and audience are constrained; portraitists could suffer the patron's wrath if they made the subject too fallible-looking.

Modernists such as Attali suggest that, with the decline of the patronage model, the artist becomes free at the same time as their work goes on sale in the capitalist market-place. Proponents of the ideological objection to artistic truth, noted earlier, maintain nonetheless that art consistently fails to locate truth. They hold that John Constable's landscapes, whose subjects are the estates of his wealthy friends, unaccountably rather devoid of peasants, are to that extent, ‘false’.Footnote ⁵⁶ An analogous and familiar criticism is that Jane Austen's novels neglect the effects of the Industrial Revolution; she herself referred to the ‘two inches of ivory’ – the very circumscribed social subject-matter – on which she worked.

John Barrell's work on landscape painting and poetry, from the period of transition between patronage and market systems, illustrates the ideological objection. Barrell argues that as the aristocracy became more committed to the economic exploitation of its estates, aristocratic taste developed an interest in more workaday – if not totally realistic – images of rural life.Footnote ⁵⁷ However, he recognises that there were constraints on how the labouring, vagrant and mendicant poor could be portrayed, to be acceptable décor for polite drawing rooms; it was only ‘discreet hints of actuality provided by tattered clothes, heavy boots and agricultural implements’ that marked the English tradition of Gainsborough and Constable from the more artificial Italianate tradition.Footnote ⁵⁸ As a periodical of the time put it:

Thus when Gainsborough ‘brought together in his landscapes of the 1750s the tradition of French rococo pastoral painting, and the more sternly georgic tradition of the Netherlands painters, this may not have been simply a happy eclecticism, but a combination that enabled him to compose [a reassuring] image of Happy Britannia’, a blend of French play and Netherlandish work in rural life.Footnote ⁶⁰

The implication, based not on direct evidence but rather on a brilliant series of critical intepretations, is that Gainsborough constructs an ideologically-driven image rather than pursuing truth. Marxist critics infer that so-called high art often aims not at truth, but at satisfying the self-image and taste of its patrons or consumers. On this view, art is essentially ideological, having the function of concealing rather than expressing truth.

Such interpretations express what Ricoeur called ‘the hermeneutics of suspicion’, that interprets all social phenomena by invoking unconscious motivation whether Freudian or Marxist – a ‘suspicion’ that Ricoeur himself endorses.Footnote ⁶¹ Barrell's claim is not that Gainsborough composed his ‘reassuring image of Happy Britannia’ fully intentionally, but rather that he did so in that not fully intentional but still meaningful way in which one is said to act on Freudian unconscious desires. A humanistic as opposed to deterministic Marxism – Lukács rather than Althusser, for instance – would allow that in some sense Gainsborough owned deep-seated commitments to the established order – commitments that could perhaps be avowed after the event – and that this prevented him from aiming at truth in his art.Footnote ⁶²

However, such interpretations are hard to prove, and tend to be as speculative as those of Freudian dream therapy. Maybe Gainsborough just saw rural life as a blend of work and play. His art is affirmative, but still high art, since we judge that although he may make concessions to patrons, he is not aiming simply to be acceptable to them. (One could also argue that since Gainsborough supported progressive bourgeois forces, in ideological terms his art may not be ‘false’.) Art that lacked such integrity clearly could not aim at truth, but whether it did lack it would be a matter of critical debate.Footnote ⁶³

In Constable's time, the ‘higher styles of landscape’ remained those more artificial or idealised ones by Claude or Poussin.Footnote ⁶⁴ That Italianate tradition, like the Pastoral tradition of Virgil in poetry, did not pretend to realism; the kind of truth it aimed at concerned not the everyday world, but an Arcadian realm. Constable or Austen did focus on that everyday world, however, and the most that one could infer from the ideological objection is that they present a partial or parochial truth. They did not aspire to a complete social picture of their society, as does a social scientist, or a realist novelist like Zola. Their concern is psychological depth, not social comprehensiveness, and in that respect their art does aim at truth. High art does not have to aim at ‘total’ truth.

8. Truth in architecture and music

The most problematic artforms for artistic truth are music and architecture. We noted, and qualified, Greene's comment that what is apprehended and expressed in art ‘cannot be translated into a conceptual medium without vital loss’. This may be felt to be especially true for music and architecture, commonly regarded as non-representational, essentially abstract arts. On this received view, literature and drama are representational and conceptual (or linguistic); painting and sculpture are representational but non-conceptual (non-linguistic); and music and architecture are neither. Some artforms have the capacity to be – or are normally – representational even if they are not always so. Hence abstract painting is non-representational, as perhaps is avantgarde modernist literature such as Joyce's Finnegan's Wake.

While artistic truth in imaginative literature might be accepted, its presence in music seems quite obscure. But Adorno suggests how it might be understood. To reiterate, he holds that all artworks, and not just those whose medium is language, possess a ‘language-character’, which he links with truth-content. By this he means that elements not meaningful in themselves are organised into a meaningful structure:

For Adorno, the truth-content of a Mahler symphony is not captured in literalist programmatic interpretations; nor are Wagner's music-dramas decoded by a process of motif-identification. In contrast, he gives the example of musical affirmation, ‘the judicious, even judging affirmation of something that is, however, not expressly stated’, such as the first movement recapitulation in Beethoven's 9th Symphony. As we saw, Adorno believes that on the strength of its similarity to language, music constantly poses a riddle, which it never answers – but then, he insists, all art does so.

Before the 18^th century, a literary model of music was dominant, and non-vocal music was neglected by theorists. During the 19^th century, the non-representational and non-conceptual nature of music became regarded as a sign of its superiority, as it ascended from the lowest to the highest of the arts. (This was not so for architecture). The view that music does not mean anything because it is not about anything – that it is non-conceptual and non-representational and does not even point to any truth – became, and remains, commonplace.

The more humanistic view that connects music and life has always had representatives, however. When Mahler visited Sibelius in 1907, they debated the nature of the symphony:

In seeking extra-musical reference, Mahler belongs with Ives and a few other modernists, in their ambitious attempt to make music reach beyond its apparent muteness, and ‘speak’. Sibelius, in contrast, belongs to the tradition of absolute music according to which the form of a work is the working-out of its musical idea, a tradition which achieved its apotheosis in the work of the first and second Viennese Schools, and Brahms, with Schoenberg its most articulate theorist. Yet even that tradition can recognise that music aims at truth, construed as the Adornian idea of ‘truth to materials’ – while like all performing arts, truth in the sense of authenticity to the work is essential.

Architecture is less abstract than music in its expression of function, and more abstract in not being a performing art. The idea of truth to materials is expressed here most famously by Ruskin, who writes that

He argues that an honest architecture avoids deception, that is, it avoids suggesting ‘a mode of structure or support, other than the true one; [such as] pendants of late Gothic roofs’; ‘painting of surfaces to represent some other material than that of which they actually consist (as in the marbling of wood), or the deceptive representation of sculptured ornament upon them’; or ‘use of cast or machine-made ornaments’.Footnote ⁶⁸ This vocabulary of truth pervaded Victorian attitudes to architecture; for instance, Richard Redgrave wrote that Pugin's works ‘deserve commendation for their illustration of truth, and as showing what one man, by earnest and well-directed attention, can achieve in the reformation of taste, and in the training …of other minds to assist in his truthful labours’.Footnote ⁶⁹

While Ruskin understands truth to materials as ‘truth to physical stuff’, Adorno regards it as truth to material that is pre-formed and shaped by history. In architecture, ‘material’ in this historical sense would include the accumulated vocabulary of Georgian door, sash window, pitched roof, or in grander buildings, Doric and Ionic columns, cornices, corbels and turrets. ‘Untruth to materials’ here would include the steeple that Sir William Chambers provided above the portico of St Martin's-in-the-Fields, ignoring the fact that steeples were never found on Greek temples.

Even in those arts that are least ‘language-like’, therefore, a case can still be made that high art aims at truth, in the form appropriate to the artistic medium. More work needs to be done, to convince those expecting something more like propositional truth, that this is indeed artistic truth in an interesting sense – a task that must remain a subject for another occasion.

9. The audience's freedom of interpretation

In this article I have been attempting to develop a more liberal concept of artistic truth than that found in didactic art – one that allows freedom of interpretation by the audience, bestowing on them an autonomy equivalent to that of the artist. This account does not naively ignore the social and historical situation of art, and is consistent with Adorno's stress on the inexhaustibility of interpretation of high art – how such art seems continually to invite new interpretation, as each generation understands it in light of its own concerns.

The idea of free interpretation has not been much-explored in the philosophical literature, and an account of it could begin by looking at those genres that undermine or deny it, such as the sentimental or sensationalist. ‘Guernica’ and Wozzeck are intended to shock, but are not merely sensationalist; violent crime fiction such as the novels of Val McDermid, or Hollywood schlock horror, in contrast, exploit a visceral reaction that gives the audience little freedom of response. An instinctive reaction to blood and gore overwhelms critical freedom; as, in a milder way, does the sentimentality indulged by romantic comedy. There is a continuum of cases. Frank Capra's popular classic ‘It's a Wonderful Life’ elevates a homely didacticism into a paean to the American dream, and is neither kitsch nor the highest cinematic art. Stephen Soderbergh's recent film Contagion, in contrast, is slick entertainment that compels attention by trading on hysteria about the danger of an international pandemic; unlike classic art, its impact dissipates with subsequent reflection.

There is political high art that does leave freedom for interpretation – examples are Joseph Conrad's The Secret Agent and Heinrich Böll's The Lost Honour of Katharina Blum. In Böll's novel, unlike the film based on it by Schlöndorff and von Trotta – still a very impressive interpretation – the message is not driven home, yet is still explicit.Footnote ⁷⁰ It is clear how the author regards, and wants us to regard, the so-called free press, but he shows rather than tells us. Political art like Böll's does not force me to believe that the press is evil – a forthright expression of a view need not be constraining. Didacticism need not imply propaganda art. In the case of Conrad, interpretational freedom – or the novel's range of concern – may lead us to say that it is not just political art. It is an artistic decision or capacity, and not a product of confusion, to make a work rich enough to be freely interpretable – although the subject-matter that the artist is tackling may itself be complex and confusing.Footnote ⁷¹

Modernist art developed increasingly sophisticated non-didactic strategies for making audiences confront issues. Implicit ambiguities or contradictions force the audience to form a view on the subject, or at least to consider possibilities. For instance, Brecht's plays ask how one can confront a corrupt system in which one is implicated. Even in works that are not music-theatre like Threepenny Opera, songs serve to undercut the spoken word. A literal-minded political interpretation – whether by the director or audience – ignores how the playwright presents and simultaneously undermines a view.

Art does many things, apart from aiming at truth. It can fail to achieve truth in many ways; by being false, or by being nonsensical, a category that can include the phenomenon described by Frankfurt as ‘bullshit’, and embraced by Lady Gaga, as noted earlier.Footnote ⁷² While high art aims at truth, and lesser art aims at entertainment – the latter, though easier to achieve than truth, is not guaranteed – perhaps art can aim at bullshit too. Certainly some artists – Salvador Dalí, Peter Greenaway, Michael Nyman, Lars von Trier, Damien Hirst – aim to impress without caring whether they say anything true, and that is one kind of bullshit.

What I am arguing may seem quite ambitious, indeed over-ambitious. It challenges the deflationary conception which regards artistic truth as referring to a varied collection of virtues partly illuminated by a contrast with narrative or representational truth. On that view, ‘artistic truth’ is like ‘moral victory’, a concept parasitic on that of ordinary types of victory; one could not explain it to someone who lacked a grasp of ordinary winning and losing, yet it differs from standard types of winning in that the moral victor loses. This is a picture that I am attempting to undermine. One should question the common assumption that there is a ‘standard’ type of truth, propositional truth.

This article leaves many ends untied, and only gestures at responses to intractable problems such as how music and architecture aim at the truth. Much further work is needed to explain the connection between truth as intended by the artist, and truth as interpreted by the audience; and I have barely scratched the surface of the ideological objection, which may turn out to be several objections. The concept of high art has been left rather open, as has the question of the extent to which its precursors aimed at truth. But I hope at least to have shown that the concept of artistic truth is a richer one than is often supposed, thereby suggesting that the very practice of artistic creation is essentially truth-directed.

Many thanks for comments from: Emma Bennett, Andy Byford, Lucille Cairns, James Clarke, Freya Carr, Arlene Keizer, David Lloyd, David Macarthur, Brian Marley, Max Paddison, Richard Read, Alastair Rennie, Mark Rowe, Barry Smith, Roger Squires, James Steintrager, Rachael Wiseman; and to audiences at the Taste Workshop at the Institute of Philosophy, University of London; the RIP lecture at UCL; and seminars at University of California Irvine, Northumbria University, and Durham University MLAC group.