1 What is art — and why is the question so hard?

In 1917, Marcel Duchamp submitted a porcelain urinal — titled Fountain and signed “R. Mutt” — to the Society of Independent Artists exhibition in New York.¹ The exhibition had announced it would accept all submissions. The committee rejected it anyway. The rejection confirmed everything Duchamp wanted to prove.

Fountain is one of the most important objects in the history of modern art, and it satisfies almost none of the traditional criteria for being art at all.

1.1 What Art Is Not

Three traditional candidates for the definition of art come under pressure from cases like Fountain. Each has serious contemporary defenders; what follows are the standard objections, not refutations.

Skilled technique. On the craft view (defended in different forms by Collingwood and, more recently, by Larry Shiner’s The Invention of Art, 2001), art requires mastery of a medium — the trained ability to produce something the unskilled could not.⁶ The Duchamp objection: a urinal, purchased from a plumbing supplier, involves no technique on the artist’s part. The craft theorist’s escape — that Fountain’s “technique” was the conceptual gesture itself, the choice, framing, and signature — works only by treating the gesture as parasitic on the tradition of craft it could not exist without. The harder objection is symmetric: much technically brilliant work (a perfectly executed greeting card illustration) is not treated as art, which suggests technical skill is at most necessary, not sufficient.

Beauty. On the aesthetic view (defended in sophisticated form by Roger Scruton’s Beauty: A Very Short Introduction, 2009),⁷ art aims at the production or experience of beauty — though “beauty” must be understood broadly enough to include the sublime, the tragic, and the formally compelling. The objection: much of the most highly regarded art in the twentieth century is deliberately ugly, disturbing, or abject — from Bacon’s screaming figures to Hirst’s rotting animals to Kara Walker’s brutal silhouettes about slavery.⁸ Scruton’s move is to broaden “aesthetic” until the abject counts — calibrated against an implicit standard of beauty it deliberately violates. Whether stretching “beauty” this far empties the term of content remains the unresolved question.

Expression of the artist’s inner states. On the expressivist view (Tolstoy in What is Art?, 1897; refined by R. G. Collingwood, The Principles of Art, 1938),⁹ art communicates feeling from artist to audience. The objection: Duchamp was communicating an argument, not a feeling, and much art is made collaboratively, by committee, or by algorithmic processes that lack “inner states” in any obvious sense. Loosen “expression” to mean the articulation of a thought or stance the work makes available — the standard expressivist move — and the objection collapses into a quibble over vocabulary. Whether that rescues the theory or merely relabels its problem is where the dispute now sits.

None of these accounts is dead. Each captures something a fully adequate definition would have to preserve. What Fountain shows is that no one of them, taken alone, will do — not that any of them is empty.

1.2 Dickie’s Institutional Theory

George Dickie, in Art and the Aesthetic (1974), proposed a different approach. A work of art, on his institutional theory, is any artefact that has been conferred the status of “candidate for appreciation” by some person or persons acting on behalf of the artworld — the loosely defined community of critics, curators, galleries, publications, and institutions that constitute the social world of art.¹⁰

This may sound like circular reasoning: art is what the artworld calls art. But it captures something important. The boundary between art and non-art is not natural; it is social. Duchamp’s Fountain became art not despite the scandal but partly because of it — the scandal was conducted by and within artworld institutions, and in being discussed, rejected, argued about, and eventually displayed in major museums, the urinal was absorbed into the artworld it had challenged.

1.3 Wittgenstein’s Family Resemblance

Ludwig Wittgenstein, in Philosophical Investigations (1953), argued that many concepts — “game” is his example — cannot be given necessary and sufficient conditions for their application. Instead, there is a network of overlapping similarities: some games involve competition (but not all), some involve skill (but not all), some involve luck (but not all). The members of the category resemble one another in the way family members do — without any single feature shared by all.¹²

“Art” may be like this. Painting, sculpture, music, poetry, architecture, dance, film — what do they share? On the family-resemblance reading, perhaps nothing beyond a loose set of practices, traditions, and institutions that overlap and cross-reference each other. Whether this is a failure to define art or the correct description of a concept that lacks an essence is contested: essentialists about art (Monroe Beardsley’s aesthetic-experience theory; Arthur Danto’s later “embodied meaning” account) argue that the family-resemblance move gives up too quickly, and that a real essence is available at the right level of abstraction.

The family-resemblance move is not the end of the conversation, however. Berys Gaut (“Art” as a Cluster Concept, 2000), Stephen Davies (Definitions of Art, 1991), and Robert Stecker (Artworks, 1997) have argued that “art” can be given a cluster account — a finite list of conditions of which any sufficient subset suffices for arthood — that is more disciplined than open-ended family resemblance.¹³ Dickie’s institutional theory, on this reading, is one such cluster among others.

The question of definition matters because it determines who gets included. If art requires a certain kind of technical training, you exclude outsider art, folk art, and children’s drawings. If it requires institutional endorsement, you exclude everything produced outside the artworld’s purview. Every proposed definition draws a line, and every line excludes something.

1.4 Money, Markets, Museums

If Dickie’s account is right that the artworld confers art-status, the next question is what gives the artworld its authority — and the answer the institutional theory is then committed to is partly economic. (Critics of the institutional theory dispute that this answer is forced, and dispute the institutional theory itself; the arguments below presuppose Dickie’s account is at least roughly correct rather than definitively so.) Cynthia Freeland, in Art Theory: A Very Short Introduction (2001), puts the question plainly:

“Museums preserve, collect, and educate the public and convey standards about art’s value and quality — but whose standards, and how?” — Cynthia Freeland, Art Theory: A Very Short Introduction, Ch. 4¹⁴

The price mechanism is not external decoration on the institutional theory; it is partly constitutive. When Van Gogh’s Irises sold at Sotheby’s in 1987 for $53.9 million, the sale did not merely report the painting’s value — it produced and ratified it.¹⁵ Provenance, authentication, museum acquisition, and auction price are the mechanisms by which an object is consolidated as canonical. Hans Haacke’s exhibitions of corporate sponsorship are a useful test case: when the artist who attacks the market is championed in Art in America and sold at Christie’s, the artworld has demonstrated its capacity to absorb its own critics.¹⁶ The Elgin Marbles, the imperial-era version of the same mechanism, show that museums are not neutral preservers but normative actors who decide whose past counts as universal heritage.

1.5 The Elgin Marbles and the Universal Museum

Between 1801 and 1812, Thomas Bruce, the 7th Earl of Elgin and British Ambassador to the Ottoman Empire, removed approximately half of the surviving sculptures from the Parthenon frieze in Athens and shipped them to Britain. They were purchased by the British government in 1816 and have been held in the British Museum ever since. The Greek government has formally sought their return since 1983.²⁰ The British Museum’s argument for retention has shifted over the decades — from claims about superior conservation facilities, to arguments about the Museum’s role as a “universal museum” serving a global public, to legal arguments about the impossibility of returning objects acquired under legally valid agreements. Each argument is simultaneously an argument about art, about institutional authority, and about whose narrative of the past is being served by keeping the objects where they are. Theodor Adorno’s critique of the culture industry focused on the commodification of art within capitalist economies,²¹ but the Parthenon case shows that the political economy of cultural objects operates through imperial and post-imperial structures as much as through markets. The Kunsthaus Zürich’s Bührle Wing (info box above) extends the same institutional argument to a different kind of asymmetric acquisition — Jewish collector flight from Nazi Germany rather than Ottoman-era imperial removal — and the institutional response has been markedly similar: provenance review, partial restitution, defence of the institution’s role as custodian for a “global public.”

1.6 Questions to Argue About

• Dickie’s institutional theory implies that the artworld could, in principle, declare anything to be art. Does this make art entirely arbitrary? Or does the artworld operate with constraints that aren’t captured by the theory?
• If “art” has only family resemblance criteria and no essential definition, what follows? Can we still make meaningful judgements about whether something is good art?
• Duchamp’s Fountain challenged the artworld and was absorbed by it. Has the artworld now become immune to challenge — since any challenge will simply be incorporated? Or is genuine disruption still possible?
• Does the question “what is art?” matter outside philosophy? Does the answer affect how you experience a painting or a piece of music?

2 What kind of knowledge does art give us?

The arts are not usually thought of as a route to knowledge. We know things through observation, experiment, testimony, reasoning. Art gives us pleasure, emotion, aesthetic experience. But this separation is too quick, and Aristotle knew it.

2.1 Aristotle on Poetry and History

In the Poetics (c. 335 BCE), Aristotle draws the sharpest distinction:

“Poetry is something more philosophic and of graver import than history, since its statements are of the nature rather of universals, whereas those of history are singulars.” — Aristotle, Poetics, Chapter 9²⁶

History, on Aristotle’s picture, records what Alcibiades did. Poetry imagines what a man like Alcibiades would do — what such a person must do, given their character, in such circumstances. History gives you the particular; poetry gives you the necessary structure underlying the particular. This is why tragedy, for Aristotle, is a more profound vehicle for knowledge than historical chronicle: it distils the essential from the contingent.

Aristotle’s contrast is sharper than modern historiography would allow. Historians from Thucydides onward have claimed access to universals — patterns of empire, structures of revolution, regularities of motive — and post-Annales historiography (Braudel, Bloch) treats long-run structural causation as the central historical object. The defensible version of Aristotle’s claim is therefore not “history deals only with particulars and poetry alone reaches universals” but “poetry can articulate counterfactual and modal possibilities — what would happen to such a person — that history, bound to what did happen, cannot construct.” The novel, which did not exist when Aristotle wrote, is the cleanest case of this counterfactual licence: Anna Karenina is not a document about a particular woman in nineteenth-century Russia. It is an investigation of what it means to be a certain kind of person — impulsive, passionate, in conflict between social convention and inner authenticity — placed in circumstances that force those conflicts to their extreme. Aristotle’s defender says you learn something about human psychology and social structure from Anna Karenina that you could not learn from any sociology or history of the same period; the sceptic replies that what you learn is what Tolstoy thought about such a person, which is a different and weaker thing. Section “Does Art Give Knowledge or Only Experience?” returns to this dispute.

2.2 Berger on Seeing

John Berger’s Ways of Seeing (1972) begins with a sentence that sounds simple and isn’t:

“Seeing comes before words. The child looks and recognises before it can speak.” — John Berger, Ways of Seeing, Chapter 1²⁷

Berger’s argument is that we bring our whole history — our assumptions, our social position, our learned conventions — to every act of seeing. A painting in a museum is not seen neutrally; it is seen through the lens of art history, cultural prestige, and the social context in which art institutions exist. The painting of a naked woman by a European master looks different if you ask: who is looking, and who is being looked at?

One claim Berger and Mulvey share is that art can deliver knowledge of how we see — the structures of assumption, power, and convention that normally operate below conscious awareness. Art can make these visible by estrangement: presenting familiar things in unfamiliar ways, or unfamiliar things in ways that demand recognition. Whether this counts as a distinctive kind of knowledge that other disciplines cannot deliver, or whether the same insights are equally available through sociology, history, and the criticism of ideology, is itself disputed.

Laura Mulvey’s essay “Visual Pleasure and Narrative Cinema” (Screen, 1975) made for film the argument Berger had made for the European oil-painting nude. In a world ordered by sexual imbalance, she wrote, “pleasure in looking has been split between active/male and passive/female. The determining male gaze projects its fantasy onto the female figure, which is styled accordingly.”²⁸ Berger and Mulvey converge on a roughly shared claim: the apparently natural act of seeing is structured in advance by who is permitted to look and who is positioned to be looked at. Critics (Camille Paglia, Linda Williams from a different angle, the cognitivist film theorists Currie and Carroll) reply that the framework over-generalises from a specific Western art-historical and Hollywood corpus and that the gaze the theory describes is one available organisation of looking rather than its underlying structure.

Linda Nochlin’s 1971 essay “Why Have There Been No Great Women Artists?” pushes the same argument one step back from looking to making: the fault, she writes, “lies not in our stars, our hormones, our menstrual cycles or our empty internal spaces, but in our institutions and our education” — the academies, patronage systems, life-drawing classes and “mythologies of the divine creator” that decided in advance who could become an artist at all.³¹

2.3 Knowing How vs. Knowing About

There is a kind of knowledge that music gives that is structurally different from propositional knowledge. You can read everything ever written about Beethoven’s late string quartets and still not know what they are like until you hear them. The knowing is not inferential; it is experiential and immediate.

Gilbert Ryle’s distinction between knowing-how and knowing-that³² is relevant here, but art knowledge isn’t exactly knowing how either. It’s something closer to acquaintance — Russell’s “knowledge by acquaintance”³³ — a direct, non-inferential encounter with a particular that cannot be fully transmitted through description.

This is the basis for what New Critics called the “heresy of paraphrase” (Cleanth Brooks, The Well Wrought Urn, 1947): the specific words of a poem, in that order, with those sounds and rhythms, embody a way of understanding something that summary replaces with a different and weaker thing.³⁴ The meaning, on this view, is not separable from the form; the knowledge is not separable from its vehicle. The view has critics: Peter Kivy and Jerrold Levinson argue that the inseparability claim, taken strictly, makes criticism itself impossible — since critics routinely paraphrase what poems mean and disagree intelligibly about it. The defensible weaker claim: paraphrase loses something specific to the form, but not everything about a poem can be said only in other words.

2.4 Glissant’s Right to Opacity

A different theoretical tradition pushes the inseparability claim much further than Brooks did, and gives it a political weight the New Critics did not have. The Martiniquais philosopher and poet Édouard Glissant, in Poétique de la Relation (1990; trans. Wing as Poetics of Relation, 1997), argues that the demand for a poem, an artwork, or a person to be transparent — fully translatable into the language and categories of the dominant cultural framework — is itself a colonial demand. Against it, Glissant proposes the droit à l’opacité — the right to opacity — the right of a work or a culture to be present and acknowledged without being made fully legible to the framework that demands legibility.³⁵ On Glissant’s account, the New Critical claim that paraphrase loses something specific to the form is conceded too cheaply if it is then read as a claim about poetic technique alone: the same observation, made about the Caribbean creole languages or about the witness of an enslaved person, becomes a claim about whose ways of knowing are admitted as ways of knowing in the first place.

The American theorist Saidiya Hartman, in Scenes of Subjection (1997) and Lose Your Mother (2007), develops the corresponding methodological move within the historical archive: in cases where the available record was written by enslavers about the enslaved (the slave-ship ledger, the auction announcement, the runaway notice), the historian who attempts to “represent” the lives behind those records will reproduce the archive’s silencing in a more sympathetic vocabulary. Hartman’s critical fabulation — a self-aware, signposted use of imaginative reconstruction precisely because the archive does not contain what the historian needs — is not an aestheticisation of suffering but the only methodological honesty available to a discipline whose evidence is itself a record of complicity.³⁶ Whether critical fabulation is best understood as a method of art (closer to the novel) or of history (subject to the discipline’s evidential constraints), or whether it crosses the boundary between the two in a way that demands rethinking the boundary, is itself the open methodological question.

2.5 Questions to Argue About

• Aristotle says poetry gives knowledge of universals, history of particulars. But isn’t history also full of patterns and universals? What exactly is the difference?
• Can a novel give you genuine knowledge about human psychology? Or does it give you a very vivid and emotionally compelling fiction that might or might not correspond to how people actually are?
• Is there a kind of knowledge that can only be communicated through art — something that cannot be expressed in any other form? Give an example, if you think so.
• Berger claims that seeing is always already conditioned by power and social position. Does this mean we can never have a “pure” aesthetic experience? Or is that concept incoherent to begin with?

3 Is beauty objective or subjective?

The standard undergraduate answer is that beauty is subjective — in the eye of the beholder, a matter of personal taste about which there is no argument. The opposing observation is that we argue about beauty all the time, treat some arguments as better than others, and acquire trained verdicts that feel different in kind from preferences for one flavour of ice cream over another. The pure-subjectivist position has to absorb that observation; the two main historical attempts to do so come from Hume and Kant.

3.1 The Rite of Spring Premiere

On the evening of 29 May 1913, the first performance of Igor Stravinsky’s The Rite of Spring — choreographed by Vaslav Nijinsky for Diaghilev’s Ballets Russes at the Théâtre des Champs-Élysées in Paris — provoked a riot.⁴⁰ Audience members booed, shouted, threw programmes, and reportedly came to blows. Within a decade the same work was being performed to reverent silence in the world’s concert halls, and it is now among the most frequently performed and recorded orchestral compositions of the twentieth century. The episode illustrates Hume’s problem: if beauty were purely subjective — merely a report of the observer’s pleasure — the fact that audiences in 1913 found the work ugly and audiences from 1923 onward found it beautiful would be no more surprising than the fact that some people prefer coffee to tea. But the shift happened across whole populations, over short periods, as critics developed the vocabulary to hear what Stravinsky was doing. Hume’s “true judge” — trained by long exposure, free from prejudice, able to compare — appears to have emerged between 1913 and 1920. Kant’s claim that aesthetic judgements carry a demand for universal agreement also seems to be vindicated by the eventual convergence, even if the timeline was longer than Kant’s model suggests. The 2017 Schutz controversy (info box above) raises a different question against the same Humean and Kantian apparatus: what if convergence is in the wrong direction, and trained taste is converging on a verdict that is wrong because of what it cannot see?

3.2 Hume’s Standard of Taste

David Hume addressed the problem in his essay “Of the Standard of Taste” (1757). He begins by acknowledging the obvious:

“Beauty is no quality in things themselves: It exists merely in the mind which contemplates them; and each mind perceives a different beauty.” — David Hume, “Of the Standard of Taste” (1757)⁴¹

If this were the whole story, there would be no more reason to argue about aesthetic judgements than to argue about flavours. De gustibus non est disputandum. But Hume thinks the relativist conclusion is wrong. We all recognise that some people have better taste than others — more discriminating, more practised, more sensitive to relevant features. Hume’s “true judge” has five qualities: delicacy of imagination, practice in a particular art, comparison with other works, freedom from prejudice, and good sense. The standard of taste is not written in nature, but it is not merely arbitrary either: it is the convergence of well-practised, unprejudiced human sensibility.

Hume holds a middle position: aesthetic judgements are not objective in the way that “water is \(\mathrm{H_2O}\)” is objective — they don’t describe mind-independent properties — but they are not purely subjective either. Some are more defensible than others, and the difference can be argued for.

3.3 Kant’s Antinomy

Kant, in Critique of the Power of Judgment (1790), gave the problem its canonical formulation.⁴² He calls aesthetic judgements (judgements of taste) peculiar because they claim universal validity without being based on objective concepts:

When I say “this is beautiful,” I am not just reporting a personal feeling (“I like this”). I am implicitly claiming that you ought to agree — I am demanding assent. But I cannot prove my case by pointing to any objective property of the object. The judgement is subjective (based on feeling) yet claims universal validity (everyone should feel this).

Kant’s solution is technical and disputed. But the problem he identifies is genuine: aesthetic judgements occupy a strange logical space between the purely personal and the objectively verifiable.

3.4 Three Objections to Pure Relativism — and the Relativist’s Replies

The claim that aesthetic judgement is purely subjective — that there is no sense in which Rembrandt’s Night Watch is a greater painting than a greeting card⁴⁴ — faces three standing objections. Each has a reply, and the replies are not negligible.

The expertise objection. If beauty were purely subjective, expertise would be impossible. But some people are better at hearing what is happening in music, more sensitive to the formal qualities of prose, more attuned to what a film is doing. Bourdieu’s reply, in Distinction (1979),⁴⁵ is that aesthetic “expertise” is the internalisation of class- and culture-specific dispositions — habitus — that feel like discrimination from inside but track social position rather than aesthetic reality. The objection does not refute relativism; it identifies what relativism still has to explain — why expertise feels like access to something rather than to a code.

The convergence objection. Different cultures, centuries, and peoples find some works compelling across contexts. Homer has been read for 2,800 years by people in radically different social worlds. Press the relativist on this and the answer is that cross-cultural persistence is rarer than the objection assumes. Homer’s persistence is partly the persistence of an institutional canon — Greek into Latin, Latin into the European university curriculum, the curriculum into the global anglophone classroom — and the same is true for Shakespeare, Beethoven, and the small set of works the objection points to. Aesthetic merit or institutional capture: that is the question, and the convergence objection cannot settle it without begging it.

The internal coherence objection. Many aesthetic judgements can be supported by reasons that appeal to features of the work itself — internal structure, formal coherence, the relationship between parts, the handling of medium. Such reasons presuppose criteria, the relativist’s standing rejoinder runs — what counts as coherence, as well-handled, as relationship between parts — and the criteria vary across traditions. The reasons are intersubjective within a tradition, not across.

The objections do not refute relativism. They show that relativism has to account for genuine phenomena — expertise, convergence, the giving of reasons — and push the relativist toward Bourdieu’s social-reproduction account or an equivalent. The dispute is then between two well-developed positions, not between a sensible view and a confused one.

3.5 Questions to Argue About

• If some people genuinely have better taste than others, how do we identify them? Isn’t that judgement itself a matter of taste?
• Kant says aesthetic judgements claim universal validity but cannot be proven. Is this a stable position, or does it collapse into either objectivism or subjectivism?
• When we argue about whether a piece of music is good, what are we actually doing? Are we trying to change each other’s feelings, or pointing to features of the music, or something else?
• The pure relativist says Rembrandt and a greeting card are equally good, aesthetically. Is that obviously wrong? What makes it wrong, if it is?

4 What does art tell us about reality that other ways of knowing cannot?

Science describes the world in terms of mass, charge, force, and frequency. History reconstructs what happened. Philosophy analyses arguments. What does art do that these cannot?

4.1 Art as a Lie That Tells the Truth

Picasso is reported to have said something close to “Art is a lie that makes us realise the truth.” The precise phrasing circulates widely but cannot be traced to a single verified source; the nearest documented version appears in an interview with Marius de Zayas published in The Arts (New York, 1923):

“We all know that Art is not truth. Art is a lie that makes us realize truth, at least the truth that is given us to understand.” — Pablo Picasso, in Marius de Zayas, “Picasso Speaks,” The Arts (May 1923)⁴⁸

The idea is genuine Picasso but the exact words vary by interviewer and translator. The setting matters: de Zayas was interviewing Picasso for a New York audience hungry for the avant-garde, and Picasso was performing the role of provocateur. The line is an aphorism for journalists, not a position-paper. The serious philosophical version of the claim that art knows what science cannot belongs to Heidegger’s aletheia (treated below) and to Martha Nussbaum’s argument in Love’s Knowledge that literature delivers moral knowledge that propositional summary cannot.⁴⁹ Picasso is good as a hook; bad as a thesis.

The fictional novel is the cleanest case of the structure Picasso names. Crime and Punishment (1866) is not true in the sense of being a historical record.⁵⁰ Raskolnikov never existed; the murder didn’t happen. But Dostoevsky’s account of guilt, rationalisation, self-punishment, and confession describes something about the phenomenology of conscience with a precision and detail that no empirical psychology has matched. The lie (the invented story) makes us realise a truth (something about how guilt actually works in a mind). Cubism, examined in the next section, makes the same move in paint.

4.2 Cubism as Epistemology

The standard account of Cubism treats it as a formal revolution in painting technique. The stronger reading — Daniel-Henry Kahnweiler’s in The Rise of Cubism (1920), revived by John Berger in The Moment of Cubism (1969) — treats it also as an epistemological claim. Single-point perspective, the dominant convention of European painting from the Renaissance to the nineteenth century, presents a scene as it appears from one place, at one moment, and produces a unified, consistent image. The Cubist charge is that this convention smuggles in a metaphysical claim: that knowledge has a single privileged viewpoint, a fixed eye from which everything is correctly arranged. Single-point perspective is not false as a picture — the eye really does see from one place — but on this view it is misleading as a model of knowing, since it presents the partial as the total.

On this reading, Cubism’s multiple viewpoints are not a failure of representation but a more honest representation of what it means to know something — that you never see it from one place only, that your knowledge is always composed of partial views that cannot be perfectly synthesised. The view has critics: Ernst Gombrich (Art and Illusion, 1960) argued that single-point perspective is a learned convention that helps the eye see, and that the Cubists were trading one convention for another, not unmasking a lie.⁵¹ The defensible claim is that Cubism makes the conventionality of perspective visible — and that visibility is itself a contribution to thinking about how images claim to know.

4.3 Heidegger: Art as the Happening of Truth

Martin Heidegger’s “The Origin of the Work of Art” (in Holzwege, 1950; translated by Albert Hofstadter in Poetry, Language, Thought, Harper & Row, 1971) makes a claim about art that is more ambitious than representation or expression.⁵² To understand it, start with a picture.

Start with the shoes. Van Gogh, around 1886, painted a pair of worn peasant shoes. Look at them: the leather is bulged by an absent foot; the laces are loosened by hands that loosened them at the end of a working day; the whole surface is dark with soil and labour. What the painting lets you see is not a pair of shoes in an inventory sense — it is the world these shoes belong to: the field, the tiredness, the woman who owns them, the hours of bent-over walking. That whole web of meaning was already there before you looked at the painting, but the painting is what made it visible to you. Nothing in a prose description of shoes does this.

“World” and “earth” as Heidegger’s terms for what just happened. Heidegger calls what the painting opens a world — not the planet, but the connected field of meaning in which an object like a pair of shoes is for something, belongs to someone, is inflected by a whole way of life. Alongside world, he names earth: the mute, resistant materiality that any work uses and cannot fully domesticate — the actual oil paint on the canvas, the grain of the wood in the peasant’s doorway, the stone of a Greek temple. Every great work, on this account, holds world and earth in tension: it discloses something (world) while keeping something else in reserve (earth), and the tension is what makes the work inexhaustible on repeated viewing.

The Greek word for what is happening: aletheia. Heidegger’s ancient-Greek term for this disclosure is aletheia, usually translated as unconcealment. For Heidegger, truth in art is not a property of sentences about the work — as it is in science, where “true” applies to a proposition. Truth in art is an event: something previously hidden comes to presence, and stays present, in and through the work. Paraphrase destroys poetry because the knowledge is not separable from the form — the form is what performs the disclosure. A painting of Van Gogh’s shoes is not the shoes; it is a specific way those shoes have been made unconcealed.⁵³

A complication worth raising at this point. Heidegger’s reading of the picture as showing “the peasant woman’s shoes” is exactly the move the art historian Meyer Schapiro pressed against in 1968 (“The Still Life as a Personal Object — A Note on Heidegger and van Gogh”). Schapiro showed by archival research that the painting Heidegger had in mind almost certainly depicts Van Gogh’s own shoes, the shoes of an urban man — Van Gogh in Paris in 1886, not a Brabant peasant. The current art-historical consensus follows Schapiro. This is not a small point: a substantial part of what Heidegger’s aletheia is supposed to disclose is the rural-labour world, and that world is being projected onto the painting rather than disclosed by it. Derrida’s The Truth in Painting (1978) argues that both Heidegger and Schapiro are restaging a deeper problem about who the shoes “belong to,” but the historical correction stands. The lesson Heidegger’s reading still teaches — that paintings can disclose a world rather than merely depict an object — survives the correction; the specific reading of these shoes does not.

4.4 What Fiction Gives That Documentary Cannot

Consider two ways of approaching the Rwandan genocide of 1994: a historical account of the political causes, the role of the Hutu Power movement and Radio Mille Collines, the failure of international intervention, the demographic scale of the killing; and a novel — like Scholastique Mukasonga’s fiction,⁵⁵ or Jean Hatzfeld’s testimonial literature⁵⁶ — that puts you inside the experience of a survivor or perpetrator.

The first kind is essential: causation, scale, political responsibility, policy failure. It is what you need to prevent recurrence. The second kind is different: what it was like, how ordinary people became killers or victims, the texture of extremity that statistics cannot convey. Neither replaces the other. Two positions on the second kind compete. The strong claim is that the imaginative, empathic, particularity-respecting knowledge fiction provides is not available through any other means. The weaker claim is that fiction and certain non-fictional forms (Hatzfeld’s testimony, oral history, ethnographic interview) are the genres where such knowledge is at home, and that propositional summary loses something specific to those forms. Whether the something lost is itself “knowledge” — and which of the two positions above tracks the right answer — is what the next section’s Forced Fork puts to the reader.

Tolstoy articulated this in What is Art? (1897). His “infection” theory holds that art is the deliberate transmission of feelings: an artist who has lived through a feeling uses external signs — words, sounds, images — to hand on that feeling to others, who then experience it themselves.⁵⁷ His account is somewhat crude (art is not just emotional contagion), but he identifies something real: art can make you understand what you could not have understood from description alone.

A warning about how this gets cited. Tolstoy’s own application of the infection criterion, in the closing chapters of What is Art?, is ferociously anti-canonical: he uses it to disqualify Beethoven’s Ninth Symphony and late period, the music-dramas of Wagner, and “the greater part of the works of Shakespeare and Goethe.”⁵⁸ (His own novels, including Anna Karenina, are not named in What is Art? itself, but Tolstoy’s Confession (1879) renounces them on related grounds.) The TOK-textbook habit of citing Tolstoy as if his criterion picks out the high points of European art history inverts what Tolstoy meant. Either the criterion is too austere (and almost no “art” is art) or it is too narrow.

4.5 Questions to Argue About

• Picasso says art is “a lie that makes us realise the truth.” Is this coherent? Can a lie be epistemically valuable? Or does this stretch the concept of “lie” beyond usefulness?
• Is the knowledge that fiction gives about human psychology genuine knowledge — true beliefs about how people actually are — or just vivid imaginative experience? How would you distinguish them?
• Can abstract art (with no representational content) give knowledge? Of what?
• Tolstoy says art infects the audience with the artist’s feeling. Is that an epistemological claim or a psychological one? Is there a difference, in this context?

5 Can art change the world?

The Romantic claim is that art is the most powerful engine of social transformation. The cynical counter is that art preaches only to the converted, decorates the walls of the powerful, and changes nothing. The truth, if there is one, lies somewhere more complicated.

5.1 Shelley’s Legislators

Percy Bysshe Shelley, in A Defence of Poetry (1821), made the strongest claim:

“Poets are the unacknowledged legislators of the world.” — Percy Bysshe Shelley, A Defence of Poetry (1821)⁶⁴

Shelley means that poets — and, by extension, artists — are the creators of the moral and imaginative frameworks within which society operates. They do not write laws; but they shape the sensibility within which laws are made and judged. The moral imagination of a society is its poetry; change the poetry and you change the moral imagination.

Shelley’s claim is consistent with several historical cases. Homer shaped the moral vocabulary of ancient Greece. Dickens’s novels coincided with reform of the Poor Laws.⁶⁵Uncle Tom’s Cabin (Harriet Beecher Stowe, 1852) is sometimes said to have been credited by Abraham Lincoln with having started the Civil War — though the “little lady who started this great war” anecdote is apocryphal, first appearing in print decades after the supposed 1862 meeting, and should not be relied on as evidence.⁶⁶ What is uncontested is the cultural saturation: the novel sold hundreds of thousands of copies in its first year and shaped the Northern moral imagination of slavery. What is contested is whether the political consequences are the art’s doing, or whether the art articulated what the political and social conditions had already made possible.

5.2 Guernica

Pablo Picasso painted Guernica in 1937 in response to the Nazi German bombing of the Basque town of Guernica during the Spanish Civil War. The painting is large (349 × 776 cm), monochrome, and depicts a scene of anguish⁶⁷ — a screaming horse, a bull, a dying soldier, a mother holding a dead child, a lamp. It contains no direct representation of the bombing itself: no planes, no explosions, no identifiable political imagery. It is entirely in the language of pain.

Guernica has been displayed prominently in locations of political significance: the United Nations headquarters in New York (where it hung for decades and where the US Secretary of State presented the case for the Iraq War in 2003, in front of a tapestry reproduction that was covered with a blue cloth for the occasion — a detail that tells you something about the power attributed to the image). It is routinely cited in discussions of political violence and war. Does it change minds, or does it become an icon that substitutes for thought?

5.3 What the World Does to Art

The Guernica tapestry episode poses a question the lesson has not yet asked: what does the world do to political art, as opposed to what political art does to the world?

A pattern often pointed to: Guernica is now in a museum. Ai Weiwei’s Remembering is reproduced in art history textbooks. Banksy’s street art sells for millions.⁶⁸ On the Frankfurt School reading (Adorno’s “Culture Industry”; Debord’s Society of the Spectacle), the art that most directly challenges the established order is most efficiently incorporated — made safe, turned into cultural capital, made to decorate the institutions it once attacked. Andy Warhol’s factory-as-art-studio anticipated this: if the artworld can commodify anything, the most radical political art will eventually become a poster in a gallery gift shop.⁶⁹ Jacques Rancière (The Emancipated Spectator, 2008) and T. J. Clark, in his work on Manet, reply that “incorporation” is too simple a story: works retain a critical reserve even after market absorption, and the assumption that commercial circulation drains political force concedes too much to the marketplace.⁷⁰Guernica’s presence in a museum did not stop UN officials in 2003 from covering the tapestry — not the behaviour of an incorporated, defanged image.

Whichever side wins the dispute, the underlying observation is not cynical but structural: the artworld (in the first lesson’s sense) is not politically neutral. It has institutional interests, commercial interests, and legitimating functions. Art that begins as critique can end as decoration; whether it always does, or whether the critical reserve survives, is what is being argued over. The question the Guernica tapestry episode raises is not only “does art change the world?” but “does the world change art?” — and “does the world change art precisely because art threatens to change it?”

5.4 Hamlet in the Gulag

During the Stalin era, prisoners in some Soviet labour camps recited and performed canonical literature — Pushkin, Shakespeare, Lermontov, Mandelstam — in conditions of extreme physical deprivation and at risk of punishment. Two memoir-records stand at the centre of the documentary case. Eugenia Ginzburg, in Journey into the Whirlwind (1967), describes her two years of solitary confinement in Yaroslavl prison (1937–1939) before transport to Kolyma: when the prison library was shut, she gave herself silent lectures on Pushkin, recited his poetry in the cell to keep her mind in working order, and on one occasion satisfied a suspicious warder’s demand for half an hour of recitation by declaiming the politically neutral Eugene Onegin — at the end of which the warder, having moved through threat and curiosity to delight, asked her to continue. Varlam Shalamov, in the late story-cycle Kolyma Tales (composed 1954–1973), records in “Athenian Nights” how, on night duty in a Kolyma camp hospital toward the end of his sentence, he and two fellow medical orderlies — all still prisoners — spent two or three hours together reciting from memory all the poetry they could recall: Akhmatova (an early version of “Poem without a Hero”), Blok, Mandelstam, and “the most important Russian poets,” together with passages of Homer, Dante, Shakespeare, and Hugo.⁷² Anne Applebaum’s Gulag (2003) notes that the practice was unevenly distributed across camps and periods, and easier in the relatively privileged “cultural brigades” and hospital details than in the killing camps of Kolyma’s gold-mining sectors. The case is suggestive rather than systematic. It raises the question of what reciting Eugene Onegin in a punishment cell, or Shakespeare in a hospital ward, could give such prisoners that justified the risk.

One answer, given by Havel and echoed by survivor-memoirists: the experience of a fully articulated subjectivity, in a world whose purpose was to deny subjectivity and reduce persons to categories. Onegin’s wasted irony, Hamlet’s indecision and consciousness of his own consciousness, the formal closure of a stanza of Pushkin recited in correct order in a cell where everything else was deformed — on this reading, these were not escapes from reality but confrontations with it. The Ginzburg warder, who began by demanding recitation as a test of suspicion and ended by asking her to continue, is the lesson’s question in miniature: art altered, in a single hour, the relation between guard and prisoner that the camp system was built to enforce. The larger claim is structural rather than anecdotal — that art preserved the constituency capable of recognising the regime as a lie, which is what eventually changes a world. A sceptical reading, advanced by some camp historians, is that any sustained collective practice — chess, prayer, recitation of mathematical proofs — would have served the same function, and that the case licenses no claim specific to canonical literature. The strong reading must show why the artistic content of the recitation, not merely the fact of organised collective practice, was doing the work.

5.5 Reproduction and the Aura

Walter Benjamin’s “The Work of Art in the Age of Mechanical Reproduction” (1936) names the condition under which all of this happens. Benjamin’s thesis: the technologies of photography and film strip the artwork of its aura — its unique here-and-now presence, its embeddedness in tradition and ritual.

“That which withers in the age of mechanical reproduction is the aura of the work of art.” — Walter Benjamin, “The Work of Art in the Age of Mechanical Reproduction” (1936), §II⁷⁴

Benjamin defines the aura, in a passage borrowed from his account of natural perception, as “the unique phenomenon of a distance, however close it may be.”⁷⁵ A reproduction abolishes the distance. You can hold Guernica in a postcard; you have not held Guernica. The democratisation of access — everyone can see the Mona Lisa in print — is also the destruction of the work’s cultic authority.

Benjamin is double-edged. He sees emancipatory possibility (mass reproduction breaks the monopoly of the cathedral and the salon) and political danger in equal measure: the same technologies, he notes, are mobilised by fascism for the aestheticisation of politics. Sontag’s On Photography (1977), invoked in the margin of the Ai Weiwei case below, extends Benjamin’s diagnosis to the photographic image as an ethical object. The contemporary extension — generative AI, deepfakes, NFTs — is ours to make: if the aura was tied to the singular here-and-now, what survives in a medium where the original is a prompt?

5.6 Ai Weiwei’s Remembering

In 2009, the Chinese artist Ai Weiwei covered the facade of the Haus der Kunst in Munich with Remembering — a work consisting of 9,000 children’s backpacks spelling out, in Chinese characters, a sentence from a mother whose daughter had died in the 2008 Sichuan earthquake: “She lived happily for seven years in this world.”⁷⁶ The earthquake killed nearly 70,000 people; a disproportionate number of the dead were children who died when poorly constructed school buildings — the product of corruption in the construction industry — collapsed. The Chinese government suppressed investigation of this corruption. Ai Weiwei’s work, by making visible what the government wished invisible, was an act of political art that led to his detention, surveillance, and eventual exile.

The work does not argue that corruption causes death (though it implies it). It does not present statistics (though the number 9,000 is itself a statistic given a human face). It uses the aesthetic resources of scale, repetition, and colour — the particular visual impact of 9,000 backpacks — to make a claim that bypasses the channels through which such claims are normally made and can be suppressed.

But the visible artwork is only part of what Ai Weiwei made. Between 2008 and 2011 he and roughly one hundred volunteers conducted a Citizens’ Investigation into the Sichuan school deaths: travelling to the affected counties, interviewing parents, and collecting children’s names from those whom the state had refused to count. They published 5,196 names on a blog the authorities repeatedly took down, and read each name aloud on the anniversary of the earthquake.⁷⁸ The 9,000 backpacks at the Haus der Kunst spell a sentence of grief; the list of 5,196 names is the work to which the backpacks point. Without the list, the backpacks are decoration about decoration; with it, the work is the act of counting the dead the state was determined not to count.

5.7 Questions to Argue About

• Shelley says artists are the “unacknowledged legislators of the world.” Is this flattering poetry, or a genuine claim about how moral frameworks are formed? What would evidence for or against it look like?
• Does political art convert anyone who doesn’t already agree, or does it mainly intensify the convictions of those already converted? Does it matter?
• Is art that is consciously made to produce a political effect — propaganda in the broadest sense — epistemically or aesthetically compromised? Can it be both politically effective and aesthetically valuable?
• Ai Weiwei’s Remembering was made at personal cost. Does the risk taken by the artist change the value or meaning of the work?

6 Who is the author of meaning in a work of art?

When we interpret a work of art, whose intentions count — the artist’s, the audience’s, or neither? This question generates one of the sharpest disputes in aesthetic theory, and the answer is not merely academic: it determines how criticism works and what interpretation is trying to achieve.

6.1 The Death of the Author

Roland Barthes, in his essay “The Death of the Author” (1967), argues that the author’s intentions are irrelevant to the meaning of a text:

“The birth of the reader must be at the cost of the death of the Author.” — Roland Barthes, “The Death of the Author” (1967)⁸³

Barthes’s argument: once a text is written, it enters a space of language that belongs to no one. The “author” who speaks in the text is not the biological person who wrote it but a constructed persona — a function of literary convention and the reader’s imagination. What a text means is not what its author “meant” to say; it is what it produces in reading, in each successive encounter with each successive reader. Meaning is not found; it is made.

Barthes’s claim is genuinely radical. It liberates criticism from biography (you don’t need to research an author’s life to interpret their work) and liberates readers from the threat of being “wrong” (there is no authoritative meaning to be wrong about). It also implies that a text has as many meanings as it has readers — which many critics regard as a reductio ad absurdum.

6.2 Intentionalism

The counter-position, developed by E.D. Hirsch in Validity in Interpretation (1967), argues that if meaning is not constrained by authorial intention, interpretation becomes arbitrary.⁸⁴ If “Hamlet” can mean whatever any reader makes of it, there is no longer any criterion for distinguishing correct from incorrect interpretations, insight from fantasy, scholarship from projection.

Hirsch distinguishes between meaning (what the author intended to communicate) and significance (what the work means to a given reader in their context). Meaning is fixed; significance varies. Criticism aims at meaning.

The practical problem with intentionalism: we usually can’t know what authors intended. Authors die, change their minds, give contradictory accounts, deny they meant what the text clearly implies. And some authors explicitly renounce their intentions as relevant: Samuel Beckett refused to explain what his plays “meant” and regarded interpretive questions with courteous contempt.⁸⁶

6.3 The Case of Forgeries

Han van Meegeren was a Dutch painter who, in the 1930s and 1940s, produced a series of fake Vermeers — paintings that experts, including the distinguished art historian Abraham Bredius, authenticated as genuine. They were bought for large sums; one was acquired by Hermann Göring during the Nazi occupation. After the war, van Meegeren was arrested for collaborating with the enemy (he had sold a “Vermeer” to Göring); to prove he hadn’t sold a national treasure, he confessed to forgery and painted another Vermeer in prison.

The philosophical question: when Christ at Emmaus (c. 1937) — one of van Meegeren’s best forgeries — was believed to be a Vermeer, it was authenticated by Abraham Bredius and acquired by the Boijmans Museum in Rotterdam as a major addition to the Vermeer corpus. Bredius wrote of “this glorious work of Vermeer” in The Burlington Magazine (1937);⁸⁷ not all critics agreed (Duveen reportedly thought it suspect), but the work entered the canon. After the 1945 exposure, the same painting was systematically downgraded — first to “good imitation,” then to “mediocre fraud” in standard art-historical reference works. The paint, the canvas, the pigments, and the composition did not change. What changed was the context of interpretation — the story that accompanied the image.

The standard reading of this case is that aesthetic experience is partly conditioned by knowledge of provenance, authenticity, and authorship in ways that show we are appreciating the idea of the work as much as the work itself. The formalist has a reply: the post-exposure downgrading reflects not the disappearance of aesthetic value but the activation of additional judgements (of historical importance, of skill given the constraints of the period, of originality) that the formal features do not by themselves determine. The painting’s purely visual qualities remained constant; the network of historical and contextual judgements around it shifted — exactly what we should expect, not a refutation of formalism. The case is therefore evidence about how aesthetic experience works in practice; whether it is evidence about what aesthetic value is depends on which reading you accept.

6.4 Gadamer: The Fusion of Horizons

Hans-Georg Gadamer (Truth and Method, 1960) offers a third position that escapes the Barthes-Hirsch impasse.⁸⁸ For Gadamer, interpretation is neither the recovery of authorial intent (Hirsch) nor the unconstrained production of meaning by the reader (Barthes). It is a conversation between the text and the interpreter.

Every text comes from a particular historical horizon — a world of assumptions, questions, and concerns that the reader does not initially share. Every reader brings their own horizon — their world of concerns and assumptions. Interpretation is not the reader imposing their horizon on the text, nor the text imposing its horizon on the reader. It is a fusion of horizons (Horizontverschmelzung): a process in which both are transformed. The reader is changed by what the text makes possible; the text’s meaning is renewed in each genuine act of reading.

This has two important consequences. First: interpretation cannot be arbitrary, because the text’s horizon exerts a genuine constraint — you cannot simply make a text mean what you like, because the text will resist. Second: interpretation is historically situated and therefore always incomplete — each generation reads its canonical texts differently, not because the texts are infinitely malleable, but because each new generation brings a new horizon to the encounter.

If accepted, Gadamer’s account explains why Homer continues to be read by communities whose horizons differ enormously from Homer’s, why Hamlet can be performed in a Soviet labour camp and remain Hamlet, and why literary interpretation is a discipline with standards even without the author’s authority to appeal to. Whether his account succeeds depends on whether the metaphor of “fusion” survives examination: Habermas and others argue that Gadamer underplays how distorted communication and unequal power can prevent fusion from happening on equal terms, and that the hermeneutic circle can quietly reproduce the dominant horizon while presenting itself as conversation.

6.5 Questions to Argue About

• Barthes says the birth of the reader requires the death of the author. Is this metaphor illuminating or misleading? Is there a version of authorial intention that survives Barthes’s argument?
• The van Meegeren case shows that our aesthetic response to art is influenced by whether we believe it to be authentic. Does this undermine the value of aesthetic experience, or does it just show that context matters?
• If two readers give incompatible interpretations of the same novel, can both be right? What would it mean for both to be right?
• Is cultural appropriation — the use of artistic traditions from a minority culture without permission or acknowledgment — a violation of authorial rights, or does Barthes’s theory imply that no one owns a tradition once it exists?

6.6 Sen no Rikyū and the Rough Tea Bowl

The 16th-century tea master Sen no Rikyū (1522–1591) transformed Japanese tea ceremony by rejecting the elegant Chinese porcelain wares favoured by the shogunal court and elevating rough, asymmetrical, locally-made bowls — often visibly flawed, cracked, or later repaired with gold lacquer (kintsugi).⁹¹ The Raku-ware Ōguro, attributed to his collaborator Chōjirō, is a stocky, uneven, matte-black vessel that looks to Western-trained eyes as if a beginner had made it badly.⁹² Rikyū held it to be a profound expression of wabi (sober simplicity) and sabi (the beauty of impermanence and weathering). His patron Toyotomi Hideyoshi preferred gold; Rikyū, compelled by Hideyoshi to perform seppuku in 1591, chose this aesthetic over his life.⁹³

Rikyū’s case puts the authorship question into a register the European examples do not reach. The bowl’s meaning is fixed neither by Chōjirō’s making (the artist) alone, nor by Rikyū’s curatorial reception (the connoisseur as second author), nor by the trained successors who came to see the bowl as Rikyū saw it (a community of practice as the author). Each of those candidates is a real authorship claim. Each is partial. The further question Rikyū’s case forces — whether wabi-sabi tracks an aesthetic truth that polished symmetry misses, or merely codifies a different cultural preference indistinguishable from Hideyoshi’s preference for gold — is the topic of the Forced Fork below.

7 What is the relationship between art and ethics?

The question is not merely theoretical. It arises whenever we encounter the work of a person whose actions we find morally reprehensible, or when a work of art depicts something that seems morally problematic.

7.1 Wagner and the Tainted Masterpiece

Richard Wagner was one of the most virulent antisemites in the history of European culture. His 1850 essay Das Judenthum in der Musik (“Jewishness in Music”) argued that Jewish composers were constitutionally incapable of producing genuine art.⁹⁷ He was Adolf Hitler’s favourite composer; the Nazi regime used his music as a cultural centrepiece of the Third Reich, with the Bayreuth Festival functioning as a state-patronised showcase under Winifred Wagner’s directorship.⁹⁸ The Israel Philharmonic Orchestra has maintained an informal ban on performing Wagner’s music from 1938 to the present.⁹⁹ On 7 July 2001, the conductor Daniel Barenboim, himself Jewish, performed an encore of Wagner’s Tristan und Isolde prelude with the Berlin Staatskapelle at the Israel Festival; a portion of the audience walked out and the controversy was debated in the Israeli Knesset.¹⁰⁰ The question the Wagner case poses is not whether his antisemitism was vile — it was — but whether that biographical fact changes anything about the Ring Cycle considered as music. Plato argued in Republic Book X that art corrupts by producing false emotional responses to false representations; Aristotle argued in the Poetics that art purifies through catharsis. Neither framework cleanly resolves whether Parsifal is a different aesthetic object when we know who wrote it and why.

7.2 Anna Netrebko and the Met Opera

The structurally identical case in popular international media: Anna Netrebko (b. 1971, Krasnodar) was through the 2000s and 2010s one of the world’s leading dramatic sopranos, recording for Deutsche Grammophon and singing leading roles at La Scala, the Royal Opera House, the Vienna State Opera, and from 2002 at the Metropolitan Opera in New York.¹⁰¹ She had been a public supporter of Vladimir Putin: 2012 election endorsement; December 2014 photograph at a Moscow event presenting a 1 million-rouble donation to the Donetsk People’s Republic opera house, posing with the DPR flag; previous Russian state honours.¹⁰² After Russia’s full-scale invasion of Ukraine on 24 February 2022, Met General Manager Peter Gelb required public artists to repudiate Putin and the war as a condition of continued engagement. Netrebko issued a statement on 26 February 2022 condemning the war but declined to repudiate Putin personally, citing the safety of her family in Russia.¹⁰³ Gelb cancelled her scheduled Met performances on 3 March 2022 — three days after Lucerne Festival’s parallel decision on Gergiev.¹⁰⁴ She filed a grievance through the American Guild of Musical Artists; the AGMA arbitrator ruled in February 2023 that the Met had breached the union contract and ordered payment of contractual minimum, while leaving Gelb under no obligation to re-engage. La Scala, the Vienna State Opera, and the Royal Opera House continued to programme her. The case is the international parallel to Lucerne’s Gergiev decision (info box above) — same week, same regime, same logic, same kind of institution decisions, slightly different individual circumstance (a soprano who condemned the war but did not denounce Putin, vs. a conductor who did not respond at all).

7.3 Separating Art from Artist

Picasso was, by the accounts of those who knew him well, a serial abuser of the women in his life. His partner Françoise Gilot, in Life with Picasso (1964, with Carlton Lake), documented the psychological cruelty, coercion, and humiliation she experienced with him — an account that Picasso’s lawyers unsuccessfully attempted to suppress.¹⁰⁵ His granddaughter Marina Picasso later characterised him in still harsher terms in Picasso, mon grand-père (2001).¹⁰⁶ Dora Maar, another partner, was psychologically destroyed by her relationship with him.¹⁰⁷ Richard Wagner was a virulent antisemite whose ideas influenced Nazi ideology; his music dramas were performed at Bayreuth under the patronage of Hitler. Wagner himself died in 1883, but the association remains troubling.

The case for separation: the work is the work, not the biography. Guernica is not diminished by Picasso’s treatment of women. The Ring Cycle is not Nazi music, whatever Wagner’s politics were. Aesthetic value is a property of the art, not of its creator. To refuse the art is to let biography determine aesthetics — a category error.

The case against separation: our relationship to art is not purely aesthetic. We attribute works to their authors; the authority and significance we grant them is in part a function of who made them. If we learn that an artist was morally monstrous, this new information changes our relationship to the work — not by changing the work’s formal properties, but by changing what the work is in our cultural economy.

7.4 The Depiction of Evil

Nabokov’s Lolita (1955)¹⁰⁹ is narrated by a paedophile who describes his abuse of a 12-year-old girl in language of extraordinary beauty. Cormac McCarthy’s Blood Meridian (1985)¹¹⁰ depicts historical mass murder with a meticulous, almost rapturous prose style. These are widely regarded as literary masterpieces. They are also, on any reading, morally disturbing.

A standard distinction here, defended in the philosophy of literature by Wayne Booth (The Rhetoric of Fiction, 1961) and refined by Martha Nussbaum: between a work that depicts evil (showing what it is like, giving it a voice, making it comprehensible) and a work that advocates evil (recommending, glorifying, or inviting the reader to celebrate it).¹¹¹ On this reading, Lolita depicts; it does not advocate. Humbert Humbert’s rhetoric is seductive, but the seduction is itself an object of critical attention — the novel shows how such men think and talk, which is not endorsing it.

The depicts/advocates distinction is contested. Plato in Republic III argues that depiction is advocacy in the relevant sense — the audience trained on vivid impersonations of vice acquires the dispositions impersonated, regardless of the author’s framing intent.¹¹² A more recent version of the worry, raised by feminist critics of Nabokov, is that the seductive prose of Lolita recruits the reader into Humbert’s perspective in ways the novel’s framing cannot fully neutralise — so the distinction between depicting his rationalisations and producing them in the reader collapses in practice. The Booth/Nussbaum reply: the cultivated reader resists this recruitment, and the novel itself supplies the resources for the resistance. Whether the cultivated reader is the only relevant reader, and whether the resources for resistance are sufficient, is the open question.

Before turning to Plato’s argument, it is worth noticing that his proposal is not merely theoretical. In 2011 the Chinese state held the artist Ai Weiwei in a small room with two guards who never left, including while he slept and while he used the toilet, for 81 days. Plato’s Republic is somebody’s working memo. The question Plato puts to art is one that working states have answered, in detail, again and again.

Plato’s critique of art in Republic Books III and X is more radical and more philosophically interesting than the emotional argument usually cited. The deeper argument is epistemological.

In Book X, Plato argues that art produces images of images — copies three removes from reality.¹¹³ The Form of Bed is the real thing (accessible only to the philosopher). The carpenter makes a particular bed according to the Form. The painter makes an image of that particular bed. The painting is doubly removed from reality: it represents an appearance of something that is itself only an instance of the real. Worse: the painter need not understand carpentry to paint a bed convincingly. Art offers the appearance of knowledge about things the artist may not understand at all.

Plato’s accusation is epistemological, not merely moral: art habituates us to mistake vivid presentation for knowledge. The poet who makes us weep for a fictional character’s suffering is not teaching us about suffering — they are stimulating emotional responses to appearances, strengthening the irrational soul at the expense of reason. The person trained on tragedy is not wiser about grief; they are more susceptible to being moved by the semblance of grief.

Plato’s verdict on the imitator at Republic 602b is blunt:

“It seems, then, that we’re fairly well agreed that an imitator has no worthwhile knowledge of the things he imitates, that imitation is a kind of game and not something to be taken seriously, and that all the tragic poets, whether they write in iambics or hexameters, are as imitative as they could possibly be.”¹¹⁴

Plato’s conclusion from this argument was not a philosophical shrug. The Republic is a detailed constitutional proposal — a specification of how a just city should be organised, down to the training of its soldiers, the regulation of its marriages, and the content of the stories children are permitted to hear. Inside that proposal, poets are banned.¹¹⁵ Not restrained, not reviewed, but excluded from the city altogether on the grounds that their work habituates citizens to respond emotionally to semblances rather than rationally to realities. The argument has a political consequence, and Plato states it: an epistemology that treats art as the cultivation of illusion leads logically to censorship. This is the fact that arts-and-ethics discussions conducted at a safe distance from political power tend to elide. The counter-argument, made by Aristotle in the Poetics, is that tragedy produces catharsis — a purgation or clarification of the emotions — and thereby leaves the audience more stable, not less.¹¹⁶ Aristotle’s reply is philosophical; Plato’s proposal is a zoning law.

Plato’s challenge should not be dismissed. The cognitivist claim that art provides genuine knowledge other modes cannot is one position; Plato’s reply — that art provides the illusion of knowledge while bypassing reason — is another, and is what defenders of cognitivism have to answer rather than ignore. Heidegger’s account of art as world-disclosure and Aristotle’s account of tragedy as catharsis are two of the standing replies. Neither makes Plato simply wrong; whether either makes Plato substantially wrong is itself disputed.

7.5 Questions to Argue About

• If we discover that an artist whose work we love was morally monstrous, should that change how we value the work? If yes, why? If no, are we committed to the view that art and biography are fully separable?
• Is there a difference between a novel that depicts atrocity in uncomfortable detail and one that advocates it? How do we tell them apart?
• Plato banned poets for educating the emotions wrongly. Was he wrong about everything, or just about which art to ban? (The catchphrase distorts: in Republic X the target is mimetic poetry that depicts vice and weakness, not poetry as such — Plato explicitly allows hymns to the gods and encomia to good men. Janaway’s Images of Excellence (1995) and Halliwell’s The Aesthetics of Mimesis (2002) are the standard correctives.)¹¹⁷
• Can morally bad art — art that glorifies cruelty, for example — nevertheless be aesthetically excellent? If so, what does that imply about the relationship between moral and aesthetic value?

8 Does music mean anything?

Music poses the hardest version of the art-and-knowledge question. Painting and literature represent something: a landscape, a person, a narrative. Music (in the Western classical tradition at least) typically represents nothing. It is organised sound. And yet almost everyone who has experienced the second movement of Beethoven’s Seventh Symphony, or the opening of Bach’s St Matthew Passion, or the moment a piece of music written for some other occasion happens to arrive at exactly the right moment in your life — almost everyone has felt that music does not merely stimulate neural pathways. It communicates. So what, if anything, does it mean? And can it give knowledge?

8.1 Hanslick’s Formalism

Eduard Hanslick, in On the Musically Beautiful (Vom Musikalisch-Schönen, 1854), made the case for pure musical formalism:

“The content of music is tonally moving forms.” — Eduard Hanslick, On the Musically Beautiful (1854)¹¹⁸

On Hanslick’s view, music is not about anything outside itself. It has no content beyond its own formal organisation — the relationships between tones, rhythms, harmonies, and structures. Attempts to read specific emotions into music are projections; we bring feelings to the music, we do not find them there. The music lover who finds grief in Beethoven’s Eroica and joy in Haydn’s symphonies is responding to formal properties (minor keys, slow tempos, certain rhythmic patterns) in ways conditioned by cultural convention, not discovering anything the music contains.

Hanslick’s position has the virtue of clarity. The standard objection is phenomenological: most listeners experience music as having emotional character — some music sounds like grief, some like exuberance, some like dread — and within a shared tradition, these responses converge across listeners.¹¹⁹ Hanslick’s reply, sharpened by Peter Kivy in Sound Sentiment (1989) and Music Alone (1990), is that the convergence is exactly what cultural conditioning predicts: shared training in a tradition produces shared response, and the expressive description (“sad music”) is shorthand for formal features (minor mode, slow tempo, falling melodic contour) historically associated with grief in Western art music.¹²⁰ What looks like the music being sad is the listener’s trained recognition of features that conventionally cue sadness. The dispute is not whether listeners reliably experience emotion in music — they do — but whether the emotion is in the music or evoked by it through learned association. The Forced Fork at the end of this section is exactly that dispute.

8.2 Tolstoy’s Infection Theory

Tolstoy’s counter (What is Art?, 1897) is as extreme in the opposite direction. Art is the communication of feeling: the artist has an experience, encodes it in the work, and the work “infects” the audience with the same feeling. Music’s great power is its directness: it can communicate emotion more immediately and forcefully than any other art form.

The problem with this account: the same piece of music communicates different things to different listeners, and even to the same listener at different times. And the composer, in composing, may not have had any particular emotional state — they may have been working out a technical problem. Mahler’s Ninth Symphony is widely received as a meditation on death — a reading first made authoritative by Bruno Walter, who premiered the work in 1912 and described its closing pages as a peaceful farewell;¹²¹ Shostakovich’s Fifth Symphony (1937) was famously subtitled “a Soviet artist’s creative reply to just criticism” — a formulation that was not, in fact, written by Shostakovich but by an anonymous critic in the Moscow newspaper Vechernyaya Moskva, and which the composer allowed to stand under Stalinist pressure while privately describing the work as something quite different.¹²² Who is right about what the music “communicates”?

8.3 John Cage and the Limits of Music

John Cage’s 4’33” (1952) consists of a performer sitting at a piano for four minutes and thirty-three seconds without playing. It was premiered by David Tudor at the Maverick Concert Hall in Woodstock, New York, on 29 August 1952.¹²³ The “music” is whatever sounds occur in the performance space during that time: coughing, rustling programmes, the ambient noise of the building and the street.

4’33” is either a profound philosophical statement (music is everywhere; the boundary between music and non-music is arbitrary; listening itself is the art) or an elaborate joke at the audience’s expense. It is difficult to hold both possibilities simultaneously — but the difficulty is itself illuminating. Cage is forcing you to examine what you think music is.

8.4 The Ineffability Problem

Suppose we try to say what music means. We find ourselves reaching for metaphors: the second movement of Beethoven’s Seventh is like grief, or winter, or yearning. But the second movement of Beethoven’s Seventh is not grief, or winter, or yearning; it is organised sound. The metaphors do not quite fit; they point toward something while failing to capture it.

Two readings of this fact are available, and they are genuinely different.

The strong reading: music gives a kind of knowledge that is structurally inarticulable — knowledge by acquaintance rather than description, irreducibly experiential, non-transferable, and structurally different from the knowledge produced by science, history, or philosophy. This is the position of writers from Schopenhauer (The World as Will and Representation, 1819, Bk III) to Susanne Langer (Philosophy in a New Key, 1942), who treats music as a “presentational symbol” of the forms of feeling.¹²⁵

The deflationary reading: the failure of metaphors to “capture” the music does not show that music carries inarticulable knowledge — only that music carries experience, and that the vocabulary of “knowledge” was the wrong one to import. On this reading (defended by Peter Kivy and, in a different idiom, by Roger Scruton’s The Aesthetics of Music, 1997),¹²⁶ music is not failing to be articulated; it is doing something other than the kind of thing that “articulation” applies to, and the inability to paraphrase a Beethoven movement is no more philosophically deep than the inability to paraphrase a sunset.

Choosing between these two readings is not the same task as choosing between the formalist and the expressivist accounts of musical content. It is the prior question of whether music belongs in the category of “ways of knowing” at all.

8.5 Questions to Argue About

• Hanslick says music is pure form; Tolstoy says it communicates feeling. Is there a third option that captures something both miss?
• 4’33” presents silence as music. What is your response — and what does your response reveal about your implicit theory of music?
• If the same piece of music means different things to different listeners, does it mean anything objectively? Or is “music means X” always shorthand for “I experience this music as X”?
• Is there something that can only be communicated by music and cannot be said in any other form? If so, what kind of knowledge is this, and how would we know if we had it?

9 Media

• John Berger, Ways of Seeing (1972) — The BBC television series and the book; four episodes of 30 minutes each. More epistemologically ambitious than it initially appears; the first episode, on oil painting and the male gaze, is a model of how visual analysis can be simultaneously aesthetic and political.
• Rainer Maria Rilke, Letters to a Young Poet (1929) — Ten letters written between 1902 and 1908 to a young man asking how to know if he should be a poet. Rilke’s answer is epistemological as much as practical: questions, he says, should be lived rather than answered. A short, serious, and beautiful text.
• Alison Klayman, Ai Weiwei: Never Sorry (2012) — Documentary film. The most direct account of Ai Weiwei’s practice as a political and artistic act; shows the cost of art as political provocation in ways that no purely theoretical account can convey.
• Andrei Tarkovsky, Andrei Rublev (1966) — A nearly three-hour film about a 15th-century Russian icon painter, made under Soviet censorship. One of the most searching explorations in film of the question of what art is for and what it costs. The final sequence — sudden colour in a film otherwise in black and white — is one of the most effective aesthetic arguments ever made.
• Zadie Smith, Feel Free (2018) — Essays on art, literature, and criticism. “Joy” and “North West” are the most directly relevant to TOK concerns; Smith writes about aesthetic experience with philosophical precision and without academic jargon.
• Ruben Östlund, The Square (2017) — A film about a contemporary art museum curator in Stockholm; a sustained, darkly comic examination of institutional art, moral complicity, and the gap between what art claims to do and what it actually does.

10 Bibliography