How Can One Tell What Is Beautiful?

Link post

The starting-point for all systems of aesthetics must be the personal experience of a peculiar emotion. The objects that provoke this emotion we call works of art [...] if we can discover some quality common and peculiar to all the objects that provoke [it], we shall have solved what I take to be the central problem of aesthetics. We shall have discovered the essential quality in a work of art, the quality that distinguishes works of art from all other classes of objects.[1]

– Clive Bell

Take any object and you will find a person who thinks it beautiful and another one who does not. Take any two people and you will find that they have different taste. A picture can be beautiful to my eyes but barren or even repulsive to yours. But some things, say the music of Johann Sebastian Bach or the paintings of Vincent Van Gogh, have the approval of nearly all who hear or see them. All well-functioning humans having some things in common, it stands to reason that our judgments should sometimes converge.

But the question is not just whether some things generally strike more people as endowed with beauty than do other things. The answer to that question is trivially yes. The question is rather whether the thing really is more beautiful than the other thing: if that is somehow written in the stars, so to put it. If I say that some or another painting by Vincent Van Gogh is beautiful, is it possible that I am stating a fact? Is what I say the sort of thing that can be true or false?

Vincent Van Gogh, "Sheaves of Wheat".

Christopher Mole makes some useful distinctions here.[2] He calls anti-realism the view that any statement such as “the Mona Lisa is a beautiful painting” means not that the Mona Lisa is objectively beautiful, but only that the speaker of said statement is enjoying experiencing it; in other words, it is not intended as a statement of fact. There are two versions of realism. Subjectivity is the view that all facts involved in aesthetic assessments are subjective, in other words they are facts about the way in which a thing appears to the subject. Objectivity is the view that some of those facts are facts about the object being perceived. (A stronger version of objectivity would hold that all such facts are about the object being perceived.)

About the anti-realist view I will only say the following. It involves making the distinction between two sorts of judgments of beauty. One is a judgment of the effect that an object has on oneself, the subject, and is by definition subjective. The other is an “evaluative appraisal of objects”, as Chatterjee & Vartanian[3] put it, and can be either subjective or objective. The anti-realist holds that the former is all there is and can be in judgments about beauty. I disagree with that. I think that saying something is beautiful means, sometimes at least, partly at least, that there is something beautiful about that thing. For example, we sometimes admit to liking things – comic book movies, Brandon Sanderson novels, app games, 80s pop music, whatever – even though they are “objectively bad”. That would not be possible if only the former kind of judgment had existed. What I am concerned with here is the latter kind of judgment, and whether all facts in it concern the subject or whether some of them also concern the object.

Who can tell what is beautiful? How can we tell what is beautiful? These are among the most debated questions in the history of aesthetics.[4] They are part of a grand conversation that has lasted for thousands of years. Let us wade clumsily into it like the uneducated buffoons that we are …

The Case for Objectivity

Clive Bell wrote that aesthetics is the search for some quality that exists only in aesthetic objects and in all aesthetic objects.[5] That is a stronger claim than the standard version of objectivity which only requires there to be one or more properties in objects that contribute to their beauty. Is there something like that? There do seem to be some properties that many beautiful things have in common. The classical conception of beauty relies on the ideas of perfect proportion, order, harmony, symmetry and so on.[6] In computer vision, there are machine learning techniques that allow us to predict fairly well how people will assess the aesthetic quality of an image based purely on the image data.[7] For example, Hayn-Leichsenring et al. took a large dataset of Western oil paintings and extracted from these features like various colour measures, self-similarity, rule of thirds conformity and so on – features that previous research had shown were relevant to aesthetic judgments.[8] They then took data from another study in which a bunch of people had rated (1) how much they liked each painting and (2) its artistic quality (corresponding roughly to our distinction between expressing one’s pleasure at experiencing a thing and making an evaluative appraisal of that thing).[9] They found that people’s ratings on both questions correlated with some of the objective features.[10] Although this looks to me like an exploratory study, it suggests that there are some objective measures in artworks that partially predict what people think of them.

Here is another example. People seem to consider harmonious colour pairs more pleasant than disharmonious pairs, but the former activate the medial orbitofrontal cortex (involved in emotion and reward) whereas the latter activate the amygdala (involved in decision-making and emotional responses like fear and anxiety).[11] In other words, aesthetic judgment of colour (or at least disharmonious colour pairs) seems to be grounded in the physical properties of colours (or colour pairs). I expect one would find similar results with consonant and dissonant sounds, balanced and irregular proportions, symmetric and asymmetric faces and so on.

You may at this point be itching to mention that beauty does not reside solely in sensory stimuli. Some things, like mathematical demonstrations for example, can be beautiful (if one understands them) without relying on sensory stimuli (assuming they are not beautiful solely because of the shapes they take in mathematical notation). In mathematicians, beautiful demonstrations and the like activate the medial orbitofrontal cortex, which is both connected to emotion and reward and also activated in more common experiences of beauty, like those involving paintings or music.[12] So if beauty is determined by facts about perceived objects, would not these facts have to be quite different for different categories of beautiful things? But we do find some salient properties both in mathematical demonstrations and in paintings, say. Examples include harmony, simplicity, symmetry and logic.

Vincent Van Gogh, "Wheatfield under Thunderclouds".

In fact, not only are there a bunch of object-properties traditionally associated with beauty, but we can also point to these when talking about why we think a thing is or is not beautiful. In fact, we can not only explain why we find something beautiful, but we can even convince others, by calling to their attention these properties, that it is so. I could, for example, possibly convince you that some musique concrète is beautiful by directing you to focus on texture, timbre, form, rhythm, humour and playfulness instead of harmony or melody.

Imagine asking a group of prehistoric humans to judge which of two similarly sized stones is the heavier. Mass is an objective measure. It is determined by an object’s density and volume. But your prehistoric people do not have access to scales. They can only feel the weight of the stones in their hands. So although they will not always be correct in judging which stone is heavier, they may perhaps be correct nine out of ten times. Is that not evidence in favour of there really being something about those stones objectively that makes so many of the prehistoric people think one of them is heavier than the other? Is that not the case even if the feeling of heaviness and the judgement of what is heavier do not arise until after the touch sensations have reached the brain? It is possible for judgments that seem largely subjective to in fact be aiming at a quantity that is rooted entirely in the external physical world. That we cannot measure an object’s beauty other than through ourselves today is not alone proof that beauty is not objective.

The Case for Subjectivity

If beauty is objective, that means we ought in theory to be able to break down that object’s constituent parts and measure its beauty. It means the object has its beauty encoded in its form and matter regardless of whether a human has ever laid eyes upon it, or even whether life exists in the first place. Maybe those rocky valleys on Mars would still be beautiful even if there never was any life in the universe – beautiful in potentia. But it is hard to shake the idea that the beauty of a valley has something to do with human cognition. David Hume contrasted sentiment with judgment, and argued that the former is always right because it is does not refer to anything beyond itself:

[A] thousand different sentiments, excited by the same object, are all right: Because no sentiment represents what is really in the object. It only marks a certain conformity or relation between the object and the organs or faculties of the mind; and if that conformity did not really exist, the sentiment could never possibly have being. Beauty is no quality in things themselves: It exists merely in the mind which contemplates them; and each mind perceives a different beauty. One person may even perceive deformity, where another is sensible of beauty; and every individual ought to acquiesce in his own sentiment, without pretending to regulate those of others.[13]

If it wasn’t obvious, there is a considerable body of research linking evaluative appraisals of artworks with various properties of the evaluating subject. That does not alone rule out objectivity (remember the example with the two stones), but the magnitude of interpersonal differences should cause us to take subjectivity seriously. People just often really, really disagree on what is beautiful! And the reasons that they do largely have to do with themselves, with their inner life.

People like paintings more if they think they come from prestigious museums.[14] People put higher monetary value on original artworks than on forgeries.[15] People who are educated in art judge it differently from people who have no relevant education.[16] There’s some evidence that various personality measures are associated with liking certain genres of pictorial art.[17] Experts pay attention to deeper features in the artworks (e.g. composition) in addition to surface-level phenomena (e.g. colour).[18] In general, our aesthetic judgments are based largely on an interaction between information – any information, not only sensory stimulus produced by the object – and our minds, I suppose maybe because meaning is rooted in connection?

Aesthetic appraisals of a wide range of sensory input (images, sounds, flavours, smells of widely varying type and quality) all activate the right anterior insula, which correlates with things like perception of inward feelings and with negative emotional experience.[19] This would seem to suggest that aesthetic appraisal of sensations is in fact a judgment of the subjective experience that the sensations produce, and not a judgment of the object being perceived. The right anterior insula is also thicker among people who meditate[20], which tracks with the folk view that meditation involves awareness and a turning inward and away from the external world. More generally, the evidence from neuroscience seems to point in the direction of aesthetic experience involving an inward focus.[21]

(I think results like these are based on studies where subjects are usually simply presented with these stimuli, without an accompanying task beyond rating the artworks. It is possible that they were only experiencing the sensory input, without at any point making a conscious evaluative appraisal of the source object of that input, but that making such an evaluative appraisal would still be possible for them, instead of or in addition to the involuntary subjective experiencing. That is, maybe they saw a painting, felt some emotion in response to it and reported that emotion, and neglected to evaluate it objectively simply because they were not prompted to do so.)

Vincent Van Gogh, "Tree Roots".

To believe that beauty is objective is to believe that any experiencing being, whether a human, a cat, a turtle or an alien, has reason to experience as beautiful any objectively beautiful thing – any thing that has properties we associate with beauty. The wildest, most different mind, so long as it experiences things, should also experience these things as beautiful. If it does not – and we even know from experience that there are always some people who do not – then it has in some way failed to recognise the beauty of the thing. This feels pretty counterintuitive.

The Case for Preference Clusters

I said previously that the question is not just whether some things generally strike more people as endowed with beauty than do other things, because the answer to that question is trivially yes. Paul Graham took this observation and built a theory of aesthetics of his own around it. He argues: all humans have things in common. Everyone has different preferences when it comes to art, but because we all have things in common, our preferences will be clustered along some axes. Good art (and beautiful things, I suppose, but Graham writes specifically about art) is simply art that caters to these preference clusters. But different subsets of people – for example people who grew up in different cultures – have different preferences, too, on average. So when people say that a piece of art is good objectively, what they really mean is that it is good at catering to the preference clusters of the set that contains all humans. To measure the objective goodness of an artwork, we would just tally all people’s opinions of it.

This seems to have some kind of sense. But I don’t think it’s what most people mean when they think or talk about the objectively beautiful. (This objection is kind of elitist, but that has never stopped me before.) Graham’s theory seems to be in conflict with what people often mean when they talk about “good art”. For example, Maroon 5 has sold something on the order of 120 million records. Assuming the majority of these people actually like the music of Maroon 5, by Graham’s standards, theirs is not only good music, it is some of the best music we can come by on Earth. Is it?

I suspect that Graham is fighting a straw man. He writes:

I wrote this essay because I was tired of hearing “taste is subjective” and wanted to kill it once and for all. [...] In fact, one of the reasons artists in fifteenth century Florence made such great things was that they believed you could make great things. [...] [This] was not just a useful illusion. They were actually right.[22]

It probably makes sense to separate two similar claims here:

  1. What is good art is subjective, but people generally find some artworks (subjectively) better than others, hence these can be said to be better objectively.

  2. What is good art is subjective and there is no way to say that one work of art is objectively better than another.

Graham argues for (1) and against (2). But I don’t think that anyone who professes to agree with (2) would ever disagree that people generally find some artworks better than others. They would just say that this does not amount to those artworks being objectively good. It almost seems as if Graham is confused about the terminology:

A lot of philosophers have had a hard time believing it was possible for there to be objective standards for art. It seemed obvious that beauty, for example, was something that happened in the head of the observer, not something that was a property of objects. It was thus “subjective” rather than “objective.”

It seems incorrect to say that philosophers have had a hard time believing it was possible for there to be objective standards for art. In fact, for a long time it was the standard view (of beauty, at least, but that is a very similar philosophical question): “Until the eighteenth century, most philosophical accounts of beauty treated it as an objective quality: they located it in the beautiful object itself or in the qualities of that object.”[23] Examples of such philosophers were Plato, Plotinus and Augustine of Hippo.

Graham goes on to write:

But in fact if you narrow the definition of beauty to something that works a certain way on humans, and you observe how much humans have in common, it turns out to be a property of objects after all. You don’t have to choose between something being a property of the subject or the object if subjects all react similarly. Being good art is thus a property of objects as much as, say, being toxic to humans is: it’s good art if it consistently affects humans in a certain way.

But importantly, this is still a relativistic notion. It still means that there is no ground truth for what is good or bad art, because that depends on who you ask and at which time you ask them. We have countless examples of artworks that have been loved by many long ago, but are not now, and vice versa. In Graham’s account, we only ever derive the goodness or badness of art from other people’s views; the grounds for our beliefs are not ultimately rooted outside the human mind. So though Graham calls this an objective account, it is not objective in the sense that most people think of it, or, for that matter, in the sense that it involves facts about the object. It involves only facts about how artworks appear to people – it does not really matter to Graham whether a painting depicts an angel or a rat – what matters is what people like more. Graham’s account does nothing to establish a permanent, eternal truth about what is good art and what is not.

I found a better (and more modest) presentation of the preference cluster view, from which I also borrowed that name, in this comment:

We say that someone has good taste when their judgment is a good predictor of others’ judgment. These kinds of preference-clusters around some objects are about the closest we can get to saying that personal aesthetic judgments can be right or wrong. Nevertheless, the ultimate seat of aesthetic judgment is in the individual – i.e., the brain that experiences an aesthetic object and determines whether I like it or not is my own, with whatever states and inputs it possesses that make up the judgment – so I do say that actual aesthetic preference is neither true nor false. [...] You and I agree that there are preference clusters around some pieces of music, but we interpret the existence of those clusters differently. To you, they suggest a kind of groping toward some as-yet-unseen aesthetic truth – what we would like if we were like we are now, only better[.] To me, they are limited in their (even hypothetical) extent by both individual difference and by cultural difference – preference clusters only crop up reliably among people who are relatively similar to one another and share a lot of cultural common ground. In my view, even if we were much, much better, smarter versions of ourselves, aesthetic judgment would continue to vary as widely as the combined variance of human cultures and the traits of individuals.

Vincent Van Gogh, "Undergrowth with Two Figures".

Michon, Roulin, Van Gogh

The Life of Joseph Roulin, Pierre Michon’s novella about the postman Joseph Roulin and the painter Vincent Van Gogh, ends with the following passage:

Who can say what is beautiful and as a result, amongst men, is deemed worthless or worthwhile? Is it our eyes, which are the same, Vincent’s, the postman’s, and my own? Is it our hearts, which a trifle can seduce, which a trifle can dismiss? Is it you, young man sitting with your hat placed next to you chez Ambroise Vollard, talking animatedly about painting with beautiful women? Or you, paintings roosting in Manhattan, merchandise whose enlightened fads nourish the dollars, doubtlessly drawing them nearer to God as well? Is it you, Browning? Perhaps it’s you, Old Captain topped in blue, looking at a little heap of Prussian blue fallen on a road; it’s you, white beasts, learned and mute, whose very volumes we caress far from here on rue des Récollettes, who know exactly what three francs is worth; it’s you, crows flying up above that no one can buy, no one can command, which do not speak and are only eaten during the worst famines, whose feathers even Fouquier wouldn’t want in his hat, dear crows to whom the Lord gave wings of matte black, a cry that cracks, the flight of a stone, and from the mouth of His servant Linneaus came the imperial name Corvus corax. It’s you, roads. Trees that die like men. And you, sun.[24]

The narrator of The Life of Joseph Roulin is not asking who can say what is beautiful precisely, but something more like, how can we tell what is beautiful? or even, how should we tell what is beautiful? And he finds possible answers everywhere.

Is it our eyes – that is, is it in the way that the thing appears to us? in the sensory stimuli produced by the object? And already we are well within the domain of subjectivity. Or is it our hearts – is it in the emotions that the thing provokes? If so, beauty seems awfully inconstant, difficult to get a grip on, difficult even to talk about, let alone argue or persuade. Or is it you, young man talking about painting – is it the critics or the artists that know what is beautiful? Or you, paintings roosting in Manhattan – is it determined by the prices given to objects by the market? Or is it you, Browning – though here I am not sure what Michon’s narrator is getting at. Earlier in the novella, we hear the narrator speculating that perhaps Adolphe Monticelli did not attain the success he could have because “he’s missing the blow from the Browning sur le motif”. On July 27, 1890, Vincent Van Gogh shot himself in the chest with a revolver, though not a Browning; he died shortly thereafter. Maybe what the narrator is asking is, is beauty located in narratives about objects and those who create them?[25] Or perhaps it’s you, Old Captain topped in blue – though again I am not sure what the narrator means; possibly this refers to Roulin, the subject of Van Gogh’s paintings, who often wears his blue postman’s uniform. Or it’s you, white beasts – it is desire or hunger, for rue des Récollettes, a street in Arles, where Van Gogh lived from 1888 to 1889, and where he befriended and painted the Roulin family, was also the place of a brothel that Van Gogh sometimes visited, as he related in one of his letters to his brother: “I attended the inquiry into a crime committed at the door of a brothel here; two Italians killed two Zouaves. I took advantage of the opportunity to go into one of the brothels in the little street called ‘des Récollets’. Which is the limit of my amorous exploits vis-à-vis the Arlésiennes.” Or it’s you, crows flying up above that no one can buy, roads, trees that die like men, sun – it is in the contents of the thing, it is written in the stars. Because these sentences seem to describe one of Van Gogh’s final paintings, Wheatfield with Crows. John Berger wrote that “[t]he silence after a felled tree has fallen is like the silence immediately after a death”.[26] He also wrote that Wheatfield with Crows was the last picture Van Gogh painted before he killed himself[27] (though apparently there is some uncertainty about this). The painting depicts crows flying above fields of wheat, a road running through it and a sun hidden behind dark-blue clouds. I take the narrator of The Life of Joseph Roulin to claim that the beauty of this painting resides in the painting itself, in the motives that constitute it, or maybe more precisely in the way that those motives reflect the natural world.

Vincent Van Gogh, "Wheatfield with Crows".


  1. ↩︎

    Bell, Art.

  2. ↩︎

    Mole, C. (2016). Real objective beauty. The British Journal of Aesthetics, 56(4), 367-381.

  3. ↩︎

    Chatterjee, A., & Vartanian, O. (2016). Neuroscience of aesthetics. Annals of the New York Academy of Sciences, 1369(1), 172-194.

  4. ↩︎

    Sartwell, C. (2012). Beauty.

  5. ↩︎

    Bell, Art.

  6. ↩︎

    Sartwell, C. (2012). Beauty.

  7. ↩︎

    Deng, Y., Loy, C. C., & Tang, X. (2017). Image aesthetic assessment: An experimental survey. IEEE Signal Processing Magazine, 34(4), 80-106.

  8. ↩︎

    Hayn-Leichsenring, G. U., Lehmann, T., & Redies, C. (2017). Subjective ratings of beauty and aesthetics: correlations with statistical image properties in western oil paintings. i-Perception, 8(3), 2041669517715474.

  9. ↩︎

    ibid.

  10. ↩︎

    ibid.

  11. ↩︎

    Ikeda, T., Matsuyoshi, D., Sawamoto, N., Fukuyama, H., & Osaka, N. (2015). Color harmony represented by activity in the medial orbitofrontal cortex and amygdala. Frontiers in human neuroscience, 9, 382.

  12. ↩︎

    Chatterjee, A., & Vartanian, O. (2016). Neuroscience of aesthetics. Annals of the New York Academy of Sciences, 1369(1), 172-194.

  13. ↩︎

    Hume, D. (2019). Of the standard of taste (pp. 25-40). De Gruyter.

  14. ↩︎

    Chatterjee, A., & Vartanian, O. (2016). Neuroscience of aesthetics. Annals of the New York Academy of Sciences, 1369(1), 172-194.

  15. ↩︎

    Newman, G. E., & Bloom, P. (2012). Art and authenticity: The importance of originals in judgments of value. Journal of Experimental Psychology: General, 141(3), 558.

  16. ↩︎

    Chatterjee, A., & Vartanian, O. (2016). Neuroscience of aesthetics. Annals of the New York Academy of Sciences, 1369(1), 172-194.

  17. ↩︎

    Furnham, A., & Walker, J. (2001). Personality and judgements of abstract, pop art, and representational paintings. European Journal of Personality, 15(1), 57-72.

  18. ↩︎

    Chatterjee, A., & Vartanian, O. (2016). Neuroscience of aesthetics. Annals of the New York Academy of Sciences, 1369(1), 172-194.

  19. ↩︎

    Brown, S., Gao, X., Tisdelle, L., Eickhoff, S. B., & Liotti, M. (2011). Naturalizing aesthetics: brain areas for aesthetic appraisal across sensory modalities. Neuroimage, 58(1), 250-258.

  20. ↩︎

    Lazar, S. W., Kerr, C. E., Wasserman, R. H., Gray, J. R., Greve, D. N., Treadway, M. T., … & Fischl, B. (2005). Meditation experience is associated with increased cortical thickness. Neuroreport, 16(17), 1893.

  21. ↩︎

    ibid.

  22. ↩︎
  23. ↩︎

    Sartwell, C. (2012). Beauty.

  24. ↩︎

    Michon, The Life of Joseph Roulin.

  25. ↩︎

    There is one other, though more far-fetched possibility. The most famous Browning by far is Robert the poet. His son, Robert Wiedeman Barrett, was overprotected as a child. Nathaniel Hawthorne wrote of him: “I never saw such a boy as this before; so slender, fragile, and spirit-like, – not as if he were actually in ill health, but as if he had little or nothing to do with human flesh and blood. His face is very pretty and most intelligent, and exceedingly like his mother’s. He is nine years old, and seems at once less childlike and less manly than would befit that age [...] I wonder what is to become of him[.]” The boy studied painting and sculpture in Antwerp and Paris, and apparently owed much of his success to his father’s fierce promotion of his work. So maybe what Michon’s narrator is wondering here is, is it facts about the artist that tells us what is beautiful? or alternatively, should we find beautiful that which would be advantageous for us to find beautiful?

  26. ↩︎

    Berger, And Our Faces, My Heart, Brief as Photos.

  27. ↩︎

    Berger, Ways of Seeing.