What is Painting?

Paintings are surfaces marked so as to represent visible objects: that was the most widely accepted idea about them until the rise of abstract art in the early 20th century, and for some people it remains the most appealing. But even if there are many paintings to which the proposition no longer applies, it applies to all but a tiny minority of photographs. The technology behind the latter medium has been through several transformations since it made its public debut in 1839; but all the while, photography as a form of imagery has steadily expanded, so much so that now it is deeply ingrained in the texture of the daily life of nearly everyone on this planet. One obvious reason for this ubiquity is that if we need a serviceable representation of a certain object, we are likely to find a photograph of it more trustworthy than a painting. We expect the mechanically transmitted image to correspond more closely to neutral fact than the handmade. Try sending the insurers a painting of the storm damage to your roof.

So is this idea that paintings represent things merely a historical hangover? Perhaps the project of painting once had this function; then the camera took over the role, and painters had to cast about for a new job description? I want in this chapter to consider whether this really was photography’s effect on painting; in the first place, by setting both photography and painting within a wider single context, the representation of what is visible.

I suggest that before looking at the matter historically, we engage with this vast and potentially amorphous topic of visual representation from two contrasting angles. On the one hand, let us consider representations of faces; on the other, representations of spaces. When it comes to format, I shall make a picture framer’s distinction: ‘portrait’ or ‘landscape’? When it comes to content, I shall make a grammarian’s – between ‘subjects’ on the one hand, and ‘objects’ on the other. The point is not to allocate every picture a place in one of these two categories; the point is to develop our understanding of representation through these alternative approaches, and then relate them to each other.

The Makapansgat pebble, from Transvaal

More than any other sort of information, our eyes seek out signs of subjects – that is, things that look at us, even as we look at them. Newborn children respond to the likeness of two eyes and a mouth in preference to any other type of image. Faces are literally the first things in their world. The earliest object taken for a likeness that archaeologists have to date discovered is a pebble, carried to a cave by hominids three million years ago, resembling a head with two eyes. In this sense, subjects – of which faces are the chief sign and indication – are the primal, biologically crucial business of visual representation. We need to know about them as a matter of love, lust, company and fear. The markings borne by animals for defence or display likewise seem to represent subjects – the eyes that stare out from a moth’s wings or a peacock’s feathers, for instance.

The wings of a bull’s eye moth

Faces and bodies are overwhelmingly the chief concern of sculpture, which in many parts of the world – notably sub-Saharan Africa – has been a far more important mode of visual representation than picture-making. In pictures, even if we do not actually assume that a face will appear, we are at least always on the alert for the possibility of one. It is an inbuilt expectation in viewing.

A whole broad swathe of images comes forward to meet this expectation literally face to face. We could call them portrayals, though it is a broader matter than simply likenesses of individuals. The subjects may be historical – Arthur Conan Doyle – or they may be fictional – Sherlock Holmes. They may be human – the manager of your local leisure centre – or they may be divine – the sun god Apollo. They may be animal – a famous racehorse or a fierce heraldic lion. The sense that any of these are conscious subjects – that they are capable of looking at us, as we at them – is what, primarily, the portrayals of them convey.

There are many functional reasons for bringing a subject to mind. One is to identify an individual, by showing distinguishing marks – the facial features or other attributes – that can be matched against the subject on encounter. Hence the ID photograph and other portrayals by means of which authorities control individuals. Perhaps still more important are the functions of portrayal that work through intuitive, pre-rational response. You want something present before you, on which you can focus your love for someone absent. You want a focus for your lust, in the absence of a partner. You want a focus on earth for your devotions to a god up in heaven. These desires for imagery can be urgent and unruly, and the authorities may in turn attempt to control them. But the same authorities may also rely on patterns of automatic response to extend their reach. Through reproduced portraits – statues, posters, coins – a ruler will seek to impose his power on the population, by seeming to be everywhere among them.

Performing all these functions, a portrayal becomes a tool in the practical psychology of a society. As a tool, it can function well or badly. A bad portrayal fails – by muddling distinctive facial features or by omitting essential attributes – to bring the required individual or type of person to mind. Excellence in portrayal, however, does something more than merely bring the subject to mind. Catching sight of a compelling likeness, we say ‘That is him’; and, intuitively, we half-feel ‘He is that’ – as if, having given them the same name, we felt that the picture and the bodily individual had the one identity, were a single person. (Persona, it is worth remembering, originally meant ‘mask’ – the mask through which a classical actor spoke: the outward likeness, that is, of a face; the sign of a conscious subject.) Because the person is the portrait – or almost – the massed images of the ruler have their effect of controlling, and for the same reason they are destroyed in times of revolution. Because of this intuitive fusion of image and individual, many people will not have their portrait ‘taken’, for fear that something of themselves will be taken away from them, or fall into the wrong hands.

This is portrayal at its fullest: representation surpassing itself, turning into virtual presence. ‘The eyes follow you round the room.’ It is a ceiling of achievement that is about as constant as the human form. If it has a history, it is not one that seems to have a systematic, progressive development. It happened, for instance, that in the 1st and 2nd centuries, stonecutters in the Roman Empire and potters in Peru were producing portrayals (see pages 44–45) that look to us today equally ‘realistic’; that is to say, showing what we expect to see when we see a head in ‘real life’. (We will return to the meanings of ‘realism’ later.)

These achievements, however, emerged from entirely separate histories, and neither attained a permanent stylistic ascendancy. The realism of the Mochica potters, like that of the ancient Roman sculptors, seemingly made way for a more formalized, rhythmically patterned art. An account that equates representation with power of portrayal does not really explain such developments: it can only protest, inadequately, that the portrayals got worse. Realistic portraits seem to stare out at us from across the centuries, bypassing them, defying them: they cut through history. If we approach visual representation as a matter of portrayal we get a display of arresting products, but we get little idea of it as a historically developing practice.

What happens if we approach representation from the alternative angle, the ‘landscape’ format? ‘Represent’, we may say, is in normal usage a transitive verb, a verb that needs an object; objects, we might go on to say, are therefore what pictures in general represent.

When we start speaking about objects, we tend to think of ‘things’ – solid, detachable chunks of material stuff. If asked to give an example of what we are talking about, we might go for ‘the table at which I sit’, or perhaps something more compact like a pebble, a pencil or an apple. All of these examples are literally ready to hand: we can imagine grasping them, feeling their firmness at our fingers. They are solid and stable, with a definite overall shape. If you bump into them they will hurt you. For such reasons the table has become the object par excellence, the stock opening counter for thinkers, from David Hume to Jacques Lacan, reaching out before them for something to oppose to human consciousness.

Setting up objects in opposition to human consciousness, we tend to think that they are free of the deficiencies of consciousness; free of partiality and distortion, they are trustworthy, the ‘objective’ truth. A way to touch on this truth is to make a mark that is determined by it. We look before us, we observe what is offered to the eyes, we record it. Recording is objective, in that it is dictated by what is outside ourselves; but it is also selective and self-interested. Using lines, expanses of colour, dots, hatchings, sweeps of the brush and whatever else comes to hand, the observer captures certain facets of the wealth that is available to the eye. The way we use the word ‘capture’ – a polite synonym for grabbing – suggests how similar observational recording is to the action of the hand closing around something and making it a possession, part of the individual’s sphere of mastery.

Marble bust of a Roman, 1st century CE

The difference is that in recording you can, it would seem, have your cake and eat it. The primary presence is left untouched. The eyes, feeding purely on light, pass from a distance over the causes of the light’s variety, leaving no more trace of their passing than a ghost. The mind gathers up relations and differences in the light – among edges, tones and hues – and relays them to the drawing hand. As it does so, the mind habitually relates these items of information to its available knowledge of the structure of objects and the way they hang together as forms. Yet it may detach itself from the other responses it usually makes to the object it is inspecting. While drawing, an observer can scan things at arm’s length from their everyday claims for attention, dispassionately concentrating on the way that what is near at hand relates to what lies behind. It is rare that one thing occupies the whole view. Things fall away and stop, as far as the eye can see, at an edge. Beyond which is another thing. Beyond which, ultimately, is the sky – the absence of things, infinity. All the world and the heavens open up to the eye, from still life to landscape, from the nearest apple to the most distant star – all related and, by this process, brought into a kind of equivalence on the drawing surface. (As, for instance, in the landscape study by Paulus van Vianen on page 46.)

Mochica head, c. 1st century CE

Observation is primarily a tool-using skill, and as with other such skills – for instance, plastering or piano playing – some people have more aptitude for it than others. But as with these other skills, training is possible. Training sharpens the painter’s ability to pick out from the given scene elements that will be recognized in the context where he or she presents the picture. It breaks in the novice’s talent and gets it to perform for an audience. As a result, training within most artistic traditions has chiefly centred on the observation of one set of objects: the works of preceding artists.

Paulus van Vianen, A Forest Scene, c. 1600

In European painting, as it grew from the time of Giotto, training involved study of the statues that were the most visible relics of classical antiquity – its paintings having nearly all perished. By drawing from a plaster reproduction of some antique sculpture, the trainee learnt the elements of the human body that required recognition – its formal structure. At the same time he would become acquainted with standards of ideal beauty. It was only after long sessions with a dusty, chipped Apollo Belvedere that the trainee would graduate to the ‘life room’, where a living naked human would very likely be repeating some pose similar to the statue. All this before the student went on to engage with the world outside the academy walls.

Throughout this process, painters were being taught to approach humans as they would objects: to consider them as 3D forms that might appear to the eye at a variety of angles, and which could be described dispassionately. They were learning to interpret the task of portrayal as the task of rendering a figure. Figures at once stood out from the 2D surface, as potentially sculptural entities, and imaginatively activated it, turning its expanses into spaces that they occupied and might move through. And when portrayal was approached as observation by means of the figure, system could be applied to it. The painter could be instructed in anatomy – the internal structure of figures – and perspective – the structure of external appearances. To the extent that these studies of coordinating principle could be criticized, analysed and improved, painting could be said to have a progressive history.

When we consider the business of visual representation in the light of this progress, the display of its assorted products starts to appear as a developing practice. To return to the distinction made in the last chapter, paintings can appear as painting – an enterprise involving many contributors and evolving with its own historical momentum. Imagine the distinction as being like that between a marketplace and a big store; between a random assortment of items vying for your attention and a complex but purposeful enterprise with developing objectives. In the marketplace, a participant’s success is purely a matter of producing individually arresting offerings; in the big store, that success is equally a matter of impact within the internal dynamics of the institution. Thus, in terms of the institution called painting, success is measured in terms of ability to shape the vision of other participants around you and of those coming in your wake.

As with any institution, this cohesive practice exists insofar as people believe it to exist. Though Western painters and historians have customarily named Giotto, working at the turn of the 14th century, as the founding father of such a practice, it seems to emerge into full self-consciousness with Alberti’s short book On Painting, written in 1435. Alberti, with a sense of venturing into virgin territory, sets out ‘to explain the art of painting from the basic principles of nature’. At once, he attempts to found the art entirely on rational premises, and at the same time to give a sense of its miraculous possibilities. His optimism for this ‘noblest of arts’ is unbounded: ‘my successors will probably make it complete and perfect’.

This optimism is boosted by two striking achievements: Giotto’s demonstration that a picture can function like a window, and Brunelleschi’s recent invention of ‘artificial perspective’, a method to systematically organize appearances in such a painted window. Alberti explains this technique by means of a technical device which he calls a velum, a transparent veil or grid held up between the eye and the world. The velum converts the world into a picture. The principle is apparent in an illustration to a German textbook of 1531. As I said in chapter 1, what matters is not how much painters used this device in practice – probably pretty sparingly – but the way it came to lend their project an underlying intellectual legitimacy. The picture was a demonstration of how the world could be known; it was a model of the way that humans could make sense of it.

Hieronymous Rodler, A Draughtsman using a Grid, 1531

The Italians who drafted theories of painting based on the principle of perspective were followed by writers in northern Europe from 1505, when Jean Pélerin produced the first French perspective primer. But before this, starting from around 1430, some painters in Flanders may have leant on an alternative new approach to picturing the world. So at least argued David Hockney in his 2001 book Secret Knowledge. He suggested that when artists such as Robert Campin and Jan van Eyck painted panels in oils with an unprecedentedly precise level of detail, they were either inspired by, or directly made use of, the fine-grain images that concave mirrors could project upon flat surfaces. Hockney’s case has an intuitive appeal if you look at Flemish paintings and the change that came over them in the earlier 15th century, a change that was certainly extraordinary and which asks for historical explanation. As the title of his book declares, however, if any such techniques were used during this period – and a majority of art historians have disputed whether they were – they remained largely ‘secret’ and did not therefore impact on publically circulating theories of painting.

Robert Hooke’s picture-box, 1694

At length, however, the look of images produced by projection did feed into discussions of painting and of how the world was to be known. The projected images in this case were made not by concave mirrors but by the camera obscura. It had been known since antiquity that light passing through an aperture into a lightless space (a camera obscura, literally ‘dark chamber’) will project an image on a surface facing the aperture. In the 11th century, the optical scientist Ibn al-Haytham (or ‘Alhazen’) had established the underlying geometry of light rays, one on which Brunelleschi’s ‘artificial perspective’ would also depend. It was only in the late 16th century, however, that devices for tracing sharply focused projections became available, when inventors in the Netherlands and Italy started placing lenses in the apertures of camera obscuras.

Certain 17th-century Dutch painters – most notably, Vermeer – explored with fascination the new qualities of vision this technology opened up. By the end of the century, the camera obscura was becoming more widespread, in various compact versions. The ‘picture-box’ depicted in an English text of 1694 envisages a topographical artist tracing a cast, inverted image from the back of a translucent screen. It looks an awkward contraption: but its image transmission demonstrated that the world could, naturally and logically, be known as a picture.

A camera obscura (drawn by the author)

The picture-box was followed on the market by the ‘camera lucida’, the ‘physionotrace’ and a host of other mechanisms offering increasingly fine-tuned transmission of natural detail on to a surface for marking. These mechanisms began to make inroads into the purely manual craft of picture-making (witness the late 18th-century vogue for the portrait silhouette). It was only a matter of time, given the simultaneous scientific advances, before a method was found of chemically fixing the camera image on to the marking surface.

The two figures who independently converged on the process of photography – Louis-Jacques Daguerre, following Joseph-Nicéphore Niépce, in France and William Henry Fox Talbot in England – had both previously painted, in an age where ‘the artist’ and ‘the inventor’ were viewed very similarly. They were each looking for means by which to develop the representation of the world in terms of close-grained observation. How did the discoveries they announced in 1839 impinge on the business of handmade picturing? To answer that, we need to retrace the intellectual implications of the idea that picture-making could be modelled on the camera.

What we have not stated is that the camera, the darkened room with an aperture casting light on a back wall, is not solely a model for what the painter does with the world and a possibility for technological research and development. It also offered, for early modern scientists, a model for the eye itself: for back wall read retina. This large-scale analogy for the small-scale organ is noted by Johann Kepler in his defining book on optics, the Paralipomena, published in 1604. It follows that the eye is, like the camera, a picture-making mechanism: and thus what the mind receives from it are pictures.

A device for the taking of silhouettes, 18th century

The mind lies behind the eye, translating those pictures into knowledge. This was the principle developed by René Descartes, writing some forty years after Kepler and shortly before Hobbes. Like Hobbes, he considers the possibility that pictures may not be resemblances at all. These pictures that the eye offers to the mind are representations of the world; but they are not its original presence. How then does the mind know anything of the world itself? If everything the mind receives is a picture, what is behind the picture?

These are problems that increasingly preoccupied European thinkers throughout the 17th and 18th centuries. Solutions to them were offered not only by Descartes, but also by John Locke and David Hume; the most radical proposal, however, is that of George Berkeley. Writing in the 1700s, Berkeley claims that ‘the pictures painted on the bottom of the eye are not the pictures of external objects’ because there are no external objects; all objects only exist as they are perceived, and the world in its entirety only exists because it is perceived by God. The arguments of Berkeley’s New Theory of Vision are at once outrageous to everyone’s intuitions of ‘what things are like’ and, given the premises from which he starts, virtually irrefutable. When Kant sets philosophy on a new basis at the end of the 18th century, it is with a full concession to the idea that the ding an sich, the thing in itself, is unknowable. Mental representations of it are the only thing we can know.

The production of images by lenses and by the human eye, from Christopher Scheiner’s Rosa Ursina sive sol, 1630

We start to see why, in the light of this, 18th-century artistic theory turned from how the painting related to the world towards how the painting related to the painter. If the world was essentially unknowable, the latter relationship was the only one that made sense. Values could only be based on what happens within the mind. It was in this context that Romantic critics theorized about expression and the mind’s creative power.

To get a measure of the complexities caused by this change in thinking, consider Joshua Reynolds – a conservative, but highly sensitive theorist – addressing the Royal Academy in London in 1786:

If we suppose a view of nature represented with all the truth of the camera obscura, and the same scene represented by a great artist, how little and mean will the one appear in comparison of the other, where no superiority is supposed from the choice of subject.

How is it that ‘truth’ has come to be ‘little and mean’? Reynolds does not exactly intend to say that in contrast with the camera obscura the great artist is a seductive falsifier: Reynolds is upholding the value of painting. He means that the element that painting inhabits is mental, ‘the imagination and the feeling’, whereas the images that the camera offers – the ‘self-representations…impressed by Nature’s hand’, as Fox Talbot would later describe them – are uninterpreted and unthought. There is a profound separation between our view of the world and the world itself.

I have suggested how painting might be seen, from the 15th century onward, as a practice demonstrating how humans know the world. The relations between this practice on the one hand and the camera obscura on the other led to two developments, both impacting on the production of painting in the early 19th century. One was the new technology of photography. The other – which in fact predated photography – was the incoming ethos of imaginative expression. The one seemed to threaten the basis of painting. The other seemed to redeem it.

In this light, we can start to answer the opening question of this chapter: how did the emergence of photography relate to painting? We can approach it by considering how much painting’s product range was affected when photographic enterprises started up across the industrialized world during the 1840s and 1850s. While its technology was by present-day standards slow, cumbersome and hard to operate, and its images were monochrome, photography quickly made inroads on the types of representational task earlier discussed: portrayal and observation. Some portrait miniaturists, for instance, were put out of business, or else, like Sir William Newton who worked for Queen Victoria, they simply switched to the new medium. In such forms of recording as topography and reportage, photographs likewise presented factually more reliable alternatives to the hand-drawn steel engravings and lithographs that were in mass circulation during the 19th century.

Millet de Charlieu, Louis-Jacques Daguerre, 1827

Carl Ferdinand Stelzner, A daguerreotype of a gentleman, c. 1848

All those media, however, were reproductive. As such, they tended to fetch lower prices than the unique, hand-crafted products known as paintings. More and more, the bourgeois wanted these valuable items on their walls, and canvases led a 19th-century image boom. Portable pictures were valuable by and large because they had, since the Renaissance, been prime demonstrations of the fullness and richness of mental activity. The outward world might be the primary and the ultimate object of thought and knowledge, but comparisons with the camera obscura encouraged people to think of the mind’s workings as internal and self-reflective. We must have more within us, when we paint, than the simple, transmitted impress of external fact: on these lines, as we have seen, Joshua Reynolds delivered painting’s riposte to the camera, some fifty years before photography was discovered. Painting’s element was the soul, the sensibility, the imagination.

All this helps explain why the alleged cry of the painter Paul Delaroche in 1839 – ‘From today painting is dead!’ – was misplaced, both in terms of actual production and in terms of principle. But there is a complicating factor to the situation. Namely, that the turn towards internal, idealist models of painting which we associate with Romanticism in fact belonged with the era preceding the arrival of the photograph. The effusive, quasi-mystical manners that went with them in painting might have been innovatory in the late 18th century, but by the 1840s they were the stuff of stale cliché. In different ways, painters such as John Everett Millais and William Holman Hunt in London, Adolph Menzel in Berlin or Gustave Courbet in Paris were looking to supplant an outmoded idealism with a new factual rigour.

The archetypal Romantic painting: Caspar David Friedrich’s A Wanderer above the Sea of Fog, from around 1818. The indefinite receding depths of landscape, opening on to the infinite sky, provide, for Friedrich, the only adequate image for the mind’s immensity – its capacity to surpass any particular object in this world. Yet as a marker of the impossibility of fully grasping the infinity he reaches for, he sets a witnessing figure between the distance and the spectator, face turned away, blocking us from the view with which it inscrutably communes.

A generation later: Courbet’s The Stonebreakers of 1849, the canvas that made him the talk of Paris when he exhibited it at the Salon just over a year later. Again we see figures, like Friedrich’s, with their backs turned away from us: but not towards the infinite. Towards grit, stones, things. This is where the word ‘Realism’ – with a capital ‘R’ – enters the vocabulary, expressing a principle rather than a quality of lifelikeness: the first of the self-proclaimed ‘isms’ that would aim to define the project of painting over the next hundred and fifty years. The critic Max Buchon used the epithet to describe Courbet, and he willingly accepted it. What did it mean?

When we say ‘real’ we often just mean ‘This is what I want to emphasize as existing.’ ‘God alone is real’, ‘You’re not in touch with the real world’ – we use the word as a rhetorical emphasis. Likewise, if I call a painting ‘realistic’, I probably mean ‘It shows things in the way that I want to assure you they exist.’ But there is a characteristic experience from which this rhetoric draws its strength. Describing observation, I said that we tend to think of objects as ‘things’. Things, by definition, are ‘real’ – the adjective derives from the Latin res, meaning ‘thing’. It is a word that makes an appeal, not so much to the mind’s eye, as to the mind’s hand. A thing is what you could grab, grasp, feel; equally, what could bump into you against your intention. It is other than the mind. Reality, by this measure, is what I can point to in the firm expectation that you will see it and that we both could grasp it, and that it is independent of our acts of pointing and grasping.

Caspar David Friedrich, A Wanderer above the Sea of Fog, c. 1818

Gustave Courbet, The Stonebreakers, 1849

One reason we might invoke the term ‘reality’ is that grasping can come unstuck from pointing. For instance a trickster painter might point you towards what cannot be grasped – a master of ‘illusionism’ such as Zeuxis with his grapes or Parrhasios with his curtain. Another cue for appeals to reality might be a philosopher who claims that all you are pointing at is another position from which to point; that grasping can never be grasped. This, crudely put, was the position that the developing philosophy of mental representations, from Descartes to Berkeley to Kant, had argued towards; it was the position that set the terms for Romantic painters such as Friedrich.

It was against such an ‘idealist’ stance that Courbet pitted his work – in common with his contemporaries in philosophy, the ‘Positivists’, headed by Auguste Comte, and to some extent (even if his pictures were far more heftily physical) in common with the newly arrived photographers, recorders of the concrete and specific. In his Stonebreakers Courbet confronted the viewer firstly with facts, as an act of political reportage. These were two labourers whom Courbet paid to pose for him after seeing them at work on a country road. As a socialist, he wanted to push their sweaty graft in the face of the bourgeoisie attending the Salon: give the viewer something hard to push against in turn, in hopes of spurring him to a more constructive engagement with the needs of another social class. Much of the Realist painting that followed Courbet in later generations, down to images of labour in the post-colonial world such as Ignatius Sserulyo’s Coffee-picking – painted in Uganda in 1965 and studded with the actual beans of the coffee crop – has turned on comparable approaches to reportage, set towards comparable aspirations.

A ‘realism’ that involves paying poor people to pretend to labour, for the sake of an exhibit aiming to engage rich people’s feelings – an exhibit that, like any Romantic work, must address them in terms of the soul, the sensibility and the imagination – is one sort of paradox. But the paradoxicality of Courbet’s realism went beyond this surface of political aspirations. Characteristically, his paintings thrust forward the thicknesses of objects, building them up from the canvas with fat knifefuls of impastoed paint. Yet wanting to paint the swelling fullness of things, he had necessarily also to paint their emptinesses and hollows. Wanting to grasp nature in all its tangibility, he witnessed the ways that so much of it slips through your fingers – water, hair, clouds. Beset by these tensions, his canvases became outstandingly intense statements of what paint could and could not do.

Realist painting, then, this conceptual cousin to photography, belied its name: it was premised on a collection of as ifs – of illusions. As if the experience of the painting were the experience of the thing itself; as if the brush loaded with paint touching the canvas were the actual finger feeling the actual solid volume; as if everything that could be seen, near and far, apple and sky, could be touched in this way. Realism brings us up against the fact that a painting can never be a simple, single thing.

The whole Realist project, in the light of the history we have been considering, might be seen as a kind of defiant struggle against the prevailing historical and intellectual conditions. The poet and critic Charles Baudelaire, at one time Courbet’s supporter, came to characterize the Realist’s paradox thus: ‘I want to represent things as they are, or else as they would be, supposing that I were not to exist.’ ‘A universe without man’, he alleged, was the logical aspiration of the Realist painter; to be a camera; to be nature’s view of nature.

Gustave Courbet, The Source of the Loue, 1863

Against the Realist, Baudelaire set l’imaginatif, the man of imagination, who said: ‘I want to illumine things with my mind and project their reflection on other minds.’ This man of imagination was pre-eminently Baudelaire’s hero Eugène Delacroix, the most literary of painters, but the formula equally applied to a range of 19th-century artists who responded to the stimulus of landscape without claiming documentary status for their responses. It could speak for the Barbizon group of painters who began work in the 1830s, such as Charles-François Daubigny; it could speak for the innovators of the 1870s, the Impressionists.

Another major critic, Théophile Thoré, writing about the Barbizon school in 1844, proposed that ‘a composition exists at that moment when the objects represented are not there simply for their own sakes, but in order to contain, beneath a natural appearance, the echoes that they have placed in our soul’. Compare Claude Monet, explaining in the 1890s that ‘a landscape does not exist in its own right…For me, it is only the surrounding atmosphere which gives objects their real value.’ For Monet, ‘atmosphere’ was not a purely meteorological phenomenon; it was the medium in which what Thoré called ‘the echoes’ occurred. Monet thought of himself not as recording observable objects, but as expressing a stimulation of his soul – what he called a ‘sensation’.

This was a word on which painters and theorists in the 19th century relied considerably in mediating between the mind and the world. Whether or not the objective world – the ding an sich – was knowable in itself, human values resided not in such knowledge but in the quality of the objects’ impact on the senses – a quality that emerged most clearly if one could ignore preconceived ideas of form and ‘simply look’. The ‘impression’ of Impressionism was just such a primary, unmediated looking: the kind of picture it offered presented not the fact as it was publically known to exist, but the way that that original had stamped or ‘impressed’ itself on an individual sensibility.

As creatures of the later 19th century, the Impressionists, unlike the earlier Realists, had arrived at a considered response to photography. This generation offered the picture-buyer not only colour, but qualities of sensibility the camera might be thought to lack. Compare the way that Jean-Auguste-Dominique Ingres, who grew up before Daguerre, portrayed Madame Moitessier in 1851 with the way that his devoted admirer Degas, who grew up after, interpreted the portraitist’s predicament at the end of the 1860s (see pages 64–65). The proposition has shifted, both to accommodate and to outbid the camera. Ingres: ‘Here is the sitter, this precious individual, represented with a formal perfection befitting her great value.’ Degas: ‘Here are two remarkable individuals caught in a moment of encounter – the alert sitter and the no less alert portraitist.’ The purchaser of Degas’s work was offered, in the formula of the writer Émile Zola, ‘a corner of nature seen through a temperament’.

Like Monet or Degas, Zola’s oldest friend Paul Cézanne was also to rely on the idea of sensation to regulate his painting, but with strikingly different results. ‘I have very strong sensations’, he claimed. A determination to tap his own desires and a hunger to possess what he saw pervaded his output, from the clumsy fantasies of lust and violence he painted as a young man to the still lifes and landscapes of his old age. The story he took as foretelling his own predicament, Honoré de Balzac’s The Unknown Masterpiece, written in 1831 at the height of the vogue for Romanticism in France, tells of a painter who yearns to paint the ultimate, definitive, living likeness of a woman – a years-long struggle that ends in delusion, for all anyone else can see are ‘colours heaped up anyhow, with a mass of bizarre lines holding them down and forming a wall of paint’.

Claude Monet, Morning on the Seine at Giverny, 1893

Cézanne’s methods were indeed singular and obsessive. When I wrote earlier about pictorial observation, I described it as proceeding backwards into space, from the object at hand to the infinite distance of the sky, picking its way into the depth by relations and differences. Cézanne, however, effectively reverses this process in his later landscapes. Setting his canvas before the view, he plucks at the sky and the far mountain as primary realities jostling for his attention; what is near, on the other hand, becomes blurry, a matter of secondary conjecture. Relating and picking out the variety of what he can see becomes no longer a method of explaining how objects stand in space, but an engrossing activity of the mind with its own momentum, overriding whatever independent qualities the objects possess.

Jean-Auguste-Dominique Ingres, Madame Moitessier, 1851

Yet at the same time Cézanne seems passionately aware of the objects’ independence from his visual ‘sensation’. He knows that they are distant from him, that they are not actually ‘in’ his eyes. It is in this spirit that he is obsessed with trying to render their solidity – coming at them from one eye position and then from another, as if to ‘get round’ them. But what he comes away with, from this tantalizing encounter of hungry eyes with elusive nature, is a clutch of surface tugs and analogies.

Edgar Degas, Victoria Dubourg, c. 1868–9

From one angle, this is a frustration. Instead of giving you the full, real world, Cézanne turns vision into a ‘wall of paint’, an impenetrable flatness. It is as if there were nothing else; as if Berkeley were right, and we merely take away pictures of what Cézanne himself calls ‘the show which the Pater Omnipotens Æterne Deus spreads out before our eyes’. Balzac’s painter, in the story, struggled in vain: trying for total likeness, he ended with no kind of likeness at all, nothing but his colours anyhow and his ‘bizarre lines’. From another angle, Cézanne’s paintings offer an achieved harmony, a bringing together of all the eye may hold – figure and fruit and tree and sky – into a single mode of existence. Either way, his wall of vision still seems, over a century after his death, as significant a juncture in the project of painting as any since Giotto showed how pictures could act as windows.

Paul Cézanne, Mont Sainte-Victoire, 1902–4

Cézanne’s later work, done in Provence, started to get attention in Paris around the turn of the 20th century (he himself died in Aix-en-Provence in 1906). It prompted metropolitan painters to turn from their late-19th-century fascinations with qualities of sensation to tougher, more structural questions of how art related to nature. Around 1908, the young Georges Braque found a way to respond to Cézanne’s work by painting landscapes and still lives that tackled nature – addressing the world of objects in all their solidity and interrelation – but which did so explicitly through the tools of art, with linear brushmarks and hatchings that could in no way be mistaken for natural appearances. Braque’s friend Picasso took the challenge further, seeking ways to float a system of marks free of a body’s volume and outlines, while keeping alive the thought of the body itself. The two artists started introducing self-contained little pointers and hooks – a nail, a pipe, a printed letter – which could tether the picture’s tangle of notations to its basis in nature while preserving its insistently non-natural surface.

For other painters, such as Piet Mondrian, the promise of this line of investigation – Cubism, as it soon became known – was that it showed how a complex of non-natural marks could stand free on the canvas, superseding any obvious reference to the visible world. We cross here one of the paths of enquiry that led to 20th-century abstraction. While Cubism was a central current in the extraordinary cascade of artistic innovation that tumbled forth in the decade after Cézanne’s death, only a few of the many streams that ran onward into the interwar period can properly be labelled as abstract. Nonetheless, it became hard from 1910 onwards for any well-informed painter not to sense that in the 20th century, the representation of the visible world had been challenged at the root.

Georges Braque, Harbour, 1909

Pablo Picasso, Les oiseaux morts (Dead Birds), 1912.

Piet Mondrian, Pier + Ocean, Composition no. 10, 1915

I can suggest four overlapping ways of describing this challenge to figuration, all, as I see it, still evident in painting over a century later. Two carry over the 19th-century themes of sensibility and realism I have already discussed, while another forms a response to painting’s broader history. But I will start with the issue nearest to the grain of Cubist painting.

(1) Space and perspective. Theodor Schoefft’s canvas of c. 1850 of a court occasion in Lahore suffers in comparison with a Sikh painter’s account of a very similar event by being less legible as well as more dowdy. It is as if perspective shackled the European artist, with his indifferently dramatic clump of figures, whereas the painter within an Indian tradition was free to put before us such figures as mattered in a happily inventive recreation of the location surrounding them.

The latter painter’s mind moved round corners. Cézanne’s mind moved round vases. The experience we generally have of locations and objects is mobile. That’s how we recognize their solidity, and for this reason painters have sometimes called a break with single-point perspective a bid for realism. Some Cubist painting could be construed as such a bid. But whether or not painters wished to commit to ‘cubist’ notations – interpenetrating planes and angles that grappled like wrenches with the represented objects – the impulse to free up the viewpoint has been very widespread in figuration of the past century.

Painting depicting the signing of the Treaty of Lahore, India, 1846, c. 1847

Theodor Schoefft, The Dasahra Festival at the Court of Lahore, c. 1845–50

Paul Cézanne, The Blue Vase (detail), c. 1885–7

Freed from perspective, the painter’s mind might become a bird, swooping in and out of windows and over the town: Peter Lanyon’s paintings of the 1950s bear witness to such an ambition. A bird might see things panoramically, map-wise. But mapping might become the next birdcage, for any painter feeling a need for constraints.

Rackstraw Downes gives us an exemplar of such a self-vigilated freedom. With other artists, what is mapped is near space. What holds the pictured scene together is not lines drawn to a perspectival vanishing point but the boundaries of ‘the visual field’, that is the extent of what can be seen from any given head position. But this takes us to a second issue that has been significant in the past century’s figuration.

(2) Sensibility interpreted as memory. If mobility is intrinsic to our real-world visual experiences, then so is duration. Even were we to paint something seen only for an instant, we should interpret what we have seen by way of our cumulative visual memories, and a painting itself is hardly ever the work of an instant (unlike a photograph). Continuous and deep-echoing personal experiences of time lend significance to the particular figures and scenes that artists happen to encounter, without which they would have little reason to paint them. Considering such a private texture of memory, we are approaching from a slightly different angle the issue of sensibility or temperament that was so important to Monet and to Zola.

Pierre Bonnard, In the Bathroom, c. 1940

Pierre Bonnard was the painter who outstandingly translated the Impressionist interest in atmosphere into a concern with memory that infuses much figurative painting to this day. In his later paintings, from the 1910s to his death in 1947, Bonnard took passing moments in his private experience – for instance, the visual field tracked in In the Bathroom – and protractedly mulled over their potential, progresssively enriching the colours until they became painted lifetimes. Slow is good, painters have read his work to say, and as a justification of their protracted handicraft it has only become the more appealing as the pace of other image media has accelerated.

Peter Lanyon, St Ives Bay, 1957

Rackstraw Downes, IRT Elevated Station at Broadway and 125th Street, 1981–3

Adolph Menzel, The Artist’s Foot, 1876

(3) Realism interpreted as estrangement. Cézanne’s painting was informed by a vivid awareness of the solidity and self-sufficiency of objects in nature, as was Courbet’s before him, albeit with different results. This broadly realist instinct – ‘realist’, lower-case – was expressed in yet other ways by the 19th-century Berlin painter Adolph Menzel. With him, however, obsessive concentration on the object may hit on a sort of mystery: what is this object? Indeed, what is an object at all? This puzzlement, which perhaps always lurks the more we surrender to staring, came upon many painters in Menzel’s wake and invigorated interwar versions of realism such as Germany’s Die Neue Sachlichkeit (‘the new factlikeness’, commonly translated as ‘objectivity’). The early work of Lucian Freud, who came to England from Berlin, was couched in it.

But the impulse was far from exclusively German. It ran through the work of Chaim Soutine, who arrived in France from the Belarusian shtetl in 1913. Soutine’s landscapes furiously reimagine pictorial space – asking at once ‘What are these objects?’ and ‘Where might I be, in relation to them?’ It informed the extremist art of the Chicagoan Ivan Albright, who spent twelve years from 1931 depicting a single studio set-up, fanatically unpicking the grain of objects until their coherence was stretched to bursting. Giorgio Morandi’s still lives, painted in Bologna between the 1920s and 1960s, remain as quiet and laconic as Albright’s are bombastic and macabre. Yet both keep the question sceptically suspended: what is there to know in what we see?

Chaim Soutine, View of Céret, c. 1922

(4) Return to the primal. The resonances of memory and the indefinite inscrutability of the object world – not to mention the riddling dialectics of Cubism – might seem to weigh down the painter’s task. Aren’t there simpler and more joyous ways to conceive it? It turned out that at the turn of the 20th century, there were. This was an era when independent, supposedly ‘unsophisticated’ painters, working at a remove from both the academies and the avant-garde, offered the latter inspiration. In common with Bustos (see chapter 1), a parish clerk in Mexico, and with the Hungarian pharmacist Tivadar Csontváry Kosztka (see chapter 4), the customs officer Henri Rousseau in Paris took to oil paints with a wholehearted belief that they could deliver whatever he wished to render, whether this was the prose of a suburban road or the poetry of a jungle.

Henri Rousseau, The Football Players, 1908

The good faith and bold lyricism of Rousseau’s paintings had a deep impact on more educated artists looking for ways out of the impasses of sophistication. Félix Vallotton, Giorgio de Chirico, Diego Rivera and Carlo Carrà were among those who felt that Rousseau had reasserted painting’s primal language. Rousseau’s stiff-standing figures and objects, represented by firmly separated units of colour and modelling, became a rhetoric of artistic reinvigoration.

They remain so to this day, although further models for non-academic figuration have been brought into the repertory. Some come from folk religion: for instance, Frida Kahlo’s prototypes for her autobiographical paintings of the 1930s and 1940s originate in Bustos’s provincial Mexico – the ‘ex voto’ paintings that worshippers place in churches, recording traumas while thanking God for deliverance from them.

Philip Guston, Studio Landscape, 1975

But more important has been the way that the marginalized, unrespectable traditions of the grotesque and of caricature have kept feeding back into figure painting. Witness the Cubist impulse to strip observed bodies to minimal sign language overlapped with cartoon reductions and distortions of the figure, as can be seen in many works of the 1910s by Picasso. The phenomenon of graphic artists suggesting new tactics to impact-hungry painters runs some way back through history (one instance being the late-19th-century craze for Japanese prints); but a significant episode for many in the art world was the way that the sometime abstractionist Philip Guston leant towards the American cartooning tradition represented by George Herriman’s Krazy Kat, when he developed a new and provocatively crude approach to figurative painting in the late 1960s.

‘Leant towards’, because this was not quotation or ‘appropriation’ from mass-circulation comics, in the manner launched a few years earlier by Pop artists such as Roy Lichtenstein (see chapter 3): Guston was drawing freehand, with decades of studio experience informing the wry, tragicomic inventions he put forward. The examples of both Pop and Guston, however, fed into ‘bad’ painting, a label launched by a curator in 1978. This new ‘bad’ figurative painting – which has in various forms carried on into the 2010s – might archly lift images from print sources, or it might attempt a more gestural dumbness, aiming to reach for some primal force.

Alice Neel, Margaret Evans, Pregnant, 1978

Good-bad or bad-bad? The question of quality inevitably arises, since crudity can be a valuable ingredient in painting as well as a handicap. At best (as in early Cézanne), crudity might register a fierce determination to make something to look at. And if the painter lacks any such drive, why should we turn our heads to look? Quality of inspiration cuts across categories. ‘I become the person for a couple of hours,’ said Alice Neel, one of the 20th century’s outstanding portraitists, reporting how draining a sitting could be. ‘When they leave and I am finished, I feel disorientated. I have no self. I don’t belong anywhere. I don’t know who or what I am.’

An equally forceful output, at the other end of 1960s painting, came from the abstractionist Bridget Riley. Where the portraitist spoke of plunging deep into psychological discomfort, Riley spoke of landscape experiences opening her eyes to ‘the pleasures of sight’. These pleasures had ‘one characteristic in common: they take you by surprise’. And thus Riley ‘had to make something which had this essential quality of arriving as a “surprise”’. But there this inventor of vital abstractions speaks for the figurative painter also. The real that fails to surprise is less than real.

Bridget Riley, Blaze 4, 1964