What backs up this presumption? The ship is on the horizon; the sun is also on the horizon (though unseen), lightening it with yellow. Sunrise, we deduce, and the ship arriving with the sun, bringing a new day, hope and rescue; the black clouds overhead (very black) will soon disappear. However, what if it were sunset? Dawn and dusk are easily confused. What if it were sunset, with the ship about to vanish like the sun, and the castaways facing hopeless night as black as that cloud overhead? Puzzled, we might look at the raft’s sail to see if the machine was being blown towards or away from its rescuer, and to judge if that baleful cloud is about to be dispelled; but we get little help – the wind is blowing not up and down the picture but from right to left, and the frame cuts us off from further knowledge of the weather to our right. Then, still undecided, a third possibility occurs: it could be sunrise, yet even so the rescuing vessel is not coming towards the shipwrecked. This would be the plainest rebuff of all from fate: the sun is rising, but not for you.

  The ignorant eye yields, with a certain testy reluctance, to the informed eye. Let’s check ‘Scene of Shipwreck’ against Savigny and Corréard’s narrative. It’s clear at once that Géricault hasn’t painted the hailing that led to the final rescue: that happened differently, with the brig suddenly close upon the raft and everyone rejoicing. No, this is the first sighting, when the Argus appeared on the horizon for a tantalizing half hour. Comparing paint with print, we notice at once that Géricault has not represented the survivor up the mast holding straightened-out barrel-hoops with handkerchiefs attached to them. He has opted instead for a man being held up on top of a barrel and waving a large cloth. We pause over this change, then acknowledge its advantage: reality offered him a monkey-up-a-stick image; art suggested a solider focus and an extra vertical.

  But let us not inform ourselves too quickly. Return the question to the tetchy ignorant eye. Forget the weather; what can be deduced from the personnel on the raft itself? Why not start with a head-count. There are twenty figures on board. Two are actively waving, one actively pointing, two vigorously supplicating, plus one offering muscular support to the hailing figure on the barrel: six in favour of hope and rescue. Then there are five figures (two prone, three supine) who look either dead or dying, plus an old greybeard with his back to the sighted Argus in a posture of mourning: six against. In between (we measure space as well as mood) there are eight more figures: one half-supplicating, half-supporting; three watching the hailer with non-committal expressions; one watching the hailer agonizingly; two in profile examining, respectively, waves past and waves to come; plus one obscure figure in the darkest, most damaged part of the canvas, with head in hands (and clawing at his scalp?). Six, six and eight: no overall majority.

  (Twenty? queries the informed eye. But Savigny and Corréard said there were only fifteen survivors. So all those five figures who might only be unconscious are definitely dead? Yes. But then what about the culling which took place, when the last fifteen healthy survivors pitched their thirteen wounded comrades into the sea? Géricault has dragged some of them back from the deep to help out with his composition. And should the dead lose their vote in the referendum over hope versus despair? Technically, yes; but not in assessing the mood of the picture.)

  So the structure is balanced, six for, six against, eight don’t knows. Our two eyes, ignorant and informed, squintily roam. Increasingly, they are drawn back from the obvious focus of attention, the hailer on the barrel, towards the mourning figure front left, the only person looking out at us. He is supporting on his lap a younger fellow who is – we have done our sums – certainly dead. The old man’s back is turned against every living person on the raft: his pose is one of resignation, sorrow, despair; he is further marked out by his grey hair and the red cloth worn as a neck-protector. He might have strayed in from a different genre – some Poussin elder who had got lost, perhaps. (Nonsense, snaps the informed eye. Poussin? Guérin and Gros, if you must know. And the dead ‘Son’? A medley of Guérin, Girodet and Prud’hon.) What is this ‘Father’ doing? a) lamenting the dead man (his son? his chum?) on his lap; b) realizing they will never be rescued; c) reflecting that even if they are rescued it doesn’t matter a damn because of the death he holds in his arms? (By the way, says the informed eye, there really are handicaps to being ignorant. You’d never, for instance, guess that the Father and Son are an attentuated cannibalistic motif, would you? As a group they first appear in Géricault’s only surviving sketch of the Cannibalism scene; and any educated contemporary spectator would be assuredly reminded of Dante’s description of Count Ugolino sorrowing in his Pisan tower among his dying children – whom he ate. Is that clear now?)

  Whatever we decide that the old man is thinking, his presence becomes as powerful a force in the painting as that of the hailer. This counterbalance suggests the following deduction: that the picture represents the mid-point of that first sighting of the Argus. The vessel has been in view for a quarter of an hour and has another fifteen minutes to offer. Some believe it is still coming towards them; some are uncertain and waiting to see what happens; some – including the wisest head on board – know that it is heading away from them, and that they will not be saved. This figure incites us to read ‘Scene of Shipwreck’ as an image of hope being mocked.

  Those who saw Géricault’s painting on the walls of the 1819 Salon knew, almost without exception, that they were looking at the survivors of the Medusa’s raft, knew that the ship on the horizon did pick them up (if not at the first attempt), and knew that what had happened on the expedition to Senegal was a major political scandal. But the painting which survives is the one that outlives its own story. Religion decays, the icon remains; a narrative is forgotten, yet its representation still magnetizes (the ignorant eye triumphs – how galling for the informed eye). Nowadays, as we examine ‘Scene of Shipwreck’, it is hard to feel much indignation against Hugues Duroy de Chaumareys, captain of the expedition, or against the minister who appointed him captain, or the naval officer who refused to skipper the raft, or the sailors who loosed the tow-ropes, or the soldiery who mutinied. (Indeed, history democratizes our sympathies. Had not the soldiers been brutalized by their wartime experiences? Was not the captain a victim of his own pampered upbringing? Would we bet on ourselves to behave heroically in similar circumstances?) Time dissolves the story into form, colour, emotion. Modern and ignorant, we reimagine the story: do we vote for the optimistic yellowing sky, or for the grieving greybeard? Or do we end up believing both versions? The eye can flick from one mood, and one interpretation, to the other: is this what was intended?

  8a) He very nearly painted the following. Two oil studies of 1818, which in composition are closest of any preparatory sketches to the final image, show this significant difference: the vessel which is being hailed is much closer. We can see its outline, sails and masts. It is in profile, on the extreme right of the canvas, and has just begun a painful voyage across the painted horizon. It has clearly not yet seen the raft. The impact of these preliminary sketches is more active, kinetic: we feel as if the frantic waving by those on the raft might have some effect over the next few minutes, and that the picture, instead of being an instant of time, propels itself into its own future, asking the question, Will the ship sail off the edge of the canvas without seeing the raft? In contrast, the final version of ‘Shipwreck’ is less active, offers a less articulated question. The signalling seems more futile, and the hazard on which the survivors’ fate depends more terrifying. What is their chance of rescue? A drop in the ocean.

  He was eight months in his studio. Around this time he drew a self-portrait, from which he stares out at us with the sullen, rather suspicious gaze that painters often assume when faced by a mirror; guiltily, we assume that the disapproval is aimed at us, whereas in fact it is mostly directed back at the sitter. His beard is short, and a tasselled Greek cap covers his shorn hair (we only hear of it being cropped when he began the picture, but hair grows a long way in eight months: how man
y extra trims did he need?). He strikes us as a piratical figure, determined and ferocious enough to take on, to board his enormous Shipwreck. The width of his brushes, by the way, was surprising. From the breadth of his manner, Montfort supposed that Géricault used very thick brushes; yet they were small compared to those of other artists. Small brushes, and heavy, fast-drying oils.

  We must remember him at work. It is a normal temptation to schematize, reducing eight months to a finished picture and a series of preliminary sketches; but we must resist this. He is tallish, strong and slender, with admirable legs which were compared to those of the ephebe restraining the horse in the centre of his ‘Barberi Race’. Standing before the Shipwreck, he works with an intensity of concentration and a need for absolute silence: the scratch of a chair was enough to break the invisible thread between eye and brush-tip. He is painting his large figures directly on to the canvas with only an outline drawing for assistance. When the work is half done it looks like a row of sculptures hanging on a white wall.

  We must remember him in the confinement of his studio, at work, in motion, making mistakes. When we know the final result of his eight months, his progress towards it seems irresistible. We start with the masterpiece and work backwards through the discarded ideas and near-misses; but for him the discarded ideas began as excitements, and he saw only at the very end what we take for granted at the beginning. For us the conclusion was inevitable; not for him. We must try to allow for hazard, for lucky discovery, even for bluff. We can only explain it in words, yet we must also try to forget words. A painting may be represented as a series of decisions labelled 1) to 8a), but we should understand that these are just the annotations of feeling. We must remember nerves and emotions. The painter isn’t carried fluently downstream towards the sunlit pool of that finished image, but is trying to hold a course in an open sea of contrary tides.

  Truth to life, at the start, to be sure; yet once the process gets under way, truth to art is the greater allegiance. The incident never took place as depicted; the numbers are inaccurate; the cannibalism is reduced to a literary reference; the Father and Son group has the thinnest documentary justification, the barrel group none at all. The raft has been cleaned up as if for the state visit of some queasy-stomached monarch: the strips of human flesh have been housewifed away, and everyone’s hair is as sleek as a painter’s new-bought brush.

  As Géricault approaches his final image, questions of form predominate. He pulls the focus, crops, adjusts. The horizon is raised and lowered (if the hailing figure is below the horizon, the whole raft is gloomily engulfed by the sea; if he breaks the horizon, it is like the raising of hope). Géricault cuts down the surrounding areas of sea and sky, hurling us on to the raft whether we like it or not. He stretches the distance from the shipwrecked to the rescuing vessel. He readjusts the positions of his figures. How often in a picture do so many of the chief participants have their backs to the spectator?

  And what splendidly muscular backs they are. We feel embarrassed at this point, yet we shouldn’t be. The naïve question often proves to be the central one. So go on, let’s ask. Why do the survivors look so healthy? We admire the way Géricault sought out the Medusa’s carpenter and had him build a scale model of the raft … but … but if he bothered to get the raft right, why couldn’t he do the same with its inhabitants? We can understand why he fiddled the hailing figure into a separate vertical, why he added some supernumerary corpses to assist the formal structure. But why does everyone – even the corpses – look so muscled, so … healthy? Where are the wounds, the scars, the haggardness, the disease? These are men who have drunk their own urine, gnawed the leather from their hats, consumed their own comrades. Five of the fifteen did not survive their rescue very long. So why do they look as if they have just come from a body-building class?

  When television companies make drama-docs about concentration camps, the eye – ignorant or informed – is always drawn to those pyjamaed extras. Their heads may be shaven, their shoulders hunched, all nail varnish removed, yet still they throb with vigour. As we watch them queue on screen for a bowl of gruel into which the camp guard contemptuously spits, we imagine them offscreen gorging themselves at the catering van. Does ‘Scene of Shipwreck’ prefigure this anomaly? With some painters we might pause and wonder. But not with Géricault, the portrayer of madness, corpses and severed heads. He once stopped a friend in the street who was yellow with jaundice and told him how handsome he was looking. Such an artist would hardly shrink from flesh at the limit of its endurance.

  So let’s imagine something else he didn’t paint – ‘Scene of Shipwreck’ with the casting redistributed among the emaciated. Shrivelled flesh, suppurating wounds, Belsen cheeks: such details would move us, without trouble, to pity. Salt water would gush from our eyes to match the salt water on the canvas. But this would be precipitate: the painting would be acting on us too directly. Withered castaways in tattered rags are in the same emotional register as that butterfly, the first impelling us to an easy desolation as the second impels us to an easy consolation. The trick is not hard to work.

  Whereas the response Géricault seeks is one beyond mere pity and indignation, though these emotions might be picked up en route like hitchhikers. For all its subject-matter, ‘Scene of Shipwreck’ is full of muscle and dynamism. The figures on the raft are like the waves: beneath them, yet also through them, surges the energy of the ocean. Were they painted in lifelike exhaustion they would be mere dribbles of spume rather than formal conduits. For the eye is washed – not teased, not persuaded, but tide-tugged – up to the peak of the hailing figure, down to the trough of the despairing elder, across to the recumbent corpse front right who links and leaks into the real tides. It is because the figures are sturdy enough to transmit such power that the canvas unlooses in us deeper, submarinous emotions, can shift us through currents of hope and despair, elation, panic and resignation.

  What has happened? The painting has slipped history’s anchor. This is no longer ‘Scene of Shipwreck’, let alone The Raft of the Medusa’. We don’t just imagine the ferocious miseries on that fatal machine; we don’t just become the sufferers. They become us. And the picture’s secret lies in the pattern of its energy. Look at it one more time: at the violent waterspout building up through those muscular backs as they reach for the speck of the rescuing vessel. All that straining – to what end? There is no formal response to the painting’s main surge, just as there is no response to most human feelings. Not merely hope, but any burdensome yearning: ambition, hatred, love (especially love) – how rarely do our emotions meet the object they seem to deserve? How hopelessly we signal; how dark the sky; how big the waves. We are all lost at sea, washed between hope and despair, hailing something that may never come to rescue us. Catastrophe has become art; but this is no reducing process. It is freeing, enlarging, explaining. Catastrophe has become art: that is, after all, what it is for.

  And what of that earlier catastrophe, the Flood? Well, the iconography of officer-class Noah begins as we might imagine. For the first dozen or more Christian centuries the Ark (usually represented as a mere box or sarcophagus to indicate that Noah’s salvation was a premonstration of Christ’s escape from his sepulchre) appears widely in illuminated manuscripts, stained-glass windows, cathedral sculpture. Noah was a very popular fellow: we can find him on the bronze doors of San Zeno in Verona, on Nîmes cathedral’s west façade and Lincoln’s east; he sails into fresco at the Campo Santo in Pisa and Santa Maria Novella in Florence; he anchors in mosaic at Monreale, the Baptistery in Florence, St Mark’s in Venice.

  But where are the great paintings, the famous images that these are leading up to? What happens – does the Flood dry up? Not exactly; but the waters are diverted by Michelangelo. In the Sistine Chapel the Ark (now looking more like a floating bandstand than a ship) for the first time loses its compositional pre-eminence; here it is pushed right to the back of the scene. What fills the foreground are the anguished figures of those
doomed antediluvians left to perish when the chosen Noah and his family were saved. The emphasis is on the lost, the abandoned, the discarded sinners, God’s detritus. (Should we allow ourselves to postulate Michelangelo the rationalist, moved by pity to subtle condemnation of God’s heartlessness? Or Michelangelo the pious, fulfilling his papal contract and showing us what might happen if we failed to mend our ways? Perhaps the decision was purely aesthetic – the artist preferring the contorted bodies of the damned to yet another dutiful representation of yet another wooden Ark.) Whatever the reason, Michelangelo reoriented – and revitalized – the subject. Baldassare Peruzzi followed him, Raphael followed him; painters and illustrators increasingly concentrated on the forsaken rather than the saved. And as this innovation became a tradition, the Ark itself sailed farther and farther away, retreating towards the horizon just as the Argus did when Géricault was approaching his final image. The wind continues to blow, and the tides to run: the Ark eventually reaches the horizon, and disappears over it. In Poussin’s ‘The Deluge’ the ship is nowhere to be seen; all we are left with is the tormented group of non-swimmers first brought to prominence by Michelangelo and Raphael. Old Noah has sailed out of art history.

  Three reactions to ‘Scene of Shipwreck’:

  a) Salon critics complained that while they might be familiar with the events the painting referred to, there was no internal evidence from which to ascertain the nationality of the victims, the skies under which the tragedy was taking place, or the date at which it was all happening. This was, of course, the point.