How Could This Be Painted 600 Years Ago?

Jan van Eyck
1 follower
17 Video Views·Aug 17, 2026
AI Generated·May contain errors
There is a face in this painting that should not be possible,
not one of the two large faces in front. Look past them, through the columns out
beyond the little garden, to a low stone wall standing above a river.
A man is leaning on that wall in a red hat, his back half turned looking out at the city.
He is a few millimetres of paint, he has a shoulder line, a posture, a direction of gaze.
He was made around 1435 by a painter with no microscope, no camera and no magnifying
instrument that anyone has ever proved he owned. The mystery is not who he is, it is how he exists
at all. The painting is called the Madonna of Chancellor Rowland and it hangs in the Louvre.
Jan van Eyck painted it in oil on an oak panel about 26 inches tall and 25 wide,
small enough to hold in two hands. It was commissioned by Nicholas Rowland,
Chancellor of the Duchy of Burgundy and one of the wealthiest men in Europe,
for his parish church in autumn. That church burned down in 1793. The panel made its way to
Paris in the years that followed and then sat in the Louvre for more than two centuries without
ever being cleaned. That finally changed between 2021 and 2024, when conservators at the Centre
for Research and Restoration of the Museums of France stripped away the oxidised varnish that
had been yellowing the surface since before anyone alive was born. Then the Belgian institute Kik
Erpia photographed the cleaned panel at a resolution of three and a half billion pixels
and put it online, where anyone can now magnify a square centimetre of it until it fills a screen.
The Louvre built an exhibition around the result and structured it as a set of open questions.
The blunt one was this, why did van Eyck paint a background landscape so miniaturised that it is
almost invisible? But before we walk out to that parapet, we first need to understand what van
Eyck was actually doing with paint. Let's start with the room. Roland kneels on the left at a
prayer desk, hands pressed together, his book open in front of him. The Virgin sits on the right with
the Christ Child on her lap, an angel hovering above her, holding a crown. Behind them, three
arches open onto a garden and behind the garden, a whole river valley. Look at the tiled floor.
It runs in a single unbroken band across the middle of the picture and it works as a border.
Roland's world on one side, the other world on the other. Now look up at the carved band of stone
directly above Roland's head. Those are not decorative shapes. From left to right they
show Adam and Eve driven out of paradise, Cain killing Abel and the drunkenness of Noah.
Behind him, the tops of the columns carry lion heads. Down at the base of the logia screen,
where column meets stone, there are rabbits squashed into the gap so tightly you would
never find them without magnification. The art historian Craig Harbison read all of this as one
connected programme. Pride, envy, gluttony, anger, lust. The seven deadly sins, or most of them,
carved into the architecture on the Chancellor's side of the painting
and entirely absent from the Virgin's side. A very rich man commissioned a picture of himself
meeting the Mother of God and the painter quietly built a catalogue of vices into the ceiling above
his skull. Out in the garden there are lilies, irises, peonies and roses, everyone identifiable
by species and magpies and peacocks that scholars are still arguing about.
Look at the book in front of Roland. It is open to a page with a large decorated letter D,
which almost certainly begins the words domine, labia, mea, aperies.
Lord, open my lips. That is the opening line of Matins, the first prayer of the day.
He painted a particular book open at the page a man would reach for first thing in the morning.
None of this was improvised and the method underneath it is the reason the parapet is
possible at all. Van Eyck started with a chalk ground, brilliant white, and drew his composition
onto it with a brush and a liquid medium. We know it was a brush and not a pen or a stick because
infrared imaging of his work shows the lines tapering and darkening at the tips where a
droplet formed. Over that drawing he built the image in translucent layers, glaze upon glaze,
using linseed oil that had been gently heated in advance so it would flow and level and leave
almost no visible stroke. Light travels down through those layers, strikes the white ground
underneath and comes back up through the colour. It is why his surfaces look lit from behind rather
than lit from in front. And the brushes were absurd when conservators cleaned Van Eyck's
portrait of his wife Margaret. They found a single hair still stuck in the black paint
at the left edge, shed from the brush he was working with. It measured about six millimetres.
Judging by its length and colour, it came from the tail of a squirrel or an animal in the weasel
family, the equivalent of a fine sable brush today. Now hold that against what happened next.
The restorers working on that same portrait had to do their retouching under a microscope.
With the smallest brushes manufactured for the profession,
to match marks Van Eyck had laid down with an unaided hand. He was not an ordinary craftsman.
By 1422 he was already a master with assistants working at the Hague for John of Bavaria.
From 1425 he was court painter to Philip the Good, Duke of Burgundy, on a salary that put
him level with senior courtiers and freed him from the rules of the painter's guild.
The Duke sent him abroad on errands that surviving records describe only as distant
and secret journeys, including a documented trip to Lisbon in 1428 to paint Isabella of Portugal
before her marriage was agreed. He signed and dated his pictures at a time when most painters
signed nothing, and on ten of them he added a personal motto, Al sich kann, as well as I can.
He wrote it out in Greek characters because he could, remember that phrase,
it comes back at the end of this. There is also a running argument about where his eye was trained.
A prayer book known as the Turin Milan Hours contains a handful of miniatures by an artist's
scholars called Hand G, painted with a delicacy that looks unmistakably Eyckian.
Many historians think Hand G is Van Eyck himself, which would make him a manuscript illuminator
before he was ever a panel painter, someone trained from the start to work at the scale
of a fingernail. Others disagree. Karel Hersel Krinsky published a full argument in 2015
that Hand G was not Van Eyck at all. The question is open, so here is the question that decides
everything. Was Jan Van Eyck the kind of painter who would put the suggestion of a man on that
distant parapet, or an actual man? Everything we can measure says an actual man. Infrared images
of this very panel show him editing. Rowland originally had a large purse hanging from his
belt, and Van Eyck took it away, presumably because a display of money was the wrong note
for a picture about salvation. The Christ child was originally drawn pointing down at the floor,
and in the finished panel, that hand is raised in blessing, turned toward the man who paid for
the picture. Even the back of the panel is painted, worked to imitate marble, on a surface no worshipper
in Orton would ever see. This is not a man who put vague marks in the distance and hoped.
So walk out there, past the arches, past the flowers, to the crenelated wall above the water.
There are two of them, two men standing at the battlements, both wearing chaperones,
the folded hood and turban headgear of the period. One faces the city, the other has turned partly
back toward us, and that second figure is wearing a red chaperone. Red is the detail. Van Eyck painted
one other man in a red chaperone, in a portrait now in the National Gallery in London, and that
portrait carries his motto on its frame and has been taken for generations to be his own face.
Which is why art historians have argued for a very long time that the two figures on this
parapet are Van Eyck and an assistant standing in the middle distance of another man's commission,
the way he tucked two reflected figures into the mirror of the Arnolfini portrait.
The Louvre did not settle this in 2024. It did something more interesting, it made those two
figures one of the questions the exhibition was built around, and handed the argument to the
public. But the two men are the smaller reveal, here is the larger one. Once you have found them,
look at what surrounds them. A river winding away, a bridge crossing it with traffic on it,
a cathedral on the virgin side of the water, a smaller church above Rowland's hands,
palaces and island, vineyards on the hillsides, fields, and then a mountain range dissolving
into haze. Every one of those things is painted at a scale where a whole building
occupies less width than the head of a pin, and every one of them obeys the same light.
The sun falls across the near garden and the far mountains from the same direction,
with the same softening as distance increases. Erwin Panofsky described the effect in a line
that has never been improved on. His eye operates as a microscope and as a telescope at the same
time, and in 2021 a researcher at the Universite de Lorraine named Gilles Simon proposed that the
eye had help. Simon is a computer vision researcher and he went after a problem that
had defeated people for a century. In 1905 the mathematician Carl Dehlmann showed that the
parallel lines in the Arnolfini portrait do not meet at a single vanishing point, but scatter
across a rough circular zone, which for decades was read as proof that Van Eyck's perspective was
intuitive rather than systematic. Simon applied a detection method built on the principle that we
immediately register what cannot be due to chance, and the scatter resolved. Four main vanishing
points appeared spaced at regular intervals along a slightly tilted vertical axis. He found
comparable structures in four other Van Eyck paintings. His conclusion was that Van Eyck
worked with an optical device, something like the perspective machine Leonardo sketched roughly
half a century after Van Eyck's death. A frame with a glass pane and an eyepiece except with several
eyepieces set along an inclined axis, letting the painter trace reality one horizontal strip at a
time and then join the strips. If Simon is right, the impossible coherence of that valley is not a
It is an instrument. However, this is where the ground gets soft, and it is worth being straight
about how soft. Simon's published analysis covers the Arnolfini portrait and several other panels.
It does not include the Roland Madonna. Nobody has run that test on this painting,
and the broader idea that 15th century painters traced projected images has been fought over for
The artist David Hockney and the physicist Charles Falco argued, from around 1999,
that painters from Van Eyck onward used concave mirrors and lenses to capture their subjects.
The response from historians was sustained and unfriendly.
The computer scientist David Stork analysed the same images and concluded they do not show
the distortions that mirrors and lenses would leave behind. Stork also pointed out that Van
Eyck's preparatory drawing for Cardinal Albergatti carries tiny pinprick holes,
which are neatly explained by a mechanical enlarging device, and not explained at all
by an optical projection. The art historian Yvon You made a documentary case. 15th century writers
discuss painting technique at length and never once describe this method. And Sara Schechner,
who curates the collection of historical scientific instruments at Harvard, examined
surviving 15th century glass and metal mirrors and concluded they were simply too crude to throw
an image anyone could trace. There is also a plainer answer sitting right there. Spectacles
had been invented in Italy in the 1280s and by 1352 a painter named Tommaso da Modena was already
painting a cardinal wearing a pair. A convex lens was ordinary equipment. Give a trained illuminator
a magnifying lens, a brush of a few squirrel hairs, oil that stays workable long enough to
be corrected and a court salary that removes any need to hurry and you may not need a machine at
all. You may just need a man who never stopped looking. And the changes underneath argue against
tracing anyway. Moving a purse, redirecting a blessing, those are the marks of a man arguing
with his own panel while the paint is still open. Two accounts of the same picture. In one, Van Eyck
built an instrument that let him slice the visible world into strips and reassemble it, which would
make him an engineer as much as a painter. In the other, he sat with a lens and a squirrel hair brush
and refused to stop, which would make him something harder to explain. Now go back to the motto,
Als ich kann. As well as I can. He wrote it on 10 paintings and for six centuries it has been read
as modesty. Stand in front of this panel and it stops sounding modest. The Louvre spent three
years cleaning it and came out the other side still asking why the landscape is small enough
to disappear. The two men on the parapet are still labelled a question rather than an answer.
We can photograph the thing at three and a half billion pixels and we still cannot tell you where
the limit of that hand was or whether anything was helping it. He told us he painted as well as
he could. Nobody has ever managed to work out how well that was. So tell me which one you believe.
A device built by a court painter with the duke's money and no obligation to explain himself.
Or an eye and a hand and a lifetime of patience with nothing in between. Leave it in the comments
and if you think the man in the red hat is van Eyck, say so.
Because I want to know how many of us think he signed this painting twice.
