Caravaggio Erased: The Wrong Name That Survived for 200 Years

ArtEcho: Exploring Art History and Critique
24 followers
16 Video Views·Aug 22, 2026
Video Summary
AI GeneratedThis video details the rediscovery of Caravaggio's *The Taking of Christ*, which was misidentified as a work by Honthorst for nearly 200 years. Through a combination of physical restoration, archival research, and provenance tracking, experts corrected a clerical error that had persisted across centuries and continents, eventually returning the masterpiece to its rightful attribution.
For almost 200 years, Caravaggio's name was missing from this painting.
Not from the paint, from the paperwork. In August 1990, Sergio Benedetti, a restorer at
the National Gallery of Ireland, arrived at a Jesuit residence in Dublin for what should have
been an ordinary cleaning visit. The dark, dirty painting waiting for him carried the name
Benedetti began to suspect that name was wrong. An older record made the problem worse.
It listed a Taking of Christ by Caravaggio, and a copy of that painting, as two separate objects.
The correct name had survived. It had simply stopped following the correct canvas.
How could the wrong name survive for almost 200 years? Could Benedetti prove that the canvas in
front of him was the original, rather than one of the copies? And what could finally defeat a
mistake that generations of records had kept alive? This is Caravaggio's The Taking of Christ.
Its subject is a betrayal. Its modern history turns on something quieter. A name copied onto a page.
So what exactly did the 1793 record copy incorrectly? The first answer is sitting in a
guidebook. Giuseppe Vasi published his Instructive Itinerary of Rome, seven years before the Mattei
family made that inventory. Vasi gave the painting to Gerardo delle Notte, the Italian name used for
The family inventory repeated him. It changed Caravaggio to Honthorst. It also described the
painting as square. The picture is not square. It is emphatically rectangular. The wrong painter
and the wrong shape travel together. Jonathan Haar's account of the archive concludes that the
inventorist relied on Vasi rather than making an independent judgment. The compiler copied a
instead of trusting the object in front of him. That gives us the entry point of the mistake.
It does not yet explain why the mistake won, because someone else looked at the
same painting in the same period and wrote down a different name.
Before either wrong name, a Roman household was keeping track of what it owned. Siriaco Mattei
commissioned the taking of Christ, and on the 2nd of January 1603, he paid Caravaggio for it.
Caravaggio gives the arrest no room to breathe. Judas presses into Christ. A soldier's polished
arm cuts across them. At the left, a disciple tears away so violently that only his head,
hands and flying red cloak remain. At the right, a man raises a lantern into a scene already lit
from beyond him. The Mattei wanted another one. In September 1626, their accounts record 12 scudi
paid to a painter named Giovanni di Attili to copy Caravaggio's taking of Christ.
Fifty years later, an inventory still kept the distinction clear.
One entry names The Taking of Christ by Caravaggio. Several pages later comes the copy.
Original. Copy. Two objects. Two entries. That is the danger inside Benedetti's hunch.
Recognising Caravaggio's composition proves very little when another painter was paid
to reproduction it. He had to identify this canvas, not merely this image.
The first physical clue appeared on a Thursday morning at the St Ignatius residence on Lower
Leeson Street. Fourteen Jesuit priests lived there. The National Gallery of Ireland was about
ten minutes away on foot. Benedetti was 47. He had spent 13 years in the gallery's restoration studio
and took occasional outside jobs to supplement his pay. Haar records that he had grown discontented
with his place at the edge of the art world. Then, an ordinary cleaning visit placed this
picture in front of him. The old frame gave him somewhere to begin. It appeared to date from the
late 18th or early 19th century. Along its inner edge ran carved gilt beads, like a string of pearls,
with some broken away. The frame looked capable of having travelled with the painting.
The name Honthorst disturbed another memory. Benedetti recalled a short note by Roberto Longhi
about a taking of Christ, sold to an English buyer under that name. If his memory was right,
the apparent mistake in Dublin was also a route back towards Rome. Then came the cloth beneath
the paint. A technical catalogue described Caravaggio's Maté, Saint John, on heavy hemp,
roughly eight to nine threads per centimetre. To Benedetti, the Dublin weave looked strikingly
similar, perhaps even cut from the same roll. A damaged frame, a half-remembered article,
a coarse piece of hemp. None could carry a Caravaggio attribution alone.
Together, they made the hunch too specific to dismiss and too dangerous to announce.
The paper trail should have stopped the mistake almost as soon as it began.
A German writer, Karl Friedrich von Ramdor, saw the painting while Vasey's era was spreading.
His guide names the scene as Caravaggio, then adds that others call it Honthorst,
which he considered less likely. There it is. A correct warning, alive beside the mistake.
But Vasey's name entered the family inventory. Von Ramdor's did not.
Once the official household record changed, the next person no longer had to inspect the painting.
The easier course was to repeat the inventory. In 1802, the original left the Maté collection
as a Honthorst. It was bought in Rome for William Hamilton Nisbet and carried to Scotland.
The sale joined the wrong name to the physical object.
A century later, an Edinburgh auction catalogue still called it Honthorst.
A handwritten Eight Guinea Mark survives beside the 1921 entry,
although the record does not securely prove who bought it that day.
That is how the error survived. A printed mistake entered the family record.
The family record governed the sale. Later catalogues repeated the sale.
But knowing how a name travelled does not tell us which canvas carried it. The Maté had owned two.
While Benedetti worked backwards from Dublin, Francesca Capaletti and Laura Tester were
working forwards from the Maté archive in Reconati. They were students, learning difficult
handwriting and rebuilding a vanished collection page by page. They did not know about the canvas
in the Jesuit house. Benedetti did not yet know how far their paper trail had reached.
The two searches were moving towards the same absence from opposite directions.
Capaletti and Tester found the 1676 inventory that separated Caravaggio's original from a copy.
In Astrubale Maté's account book they found the 12 scudi payment to Giovanni di Attili for making
that copy. They followed the original out of Rome and into the Hamilton Nisbet papers in Scotland.
Then they reached the 1921 auction and the painting disappeared. Benedetti had reached
the same auction from the Dublin side. His paperwork disappeared there too. When Benedetti
found their published research he contacted Capaletti without telling her which painting
he was examining. Then he worked through the article line by line, footnote by footnote.
Capaletti and Tester had followed the original until they lost the picture.
Benedetti had followed the picture until he lost the record.
They had the history without the canvas. He had the canvas without its history.
Now the Dublin painting had to survive more than resemblance.
Cleaning exposed decisions beneath the dirt. Study of its surface, support and materials
could strengthen Benedetti's case or dismantle it. The comparison with the Maté St John mattered
because both supports were single sheets of good hemp. The similar weave did not prove authorship
but it gave the investigators a physical relationship they could test.
The paint supplied another kind of evidence. Caravaggio builds the scene through pressure
and overlap. Judas' arm and the soldier's arm double one another around Christ. The fleeing
disciple's cloak makes a red arch over the kiss. The figures have nowhere to withdraw.
Examination also revealed changes made during painting including a revision around Judas' ear.
This is the kind of evidence expected from a painting being worked out
rather than simply transferred from a settled design. Benedetti's hunch survived but the tests
could only support a Caravaggio attribution. They could not identify this canvas as the Maté
original. For that, the physical evidence still needed the broken paper trail.
Now the two dead ends could be placed against one another. The archive followed the original
from the Maté family to Scotland and lost it in 1921. The Dublin investigation followed a
honthorst backwards to the same sale. Beyond that auction, the link to Dr Marie Leah Wilson
remains incomplete. The film leaves it incomplete. But Leah Wilson later owned the painting and gave
it to the Dublin Jesuits. The gap remained but it no longer stood between two competing canvases.
It stood inside the history of the Dublin one. Benedetti could give Cappelletti and Testa the
canvas their archive had lost. They could give his canvas the original and copy history that his eye
could never supply. The case did not become stronger because one expert defeated the others.
It became stronger because three people could test it from different directions.
For three years after Benedetti's first visit, research, analysis and consultation tried to break
that agreement. No frame, thread, inventory or remembered article had to carry the attribution
alone. If the case survived, the Jesuit community would not merely own an old honthorst.
It would own a Caravaggio and that answer would create another decision. What finally defeated
the wrong name was not another name printed with greater confidence. It was evidence that
could be checked from both directions. Paint against paper, object against root, one investigation
against another. On the 16th of November 1993, the National Gallery of Ireland made the discovery
public. The name that had survived in the archive was attached once more to the painting.
Caravaggio was back. A recovered Caravaggio is more than an art historical answer. It is an object
worth an immense amount of money, owned by a religious community that kept it while the art
world did not know what it was. The Jesuits had received the painting as a gift from Marie-Lea
Wilson. Now they had to decide whether recovering its name also meant selling it. Father Noel Barber
argued that they had received it freely and should make it freely available. The Jesuits
retained ownership and placed the taking of Christ on indefinite loan to the National Gallery of
Ireland. The painting had travelled from a Roman palace to Scotland, from Scotland to a Dublin
doctor and from her to a Jesuit house. Its final short journey was the 10-minute walk to the gallery.
This time the right name travelled with it. The recovery needed all of it. A damaged frame,
a remembered article, coarse hemp, a payment in 1603, another payment for a copy in 1626,
the inventory that kept the two apart and the later inventory that let the wrong name in.
Benedetti made the painting answer questions about its materials. Cappelletti and Testa made
the archive answer questions about its identity. Neither search was enough until paint and paper
met on the same object. Caravaggio painted an instant in which one gesture fixes a name to a
The painting's later history turns on another act of naming, this time in an inventory.
That name held for almost 200 years. One copy helped make this confusion possible.
But what happens when copying stops being one clue and becomes the whole business?
That is the question inside the Bruegel copy factory.
