A nearly nude muscular man holding a stringed instrument and bow. A two-headed dog stands before him.
Portrait of Cosimo I de' Medici as Orpheus, c. 1537-1539, by Agnolo Bronzino, Italian (1503-1572), 1950-86-1

When Conserving a Painting Reveals a Different Picture

Mark Tucker

Considering the layers of paint—and meaning—in a Medici portrait by Agnolo Bronzino

One of the most exciting moments in the examination of a painting is discovering that there is something hidden beneath the image normally visible. We may find drawing an artist made to set out their composition before painting, known as underdrawing. Or the artist may have made changes to their initial idea as they worked (pentimenti in Italian). Sometimes there is an altogether different painting hidden underneath. Aided by imaging techniques—x-ray radiography and infrared reflectography (IRR) being the most common—we often come across things artists assumed no one but they would ever see.

The discovery of underdrawing or of pentimenti gives a fascinating window onto the creative process and mindset of the working artist. Were they freely spontaneous, or deliberate and methodical? Self-assured and direct, or self-questioning and searching? Were the reasons for changes from first idea to finished painting the artist’s own, or motivated by the influence of others? 

When close examination or imaging techniques reveal an underdrawing or earlier stages in the creation of a painting, it’s almost like being able to look over an artist’s shoulder as they worked. Such discoveries take us back to moments of inspiration, excitement, hesitation, or struggle as artists made their ideas, sensations, and emotions visible, bringing their work ever closer to the perfection of form and expression they sought. Any large paintings collection will have a number of examples of images hidden beneath images, and the Philadelphia Museum of Art is no exception.

Let’s look at a particularly striking example of an artist re-working a painting that was already either completed or nearly so: Cosimo I de’ Medici as Orpheus by Agnolo Bronzino. In 2021 we lent this work to the Metropolitan Museum of Art, where it figured prominently in the major exhibition The Medici: Portraits and Politics 1517-1570. Returned to our museum in late 2021, it’s on view in Gallery 350.

I had the experience of studying, cleaning, and restoring this painting a number of years ago. Thought to have been painted (and as we’ll see, re-painted) sometime between 1537 and the early 1540s, it depicts a very young Cosimo I de’ Medici, around age 17, at the very start of his rule as duke of Florence. The head is Cosimo’s, anyway; the massive, heroic body is straight out of ancient Greece. In fact, it’s Bronzino’s direct visual quotation of the Belvedere Torso, a famous ancient sculpture admired and studied by many artists of the Renaissance and later periods.

A woman making sketches of a large marble torso in a room with grand pillars
Painting of an artist studying a cast of the Belvedere Torso positioned and lit very much like the torso in the Philadelphia Bronzino. Design, 1778-80, by Angelica Kauffman (British, 1741-1807) ©Royal Academy of Arts, London; photographer: John Hammond

We’re still not sure for whom Bronzino made this unusual portrait, or exactly why he painted the young duke as Orpheus, the poet, peerless musician, and prophet of Greek myth. Considering Cosimo’s complicated personal and political circumstances at the time, and various versions and possible interpretations of the Orpheus myth, the painting’s purpose is obscure, its tantalizingly ambiguous symbolism debated by scholars to this day.

Most basically, though, what we see is a scene from the story in which Orpheus descended to the underworld—you can see a lake of fire in the upper-right background—hoping to bring his beloved dead wife Eurydice back to the world of the living. On his descent, he had to get past Cerberus, the vicious three-headed dog-monster that guarded the world of the dead. He accomplished this by calming the beast with his enchanting music. In the moment shown he’s just done that. The hellhound’s heads are calmed (the third one is barely visible at the far left) and Orpheus has stopped bowing the stringed instrument (not the lyre of myth here, but an update by Bronzino to a Renaissance lira da braccio). That’s the moment we see now. But Bronzino started with a different idea.

X-radiograph, showing changes to the figure’s pose and costume, as well as some later repairs to the wood panel on which the portrait was painted. Image: Conservation Division, Philadelphia Museum of Art
In this reflected-infrared detail, the muzzle and teeth of Cerberus’s leftmost head can be made out just above where the neck and body of the musical instrument meet, and the former position of the bow is visible where it crosses the figure’s right hand. Image: Conservation Division, Philadelphia Museum of Art

When we took x-radiographs and infrared photographs, they showed that he had completely repainted parts of an earlier version to arrive at the one now visible. That first version showed a slightly earlier point in the story, before Cerberus had been calmed completely. Cosimo’s head stays the same, but there are a number of major changes from version one to version two (the one we see now).

In version one, the near head of Cerberus is shown prominently, still baring its teeth; Cosimo-Orpheus is still bowing the instrument, he has a strap across his shoulder and back, and he’s much more covered-up by drapery. (In fact, if you look closely at the painting, you can still see the pink strap showing through the repaint.)

In version two, Cerberus’s docile head is barely glimpsed in shadow at the far left.  Cosimo-Orpheus has stopped playing and is holding the bow placed between his legs. His left leg is lowered to make more room for the new position of the hand holding the bow, and the position of the instrument and the shape of its pegbox have been altered.

In 1985 I wrote an article for the PMA Bulletin about the discoveries I made while working on the painting, and I wanted to make an illustration that would give a clearer suggestion than the x-ray and IR images could of how Bronzino’s first version might have looked. I started to paint a small black-and-white watercolor based on what could be made out in the technical images, but the publication deadline was tight and I was only able to complete Cosimo’s head and some other parts of the figure.

I set the little unused illustration aside at the time, but finished it a few years later as a gift for a friend. For that non-scholarly purpose, I represented some elements, like the clasp on the shoulder strap, the drapery over the thigh, and the figure’s right hand in more specific detail than was actually visible in the IRR and x-ray images. Still, the main elements of the changes described above are represented.

Author’s watercolor suggesting the general form and arrangement of elements before Bronzino repainted areas to arrive at the final composition

We really don’t know what motivated Bronzino to change the moment depicted. But the difference between the Cosimo-Orpheus in version one, who has yet to overcome a challenge, and in version two, who already has, would certainly have had major symbolic significance in a portrait of the young duke as a rising political figure. There is also the question of why Bronzino removed drapery from the first version to bare more of the figure. This, along with the specific repositioning of the bow and re-shaping of the instrument’s peg box, were all pretty bold moves to make the painting more sensually provocative.

The art-historical research and interpretation around these questions is fascinating, but the questions were only raised because of the new information that came to light though the kind of close investigation of materials, technique, and process we undertake all the time as conservators in our work with the museum’s collection. Bringing finds like these into the conversation—revelations of the artist’s mind and hand shaping a work of art—is truly one of the great satisfactions of our profession.

Mark Tucker is the Neubauer Family Director of Conservation. He joined the museum in 1980 as an intern in paintings conservation and was in charge of the museum’s Paintings Conservation department from 1985-2015. He has authored and co-authored publications on Rogier van der Weyden, Antonello da Messina, Masaccio and Masolino, Pontormo, Bronzino, Rembrandt, Jacques-Louis David, Edvard Munch, and Thomas Eakins. He has co-curated PMA exhibitions on the Master of the Morrison Triptych (2004) and Thomas Eakins’s The Gross Clinic (2010), and collaborated with curatorial and conservation colleagues to organize the 2017-18 exhibition Old Masters Now: Celebrating the Johnson Collection.