The Velázquez in the basement

Yale University Art Gallery

When it was donated to Yale in 1925, The Education of the Virgin (ca. 1617) had been marred by areas of lost paint, damaged by abrasion over an old stretcher bar, and cut down—leaving the angel at the top headless. But the style and intelligence of the Spanish master Diego Velazquez shine through, says the former Yale art curator who discovered it. View full image

In late 2004, I came across a dirty and damaged painting in a storage facility at Yale. A large canvas, more than five feet tall and four feet wide, the painting had been badly abraded, and its surface was disfigured by areas of paint loss. It had been cut, so that at upper left one could see the hands and arms of a now-headless angel. Despite that, the picture struck me as a powerful work: boldly conceived, impressively painted, filled with brilliant details, and characterized by a serene power. Depicting the scene commonly identified as "the Education of the Virgin"—that is, the young Virgin Mary being taught to read by her mother, Saint Anne, and her father, Saint Joachim—it had been at Yale for nearly a century but had never been catalogued as anything other than "Anonymous, Spanish School, seventeenth century."

This summer, after six years of questioning, research, and discussions with other art historians, I published an article in the Madrid-based arts journal Ars, presenting the scholarly arguments for my conviction that the painting is a previously unknown work by Diego Velázquez (1599–1660), one of the greatest painters of the European tradition. I fully expected academic debate to follow, but debate that would slowly stretch over years before arriving at any common consensus.

I was thus unprepared for the front pages. "A New Velázquez Found in the Basement of Yale University" proclaimed the headlines of Spain's El País on July 1. The story was accompanied on the front page by a large photograph and continued inside the newspaper with a double-page spread. This was just the beginning. Over the next few days the story came to be covered by virtually every newspaper in Spain, by the Guardian in Great Britain, and by newspapers from Argentina's Clarín to Zimbabwe's NewsDay.

One never anticipates that long, careful, academic research will become headline news. I'm sure that I've worn a rather bemused expression over the past two months, as I have been overwhelmed with requests not only for comments about the painting, but also for photographs of myself; the story of someone finding a Velázquez seems to capture the public imagination as much as the painting itself. This is my own version of that story: not the scholarly arguments for the attribution, but rather the more personal tale of the discovery.

The Education of the Virgin has never been on public view. Professors, curators, and conservators are known to have looked at the painting from time to time, but it seems not to have generated very much curiosity. To be fair, the basement storerooms of the Old Art Gallery building were a bit dark and crowded, hardly the best place to study a large, dark, damaged painting. (Some media accounts have implied that Yale kept the painting in some kind of dark closet where high humidity caused the paint to fall from the canvas, but this is pure fabrication. All the damage happened long before the painting arrived at Yale, and the basement was a proper museum storeroom.) In 1963, the Education had been sent to a conservator who cleaned off previous restoration work, but either he or the Art Gallery curators of that time decided not to undertake an extensive restoration to make the still-anonymous painting suitable for public exhibition. In my years as a graduate student at Yale, from 1993 to 1999, I never saw the work, even though early Baroque painting was one of my main interests.

I returned to Yale to work at the Art Gallery in 2002, and soon thereafter, the gallery started preparing for the renovation of its buildings (still under way). The paintings were moved to a new storage facility, and the curators began planning a complete reinstallation of the galleries. I thus came to theEducation of the Virgin. I cannot say that I knew what it was at first glance, but I knew that it was worth puzzling out. In the parlance of curators and art dealers, the work was by someone; it was attributable. Not only was it a work of great quality, but the painting was so confidently executed that it seemed to bear the signature style of a particular artist. I was looking at many paintings at the time, so the Education did not immediately monopolize my attention, but every few weeks, I would return and stare at it. A few ideas came to mind—both Spanish artists and artists working in Naples under the influence of the Spaniard Jusepe de Ribera—but nothing really held up.

Until one day, as I stood in front of the painting, it all fell into place in an instant. I distinctly remember the moment and my thoughts: "Wait. No. I know exactly what this is! This is early Velázquez! This is just like the Tavern Scene at the Hermitage, the picture in Edinburgh, the Saint Thomas in Orléans, and the group in London. This is just like everything in the Velázquez in Seville exhibition catalogue." Everything was familiar: the drapery arranged in heavy folds, the naturalism of the figures and still-life elements, the way the figures emerged from the dark, the color palette. Even some of the faces seemed to be those of the models used for other paintings.

I immediately told myself I was insane. It couldn't be. It was too implausible. There was no way that I'd discovered a Velázquez. How could it have failed to attract notice, given that it had been at the university for close to a century?

Over the next weeks and months I tried every alternate hypothesis I could imagine. We brought the painting into the conservation laboratory and analyzed the materials. (At first, I did not even let on to the conservator, Patricia Garland, that I had so grand a name in mind but merely indicated that I thought it might be "from Seville, around Velázquez's time.") The technical study nonetheless revealed that the materials—the colored pigments, ground layer, and canvas—were those found in other early works by the artist, and an X-ray demonstrated that the painting's execution, in long, confident strokes, was likewise consistent. I also began reading everything I could find about early Velázquez and Sevillian painting of the time. My initial hope of finding some reference to a missing work by the artist quickly dissipated as I learned that there was virtually no documentary evidence for any of his early paintings done in Seville. Yet I grew ever more confident. I found links to works by other Sevillian painters and further comparisons to Velázquez's own first paintings.

In addition to all the small bits of supporting evidence, moreover, I came to see that the mind behind the painting was the same that would, decades later, paint Las Meninas. The young Virgin Mary looks directly out of the Education, and her gaze signals to us that there is more to the story than immediately meets the eye. (That "more" involves a complicated theological debate about the Immaculate Conception then current in Seville.) In making us complicit spectators, the young Velázquez—only 18 or 19—already here demonstrates the same pictorial intelligence, the same probing psychology, and the same questions about the interaction of viewers and paintings that underlie his later masterpieces.

By the spring of 2005, I thought it time to begin showing the picture to a few more people. I sent an e-mail to a friend and colleague, Salvador Salort-Pons, a Velázquez specialist and associate curator of European art at the Detroit Institute of Arts. (He was then a curator at the Meadows Museum in Dallas.) We had corresponded about a number of seventeenth-century paintings, and my first e-mail about the Educationstated simply that I was sending a photograph of what I thought was an important painting, but that I did not want to prejudice his opinion by saying anything more. A few minutes and his reply arrived: "I am trembling!!!! That's a very important painting. I need to see it. No doubt: Spanish, Sevillian … But I am afraid to say." His two subsequent visits to New Haven, during which we studied the painting detail by detail, only confirmed his immediate reaction: this was Velázquez.

Not long after sending the photograph to Salvador, I deliberately left the photo on my desk when another visiting colleague asked to use my office to make a call, though I had not mentioned the painting to him up to that point. I stepped out of the office. I heard him pick up the phone, then put it down. Then his voice carried through the open door as he said to himself, thinking out loud: "This is not Giovanni Dò. This is not the Master of the Annunciation to the Shepherds. THIS is a BIG PROBLEM." Giovanni Dò and the so-called Master of the Annunciation to the Shepherds are two artists whom I too had considered as candidates for the attribution—but in an instant he had moved past them and was already sure that he was looking at Velázquez. Yet he also immediately realized that I was going to spend a lot of time trying to deal with the picture, and that it would be a controversial addition to the canon.

He was right. That was five years ago, and there was still much more work to do. One of the first problems was to establish the work's provenance. Its identification number at Yale, 1900.43, would normally have indicated that it came to the university in 1900, but I quickly discovered that that number had been assigned in 1959, when the picture turned up in an inventory and its donor was unknown. Around the same time that I learned this, I found documents indicating that in 1970–71 a graduate student, Humberto Rodriguez-Camilloni '67, '71MArch, '81PhD, had studied the painting. He sent photographs to one or two scholars and suggested that the work was not unlike the early Velázquez, but he received rather curt responses and appears to have pursued his ideas no further.

After further research, I discovered that in 1925, the School of Art had recorded a gift from Henry Hotchkiss Townshend, Class of 1897, '01LLB, and his brother Raynham, Class of 1900S, of "two paintings, oil on canvas, framed, Spanish, style of Murillo, religious subjects." There are no other paintings in Yale's collection that could fit this description, so the Education must have been one of them. Later board minutes and photographs confirm this provenance and document the work's transfer into the Art Gallery's permanent collection.

In 1925, the Townshend brothers had come into possession of the large house built by their grandfather and started renovating it. The Education of the Virgin, a large and damaged picture, and one that must have been at odds with the décor of the Gothic Revival house, was apparently given to Yale as that renovation proceeded. There is no record of how the painting had come into the possession of the Townshend family, but Henry and Raynham were the sons of Captain Charles Hervey Townshend (1833–1904). He was one of the great American merchant sailors of the nineteenth century and sailed repeatedly from New Haven to the Mediterranean; the most reasonable assumption is that the painting came from Europe aboard one of his ships.

The provenance, however, is just one piece of the puzzle, and it does nothing to prove the attribution. Similarly, technical study of the painting can confirm that the materials are those used by Sevillian artists of the early seventeenth century, and that it was painted in the same way that Velázquez is known to have worked, but material evidence cannot prove the case. Research adds support to the hypothesis that the work is by Velázquez, but there is not—nor should we expect there to be—a paper trail that connects the Yale canvas to a contract signed by the painter. The final judgment about the Yale painting requires a leap of connoisseurial faith. Five years of research simply confirmed what I felt that I already knew in a moment back in late 2004. At the end of the day, what really matters is that the work just is, in every stroke and every detail, comparable to the other early paintings by the artist.

When the journal Ars, which had run a series of articles on works reattributed to Velázquez, decided to publish my study this year, I expected that its appearance would begin the long process of academic debate. At 11 p.m. on June 30, however, the day before my article was to be published, I received the following e-mail from the news desk at El Mundo, one of the main Spanish newspapers: "We are hearing from another newspaper here in Spain that Yale University has just found an unknown piece of Velázquez at a basement. We will be really grateful if you can assure the info and give us further details on the discovery."

I still did not grasp that this was major news. I sent a short polite reply explaining that yes, I had been a curator at Yale, that I had found a Velázquez, and that my article on the painting was due out in Ars that week. I offered to answer further questions the next day. I then went to bed, with no clue of what was erupting in Spain. By the time I woke the next morning, my inbox was full. One of the editors at Ars sent me the front-page story in El País from her personal e-mail, because the Ars server had crashed.

The front-page attention seems finally to have passed, but the story is far from over. There have been calls for some sort of international symposium to evaluate the attribution, but these miss the point: the attribution will only be decided over time, as scholars have a chance to study the painting and give it their own long consideration. One major scholar in the field, for example, was initially convinced that the Education was an early work by Velázquez's slightly younger contemporary, Francisco Zurbarán; some months after telling me this, however, he called to say that after much more thought he had decided that Velázquez was the only possible attribution. These things take time.

Moreover, the painting still remains to be restored, and this too will be a process carried out in consultation with experts from around the world. There are some elements of the work that can never be repaired. There is no way to reconstruct, for example, the angel originally at the top of the canvas. Yet many other parts—the horizontal line of damage from the old stretcher bar, or the areas of lost paint in the faces and draperies of the figures—can be pulled back together, in the way that they are in so many other paintings of this date. Unfortunately, this conservation will move slowly, and the painting will not be on public view for at least two years.

Eventually, though, the Education of the Virgin will take a permanent place in the Art Gallery's installation. While it will never be possible to re-create the thrilling moment of first recognizing the artist responsible for the work, the first glimpse and long study of the painting's serene power are still experiences that await future visitors to Yale.