Leonardo's Lost Princess

by Peter Silverman

  Cotte began by digitizing the portrait at a resolution of 1,570 pixels per millimeter, an extraordinary level of definition. At this resolution, the slightest nuances and tiniest details––the craquelure (network of fine cracks on the surface), the grainy surface texture of chalk or graphite, and even fingerprints––were perfectly visible.

  In a single scanning session, which lasted only one hour, the work was measured and captured in thirteen spectral bands, and the multispectral camera recorded and generated approximately 24 gigabytes of digital data. Cotte’s goal was to produce an image that contained additional information: what could be seen beneath the various paint layers, since many pigments that are opaque to ordinary visible light are transparent to infrared. Cotte explained:

  The study or analysis of images of such ultra high resolution and high definition gives the researcher a considerable weapon––a trump card, if you will. A few hours spent before a large computer screen, armed with images produced by a multispectral camera, can be invaluable, complementing the evidence obtained in a traditional conservation studio. The high degree of sophistication and extreme precision of multispectral images enable an in-depth study of the object’s physical characteristics, reducing the need for further direct contact, since the work is handled only once in the initial scanning session.14

  The results:

  Perfectly clear, ultra-high-resolution images

  Normalized (standardized) colors with a unique level of accuracy

  A broad spectral range, with infrared and raking light infrared images extended to 1,050 nanometers (instead of the 850 nanometers possible from film)

  Information that is more pertinent and discriminating

  With this new method of investigation, the colorimetric print was born. Works of art can now be compared on an accurate scientific basis. With digital reconstruction, multiple combinations of data can be explored from different points of view and for different purposes.

  Recalling his initial impressions of our portrait, Cotte wrote,

  When the portrait was first presented to us, it was described as a watercolor, a colored drawing, or, alternatively, as a drawing in wax crayons, chalk or pastels. No one was exactly sure of its technique. That the technique had not previously been accurately described is almost certainly due to the later restorations, which masked the media of the original drawing . . . the portrait is drawn on the smooth hair side of the vellum rather than the rough side. This provides an important clue, for some techniques can only be carried out on the flesh side, which results in an entirely different appearance.

  The multispectral images enabled us to identify accurately the media of the portrait. It is executed in the technique generally termed à trois crayons, that is to say, with finely sharpened pieces of natural black, red and white chalks. This mixture was combined with pen and ink, especially in the areas of hatching. Pen and ink was a favorite technique of Leonardo’s, as was red chalk (of which he may have been an early pioneer). Although generally associated with later periods (especially the 18th century), the trois crayons technique has been used for portrait drawings since the Renaissance. To be completely effective, it requires a toned or colored support to set off the white highlights. In early periods, artists often exploited the natural flesh colour of vellum or parchment, mixing red and white chalks on top of it to reproduce the sitter’s complexion.15

  Aided by his illuminating camera, Cotte was able to present a clear view of the media used in the portrait. He found that the basic contours and shadows were executed in black chalk, strengthened in pen and dark brown ink—once again, a typical technique of Leonardo’s. A detail at the junction of the subject’s bodice and neckline showed traces of strokes of all three chalks against the flesh-colored midtone of the vellum. The way in which the artist exploited the support is particularly clear in another detail: the eye, where the untouched golden tone of the vellum conveys the amber color of the subject’s iris.

  Cotte and his digital detectives carefully studied the innovative media technique, analyzing the chalks used (black, red, and white) and their mixture with pen and ink; examining the stylistic parallels with known Leonardo works; and cataloguing the period details of style and dress. They found that the interlace ornament in the costume corresponded to patterns that Leonardo explored in other works. They easily determined that there were noteworthy similarities between our portrait and Leonardo’s Lady with an Ermine. These included the modeling of flesh tones using the palm of the hand, the intricacy of the patterns of the knot-work ornament, and the treatment of the contours.

  Each of these pieces of evidence was significant in the slow building of a case. But another test of authenticity when studying Leonardo is whether there is mathematical precision in the anatomy. Leonardo’s rules of anatomy are well known and described in his notebooks and teachings. Most famous is the drawing Vitruvian Man, which depicts a male figure in ideal proportions. Cotte’s digital equipment allowed the level of high resolution that made this examination possible.

  It was Leonardo who first called the eye “the window of the soul,” and his fascination with the eye’s anatomy and transcendent function is obvious in all of his portrait works. For this reason, Cotte paid special attention to the eye of the lady in profile. “Some of this might seem obvious or universal,” he explained, “but the way each artist handles these features is more individualized than one might imagine, and their combination is unique. Leonardo, for example, consistently made the bottom of the eye’s iris coincide exactly with the edge of the lower eyelid, and it is surprising how few portrait artists took the time to render the delicate lower eyelashes of their sitter.”16

  Once again, Cotte was aided in his comparison by an earlier digitalization of Lady with an Ermine, Leonardo’s rendering of Cecilia Gallerani. Scrutiny of the two pieces showed a thrillingly identical treatment of each detail, including the outer corner of the eyelid, the fold of the upper eyelid, the contour of the iris, the lower eyelashes, the upper eyelashes, and the juxtaposition of the edge of the lower eyelid with the bottom edge of the iris.

  Recall that in her evaluation of the drawing (see chapter 2), Mina Gregori had also noted the eye, stating, “In my view, exact parallels in the brightness and transparency of the girl’s eye are only to be found in other examples in the drawings of Leonardo.”

  The anatomical details offered further proof through the unique Leonardo attention to measurement in the construction of his subject. Cotte found that the proportions of the head and face reflected the strict rules that Leonardo outlined in his notebooks. Yet even though these measurements added to the body of proof, there was much work still to be done. This magnificent process would open up the discussion between old school and new school. Old school uses connoisseurship; new school brings in technology as well. New school methods merely confirm or disprove; they don’t change the art or replace the eye. The eye still comes first.

  One of the most innovative aspects of Cotte’s camera technique was its ability to track restorations, separating them from the original and from one another. In the course of its existence, the portrait had undergone extensive restoration—not uncommon for works of this age. Cotte observed, “Because Leonardo was so innovative and adventurous in his exploration of new techniques, many of his works have suffered more than those of his contemporaries. (One thinks of the wall painting of The Last Supper or the lost Battle of Anghiari.) Even the portrait of Cecilia Gallerani reveals condition issues.”17

  With the aid of the standardized multispectral images and a cross-comparison of the false-color ultraviolet, and infrared, Cotte was able to construct a map of restorations, allowing a clear distinction between the original media and the pigments added later. He found that “in this case, the restorations––the retouching and reinforcement of the original lines and hatching––are, by comparison with Leonardo’s own handling, heavy and overemphatic; in a few areas, they compromise the reading of the work.”18

  Among Cot
te’s findings was that a thin layer of pink pigment had been applied by brush to much of the cheek area and forehead, using a system of hatching. “The restorer obviously aspired to be consistent with Leonardo’s handling, but, alas, no restorer could achieve the same degree of subtlety.”19

  Reinforcements in pen and ink carried out during restoration were easy to recognize, and Cotte was able to identify the ink used by the original artist (purportedly Leonardo) and the restorers. In the knot patterns on the shoulder, the restorer’s efforts to reinforce contour lines, details, and hatchings were likewise easily distinguished. Cotte saw that occasionally, the transparency of the ink in the redrawn areas allowed the original black chalk strokes underneath to show through, again highlighting the distinction between the artist and his restorers.

  “The restorer’s stroke is hesitant, sometimes wavers, and is irregular in its thickness,” he noted. “It is not always easy to read. The contrast between the labored, less coherent, and less logical hatching of the later intervention and the lively, correct, refined, and harmonious hatching of Leonardo, in which each stroke seems to have a precise role or significance, is abundantly clear from the multispectral images.”20

  Cotte also discovered that the later restorations were the work of a right-handed person. “The movement of the strokes, in contrast to those of Leonardo, starts with a rather timid––even hesitant––placement of the drawing instrument (brush or pen), then continues with a thicker line, with its several nervous wobbles, only to taper off again,” Cotte observed.21

  As part of his investigation, Cotte studied the ink. Although he acknowledged that ink cannot be scientifically dated in the same way as vellum, since the same recipes continued to be in use for centuries, he was able to determine that the ink was compatible with a fifteenth-century date.

  In the end, Cotte attempted to establish a firm distinction between the original and the restoration, creating a virtual reconstruction of the original based on a number of educated assumptions:

  The vellum, naturally soiled by the wear and tear of five centuries, would have been brighter.

  The colors throughout were probably less dirty, less grayish.

  The green was made from a piece of ampelite (a black substance like coal) on a yellow background.

  The red of the bodice was achieved with pigment from hematite on a yellow background.

  The entire bodice would have been rendered with the same red.

  Cotte also noted a tantalizing clue as to the portrait’s original usage. The remains of three needle holes along the left edge of the vellum showed that it originally came from a book or a manuscript. The piece of vellum was evidently cut from its codex with a knife. Small cut marks were visible a few millimeters in from the lower left edge. He judged that they could not have been made with a pair of scissors and that they were situated exactly at the fold of the gathering of the codex, below the lowest needle hole. It was one more piece of the mystery we were destined to pursue: What was the purpose of the portrait, and where had it originally been displayed?

  We sadly realized that the portrait could no longer hang on our wall for our private enjoyment. “This is scary,” Kathy worried. “It’s an enormous responsibility, and I don’t want it. Isn’t the purpose of a work of art to enjoy it? Art is made to be displayed, not to put away. But the whole thing makes me very nervous.”

  As we were deciding what to do, a friend told a story about Bill Gates and his Codex (a collection of scientific writings by Leonardo that Gates acquired at an auction in 1994). There was to be a show in France, and someone involved in the show described the security as intense, saying the Codex was always accompanied by a team of former Special Forces heavies. Furthermore, Gates’s home, outside Seattle, Washington, was a fortress with advanced electronic security systems. Each guest entering the property carried a microchip, which identified him or her throughout the house.

  When Kathy heard of these extreme security measures, she looked at me meaningfully and said, “Peter, the Codex is highly significant, but so is our portrait, in its own way.”

  At the urging of Kathy and others, I placed the drawing in a secure vault in Zurich, protected by armed guards and alarm systems. Until I knew for sure whose work it was, I wasn’t going to take any chances.

  In any case, Cotte no longer needed the original. He had all of the digital information required. Multispectral imaging had digitally stripped away the “noise” built up through centuries of aging and restorations. So many of the critical pieces fit together to form a high probability that it was the work of Leonardo. But high probability was not enough.

  Sitting at his computer, lovingly reviewing the photo enhancements of the portrait, Cotte’s curiosity grew. He needed to look into questions of style and period that were beyond the range of his camera. He could make the details visible, but he could not know all of the meaning. For that he needed another eye. It was time to bring in a big gun, a Leonardo expert of unquestionable authority. We discussed names and were inevitably led to Martin Kemp.

  6

  A Scholar’s View

  It had long since come to my attention that people of accomplishment rarely sat back and let things happen to them. They went out and happened to things.

  —Leonardo da Vinci

  Martin Kemp, emeritus professor of art history at Oxford University, was a leading Leonardo scholar. Slender, soft-spoken, and youthful in his sixties, he lived in something of a Renaissance cocoon: a historic eighteenth-century house near Oxford. It was there that he burrowed in to do the laborious work of examining the numerous pieces of art and photographs of art that were sent his way.

  Kemp was a star in the art history arena, and he had written extensively on imagery from art and science in the Renaissance. He had recently published, in 2004, Leonardo, in which he probed the real meaning behind Leonardo’s greatest masterpieces.

  Kemp’s passion was the relationship between art and science, and he had an unusual background for an art historian. He was trained in natural sciences and art history at Cambridge University and at the Courtauld Institute of Art at the University of London. In his career he has curated many exhibitions and has been the favored consultant for those exploring the intricacies of Renaissance style.

  Kemp’s explorations have also taken him in intriguing directions. Always fascinated by Leonardo’s scientific sketches, in early 2000 he was visited by daredevil skydiver Adrian Nicholas, who wanted to consult him on the construction of a parachute according to the specifications of Leonardo’s drawings, using authentic period materials. In 1485, Leonardo had scribbled a simple sketch of a four-sided pyramid covered in linen. Alongside it he had written, “If a man is provided with a length of gummed linen cloth with a length of 12 yards on each side and 12 yards high, he can jump from any great height whatsoever without injury.”1 Nicholas wanted to test the concept, and he figured the best person to help him was a bona fide Leonardo scholar.

  Kemp agreed to serve as a consultant, and in June 2000 the contraption was ready. The chosen destination was South Africa, where Nicholas launched himself from a hot-air balloon 10,000 feet high. He parachuted for five minutes as a black box recorder measured his descent, then he cut himself free of the contraption and released his Leonardo-inspired parachute. He made a slow, graceful, uneventful descent.

  Leonardo’s parachute design had worked flawlessly.2 Nicholas and Kemp were elated. “It took one of the greatest minds who ever lived to design it, but it took 500 years to find a man with a brain small enough to actually go and fly it,” Nicholas told the media, adding that “all the experts agreed it wouldn’t work—it would tip over or fall apart or spin around and make you sick—but Leonardo was right all along. It’s just that no one else has ever bothered trying to build it before.”3

  It was an exhilarating adventure for Kemp, who would grieve deeply for Nicholas when he died in a skydiving accident only five years later.

  Kemp’s fascination with the intuitive in
tegration of art and science made him the perfect choice to weigh in on the authorship of our portrait. But getting his ear was no easy task.

  “Since the publication of The Da Vinci Code, I was getting e-mails from people saying they owned the next Leonardo,” Kemp told me. “I call them the ‘Leonardo Loonies.’ Leonardo attracted lunatics more than any other figure—Shakespeare, Dante, Newton.”4

  Kemp liked to recount some of the wilder stories, including the theory that Leonardo had faked the Shroud of Turin and that the image was not Jesus Christ but Leonardo himself. But Dan Brown’s huge bestseller The Da Vinci Code, and the fantasizing it encouraged, drove him around the bend. In Kemp’s experience, especially in the modern media-saturated environment, if there was a conspiracy theory or a secret society, Leonardo was bound to make an appearance.

  So Kemp was naturally skeptical of all claims. When I first e-mailed him about a possible Leonardo find, he later told me, he thought, “Oh, dear, another bout of painful correspondence.” But after the digital file came through, he saw that it warranted a closer look.

  Kemp’s first task was to view the original, which was locked away in Zurich. He told us that the viewing would have to wait several months, when he planned to be in Zurich to make a television program. He would not go beforehand; it was his policy to accept no money, even travel expenses, for giving an attribution. He believed unequivocally that once one accepted money of any kind, one’s opinion could be bought.

  True to his word, some months later Kemp sat in a large viewing room at the bank where the portrait was being kept and waited for the two security guards (whom he humorously described as having big shoulders and no necks) to retrieve the portrait from storage. They laid it on the table in front of him and retreated to a corner of the room to wait in respectful silence.

  There is a mythology about art experts like Kemp that they have an unerring instinct, like an energy vibration, that tells them they are in the presence of a Master. Would Kemp have that moment of recognition? He himself had once described a similar sensation when viewing the real thing. “It was what struck you—that first jump of recognition,” he said reverently. “If you didn’t get that interior feeling that grabbed you, perhaps you would not pursue the matter. But the process was also more organic—involving other criteria and contextual support to ask, ‘Does this belong, does it tell us about Leonardo?’ ”5