Did the National Gallery get it wrong?

Leonardo da Vinci, The Virgin of the Rocks

“Leonardo da Vinci (1452-1519) was an Italian polymath, painter, sculptor, architect, musician, mathematician, engineer, inventor, anatomist, geologist, and writer. His genius, perhaps more than that of any other figure, epitomized the Renaissance humanist ideal. He is widely considered to be one of the greatest painters of all time and perhaps the most diversely talented person ever to have lived.” 1

For these reasons, every painting attributable to Leonardo is considered a masterpiece among the World’s artworks, and any time a new exhibition of his work is put together, it generates a huge public interest. As 2019 is the 500 anniversary of Leonardo’s death, it has generated a number of exhibitions with one in Florence focussing on Leonardo’s relationship with the workshop of Andrea del Verrocchio; a retrospective at the Louvre, which owns five paintings; Leonardo’s drawings from the Royal Collection at Buckingham Palace, and an exhibition focussed on a single painting, “The Virgin of the Rocks” at the National Gallery, London.

This is the National Gallery’s second exhibition featuring Leonardo in a period of eight years. The exhibition of 2011, coming shortly after a lengthy examination and restoration of the gallery’s “Virgin of the Rocks”, provided a unique opportunity to examine this painting alongside its namesake, and similar composition from the Louvre Museum. Since 2011, further investigation of the National Gallery’s painting has been undertaken, and the new exhibition of 2019 reveals more details of what lies beneath its painted surface than was previously made public. The revelations are exciting and add further support to a case, which, unfortunately, the developers of the National Gallery’s exhibition, have, once again, not considered, or developed.

Leonardo’s Virgin of the Rocks is one of the National Gallery’s most important works. It is only equalled in importance, within the collection, by Michelangelo’s Deposition, and Piero della Francesca’s Baptism of Christ. And yet, the full importance of this painting has not been entirely recognised and acknowledged by art historians at large or assumed by the National Gallery in particular. So where is the problem?

The problem lies in the dating of the painting.

On the 25th April 1483, the Brothers of the Confraternity of the Immaculate Conception commissioned Leonardo da Vinci and two assistants to provide the paintings for an altarpiece for their chapel in the church of San Francesco Grande in Milan. The commission was for a central image showing the Virgin adoring the Christ Child. Leonardo responded with a composition showing the Virgin and Child with John the Baptist and an angel set in a landscape of rugged rocks, and to become known as the Virgin of the Rocks because of this feature. At some later date a copy was made, and seemingly both are by the hand of the master.

How there happens to be two versions of the same painting, one in the National Gallery and the other in the Louvre Museum, and which is the earlier of the two are questions the answers to which are unknown and have been subject to much speculation. The two paintings are very similar but very different. They are almost the same in form, and in composition. They differ in colour, in details, in lighting and in symbolism. These differences have never been successfully accounted for.

All that is known beyond doubt is that the painting of the Virgin of the Rocks now in the Louvre was in France by 1625 and that the painting of the Virgin of the Rocks now in the National Gallery London was sold by the successors of the Confraternity to Gavin Hamilton in 1785.

It is considered by art historians that the Virgin of the Rocks which usually hangs in the Louvre Museum is the earlier of the two and is dated at around 1483. The National Gallery’s painting is believed to be copy made by the artist and assistants and to date from about the 1495 to 1508.

This theory appears to have been universally accepted even though the Virgin of the Rocks which is now in London can be identified as the one that came from Milan, where you would expect the Confraternity’s painting to be, and the version in the Louvre has apparently always been in France, where Leonardo spent his last years as the guest of the French King.

In order to support the theory that the Louvre painting is the earlier, an elaborate hypothesis has been formed involving Leonardo delivering the Confraternity’s painting, then removing it, selling it to a private client and painting the London version as a substitute.  None of this is supported by documentary evidence.

All the evidence points clearly to the fact that the London painting is the earlier by ten years, and that the Louvre painting was created for a different client. This is supported by the documentary evidence, the stylistic evidence, the symbolic evidence, by basic mechanics and simple logic.

The notion that the Louvre Virgin of the Rocks is “the original” and the picture in the National Gallery is a copy made as a substitute has become a statement of provenance for the two artworks, and is repeated without challenge by the British custodians of the painting in the National Gallery.  The fabrication that Leonardo sold the original painting in a pique of annoyance over the small size of the bonus awarded him has been seized by almost every member of the media who has done a story on the exhibition. The hypothesis is simply accepted as fact. 

Both art historians and the media continue to build upon the hypothesis, searching for symbolism within the paintings to support various theories, while at the same time selectively ignoring iconography that indicates so clearly which painting was done for the Confraternity of the Immaculate Conception and which was done for a different client entirely. 

The Virgin of the Rocks : Why the painting in the National Gallery London is the earlier of the two.

1. When the painting was commissioned by the Confraternity of the Immaculate Conception, they commissioned a team of three, Leonardo da Vinci, Ambrogio de Predis and Evangelista de Predis.
The painting shows signs of a hand other than Leonardo’s. Generally, signs of a hand other than that of the master’s would lead to the work being considered a second or workshop copy. In THIS case, it indicates the opposite. It indicates that this is the painting that fulfilled the original commission.

2. There is a trail of paperwork that supports this being the earlier work (“original”). Leonardo and his team pursued the Confraternity for payment. The Confraternity did NOT pursue Leonardo and his team for delivery of the work. The Confraternity complained that the painting was unfinished upon delivery, and after much negotiation, de Predis worked on it further, in order to be paid.

3. A story has been fabricated to support the idea that this is the second painting. This story is that Leonardo took the original and provided another, this being the second work. The National Gallery’s current version of this fiction is that the Confraternity “rejected” the first painting, for reason unknown. There is no evidence to support this claim. There IS evidence that Leonardo negotiated for twenty years to get the balance of the money owed by the Confraternity, and that he suggested that he might take the painting away. There is no evidence that this took place. On the contrary, the evidence is that the Confraternity retained the painting, and that negotiation continued.

4. The likelihood of Leonardo painting the work a second time to get a little more money is ludicrous. He claimed that the sum of money he was originally paid was spent mostly on the materials and certainly did not cover the completion of the painting, let alone painting a second version.
Leonardo would not have painted a second version unless he was absolutely guaranteed of getting paid for it. In other words, the second painting was second commission.

5. The claims that the painting in the Louvre was the “original” and earlier work were made at a time when the two paintings had not been seen together and good photographic reproduction was not available, let alone the scientific photo analysis available today.
The claims were reinforced by Martin Davies, Director of the National Gallery, from assessment made during during WWII, at a time when the painting in the Louvre was not available for examination. Martin Davies’ opinion was based, of necessity, on paintings within the National Gallery collection.
The works available for comparison included paintings from the workshops of Verrocchio, Ghirlandaio, Perugino and Botticelli. It was clear that the Leonardo had characteristics that differed from those paintings. When it is hung on the wall adjacent to them, as it was in 2014-15, it looks out of place.
Martin Davies, probably working from prints of the painting in the Louvre, and from memory, determined that the sweeter, prettier, more delicate painting was more in keeping with the Florentine style, and that the painting now in the Louvre may have been done before Leonardo left for Milan.
This is contradicted by the fact that we know that the painting was commissioned in Milan. We also know that the altarpiece to receive it had already been constructed before the painting was commissioned, and the painting had to fit.
Kenneth Clark did not agree with the Florentine date, but hesitated to directly contradict his successor at the National Gallery in print, over which was the earlier picture.

6. The 21th century examination of the painting showed two underlying drawings.
The lower drawing shows the Virgin Mary with her head turned to the left, her right hand outstretched as if in wonder or adoration and her left hand on her breast in a gesture of love. Her gaze is turned towards the Christ Child, the figure of whom was identified in a more recent scan of about 2018.
The second drawing relates directly to the present composition.

Changes to the composition, as well as pentimenti of a more minor nature, are generally considered as indicative that the work which bears them is earlier and more “original” than a painting which does not show these signs.
In this case, since the recent discovery of the underlying drawings, the National Gallery’s position has been that Leonardo “returned to the composition he had already created”. In the case of the pentimenti such as changes to the fall of the Angels curls, the NG has stated that Leonardo discarded these “changes” and returned to the arrangement of the “earlier” painting i.e. that in the Louvre.
This is nonsense. Both the underlying draught of a different composition and the pentimenti of the second underlying drawing indicate that the NG painting is the earlier of the two, and that the composition of the painting in the Louvre, and the fall of the Angel’s curls are dependent on the final arrangement of the National Gallery painting, not the other way around.

7. Three parts of the Louvre painting differ significantly from the National Gallery painting. These are the face of the Infant John, who looks younger than the Christ Child, the face of the Angel with eyes turned towards the viewer rather than to the Christ Child, and the hand of the Angel, pointing at John. The National Gallery’s position has been that these changes were made in the painting of the NG picture.

By a remarkable coincidence, a detailed drawing exists for each one of these three changes.
a) There is a draft of the baby’s face which fits the Louvre painting exactly.
b) There is a draft of the outward looking angel’s face of the Louvre painting.
c) There is a draft of the angel’s pointing hand for the Louvre painting.

The existence of these drawings means that it really is unbelievable nonsense to suggest that the three changes were made in the creation of the National Gallery painting.
The drawings confirm in the most convincing way imaginable that the Louvre painting is the later of the two.

8. Drawings exist that are directly related to the National Gallery painting’s two compositions. On a sheet of drawings in the Metropolitan Museum there are four drafts which indicate the develop of Leonardo’s ideas.
We can see him alternating between a Madonna Misericordia showing the Virgin with outstretched arms enclosing the adoring John, and the Madonna in Adoration of the Christ Child with one hand on her heart. Various element that occur in the final painting are present in these drawings, including the outstretched arm, the hand held above the Christ Child as if signifying that he is her son, the distant landscape, and the enclosed space suggested by a stable but transformed into rocks.

9. Martin Kemp has suggested that this sheet with compositional drawings is dated to the 1490s, because of the presence of an optical diagram, which we both agree is probably the earliest drawing on the page.
However, I disagree with Kemp that Leonardo’s interest in all aspects of optics began at such a late date. It is plain from the Adoration of the Magi that he had a strong interest in the closely related discipline of perspective drawing before leaving Florence.

When one rotates the page so that the optic drawing, which suggests someone gazing downward, is turned by 180* so that it appears at the top left of the page rather than at the bottom right, then one has a drawing suggesting someone gazing upward, perhaps into a painting above the altar in a chapel. I believe that the optical drawing is directly associated with the creation of the first picture i.e. that in the National Gallery, and is of a simple nature, not necessarily associated with a more detailed study of optics such as developed by Leonardo at a later date.

10. Style.
We now have the advantage of highly detailed colour photography to assist in the comparison of paintings. We can look at all of Leonardo’s known works in a matter of minutes.
While The National Gallery Virgin of the Rocks sits discordantly in the NG’s collection of similarly dated Florentine works, it forms a continuum with Leonardo’s other known paintings.

a) Ginevra de Benci c 1474-78 The face is pale, without strong shadow, but has the surface features of a painting from the School of Verrocchio- every concavity and convexity of the face is defined by slight shadows. There is a drawing by Verrocchio, Head of a Woman, in the BM that demonstrates this method of articulation of the facial contours. This characteristic appears in Leonardo’s earlier works, but not his later works. The other characteristic of this painting is the setting of the subject against a dark background.

b) The Benois Madonna c 1478, has taken a step towards the depiction of daylight.

c) Madonna of the Carnation c 1478-80, is notable for the strengthening of the light and shade that falls on the two figures. This is uncharacteristic of Florentine painting.

d) The Virgin of the Rocks, National Gallery, 1483, completed 1507-8. The subject matter has moved from an interior to a setting in nature. The play of natural light on the figures is a direct development of the intensified use of light and shade that Leonardo was developing in the two previous small Madonnas. The light and shade defines the facial contours, as per Verrocchio, rather than disguising them, as per Leonardo’s later use of sfumato.

e) Portrait of a Musician c 1490 uses the lighting to increasingly dramatic effect.

f) Last Supper c. 1495 – Leonardo has applied a less dramatic approach to lighting the figures in this painting, probably because of its size, position and purpose.

g) The Madonna and Child with St Anne (Louvre) c 1503 The Virgin of the Rocks in the Louvre is closer in treatment to this painting than to the Virgin of the Rocks in the National Gallery.

h) The Mona Lisa c 1503 Leonardo has moved from using light and shade to define the facial features to using sfumato to soften and disguise them.

11. Gesture.
We know that the subject matter of the Confraternity’s painting was intended to represent the Immaculate Conception- i.e. the One Immaculately Conceived. This is indeed the focus of the National Gallery’s painting, as Mary presents and blesses the Christ Child, an Angel supports Him, and John the Baptist worships him. The National Gallery’s painting fulfils the commission.

The painting in the Louvre, however, provokes a great number of questions, and suggests eccentricity and even heresy. Why is John the focus of the painting, when plainly, the focus in the Confraternity’s picture should be the Christ Child? Why does John appear younger and smaller? Why is the Angel directing us to the “wrong” child?
We know from “the Last Supper” that Leonardo used gesture for narrative, so this can hardly be ignored.

The simple and obvious explanation is that the Louvre painting is not the painting done for the Confraternity of the Immaculate Conception, but was created for an entirely different client, who wanted the emphasis to be upon the second child in the picture, not the Christ Child.

12. If we accep, on stylistic terms, and in the terms of the contract, that the National Gallery painting is in fact the earlier of the two, and painted around 1483, then we known of one wealthy prospective client who would have had opportunity to see the work, and commission the second.

This is King Charles VIII of France who was in Milan in 1494 and whose wife, Anne of Brittany, had given birth to the Dauphin, Charles Orlando, 11th October, 1492.

13. What evidence supports the painting now in the Louvre having been done as a commission for the King of France?
Firstly, while the Christ Child is clearly identifiable by the blessing that he gives, by His Mother’s gesture and the Angel’s support, the second child cannot definitely be identified as John the Baptist. He does not have the trappings of cross and scroll. He is smaller, not larger than the Christ Child.
But he is plainly a little person of great significance, as indicated by the pointing Angel.

14. Leonardo da Vinci employs symbolism in many of his paintings. The symbolism is not mysterious. It is enlightening. Ginevera di Benci is set against a juniper bush, Mona Lisa Gioconda identifies herself by her smile. Similar telling symbolism is to be expected in this commissioned work, and indeed, it is there to be found.
In front of the kneeling baby, in place of the lilies that would normally accompany an image of the Madonna, there grows the blue iris, the Fleur de Lys of the French Kings. A second plant can be just as easily identified- the violets, favourite flower of Anne of Brittany, sprout in front of the Virgin Mary.


When the National Gallery’s painting of the Virgin of the Rocks is accepted as the work commissioned and painted for the Confraternity of the Immaculate Conception, then it stands as Leonardo’s great statement of God’s spiritual Creation set in the context of His natural creation- the grandeur of the Earth. It also represents a forward move in composition, with a trapezoid shape dominating and containing the figures with an energy greater than the simple triangular form.

The National Gallery’s painting introduces light that has only been hinted at in Leonardo’s earlier works. The figures are bathed in the clear cool light of day. It is a light more intense and directional than Leonardo has employed before. It emphasises and models the features.
This characteristic of this particular painting was to influence not only all Leonardo’s students, but many other artists for centuries to come- Raphael, Andrea del Sarto, Carravaggio, Velazquez, Rembrandt, and onwards into the 21st century.

If we accept that the National Gallery painting was subject to the process of development, for which we have evidence, was worked on by Leonardo and his team, was delivered but not quite finished, and was the subject of dispute over the payments, then it becomes the work which was ALWAYS in possession of the Confraternity, until brought to England by Gavin Hamilton.

The painting in France has ALWAYS belonged to France. It was commissioned for the King of France, it represents the Dauphin of France, it contains the symbols of France, and, since its delivery, has never left the ownership of France.

Leave a Reply

Your email address will not be published. Required fields are marked *