Nevertheless, despite the quality of the works on display, the exhibition is a triumph even from a purely curatorial point of view. I’ve not been to a show where everything is so clearly laid out, properly - yet accessibly - researched, and just plain enjoyable. The catalogue is excellent, a model for all curators (and ideally you should read it first before seeing the exhibition). Faced with the challenge of so few Leonardo paintings in existence, the curator here, Luke Syson, has resisted padding out the exhibition with too much peripheral work. Instead he has allowed Leonardo’s brilliance to speak for itself. Each room of the exhibition is focused around one or two paintings by Leonardo, and each has a collection of related drawings and paintings by both Leonardo and those in his immediate circle. Contrary to Richard Dorment’s rather odd complaint in The Telegraph, those few paintings by other artists, such as Leonardo’s highly talented pupils Giovanni Antonio Boltraffio (d.1516) and Marco d’Oggiono (d.1524), are not a ‘major flaw’ in the exhibition, but help frame our view of Leonardo’s skill, since his own paintings tower over everything else from the period. 

In part, the success of the exhibition is due, ironically, to the scarcity of Leonardo’s paintings (he is thought to have painted only about 20, and 9 are included here), for his drawings play an equally starring role. Too often in monographic exhibitions an artist’s drawings are relegated to a separate area or room, tacked onto the display of paintings like an awkward relative. But in Leonardo, the drawings are arranged as the ultimate appetiser to the paintings. And thanks to the large number of Leonardo drawings that survive, along with his copious notes, we can chart the creative development of each painting, from an early doodle, to preparatory drawing, then cartoon, and then finally the painting. This is possible with very few artists, and probably Leonardo is the only case where we can chart the evolution of genius so fully. 

Picture: Leonardo, 'The Musician', oil on walnut, 44.7 x 32 cm, Veneranda Biblioteca Ambrosiana, Milan (99), De Agostini Picture Library. 

So, onto the works themselves. The exhibition begins appropriately with Leonardo’s ‘The Musician’ [above, Pinacoteca Ambrosiana, Milan] (his only painting of a male sitter), dated to 1486-7. This unfinished portrait (only the face and some hair can be considered complete) was ground-breaking in its day because it turned away, literally, from the convention of painting portraits only in profile. A profile portrait was considered to best show the nobility and breeding of a sitter, and was valued by rulers (particularly in Leonardo’s case Ludovico Sforza) for its instant recognisability, rather like a monarch’s profile head on a coin (how Leonardo, who himself painted profile portraits, though none survive, must have felt so limited by such a turgid convention). And The Musician, despite its unfinished state, bristles with originality, not least in its presentation of light and shadow on the sitter’s strangely alert face. The sitter’s hand and the musical score have sadly been abraded, due to having been over-painted at some point. The over-paint and unfinished state may explain why the picture has not always been accepted as a work by Leonardo - though it is hard to see who else could have painted it – and this is lucky, for when Napoleon’s army took away their haul of art from Italy they left The Musician behind, thinking it a mere Luini. 

Picture: Leonardo, 'La Belle Ferronniere', oil on walnut, 63 x 45 cm, Musee du Louvre (778) RMN/Frank Raux.

Inevitably, the question of attribution rather haunts the exhibition, as I shall explain below. Almost all of Leonardo’s paintings have at some point been debated back and forth by art historians. La Belle Ferronniere [Louvre] has often been ‘given’ (as Renaissance art historians are fond of saying) to Leonardo’s followers, though again it is hard to see why since it is evidently so much better than any work of, say, d’Oggiono. As so often with paintings from the Louvre, which has an aversion to cleaning pictures, one wonders whether the portrait’s darkened and dirty state may have led to doubts among scholars. The flesh tones are now practically non-existent beneath the yellowed varnish. A slight damage and crude retouching to the bridge of the sitter’s nose breaks the gentle rhythm of her features. Equally, the picture, like all Leonardo’s paintings, is overwhelmingly better in the flesh than in reproductions. I’m ashamed to say that even I have occasionally looked at photographs of La Belle Ferronniere and wondered what all the fuss was about. The craquelure, which seems so disfiguring in photographs, is hardly noticeable in real life, while the contours and delicate lighting of the face work in a way that the photograph can never relate. It is a testament to Leonardo’s skill that even the most advanced forms of photography cannot capture the brilliance of his work. (Doubtless Leonardo would have been intrigued by this, and set about inventing a new type of camera.) There are many suggestions in the catalogue as to who the sitter is, and it is one the ironies of history that the two most plausible are Ludovico’s wife, Beatrice D’Este, and his mistress Lucrezia Crevelli. 

Picture: Leonardo, 'Portrait of Cecilia Gallerani', oil on walnut, 54.8 x 40.3 cm, Princess Czartoryski Foundation, on deposit at the National Museum, Cracow.

Hanging in the same room as La Belle Ferronniere, and, if it is possible, entirely overshadowing it, is Leonardo’s Cecilia Gallerani, or ‘The Lady with an Ermine’ [Czartoryski Foundation]. It is not hard to see why this was the only exhibit which arrived under armed guard. The sitter, another mistress of Ludovico, is an undoubted beauty, and Leonardo would have been pleased to hear Luke Syson remark ‘it is hard not to fall in love with her’. And yet Leonardo manages to elevate her from a mistress (and how often do we see them portrayed as purely sexual beings) to an engaging and astute figure in her own right. A perceptive contemporary, Bernardo Bellincioni, wrote that Leonardo ‘makes her appear to listen…’ Much has been written about the symbolism of the ermine, but I want only to note how well it is painted. Again, a number of studies of the ermine are hung nearby, allowing us to see how much care Leonardo took to convey details such as the beast’s claws sinking into the sitter’s dress. (It's worth noting that Leonardo liked this portrait so much, he recycled part of it - the hand can be seen in The Last Supper, for St Philip.)

One omission from any discussion about Cecilia, indeed, throughout most of the exhibition and the catalogue, is any mention of condition. When dealing with an artist such as Leonardo the question of a painting’s condition, of what original remains and what is added by a later hand, must be of the greatest importance. But here it is almost entirely lacking. Surely, viewers need to know that the background of Cecilia is entirely over-painted, making the sitter’s profile seem unduly hard and unnatural, and the turn of her shoulders stiffer than the artist intended, or that the tips of her two lower fingers have been repainted. 

Picture: Leonardo, 'The Madonna of the Rocks'. Left, Musee du Louvre (777) RMN/Frank Raux, oil on wood, transferred onto canvas, 199 x 122 cm,  and right The National Gallery, London (NG 1093), oil on poplar, 189.5 x 120 cm. 

The central room in the exhibition shows the juxtaposition – for the first time ever, it is believed – of the two versions of Leonardo’s Madonna of the Rocks. They are hung opposite each other, which does not allow for close comparison of the various details, but, surprisingly, helps emphasise their differences and respective qualities. The version in the Louvre, dated to 1483-c.1485 generates more impact, despite the fact that it is hidden beneath the ubiquitous layers of dirt and yellowed varnish. It is hard to say why this is, and the respective merits of the two have been argued about endlessly by scholars. The National Gallery’s later version, completed in fits and starts between 1491/2 and 1506/8, has always been perceived as slightly inferior, and perhaps the work of Leonardo delegating areas to his assistants. The recent cleaning of the National picture, however, has seen the gallery declare that it is ‘a Leonardo’. I do not doubt this, for whether parts of it were painted by assistants or not (and there is plenty of documentary evidence to confirm that Leonardo used trained artists to help him complete this and other commissions, one source describes ‘two of Leonardo’s pupils […] doing some portraits and he from time to time put a touch on them’) Leonardo considered it ‘a Leonardo’, and so did the people paying him. But one thing is clear from this exhibition – the Louvre version is better by several degrees. 

In the Louvre’s Madonna of the Rocks the drawing is as close to perfection as one sees in a Leonardo painting. Each finger, join, muscle and glance seem to be inspired with life. In the London version, however, some of the details, most particularly Christ’s right hand (below), are executed so cursorily as to be almost painterly, not a practice we normally associate with Leonardo. The lower part of the blue drapery worn by the Madonna in the London Rocks is so perfunctory it seems she is wearing a sleeping bag. 

Picture: National Gallery, London, detail from 'The Madonna of the Rocks'.

It seems evident from the comparison of both works that Leonardo could not muster the same level of interest or precision in the second version. And who can blame him, for a genius obsessed with detail and innovation there can be no more tedious task than repeating something you have already done to perfection. It is telling that not long after the London Madonna of the Rocks was painted, Castiglione described Leonardo as ‘one of the world’s finest painters, [who] despises the art for which he has so rare a talent…’ To me, there is no more obvious indicator that Leonardo, could, in effect, simply not be bothered to complete the London version of the Rocks to the same degree as the Paris version than in the omission of the angel’s pointed hand, an element which forms such an important part of the narrative in the first version. 

Unsurprisingly, the catalogue entry for the National’s version of The Madonna of The Rocks sees a spirited defence that the picture is almost wholly by Leonardo. Another picture receiving a spirited defence that it too is a Leonardo is the Madonna Litta [Hermitage, Cat.57]. This picture, on loan from the Hermitage in St Petersburg, was lent only on condition that it was called ‘a Leonardo’, and that the Hermitage could write the catalogue entry. They duly have, and consequently it reads a little like a Soviet propaganda piece, a curious blend of nationalism and art history. Perhaps this approach is understandable, when you consider how high the stakes are – the prestige to a museum of having ‘a Leonardo’ is priceless. It is telling that Syson himself, however, in a prelude to the Madonna Litta catalogue entry, clearly conveys his own doubts over the picture. I found it equally telling that, in his speech at the beginning of the press preview this morning, instead of saying ‘we have nine Leonardos’, he used the phrase ‘we have nine pictures lent by their institutions as Leonardos’.

Picture: Leonardo, 'The Madonna Litta', about 1491-5, tempera on canvas, transferred from wood, 42 x 33 cm, The State Hermitage Museum, St Petersburg (GE-249).

The Madonna Litta was celebrated in its day. But scholars have long had doubts over whether it is by Leonardo, or based on his design. And here we come back to the question of Leonardo’s studio, a question which is dealt with clearly in both the catalogue and exhibition. I won’t go into those details here, save to say that the Madonna Litta, as we see it today, is extremely hard to quantify. Again it comes down partly to condition. It is painted in tempera (unusual for Leonardo at this date) on a panel which has been thinned down and laid onto canvas. As a result, the picture’s surface has lost the delicacy of the original. It appears flat and uniform. Kenneth Clark described the picture as looking ‘like an oleograph’, and he was not far out. I found it an underwhelming picture, too severely painted and unsubtle for Leonardo, even accounting for the different medium. 

Picture:Leonardo, 'Head of a Woman', metalpoint heightened with white on grey paper, 17.9 x 16.8 cm,  Musee du Louvre (2376). 

For me, the most telling clue as to whether the Madonna Litta is ‘right’ or not could be found in Leonardo’s drawing for the Madonna’s head [above, Louvre, Cat.59] hung next to the painting. The comparison is a cruel one for advocates of the Litta. All Leonardo’s figures are characterised by the expressiveness of their faces, even in the most trivial doodle. In the drawing of the Madonna the characterisation is so acute that we cannot fail to notice at once both her pride as a mother and also her fear at Christ’s ultimate fate, as symbolised in the painting by him holding a chaffinch, a symbol of his torture and death. And yet in the painting, the Madonna’s delicate expression is noticeably lacking, with deadened features and a less lifelike gaze. One is constantly reminded when looking at the Madonna’s flesh tones of the more uniform and silvery tones of Leonardo’s immediate followers, most notably Boltraffio, two of whose studies for the Litta hang nearby [Cats. 61 & 62]. 

Picture: Giovanni Antonio Boltraffio, 'Portrait of a Young Woman as Artemisia', about 1494, oil on wood, 49 x 37 cm, Mattioli Collection, Milan, photo, Vittorio Calore, Milan. 

A number of paintings by Boltraffio are in the exhibition, and he seems clearly to have been the most able of Leonardo’s followers (better than Marco D’Oggiono for example). Of particular note is the Portrait of a Young Woman as Artemisia [Cat.18, Mattioli Collection, Milan]. This is dated in the catalogue to c.1494. It was discovered in the mid-twentieth Century, and despite an attempt then to attribute it to Leonardo, was given to Boltraffio. If it is by him, then it is one of his masterpieces, for I find it difficult to accept that the startling fidelity of this fresh and modern face could be by the same artist as some of the other Boltraffios in the exhibition, with their stiff and mannered expressions. Particularly noticeable is the way the face is lit, deftly and with great control, especially around the eyes. The fine shadow cast by the sitter’s right eyelash is worthy of Leonardo himself, as indeed is some of the drawing in the hand. As heretical as it may sound, the portrait is so good that I found myself wondering if this picture might be by someone better than Boltraffio... Again, condition governs our view of this curious painting (which Richard Dorment, incidentally, calls 'second rate', and which photographs horribly). The picture has been severely damaged and very badly retouched. Little of the original veil remains, and there is extensive damage in her neck, forehead and hair. The bowl is largely reconstructed, along with the hand too. But there is enough modelling in the hand, especially in the sinewy stretch of flesh between thumb and index finger, to suggest the highest skill. Is this one of the lost portraits Leonardo made by his assistants, but which he ‘put a touch on’? Go and see it and let me know what you think.  

Picture: Leonardo, 'Christ as Salvator Mundi', about 1499 onwards, oil on walnut, 65.5 x 45.1 cm, private collection, (C) Salvator Mundi LLC/Tim Nighswander/Imaging4Art.

Let me turn now to the two pictures in the final room, the rediscovered Salvator Mundi and the Duke of Buccleuch’s version of the Madonna of the Yarnwinder. The two make good bedfellows, being dated to c.1499, and similarly painted in areas. Much has already been written about Salvator Mundi, but I will say that it is far better in the flesh than the photographs suggest. And nor is it as badly damaged as I had originally feared. Most importantly, it works as a picture, and still delivers an ethereal vision of Christ, just as Leonardo would surely have intended. It is hard to overstate the significance of its inclusion in the exhibition. There are still those who mutter about the National Gallery including a picture that was found by ‘the trade’. But if it wasn’t for the trade, the picture would never have been found (in a minor US auction in 2005). Who cares who owns it, or how much they paid for it? It dramatically adds to our knowledge of Leonardo, and that, as far as this exhibition is concerned, is all that matters. The Madonna of the Yarnwinder is a picture that I was prepared to be sceptical about from the few reproductions of it that I have seen. But again, in the flesh, this is an exquisite work, albeit one that has been finished in the background by someone else. 

Picture:Leonardo & a later unidentified artist, 'The Madonna of the Yarnwinder', about 1499 onwards, oil on walnut, 48.9 x 36.8 cm, Private Collection (C) The 10th Duke of Buccleuch and the Trustees of the Buccleuch Living Heritage Trust.

Finally, the Last Supper. This section of the exhibition is upstairs in the main gallery, in the Sunley Room. At first it seems a little disappointing – the photographic replica of the original Last Supper is not full-scale, while Giampetrino’s scale copy is badly lit. But around it are some of the finest drawings in the exhibition – all of the surviving drawings by Leonardo for the Last Supper. My favourite are two studies which show how hard he strove to achieve the calm harmony and unity of grouping Christ and the Twelve Disciples convincingly. These drawings (Cats. 69 & 70) show some of the disciples on the near side of the table – and it is clear from the problems of scale why Leonardo chose to abandon this concept in favour of arranging the figures all on one side (thereby inspiring my favourite art history cartoon). 

So what then can we make of Leonardo? This exhibition shows how we are left with only a fraction of his bewildering brilliance. We cannot hear his exquisite and much remarked upon music, nor his celebrated conversation, we cannot see his sculptures (although we know he made them), nor marvel at any of his engineering schemes (nor even his own beautiful good looks). As Vasari states:

It says something of our era that we still see Leonardo as the defining figure of art. We will queue round the block to pay homage - in 'a solemn festival' - to a handful of paintings made by a curious man who lived 500 years ago. How should we feel that we, today, have nothing to match him? Indeed, will there ever be anyone to match him? Or do the unique circumstances which led to such a creative force no longer exist? Suffused throughout the exhibition is Leonardo’s own sense of faith. He was clearly painting from both a sense of personal belief and scientific exploration, and it was this combination of curiosity restrained by faith that allowed him to create objects of startling originality. Was this era of great geniuses the high point of creativity, the confluence of the two great drivers of artistic expression in western art, religion and the exploration of humanity? Today, we have the answers to more or less everything, and yet nobody is as creative, artistically, as Leonardo. The two giants of contemporary art make millions either by casting enlarged jelly babies or painting spots on canvas (in fact they get other people to do it for them.) And certainly, there is no shortage in the world of people of faith, but it tends to be of the fundamental kind, unquestioning, and binary - either on or off – the type that excludes a Leonardo-esque curiosity. The contrast between Leonardo and, say, Koons suggests that great art grows out of both observation of the outside world, and a combination of sensory impulses, be they religious or otherwise emotionally driven, from within. Leonardo, as this exhibition shows, had both. For while today we make art with the human hand, Leonardo made it with the human spirit.