The Last Leonardo: The Making of a Masterpiece. Ben Lewis

if it was placed in a corridor, on the stairs, or worse, in a storeroom, it was probably thought of as a second-rate work.

At the heart of this research are the inventories, handwritten on parchment or thick crusts of paper. These can be catalogues of collections compiled by the owner’s clerks or ‘keepers of pictures’. They may be itemised lists of works of art available for purchase from a hard-up merchant, or troves of freshly acquired art that are about to be shipped out of Italy, or the valuables that a bride took with her to her new home. In an age when money was not kept in banks, you were what you owned. You didn’t want your possessions, spread across your many-roomed mansions, to slip away. Largely ignored as a source of information until the late twentieth century, inventories have become the coalface of one of the most fashionable fields of art history today, rich seams of data from which deductions, speculations and occasionally conclusions can be extracted.

      While inventories are vital to building a case for attribution for thousands of Renaissance paintings, the raw material is challenging. The fragmentary nature of the records means that most histories have gaps. The names of artists are spelt in many ways, and attributions can change from list to list. Descriptions of the paintings are, until the late nineteenth century, only textual, with scarcely a visual reference. That is an immense problem, because the range of subjects – especially biblical and classical – was limited, the titles are often similar, artists often made several paintings of the same subject, and the descriptions in the inventories are brief. Dimensions are rarely supplied, and sometimes there is only a title without a painter’s name attached. The result is that provenance histories for works of art from before the nineteenth century are frequently assembled from a range of probabilities, which reinforce each other. Such structures can be precarious, wobbling between the likely and the hypothetical. The evidence is often circumstantial, but art history is a discipline that studies the products of the imagination; a certain flexibility is permitted, while the marvellous objects themselves have been known to inspire the most rigorous of academic minds to meld fact with fantasy.

Since the author and the date of Robert Simon’s painting were unknown, he began to research its provenance within weeks of acquiring it. By his own account he spent an hour a day, every day for six years, studying the Salvator. Every Old Masters dealer has to present an account of the provenance of artworks they wish to sell. Most of them subcontract this work to specialists and academics, but Simon is a particularly scholarly dealer, who enjoys the archives and takes pride in his abilities. He had done provenance research many times before the Salvator arrived in his gallery.

      The first clue was two initials and a number on the back of the painting: ‘CC 106’. Simon traced that back to the important nineteenth-century Cook Collection, belonging to a British cloth merchant. Some claim that Sir Francis Cook assembled the greatest art collection in private hands in Britain at that time, with the exception of Queen Victoria’s. A three-volume catalogue of his treasures was published in 1913. There, Simon discovered his painting, listed as ‘cat. number 106’, on page 123 of Volume I, which was entitled ‘Italian Schools’. However, it was not attributed to Leonardo but described as a poor copy, and there was no photograph of the painting in the catalogue. Simon turned to the photo archives.

      It was the technology of photography that made modern art history possible. From the mid-nineteenth century specialised photo studios, most famously Alinari in Rome, methodically, accurately and beautifully photographed every notable work of art they could find, supplying an ever-growing market with perfect images, albeit in black and white. Museums and institutions built collections of thousands of photos, while art historians and connoisseurs amassed their own private stockpiles – it was a way for them to keep images of all the art they studied and loved close to them, in their homes. The previously uncontainable – a vast sea of images spread across many thousands of kilometres, too large and diverse to be committed to memory – could now be held in one’s hands, spread out on a table or stored in a cupboard.

For art historians, photography was like the spear that enabled cavemen to hunt woolly mammoths. The scholars scribbled notes on the backs of their images, with dates, authorship, and what they knew of the ownership of the painting. By laying out a selection of photographs on their desks they could study, for example, the drapery folds or facial features in a hundred anonymous Renaissance altarpieces and group them according to stylistic traits. They could then associate those stylistic traits with documented works by this or that artist, thus defining his or her oeuvre and, if they had some dates, stylistic development. They could pull out all the pictures they had on a particular subject, like the Last Supper, or the Madonna and Child, or indeed the Salvator Mundi, and, if they knew the dates, arrange them chronologically to see how the treatment of that subject evolved over time, which artists innovated, and which copied those innovations. Thus was born an art history of style and symbol. Acquired over many decades, many of these photographs have survived while the pictures they record have been destroyed or gone missing. They are not just a record of the art we have; they are also a record of the art we have lost.

      Robert Simon visited the Witt Library photo archive in the basement of the Courtauld Institute of Art in London. There he accessed all the folders of images marked ‘Salvator Mundi’. Soon he found a photograph of the Cook Collection’s Salvator, where once again it was listed as a copy. Simon was not surprised by that. Sleepers were almost always miscatalogued, otherwise they would not have ‘slept’ so long. On the bottom right of the photograph was a typed text reading ‘(Cook Coll. Richmond)’, and underneath, handwritten, ‘Whereabouts unknown (1963)’. So nobody had known where this picture was in 1963.

Parts of the Cook Salvator looked different from the painting Simon had bought. In the Cook photo, Christ had a moustache and facial hair that made him resemble a Mexican bandit in a 1950s B-movie. That indicated that Simon’s Salvator had been restored or repainted in some way since the date of the photo. However the blessing hand, embroidery, orb and other features of the Cook were identical with Simon’s painting. Now he could narrow down his search. Before this discovery the painting could have come from any European country, but Simon now had a focus: Britain.

      The second clue led back to Britain as well. Everyone in the Old Masters business knows of an etching made by the seventeenth-century print-maker Wenceslaus Hollar which bears an inscription by the artist, ‘Leonardus da Vinci pinxit’, the word ‘pinxit’ testifying that the print was a copy of a painting by Leonardo. It is an image of Christ as saviour of the world, orb in one hand, the other raised in blessing, with flowing curly hair remarkably similar to that in the Simon and Cook painting. The original Leonardo had long been presumed lost. Simon compared his painting to this print. It looked so similar in significant clues – its drapery and its blessing hand, even if – a significant clue in the contrary direction – Hollar’s Christ had a curly beard with a central parting, and his didn’t.

      Simon knew where to go next. There was a particular volume on his shelf which many dealers have, and which is often a first port of call for researching the history of potentially important unknown paintings. One day in 2006 or 2007 – he can’t remember which year exactly – Simon pulled out his copy of the Walpole Society Journal, 1972. In it, the keeper of the British royal collection, Oliver Millar, had published an inventory of King Charles I’s art collection, meticulously turning a few slightly differing seventeenth-century handwritten manuscripts into a hundred-odd pages of neat type. Simon soon came across a description of a work that might match his painting: page 63, item number 49, a ‘Peece [sic] of Christ done by Leonard’. Now he had found a record that Charles I had owned a painting of Christ most likely by Leonardo, and that painting was, in all probability, the one he had bought a 50 per cent stake in for the decidedly unprincely sum of $587.50.

At the recommendation of Martin Kemp, Simon contacted a young art history graduate, Margaret Dalivalle. She had been a student of Kemp’s at Oxford and was writing a PhD about notions of the copy and the original in seventeenth-century painting. Simon asked whether she could, as she recalls it, ‘contribute to the research into the provenance history of a newly discovered painting’.

      Dalivalle was born in Ayrshire in Scotland, and showed an artistic bent from an early age, encouraged by her godmother, who worked