Can the leaked Louvre book give us the final word on the much-maligned painting that we all need? This is, or could have been, a review of a book that was never published, the existence of which has never been officially acknowledged. Its non-publication by Hazan for the Musée du Louvre coincided with the non-appearance of the Salvator Mundi in the French museum’s remarkable exhibition, Leonardo da Vinci, in 2019-20. The authors, who miss this significant item from their lists of publications, are Vincent Delieuvin, the curator responsible, and two members of the national Centre de recherche et de restauration des musées de France, Myriam Eveno and Elisabeth Ravaud. The non-book seems to have flitted momentarily across the shelves of the Louvre bookshop on 18 December 2019. (It should be understood, in relation to what follows, that I am not confessing to owning a copy of the non-book.)

Its non-publication also coincided with one of the many missed publication dates for the full-scale monograph, , by Robert Simon, Margaret Dalivalle and myself. It is unfortunate that neither publication could draw on the other. However, in retrospect it is good to have two fully independent studies of the now infamous $450m painting, not least for their high-level scientific examinations. The techniques range across multispectral scanning, infrared reflectography, X-rays, ultraviolet light, optical coherence tomography, X-ray florescence, microscopy and macro-photography. The results are gratifyingly compatible.

Delieuvin’s careful review of all aspects of the painting, not least its many close variants, leads him to conclude that the painting is indeed by Leonardo. “The results of the historical and scientific study presented in this book confirm the attribution of the work to Leonardo da Vinci,” as Jean-Luc Martinez, the director of the Louvre, said in the preface. Delieuvin’s assessment of the many variants by followers was judicious, although he underrated the pedantic version once in the Yarborough Collection, which was sold in Paris in 1962 and is no longer traceable. Alone of the versions, it reflects Leonardo’s design for Christ’s right sleeve in the wonderful red chalk study at Windsor. Such leaking of abandoned motifs into works by followers happens elsewhere, most notably in the two versions of the Madonna of the Yarnwinder.

Ill-informed controversies

It is the scientific examinations that warrant most attention in the light of current ill-informed controversies of the autograph status of the painting in various public media. The broad issue is what Delieuvin calls “the decisive contribution of the scientific examinations”. But what is the status of science in the process of attribution? I have argued that this question has generally been ill-answered in the profession. Much of the time, the scientific analyses play a permissive role; that is to say, not placing obstacles in the way of a proposed attribution. For some artists, and for some paintings, science can play a more determining role. This is particularly the case with Leonardo, whose idiosyncratic preparatory and technical procedures become fully apparent, above all through infrared reflectography.

Even after the making of a preparatory cartoon, for which there is definitive evidence in the Salvator Mundi, Leonardo restlessly fiddled with motifs. In the technical images, we see pentimenti (adjustments) in both of Christ’s hands, the interlace pattern on the drapery bands, and the plaque at the centre of his chest. The French authors also see a major change in Christ’s blessing hand and argue that it was not originally in that position at all. Much as I would welcome such a major pentimento, there is evidently a reserve for the wrist and hand in the thick blue drapery. We also know from the Windsor drawing and the Yarborough copy that Christ’s left arm was in that position at an early stage. Pentimenti are absent in the copies, not least in the De Ganay version, which had been proclaimed as the original. The underdrawing exhibits the niggling precision typical of a copy.

Blending technique

Leonardo’s technical procedures in the Salvator Mundi are characteristically refined and optically subtle. His use of grains of glass of at least two different densities in paint layers is wholly characteristic. The edgeless gradation of tonal transitions from one plane to the next is aided by his hand-print technique of blending. As Delieuvin argued, the fact that X-rays of the painting only produce a pale ghost is common to all the late paintings. Leonardo deploys limited quantities of opaque white lead after 1500. Throughout, the technical examinations allow us to witness Leonardo at work in his highly individual manner.

What is needed now, in the face of continued “stories”, is direct scrutiny of the painting’s magical qualities– based on open opportunity to see the real thing. An open publication of the technical analyses is also highly desirable rather than scans of the “non-book” being widely leaked to the media in a deliberate ploy to counteract the bogus claims made in a recent French film documentary. Instead, we are likely to be faced with the continued sound and fury of sensation-seekers.

The painting deserves better. It is not to blame for its ownership, past and present. Being owned by dealers, a Russian oligarch and a secretive Saudi prince tends to attract unfavourable attention. Poor painting! It is time the abuse stopped and the looking started.

• Martin Kemp is emeritus professor of the History of Art, Trinity College, Oxford. His latest book is Heavenly Visions: Dante and the Art of Divine Light