It is my hunch that “deaccession” is not a word you often hear down the pub (always assuming anyone can still remember what a pub is), but all it means — in the words of the great god Wikipedia — is “the process by which a work of art or other object is permanently removed from a museum’s collection”. Some people think there is nothing particularly sinister about deaccessioning, but I am not one of them. At least potentially, given the immense financial battering museums have received as a result of the lockdown, some of them may contemplate deaccessioning as a financial panacea, since the collapse in income will continue until routine air travel is restored and social distancing is a thing of the past. However, there are all sorts of reasons why to do so would be — in the words of 1066 and All That — A Bad Thing, even in a changed world. I will begin by arguing the case more generally, and conclude by returning to the here and now.

Before getting on to the main issue, however, it seems important to stress the fact that — especially in the case of old masters and their ilk — it is by no means invariably straightforward to determine who did what, for all that the wonders of modern science now make it much easier to do so than used to be the case. Inevitably, this can mean that people contemplating deaccessioning particular works do not necessarily know what it is they will be kissing goodbye to.

The simple answer to the question “Is deaccessioning ever to be encouraged?” is “No”. The principal reason is that neither of the two main forms of deaccessioning, which may be defined as bottom-feeding and top-lopping, is a good idea.

In the former, bottom-feeding scenario — so the theory goes — all sorts of at best unimportant and at worst downright hideous items are painlessly removed from a museum’s collection, which brings in a little money, but more positively frees up a combination of storage space and staff time.

The first point to be made is that it is almost axiomatic that these pieces, which are unloved by their institutional owners, will be equally unloved by the market. If the opposite were to prove to be the case, then it stands to reason that it might well have been better to hold on to them. Moreover, the fundamental problem with this kind of housekeeping exercise is that it necessarily presupposes that the experts in charge know what they are doing.

The National Gallery, like all other national museums in the United Kingdom, is not permitted to deaccession. An exception to this otherwise unbending rule — not relevant to the NG, but to other museums such as Tate and the British Museum — allows for the disposal of duplicates, such as two identical versions of a print.

However, it is not hard to imagine that under another dispensation the powers that be in Trafalgar Square might have felt tempted to part with NG27 at some point between its arrival in the Gallery in 1824 as part of the founding bequest of John Julius Angerstein, which explains its very low inventory number, and its change of status in 1970.

That was the year it was published as the original of Raphael’s Portrait of Pope Julius II, having been — in the words of the relevant National Gallery catalogue — “hitherto catalogued as an early copy of what was considered by his contemporaries to be one of the most celebrated and epoch-making of Raphael’s portraits”. Raphael’s Portrait of Pope Julius II of 1511-12 is first recorded on 12 September 1513, not long after the sitter’s death earlier that year, in Santa Maria del Popolo in Rome. It was then displayed on the high altar of the church as one half of a species of diptych, together with a Holy Family which has come to be known as the Madonna of Loreto.

People disposing of artworks may not know what they are kissing goodbye to

However, at the time of the rediscovery of the portrait in 1970, the original of the Madonna of Loreto still remained untraced. It was not until October 1979 that an exhibition at the Musée Condé at Chantilly unveiled the newly-restored version of the composition, which had been part of the museum’s collection since its opening in 1870, as the hitherto missing Raphael.

In both cases, the incontrovertible evidence that the works in question were indeed originals as opposed to old copies involved a combination of scientific analyses and archival digging. The former was above all represented by evidence in the form of x-rays — which have since been supplemented by infra-red reflectography — revealing various major changes of mind beneath the surfaces of the poplar panels on which they are both painted. At the same time, research into their respective provenances allowed them to be traced back all the way to Santa Maria del Popolo, from which they were bought by Cardinal Paolo Emilio Sfondrati in 1591.

Bernini’s patron, Cardinal Scipione Borghese, then acquired Sfondrati’s collection of 71 paintings, including these two, from him in 1608. They duly appear in a later Borghese inventory of 1693, at which point they were both inscribed with inventory numbers which are still visible in their bottom left-hand corners. Their subsequent histories up to their arrivals at their final resting-places have also been established.

Since 1970, as an unimpeachable and stunning Raphael, the Portrait of Pope Julius II would potentially have been a prime candidate for top-lopping, and selling a painting of this importance would net its owners an extremely tidy sum. By a piquant coincidence, for the last decade or so the world record auction price for an old master drawing has been held in succession by two head studies by none other than Raphael, with the Head of a Muse for his fresco of Parnassus selling for £29.2 million in December 2009, only to be overtaken three years later in December 2012 by the Head of an Apostle for his Transfiguration, which made £29.7 million. The National Gallery’s annual grant-in-aid from HM Treasury costs the public purse a few million less.

The amount paid for the Apostle has still not been surpassed in the world of drawings, but the prices achieved by old master paintings in recent years make these numbers smack of the bargain basement. The ultimate high came on 15 November 2017 in New York, when Christie’s sold Leonardo da Vinci’s Salvator Mundi for an eye-watering $450 million. There are all sorts of intriguing mysteries surrounding that work, its purchaser, and not least its present whereabouts.

What is more, some people — although I am absolutely not of their number — refuse to accept that it is by Leonardo, but the sale price is not a matter of opinion.

Happily, and even in our present plight, I am confident that there is not a snowball’s chance in hell of any British government contemplating removing the government grant and asking the National Gallery — or anyone else — to fend for themselves by flogging off the odd masterpiece every now and then. But it is certainly not the case that such things have never happened in other places at other times. What is more, and even allowing for what might be described by Dan Brown as the Da Vinci Effect, Raphael’s Julius II would no doubt fetch a pretty packet.

In the history of deaccessioning, the ultimate bonanza is what is often referred to as the Hermitage deal. In the 1920s, not long after the fall of the House of Romanov and the establishment of the Soviet Union, rumours began to circulate in the West about plans to solve the problem of a perceived need for foreign currency by selling off works of art. On one level, the idea must have appeared to make a lot of sense, since private property was virtually outlawed in what Sidney and Beatrice Webb were to dub “A New Civilisation”, and all works of art, even in museums, could potentially be denounced as unnecessary luxuries.

Better still, portraits of the rich and mighty, not to mention religious paintings and sculptures, must have seemed particularly suspect ideologically. The challenge was the perennial one faced by potential deaccessioners: as I have outlined, the sums of money that can be raised by selling off minor items, even in bulk, are disappointingly small, while the other obvious possibility is the undesirable one of disposing of the crème de la crème. In the event, the Soviets pursued both options at different times, but the latter is the one that makes the blood run cold.

In 1925, a body called the Central Office for State Trading of the USSR for the Purchase and Sale of Antique Objects, which came to be known as the Antiquariat (or Antikvariat), was created under the leadership of Nicholas Ilyn (or Iljin). Above him in the hierarchy was Anastas Mikoyan (1895-1978), from 1926 the People’s Commissar for External and Internal Trade, who would have oversight of all the Antiquariat’s activities. It was not until around 1929-30, however, that significant progress was made by it on the deaccessioning front, and even then the utmost secrecy was called for, since the Soviets were desperate to avoid their activities being made public.

The Antiquariat began to make lists of what they deemed the most important works of art in the Soviet Union, but naturally the pre-eminent potential source of treasures was the Hermitage, which housed the collections of the Tsars. The French art dealer Germain Seligman was invited to visit the Soviet Union in 1927, and then contacted again in Paris the following year and asked to organise the sale of various works of art, but the advice he received from the Quai d’Orsay was to walk away from the deal of a lifetime. Moreover, it would appear that in 1928 Armand Hammer and his brother Victor, who were living in Moscow at the time, were offered paintings from the Hermitage, and then tried to involve Sir Joseph Duveen, but nothing came of their schemes.

What is certain, however, is that Calouste Gulbenkian, who was often said to be the richest man in the world, had first pick of the bounty of the Hermitage, possibly through the intervention of his fellow Armenian Mikoyan, and made no fewer than four separate agreements between 1928 and 1930. Gulbenkian’s passion for art was omnivorous, and he was at least as gripped by the so-called decorative arts as by the so-called fine ones.

The first contract involved 24 pieces of silver and gold items, and only three paintings — two Hubert Roberts of Versailles, and Dirk Bouts’s Annunciation. The second added 15 pieces of silver, and a splendid full-length portrait by Rubens. The third tranche comprised Houdon’s incomparable marble sculpture of Diana and five pictures, including Rembrandt’s Pallas Athene.

However, Gulbenkian only kept the Diana and the Pallas Athene, and sold the four other paintings on to Wildenstein, presumably as part of some kind of non-interference deal. Finally, the last negotiation concerned the Portrait of an Old Man of 1645 by Rembrandt, which — like the rest of his extraordinary collection — is now in the Calouste Gulbenkian Museum in Lisbon. The Wall Street crash of October 1929 might have deterred many a collector, but not Gulbenkian, who had simply had enough, and was at the same time happy, in letters to the Soviets, to deplore with considerable eloquence the sales from which he was benefitting.

Gulbenkian’s last transaction was concluded in late 1930, but by that date a second front had opened up. Berlin was the conduit through which the items reached the West, and Berlin continued to be of crucial importance in the next phase of the story. As early as January 1930, Charles Henschel of Knoedler in New York summed the position up when he wrote: “There has been a lot of talk for the last year of some of the Hermitage pictures being sold; this now seems to have come to a definite head.”

Knoedler were the crucial link to Andrew Mellon, but they would not have been able to broker a deal with the Soviets had it not been for a whole network of alliances and associations.

The most crucial of these at the Soviet end was the young art dealer Francis Zatzenstein (1897-1963) — also sometimes known as Katzenstein — who adopted the surname of his first wife, which was Matthiesen. He was based in Berlin, where his establishment, the Matthiesen Gallery, also traded under the same name. He had been approached by the Soviets in November 1929 and asked to value what were regarded as their hundred best paintings.

Their protestations that they had no intention of selling must have seemed particularly hollow if it is true that Zatzenstein saw a number of paintings from the Hermitage when he visited Gulbenkian in Paris a few months later in 1930.

Zatzenstein naturally realised that more sales might be possible, but he lacked the wherewithal to broker the deal. He contacted Colnaghi in London, and they then brought in Knoedler, with whom they enjoyed a good working relationship. In their turn, Knoedler involved Andrew Mellon, who was not only one of the richest men in America, but also a considerable public figure, having served as Secretary of the Treasury under three successive presidents since 1921.

Mellon’s immense wealth financed the whole project, but like Gulbenkian he was both intensely publicity-shy, which would have seemed to guarantee discretion, at least to begin with, and unused to making major purchases of works of art he had not seen in the original. Perhaps above all for this reason, he made an arrangement with Knoedler which would allow him to pass over any painting he chose not to retain.

By April 1930, Mellon had already received Van Dyck’s Portrait of Philip, 4th Lord Wharton, and it was to be followed by a further 20 paintings, all of which he kept. Dividing them up in terms of art history, as opposed to the dates of acquisition, it emerges that they were all but exclusively — if unevenly — split between the renaissance and the baroque, with a single Chardin House of Cards as a kind of chronological postscript. In the former category, there were seven fifteenth- and sixteenth-century paintings, all of supreme quality.

Conversely, the seventeenth-century representation — with the exception of the supposed Velázquez’s Innocent X — was all Dutch and Flemish, comprising six Rembrandts and two portraits by Hals from the former school, and four Van Dycks, including his Portrait of a Flemish Lady (arguably the most shocking loss of all, since it involved splitting up a pair of portraits), from the latter. Of the lot, only one of the Van Dycks was a blatant mistake even at the time, while the Velázquez has since fallen by the wayside.

If the works of art were almost all superlative, then it has to be admitted that the human side of the story was frequently at best comic and at worst unedifying. Predictably, the Westerners did not relish their visits to Leningrad and Moscow, for all that they subsisted on a diet of vodka, caviar, and sturgeon, whereas the Soviets — in the form of Illyn and Boris Kraevsy from the People’s Commissariat of Foreign Trade — ended up enjoying themselves mightily in New York.

Having sealed the deal, they moved from an obscure little hotel on West 23rd Street to the Biltmore, and returned across the Atlantic in first class, having travelled steerage on the way out. Their three exact counterparts in Ernst Lubitsch’s enchanting film of 1939, Ninotchka, which concerns Five Year Plans, Soviet agents in Paris and a Grand Duchess’s jewels, might almost have been taking their correspondence course.

The astonishing success of the Mellon deal could have been followed up by an even more extraordinary coda, had not the prices seemed too high. In 1931, like many another very rich man before and since, Mellon turned down an offer that would hardly have bankrupted him, in this case for Leonardo’s Benois Madonna and Madonna Litta (a work whose authorship remains controversial), and Giorgione’s peerless Judith.

That seems almost more regrettable for Washington than the fact that — but for his general aversion to nudes and religious subjects — he might have purchased Rembrandt’s Danäe and Prodigal Son. Conversely, the universal as opposed to specialist revaluation of Caravaggio that might have secured his Lute Player was yet to come. Colnaghi were involved in various slightly later operations that extracted further masterpieces from the USSR, including what is surely the greatest European painting in the southern hemisphere, Giambattista Tiepolo’s monumental Banquet of Cleopatra, which reached the National Gallery of Victoria in Melbourne in 1933.

It is of course impossible to know if there will be a deaccessioning frenzy in the wake of the Covid-19 pandemic, but my fingers are crossed that there will not. One encouraging recent precedent is the case of the Detroit Institute of Arts. In May 2013, Kevyn Orr, who was the man who had the unenviable task of sorting out the mess in bankrupt Detroit — the city had a $15.6 billion debt — suggested that the collection of the Detroit Institute of Arts might be sold to help balance the books.

In the event, nothing of the sort occurred. As a matter of fact, far smaller raids in those places where they are not illegal are entirely commonplace, and — since they almost invariably ride roughshod over the intentions of donors — can hardly fail to act as a powerful deterrent to people thinking of giving their works of art to the museums concerned, which should give anyone contemplating a spot of deaccessioning pause for thought.

The late — and great — Sir Denis Mahon (1910-2011) made elaborate provision to guarantee that the various institutions he wished to be the recipients after his death of his peerless collection of Italian baroque art would not be able to hold on to them if they charged for entry. How much more furious would he not have been at the idea of them being sold off?