Abstract Combining Impressionist, Barbizon and Dutch seventeenth-century masterpieces with eighteenth-century French and English portraits and Venetian veduti, the collection of paintings formed by the Anglo-Armenian financier and oil magnate Calouste Gulbenkian (1869–1955) and now housed in the Gulbenkian Museum, Lisbon closely resembles collections formed by contemporaries such as Henry Frick and Jacques Doucet. The manner of its acquisition, however – most notably purchases from Russia’s Hermitage Museum – displayed a ‘buccaneer’ element. New research shows how Gulbenkian leveraged his oil interests as well as promises of a bequest to influence the leading British, French and American agents, dealers and curators of the twentieth century. In February 1956 the Directeur des Musées de France, George Salles, wrote a short report on Calouste Gulbenkian and his collections. The Anglo-Armenian oil magnate had died the previous year in Lisbon, leaving his collections to a charitable foundation bearing his name. As noted by Salles, Gulbenkian had been, ‘along with Sir Basil Zaharoff, probably the richest man of our generation’. ‘This cosmopolitan has been the talk of the financial world and of international politics for so long’, he went on, that Gulbenkian needed no further introduction. Gulbenkian’s collection was ‘both renowned and at the same time, so to say, unknown.’ Of the staff of the Réunion des Musées Nationaux, only Salles’s predecessor Henri Vierne and Paul Vitry of the Louvre’s department of sculpture had been inside Gulbenkian’s hôtel at 51 Avenue d’Iéna.1 Both had since died. If the authorities allowed the collection to leave France, Salles wrote, ‘we would leave ourselves open to the same reproaches meted out to those of our predecessors who let the famous Wallace Collection emigrate to London.’2 Born in Istanbul in 1869, Gulbenkian (Fig. 1) had settled in London in 1897, a few years before the aforementioned Wallace Collection was translated from Paris. The Wallace Collection influenced his taste, much as it did that of J. P. Morgan and Henry Frick.3 In its outlines Gulbenkian’s collecting strongly reflected a Belle Époque fascination with the eighteenth century inspired by the Goncourt brothers, a copy of whose L’Art du dixhuitième siècle (1880–82) he acquired in May 1899.4 It would be easy to imagine Gulbenkian’s museum standing somewhere near the Parc Monceau in Paris, alongside other ‘collection museums’ established by contemporaries equally at home among la haute banque, such as Nélie Jacquemart (1841–1912) and Moïse de Camondo (1860–1935).5 Seventeenth-century Dutch and Barbizon works hung easily alongside Nattiers, Gainsboroughs and Fragonards in such collections, just as they did in Henry E. Huntington’s collection in San Marino, California.6 Impressionists could also be accommodated, as Jacques Doucet did in his hôtel in the Rue Spontini.7 Fig 1. View largeDownload slide Charles Joseph Watelet, Calouste Gulbenkian (1912, Calouste Gulbenkian Museum). Fig 1. View largeDownload slide Charles Joseph Watelet, Calouste Gulbenkian (1912, Calouste Gulbenkian Museum). Although unstinting in his pursuit of quality, Gulbenkian sought characteristic examples of a familiar and fashionable canon. The manner in which he acquired these works, however, was unusual. Gulbenkian sought to keep the wide range of dealers and curators he consulted on their toes by drawing on his own ‘Intelligence Service’, as well as on his insider’s knowledge of the stock exchange.8 This mingling of collecting and oil finance was crucial to pulling off the great coup of 1928–30, when Gulbenkian became the first to acquire works from the Hermitage. It epitomized what one curator dubbed ‘the element of the buccaneer in his character’.9 Although the Gulbenkian Foundation archive in Lisbon contains a wealth of invoices, correspondence and notes (as well as a memoir written by his curator10), Gulbenkian’s activity as collector has been little studied.11 In the absence of scholarly research, recourse has all too often been had to stereotypical ‘oriental’ tropes of possessiveness, intrigue and mystery. The collector viewed his treasures as a ‘harem’, we are told; hence his supposed unwillingness to allow outsiders to view it.12 The ‘harem’ analogy fails to acknowledge Gulbenkian’s generosity in lending to temporary exhibitions as well as museums. Nor does it explain why he allowed so many of his cherished acquisitions to remain in storage with dealers, without any urgent desire to set eyes on his new possessions. Sardanapulan visions of Gulbenkian immolating himself with his ‘harem’ also overlook the considerable thought he devoted to the question of where his collections would be reunited after his death.13 This essay is divided into five sections. The first considers Gulbenkian’s early influences and tentative first steps. The second considers the 1920s, the most acquisitive phase of Gulbenkian’s long collecting career, focusing on his collaboration with the commission agent Arthur Ruck and his relationship with the Duveen family. The third section addresses the Hermitage acquisitions, drawing on new sources from Russian state archives. The penultimate section considers Gulbenkian’s ‘de-centralizing’ of his collections after 1936, and his collaboration with Kenneth Clark, David Lindsay (28th Earl of Crawford) and John Walker. The final section offers an overall assessment of Gulbenkian as collector. ‘Convalescent pictures’: first steps, 1898–1917 Gulbenkian was born into a wealthy Ottoman Armenian family belonging to Istanbul’s amira or merchant élite.14 There is little evidence to suggest that his family were collectors, although they did trade in Caucasian and Persian carpets, alongside a wide range of commodities such as mohair, wool, opium and kerosene. After three years’ study at the École de Commerce in Marseille and King’s College London, Gulbenkian was sent by his father on a short trip to Baku in 1888. While success in carpet trading required a discriminating eye, the attention devoted to the history of the Caucasian carpet-weaving manufacture as well as the landscape descriptions found in Gulbenkian’s published account of this trip, La Transcaucasie et la Péninsule d’Apchéron (1891), suggest a particular sensitivity to craftsmanship and colour.15 In the late 1890s Gulbenkian joined forces with the fellow Ottoman Armenian Hagop Kevorkian, exporting Persian rugs, Iznik faience and other antiquities to London. This activity was not welcomed by Calouste’s younger brother Karnig, who suspected Calouste was more interested in forming his own collection than in making profits.16 In 1897 Gulbenkian settled in London, moving into a Bayswater terrace house, 38 Hyde Park Gardens, in 1900. He had all his possessions and those of his wife Nevarte (née Essayan) moved there from Istanbul. Unfortunately no packing lists or inventories survive, and so the earliest record of Gulbenkian’s collecting of Western art is a small black ledger he opened in 1898.17 Here he recorded his purchases (mainly from Agnew’s), exchanges and sales, up until 1903, when he sent many of the items to be auctioned at Christie’s. The ledger shows an interest in ‘fancy’ pictures of girls and nymphs, including Charles Joshua Chaplin’s In Disgrace, Jean-Jacques Henner’s Nymph Reclining by a Pool, Joshua Reynolds’s Nymph and Cupid and a Jean-Baptiste Greuze Girl Listening at a Door.18 Gulbenkian purchased Madame Du Barry from Agnew’s in 1902 (as a work by François-Hubert Drouais) for £2,205, selling it to a Madame ‘C. Heriot’ for £2,977 a year later. This may have been Cyprienne Hériot (née Dubernet), a shop-girl at the Grands Magasins du Louvre who married its owner, Olympe Hériot, becoming a leading socialite and collector. In 1910 Hériot decided that it was not a Drouais, and sought redress. Gulbenkian persuaded Agnew’s to take the blame for the misattribution and pay her the price she had paid Gulbenkian. Gulbenkian quietly reimbursed Agnew’s and took the painting back.19 At the 1903 auction the Chaplin, Greuze and the Reynolds (now downgraded to ‘after Reynolds’) all sold at heavy losses.20 A Guardi Venice from the Lagoon which Agnew’s had sold to Gulbenkian in 1901 for £420 (more than double what Agnew’s had paid in 1896) was bought in at £21.21 Gulbenkian was a wealthy man, but this was not a promising start: he told Agnew’s that he was tired of acquiring ‘convalescent pictures’.22 The ledger, as well as Agnew’s engraving stockbooks, show Gulbenkian purchasing mezzotints after eighteenth- and nineteenth-century English paintings by Joshua Reynolds (including Mrs Payne Galwey), Thomas Lawrence and John Constable.23 It would be easy to interpret this simply as an inexpensive first step towards acquiring paintings by ‘matching’ artists, a way of possessing (by proxy) works by favourite painters without risking painful discoveries such as those Gulbenkian had made in the case of his ‘Drouais’ and his ‘Reynolds’. Having bought two Constables24 from Agnew’s in 1900–1 (again, at big mark-ups) only to return them, Gulbenkian spent the rest of his life vainly endeavouring to acquire ‘a very fine Constable’, being underbidder for The Young Waltonians in 1946.25 Yet Gulbenkian clearly saw mezzotints as much more than proxies, valuing them as works of art in their own right. Alfred Whitman’s guides to print collecting feature among the many museum catalogues, art monographs and periodicals which Gulbenkian began buying around 1900.26 Gulbenkian was admittedly following an established fashion for English portrait mezzotints that saw prices rise to ‘astonishing’ levels from the 1880s until 1914.27 In 1920 he paid £1,100 for a proof of Mrs Carnac (J. R. Smith after Reynolds) and a positively eye-watering £2,000 for a ‘very fine’ first state of Lady Catherine Pelham Clinton (J. R. Smith after Reynolds).28 Gulbenkian began to express concerns that his activity might be inflating a bubble.29 In the years after the 1929 crash, mezzotint prices went into steep decline but Gulbenkian’s interest did not wane.30 Contemporaries attributed ‘the glory and the shame’ of the mezzotint boom to eighteenth-century mezzotints’ appeal to ‘sweetness’ and ‘conventional charm’.31 Gulbenkian must have been aware of such charges of meretriciousness, which could just as easily have been levelled at many of the paintings he acquired. In later life he told George H. Davey of Knoedler’s that he disliked female portraits which were ‘gnan-gnan [namby-pamby] or too artificial.’32 It was nonetheless important that Gulbenkian feel a connection with the lady depicted. In 1914 Gulbenkian had some doubts on this point with regard to a Hoppner he acquired from Agnew’s in 1900.33 ‘When she arrived I did not think we should become great chums’, he wrote to Charles Romer Williams of Agnew’s; ‘I hope that in the end she will prove a devoted friend and that we shall not divorce. I must say she is daily more flirting with me.’ It was not enough. After an eighteen-year marriage, the pair divorced in 1918.34 These early years also saw Gulbenkian’s only recorded attempts to acquire paintings direct from the artist.35 He approached Edward Poynter regarding the latter’s Skirt Dance (1898, private collection), in which a girl in a diaphanous dress performs for her wealthy Roman mistress. Balking at the £2,000 price, Gulbenkian noted that Poynter had sold his Visit to Aesculapius (1880, Tate) for just £1,000.36 Poynter was taken aback to discover that Gulbenkian knew this, and reluctantly explained why he had sold the work below price.37 Gulbenkian also approached John William Waterhouse enquiring after his Ariadne and Flora and the Zephyrs (both 1898, now in private collections), only to find both had already been sold.38 Gulbenkian admired these painters as well as Charles Joshua Chaplin for their ‘decorative’ depictions of the female form.39 As he told Davey in 1942, ‘apart from top-notch Old Masters it is a relaxing distraction for me to acquire the odd decorative painting of the modern school, which I keep for a time, then dispose of by giving it away.’40 He particularly wanted Davey to look out for an Alma-Tadema. It should include ‘nymphs, featuring the striking and harmonious tones of water, marble, flesh and flowers’ – but on no account horses or chariots. ‘I know this kind of painting is out of fashion,’ he conceded, ‘but I don’t care, as when I get tired of it I can dispose of it as I wish.’41 At his death a few such works remained on the walls at 51 Avenue d’Iéna, including Chaplin’s The Star (Fig. 2).42 Fig. 2. View largeDownload slide Charles Joshua Chaplin, The Star (c.1870, Calouste Gulbenkian Museum). Fig. 2. View largeDownload slide Charles Joshua Chaplin, The Star (c.1870, Calouste Gulbenkian Museum). ‘Friendly prices’, 1918–1927 Up until 1914 Gulbenkian mainly patronized Agnew’s in London, and Graat et Madoulé in Paris. After the war he spread his custom among several dealers: P. & D. Colnaghi, Duveen, Knoedler, Paul Rosenberg, and Georges Wildenstein, to name just a few. As early as 1915 Gulbenkian was writing to Joseph Duveen in New York that he missed him; Duveen was ‘very flattered’ to hear this. ‘I know that you mean this,’ Duveen replied, ‘and I can assure you that the sentiment is cordially reciprocated.’43 The following year Gulbenkian wrote that several ‘very big names’ had told him that they were interested in selling works of art, promising to pass these names to ‘Joe’ when he was next in London.44 When Joseph’s brother Ernest threatened to leave the firm and set up independently, the former wired Gulbenkian from New York. ‘As my kind friend I look to you during your conversations with him [to] gradually induce him stay with us but naturally [on] no account tell him I telegraphed [asking] you to intervene.’ Gulbenkian clearly did his best to oblige, and Ernest remained.45 Duveen’s sold a great deal of eighteenth-century French porcelain and furniture as well as a number of mosque lamps to Gulbenkian in the years between 1901 and Joseph Duveen’s death in 1939. With a few remarkable exceptions such as Domenico Ghirlandaio’s Portrait of a Young Woman (Fig. 3), Gulbenkian bought relatively few paintings from the firm, however.46 This was probably a result of Gulbenkian’s refusal to pay the high prices Duveen’s American clients accepted without a murmur, or perhaps even viewed as a reassuring sign of quality. Though Gulbenkian was clearly close to several members of the Duveen family and enjoyed their company, as a collector he may also have been put off by the firm’s high profile. On a visit by Edward Fowles of Duveen to 51 Avenue d’Iéna in 1931, Gulbenkian told Fowles of his pride at assembling a collection of masterpieces without Joseph Duveen’s help.47 The ‘element of the buccaneer’, the competitive side to Gulbenkian’s personality admired the Duveens’ success, but also viewed it as a challenge to be met. Fig. 3. View largeDownload slide Domenico Ghirlandaio, Portrait of a Young Woman (c.1490, Calouste Gulbenkian Museum). Fig. 3. View largeDownload slide Domenico Ghirlandaio, Portrait of a Young Woman (c.1490, Calouste Gulbenkian Museum). For paintings, therefore, Gulbenkian was more comfortable doing business with Colnaghi. Between 1918 and 1925 Gulbenkian acquired twenty-six paintings through the firm, for a total of over £45,000. Twenty-one were purchases on Gulbenkian’s behalf at auctions (mainly at Christie’s). They included his first Netherlandish masters: two small heads by Rogier van der Weyden, a skating scene by Aert van der Neer, landscapes by Jan van der Heyden (ex-Peel collection) and Jacob van Ruisdael as well as two paintings by Rubens (including Loves of the Centaurs).48 In the firm’s letterbook for 1924–7, Gulbenkian’s name appears more often than any other client. Known as ‘Gulby’ among the staff, Gulbenkian’s close relationship to the firm is evident from a 1926 letter Herbert Gosling of Colnaghi’s London office wrote to Gulbenkian’s wife, Nevarte, accompanying two passes to a Royal Academy private view. Gosling explained that Otto Gutekunst had told him ‘to suggest that for his “annual kindness” he will expect you to send him at least two Millionaires per annum who will spend a few hundred thousand in Pictures.’49 Gutekunst encouraged his client to aim high, advising him when the Six collection’s Street View in Delft by Vermeer was coming up for auction ‘That, a fine De Hoogh, Metzu and perhaps Terburg would give you, besides Remb[rand]t, Hals, Hobbeme [sic] and Ruysdael, about the best Dutch Art has given us in the 17th century and which has never been equalled.’50 When George Romney’s Mrs Davenport came up five years later, Gutekunst wrote that ‘If you do not let me buy it for you, at my discretion as far as price is concerned (and it may fetch up to £20,000 or even a bit more) I must wash my hands of the whole business!’ Having purchased a similar Romney (Miss Sarah Constable) in 1914 for £8,316, Gulbenkian decided not to pursue it. Mrs Davenport was purchased by Duveen at a ‘fancy price’ (£60,000), and sold two years later to Andrew Mellon.51 As Gulbenkian’s knowledge increased he turned to agents to enquire after desiderata which remained in private hands. Of course, Colnaghi and the Duveens had their own agents on hand to make enquiries. Gutekunst’s unnamed agent approached the Earl of Tankerville to ask after one of Romney’s best-known depictions of Emma Hamilton, Emma as a Bacchante. It was not for sale.52 Gulbenkian never relied on one source of information on what works were or were not available, and had regular, direct recourse to a large number of agents, who included the archaeologist Howard Carter and the art historian Frederik Schmidt Degener, but not the notorious dealer Georges Joseph Demotte (whose offer of services Gulbenkian ignored).53 Gulbenkian’s chief agent in the 1920s was Arthur Ruck of 4 Berkeley Street, London, ‘Agent for the Private Sale of Important Works of Art’, who seems to have begun advertising his services in art periodicals in 1916, and who went bankrupt in the crash of 1929.54 Although Ruck was also employed by Duveen, helped found the journal Apollo in 1924, and handled many important works from British aristocratic collections, little is known about him.55 Ruck first approached Gulbenkian in 1918, proposing a set of five Gobelins tapestries which ‘belong[ed] to a well-known noble gentleman.’56 Soon Ruck was on the hunt for Sèvres porcelain, Hoppners, Turners as well as the MacGregor collection of Egyptian antiquities, courting Lady Wantage, Lord Yarborough and other owners. Ruck’s role was to serve as a discreet intermediary to sometimes reluctant aristocratic sellers, taking a 10 per cent commission. Which side paid the commission is not clear, but in one letter Ruck asks Gulbenkian to ‘cover me’ (pay his commission, presumably) in the case of a Mrs Dowling and a Monsieur Gombrich, as ‘I have no commission arrangement [with them].’ Gulbenkian prized discretion, and had little time to waste travelling to, say Attingham Park on the off chance that Lord Berwick was willing to negotiate a ‘reasonable price’ for his Carpaccio Adoration, or was still asking for something around £60,000.57 Gulbenkian particularly disliked being shown artworks by their owners, as Lord Darnley insisted on doing as ‘no one else knows anything about the pictures.’58 Ruck helped Gulbenkian acquire several of his most important paintings, including the aforementioned Carpaccio and Turner’s Wreck of the Minotaur (ex-Yarborough collection). He was a shrewd tactician. After negotiating an Edmé Bouchardon bas-relief down to £600, Ruck wired Gulbenkian asking him to reply offering £500, so that Ruck could show the telegram to the owner (Henry Harris), further beating Harris down. The ruse worked: Gulbenkian got the relief for £500, commission included.59 Ruck did not have the space to store items, but Duveen and Colnaghi were always willing to help Gulbenkian. In 1926, shortly before 51 Avenue d’Iéna was completed, Colnaghi had thirty-two paintings in storage; Duveen held thirty items at 4 Grafton Street, including Thomas Gainsborough’s Mrs Lowndes-Stone, a Caffieri commode, Falconnet’s The Bathers and a pair of Kangxi vases.60 Although Ruck went bankrupt in 1929, Gulbenkian took care to pay subsequent commissions owed him for acquisitions such as Gustave Courbet’s La Neige in such a way that Ruck (rather than his administrator) received the money.61 Alongside his heavy purchases of art books and subscriptions to art periodicals, in the early 1900s Gulbenkian educated himself at the feet of curators. He hired the Louvre’s Camille Benoît as tutor in 1903. Their ‘conférences’ worked their way chronologically through the museum’s Italian and Dutch paintings, beginning with the Italian primitives, which Benoît conceded were ‘the most difficult’. Part of these afternoon sessions were ‘devoted to quizzes and exercises in which I will pose you questions.’62 Faced with ‘difficult’ schools Gulbenkian could be surprisingly humble. When the first Degas sale was drawn to his attention in 1918 by Charles Carstairs of Knoedler’s, he observed that ‘my first impression of Manet and Degas is one of curiosity, as I am not accustomed to that school, and being ignorant I must be educated.’63 In 1919–20 Gulbenkian bought two Degas at auction, both through Colnaghi: The Artist in his Studio (£2,205) and Les Trois Danseuses (£861).64 Paul Rosenberg’s gallery at 21 Rue La Boétie was aimed at just such autodidacts, and was arranged so as to encourage collectors to move from Barbizon works into more adventurous territory. In 1918 Rosenberg sold Gulbenkian Jean d’Aire, a Rodin bronze from the famous Burghers of Calais. Corots and Impressionists such as Claude Monet’s Les Bateaux (46,000 francs) followed.65 Seeking to maintain his ‘independence’, Rosenberg informed Gulbenkian that he did not take any commission when bidding on behalf of clients at auctions.66 He did, however, try to co-opt those advisers Gulbenkian sent to his gallery to inspect works. When Rosenberg invited Madoulé (of Graat et Madoulé) to enter some sort of collaborative arrangement for sales to Gulbenkian (a cut of sales, one presumes, in return for favourable reports to Gulbenkian) Madoulé refused. ‘I don’t need to tell you’, Madoulé informed Gulbenkian, ‘that Graat and I are above those kinds of practices’ – ‘[nous] ne mangeons pas de ce pain là.’67 Gulbenkian was a difficult client for dealers and agents alike, regularly demanding ‘friendly prices’ and expecting special terms: for Edward Fowles of Duveen, he was ‘one of the richest, most inscrutable and – at the final count – least profitable of our clients.’68 Lockett Agnew was angered when he discovered in 1915 that his adopted son, Charles Romer Williams, had agreed to allow Gulbenkian to return pictures in exchange for a full reimbursement, including commission. This was something ‘unheard of’ for Lockett, and the episode lived long in the firm’s institutional memory. Gulbenkian tried to assuage Agnew with flattery, writing that ‘Charlie’ had the makings ‘of a great business man’.69 For Gulbenkian, however, such a response was unusual. Any evidence or even suspicion that Gulbenkian was not being treated in a ‘friendly’ manner usually resulted in him sending a fiery cable or letter demanding a full explanation. Ruck came in for more of this treatment than anyone. In 1923 when Gulbenkian tried to blame him for his supposed failure to acquire Lord Glenconner’s Hoppner, The Frankland Sisters, Ruck did his best to defend himself: Please do not say I ‘bamboozle’ you and lose your time, and wire me saying I am neglecting your interests. I had nothing to do with the portraits of the Frankland Daughters. I secured a short option on the two Reynolds pictures [The Young Fortune Teller and Diana, Viscountess Crosbie] and had to sell them immediately. I felt quite sure you would not pay the price . . . I am pleased to offer you things, but in most cases I find it difficult to get you out of London to see them. The Berwick Carpaccio [Adoration] and Penryhn [sic] Rembrandt [Catrina Hoogshaet] for instance. You have promised time and again to arrange to see these, and one of these days you will be disappointed to hear someone has snapped them up, and you will no doubt blame me. I held the Turner ‘Marriage of the Adriatic’ for you for a long time, waiting for you to arrange to see it. The Trustee’s [sic] [of Ralph Brocklebank] got annoyed with me for not being able to make an appointment, and sold it to Joe Duveen through Agnews. I could have sold it to Joe Duveen, but wanted to keep it for you. Mr Andrew Mellon of America bought the Hoppner and paid a tremendous price for it, before I knew Glenconner would sell.70 As it happened, the two Reynolds and the Turner went to Henry E. Huntington; the Penrhyn Rembrandt (which Gulbenkian sought in vain for decades), was sold only in 2015.71 There are several underlying factors which explain why dealers tolerated such behaviour. Firstly, Gulbenkian was skilled at securing special terms today in exchange for promises of bigger deals tomorrow. After Gulbenkian persuaded him to sell two Fragonards at low prices (and throw in a Watteau self-portrait he had not wished to sell), Georges Wildenstein took comfort in the future. ‘I am sure that I will not regret the sacrifice I made, because I am certain of the big deals [‘grosses affaires’] we will make together later on.’72 As Gosling’s letter cited above humorously notes, Gulbenkian also helped by providing other millionaire clients. Ruck thanked Gulbenkian for bringing Walter Samuel (2nd Viscount Bearsted, chairman of Shell) to him, for example.73 Gulbenkian also lent money to dealers on easy terms; lots of it, in the case of Duveen. In 1921 Gulbenkian placed 1.2m francs (c.£900,000 at today’s values) at the London branch’s disposal.74 The most important thing Gulbenkian could offer in exchange for special treatment was investment advice. The shares of Mexican Eagle, Royal Dutch and Shell Transport and Trading (known simply as ‘Eagles’, ‘Royals’ and ‘Shells’) were the most talked-about equities on the London Stock Exchange in the early 1920s – and Gulbenkian was the most talked-about insider. As the Figaro noted in 1920, ‘Everyone claims to know Monsieur G . . . or the friend of a friend of Monsieur G., one of the today’s highest-profile personalities.’75 In 1920 Gulbenkian wrote to Duveen remarking that he knew exactly how much profit the latter had made following his investment advice. In return he expected ‘a very prominent Work of Art, one in which I shall see increasing value and in any case, on which, when I look at it I shall exclaim with great joy “this is from Joe”.’ At the end of the letter he put it more simply: ‘Stick to your shares and find something very fine for me.’76 Otto Gutekunst and Alec Martin of Christie’s had also invested in these shares. ‘What about M[exican] Eagles!’ expostulates Gutekunst at the foot of one 1921 letter. Martin sometimes found it difficult to write about anything else. ‘You may think I am over nervous,’ he writes in one letter, ‘but I have put all my savings into these shares,’ exposing ‘my wife and six little ones, if any thing happened to me.’77 Had these dealers cut ties with Gulbenkian they would have struggled to make informed investment decisions: Royal Dutch Shell published very little information about its subsidiaries. As Martin’s letter indicates, the stakes were much higher than the sums expended (or not expended) on this or that work of art. A similarly hidden set of non-art related costs and rewards lay behind Gulbenkian’s coup in Russia. ‘The Russian Art Treasures bogey’, 1928–1935 Gulbenkian had shown an interest in acquiring works from Russian collections as early as 1922, when he asked Gutekunst for his view. ‘The Russian Art Treasures bogey has been cropping up on and off during the last two years’, the latter noted, adding ‘We do not believe in it. I think the Allies would taboo the thing and tell all the world that it w[oul]d buy stolen goods at their own risk and peril and that they w[oul]d eventually have to be given back.’78 Martin Conway’s well-publicized trip to Leningrad two years later, however, only served to whet appetites further.79 Marked-up Hermitage catalogues show the care Gulbenkian took in preparing his lists of desiderata, taking advice from Schmidt Degener and the jeweller and goldsmith Marcel Aucoc, who visited St Petersburg and Moscow in October 1928 to reconnoiter.80 Gulbenkian negotiated in Paris with a number of representatives from Antikvariat (the Soviet Art Export Agency), including a M. Birenczweig, Grigor Samuell, and A. M. Ginzburg. They declared some items on Gulbenkian’s list to be unavailable and added new ones. Lists were repeatedly altered on the fly. As one might expect, Gulbenkian’s first list of September 1928 was too long, and he was told to restrict himself to eight paintings and fourteen items of silverware (see online Appendix 1). Gulbenkian wanted these as a job lot, but was prepared to deal for Giorgione’s Judith or the Rembrandts separately. Aucoc travelled to Berlin in April 1929 to receive the first consignment, purchased for £54,000, which included twenty-four pieces of silverware by François-Thomas Germain and other eighteenth-century French masters, a Louis XVI bureau plat and three paintings, none of which had been on Gulbenkian’s lists: Dieric Bouts’s Annunciation and two Hubert Roberts depicting Versailles (including Le Tapis Vert, Fig. 4). Fig. 4. View largeDownload slide Hubert Robert, Le Tapis Vert (c.1775–1777, Calouste Gulbenkian Museum). Fig. 4. View largeDownload slide Hubert Robert, Le Tapis Vert (c.1775–1777, Calouste Gulbenkian Museum). When he was shown Gulbenkian’s list Hermitage director Sergei Troïnitsky objected, noting that if this sale went ahead the Hermitage could drop to the second or third rank of museums.81 In the past curators had used discussions over valuations to play for time, and the makeup of the first Gulbenkian consignment suggests that these tactics worked. But not for long. The selection and valuation of works was entrusted to ‘shock brigades’ sent by Narkompros (the Education and Culture Ministry). Works actually on display were no longer exempt.82 Although the first consignment had taken seven months, the following three consignments were negotiated quickly. In February 1930 Gulbenkian secured Rubens’s Helena Fourment and fifteen further pieces of silver for £155,000, taking delivery the following month. In June came Jean-Antoine Houdon’s marble Diana along with two Rembrandts (Titus and Pallas Athene, Fig. 5) and three other paintings (by Lancret, Watteau and Ter Borch), for £140,000.83 Bought directly from the artist by Catherine the Great, Diana had stood as ‘a kind of greeting card and invitation’ to the Hermitage since its move there from Tsarkoye Selo in the early nineteenth century.84 The Ter Borch was a last-minute change; Hermitage curators had sent it in lieu of Pieter de Hoogh’s Lady and Kitchen Maid.85 The fourth and last consignment consisted of just one item: Rembrandt’s Old Man (£30,000). The third and fourth consignments included Wildenstein as a third party, who paid £95,000 for options on five works. He duly received Titus, the Lancret and the Watteau, and had an option on the Rembrandt Old Man as well. Gulbenkian decided to keep this Rembrandt for himself, reimbursing Wildenstein £40,000.86 Fig. 5. View largeDownload slide Rembrandt Harmensz van Rijn, Pallas Athene (c.1657, Calouste Gulbenkian Museum). Fig. 5. View largeDownload slide Rembrandt Harmensz van Rijn, Pallas Athene (c.1657, Calouste Gulbenkian Museum). In the end Gulbenkian acquired three out of his original list of nine masterpieces, as well as a third Rembrandt, plus a range of items which greatly enhanced his existing collections of silver and furniture. All purchases were delivered to 51 Avenue d’Iéna, if not literally by the diplomatic bag, then at least under the seal of the Persian legation. Thanks to Gulbenkian’s status as conseiller financier to the Persian legation in Paris, customs officials were barred from opening the cases, and no duty was paid.87 It would not be the last time Gulbenkian used his diplomatic status to evade customs as well as media coverage. Having sketched out the phases of Gulbenkian’s campaign of acquisitions we can now consider the personal and business relationships which made them possible. Gulbenkian’s friendship with Alfred Chester Beatty laid the foundation of these purchases in the mid-1920s, when the pair joined forces to secure control of two mining concessions granted by the Soviet Union in 1925, under the New Economic Policy: Lena Goldfields and the Tetiuhe Mining Corporation.88 Beatty also tried to serve as honest broker in the very public dispute between Gulbenkian and Henry Deterding, chairman of Royal Dutch Shell, for control of Venezuelan Oil Concessions Ltd. A well-informed collector of manuscripts as well as wealthy mining magnate, Beatty was conspicuous by his habit of always dining alone, a custom Gulbenkian shared.89 He was the only one of Gulbenkian’s circle who knew what it was like to juggle business and the many demands of serious (as opposed to trophy) collecting: scouting for acquisitions, keeping up with the literature, cataloguing one’s collections and, last but not least, enjoying the contemplation of them. Gulbenkian recognized that Beatty was better at this juggling than he was, and regularly turned to him for advice.90 Western-owned, profit-seeking concessions were very few in number inside communist Russia. Tetiuhe and Lena faced difficulties moving equipment in and profits out of the country; western staff were occasionally arrested on charges of espionage. In an attempt to resolve the situation Gulbenkian and Beatty met in March 1927 with a representative of Narkomtorg (People’s Commissariat for Foreign and Domestic Trade), Georges Piatakoff. Piatakoff became vice-president of Gosbank (State Bank of the USSR) the following year, rising to president in 1929. The importance of their relationship was as evident to the the People’s Commissar for Foreign Affairs, Georgy Chicherin, as it was to Gulbenkian. ‘Gulbenkian is a very important figure’, Chicherin observed, ‘a former ally of Deterding, who subsequently turned against him. He is the best card we hold. All those who know him from England are talking about him as a strong financial and political power.’91 Alongside the mining ventures Gulbenkian sought to persuade the Soviets to appoint him their intermediary in negotiations with Royal Dutch-Shell, Anglo-Persian and Standard Oil. Since 1920 the Soviets had sown dissension among these oil majors, stymying efforts by a ‘united front’ (Front Uni) to impose an embargo on Russian oil. To the chagrin of foreign investors who had seen their Baku- and Grozny-based oil companies nationalized in 1920, these firms struck a series of deals with the Russians, buying what many considered to be ‘stolen’ oil. Russia’s share of the world oil market, however, remained tiny relative to its reserves. Gulbenkian proposed to give Russia a seat at the big table, allowing the Soviets to enter into long-term cartel agreements with the majors and secure a larger share of world markets.92 The question of Russian oil and of foreign investment in Russia more generally was a complex and controversial one in the 1920s, when many countries had yet to establish diplomatic relations with the communist regime. The detail is of little concern to the story of Gulbenkian’s collecting, but the hitherto unacknowledged relationship between the apparently distinct activities of business and collecting is significant. Gulbenkian’s mining interests and his offer to help sell Russian oil abroad provided him with high-level contacts in Moscow as well as the goodwill of the Soviet regime. Although Piatakoff repeatedly reminded Gulbenkian that art and museums did not lie within his bailiwick, Gulbenkian regularly appealed to him when negotiations with Antikvariat bogged down. Gulbenkian told Piatakoff that he considered the supposedly ludicrous prices asked of him ‘a personal insult’. He warned that if any of his desiderata were sold to someone else ‘I would consider that as an “unfriendly act”.’ Piatakoff passed these comments to Commissar for Trade Anastas Mikoyan: ‘We really need to act more decisively’, Piatakoff observed, or else Gulbenkian might lose interest in the oil discussions.93 This tactic served Gulbenkian well: the head of Antikvariat’s Paris office, A. M. Ginzburg, found himself facing Gulbenkian’s entreaties on one side, and on the other, pressure to oblige him from Piatakoff, Gostorg (the main USSR export body), the Foreign Ministry and the Concessions Committee.94 Once the Soviets decided not to take up Gulbenkian’s aforementioned offer to represent their oil interests, however, the art sales stopped. A Trotskyite, Piatakoff’s influence diminished as Stalin began rooting out rivals.95 Gulbenkian endeavoured to secure Greek coins and Manets through Zatzenstein of the Berlin gallery Mathiessen in 1931–2, but the window of opportunity had closed.96 Beatty and Gulbenkian divested from Tetiuhe Mining and Lena Goldfields.97 Sales of art to others, however, continued until 1935. By that point more than 24,000 works had been deaccessioned from The Hermitage, including 2,880 paintings.98 Newly-discovered files from the Russian State Archives shed light on the internal squabbles among competing export agencies, central committees and curators such as Troïnitsky, who did everything they could to prevent sales. Together with Gulbenkian’s papers they paint a picture of confused rivalry among a galaxy of Soviet agencies inside and outside Russia. Gulbenkian noted that Antikvariat’s Berlin and Paris agents worked at cross-purposes.99 When these agents claimed that Gulbenkian’s offers were too low he appealed to auction records or proposed consulting third-party experts.100 One Soviet official criticized Ginzburg for being too tough a negotiator: The difficulty is that the people who know their way around this business are opposed to selling, and put up as much resistance as they can to prevent it. Me, all I’m interested in is results, that we get more money. Of course, it is better to take 50 than 15, but if you want to get 50 for the painting which costs 15, you get nothing. This Gulbenkian I spoke to is the greatest obsessive of them all. I met with him several times and he kept telling me ‘For God’s sake, sell me a painting.’ He wanted to buy all our junk. These paintings – they’re nothing but a load of rubbish. Negotiations should, he continued, only be entrusted to people who knew nothing about art. Otherwise ‘sabotage by the intelligentsia’ would continue. It was worrying that Ginzburg’s knowledge appeared to be growing on the job: ‘he can now tell his Raphaels from his Rembrandts.’101 Gulbenkian’s view of the ethics of these sales was laid out in a memo to Piatakoff of July 1930, a document reproduced in full in an online appendix to this article. His argument is summarized in another letter to Piatakoff: ‘I keep on repeating to your representatives that as a friend of your country I certainly do not like the idea of its works of art being sent abroad’, he wrote, ‘but if, on the other hand, they are going to be sold, I should have the first opportunity. Can there be anything fairer?’102 Gulbenkian drew a distinction between works in public museums and those which had been in private collections, arguing that the latter should certainly not be sold. He noted that sales to Mellon had been very high-profile, and argued that such sales therefore came at far greater cost in terms of ‘prestige, propaganda and publicity’ than sales to Gulbenkian himself.103 Gulbenkian’s purchases from the Hermitage in 1928–30 epitomized the ‘buccaneer’ element of his collecting persona. His acquisitions, like his desiderata lists, were conservative in their taste. Even with all the riches of the Hermitage at his command, Gulbenkian still left room on his list for a Hoppner. He did, admittedly, include earlier Italian works by Giotto and Perugino on the final list (online Appendix 1), but by that point (November 1930), the exercise had become somewhat academic, in view of the deteriorating situation in Russia. Although he did acquire one or two works by fifteenth-century Netherlandish and German artists, Gulbenkian never showed much interest in Italian art before Raphael.104 Duccio, Uccello, Perugino, Mantegna, Botticelli and the Bellini are hardly mentioned at all in Gulbenkian’s correspondence. He conceded in 1950 that a Giovanni Bellini (Madonna in a Red Cloak) offered to him was ‘beautiful, and the master a prestigious one’, but it failed to rouse that ‘emotion’ which he ‘always look[ed] for when acquiring a work of art.’105 If nothing else, the Hermitage acquisitions represented a double triumph over ‘Joe’: not only had Gulbenkian won the race to have first pick (at prices which appear modest in comparison to those paid by later purchasers), but the Hermitage acquisitions formed the basis of his boast to Fowles, of having formed a collection of masterpieces without recourse to Duveen. Our understanding of this episode has certainly been deepened by recent scholarship on Soviet-era sales.106 Without the broader context, reconstructed here from previously unexamined material among Gulbenkian’s papers and in Soviet state archives, it has nonetheless been difficult to explain why Gulbenkian won this race. We can now appreciate how his approach, leveraging mining and oil interests to acquire a specific group of masterpieces, represented a via media between that of Averell Harriman and Armand Hammer (fellow mining magnates who exported artefacts in bulk) and that of Knoedler and Duveen (high-end art dealers who sought quality, not quantity). The Hermitage acquisitions marked a shift in the character of Gulbenkian’s collecting. After 1930 his collecting of ‘decorative’ paintings was largely restricted to works by Stanislas Lépine and Henri Fantin-Latour. In 1931 he acquired Paul Émile Chabas’s September Morn (Metropolitan Museum of Art) from Leon Mantacheff; this depiction of a nymph standing in Lake Annecy was unremarkable, but had achieved notoriety in the United States, where reproductions of it enjoyed a succès de scandale.107 Though Gulbenkian gave away September Morn, at his death his collection included seven of Lépine’s Seine views, and four Fantin-Latour still lives. The pace of collecting declined dramatically after 1930. What with Gulbenkian’s fondness for giving away works as well as his insistence that he would live as long as his grandfather (i.e. to 106), it is potentially misleading to presume that the collection as left at his death in 1955 was complete. Francisco Goya’s Countess de Chinchón, the Penrhyn Rembrandt, Constable’s Salisbury Cathedral and several other desiderata of the kind Gutekunst called ‘big-big, big game’ remained at large in 1955, while other works remained in the collection which Gulbenkian would probably have disposed of, had he had more time in Paris.108 These caveats aside, the figures are conclusive: of the paintings and pastels in the collection in 1955, ninety-one were acquired in the fifteen-year period 1914–29, forty-five in the years 1930–45. Even before its completion in 1927, Gulbenkian had expressed the desire to make 51 Avenue d’Iéna (Fig. 6) into a house museum, intended to preserve his taste for posterity. As his curator, Marcelle Chanet, noted in her unpublished memoirs, the house was anything but a family home. Though Gulbenkian had given considerable thought to every aspect of the building’s design and the arrangement of his art, the needs of his family took second place to those of the collection. When in Paris Gulbenkian slept at the Ritz, his valet bringing over clothes from Avenue d’Iéna each morning.109 Fig. 6. View largeDownload slide The Picture Gallery at 51 Avenue d’Iéna, Paris (photo c.1955, Calouste Gulbenkian Museum). Fig. 6. View largeDownload slide The Picture Gallery at 51 Avenue d’Iéna, Paris (photo c.1955, Calouste Gulbenkian Museum). A draft (unexecuted) 1924 codicil to his 1916 will directs that in the event of Gulbenkian’s death construction of the hôtel should continue, and that it ‘should remain as a kind of family museum with all its contents,’ passing to his son Nubar (who would have to right to live in it) and then on to other ‘male issue’. If that issue died out it was to become a ‘private museum placed under national supervision.’ This ‘Gulbenkian Museum’ would be administered by a committee made up of ‘efficient and eminent scholars and critics of art’, who would advise a board of trustees on acquisitions. ‘I do not desire that it should be freely open to the public as [a] public museum but only to scholars and to genuine lovers of art.’110 As Salles’s comment cited at the beginning of this essay demonstrates, Gulbenkian’s collection was notoriously difficult to access in the 1930s. This codicil suggests that the collector intended to keep it that way. ‘Happy surroundings’, 1936–49 In December 1935 the director of the National Gallery, Kenneth Clark, wrote to Gulbenkian asking to see his collection. ‘I know that you are naturally unwilling to show it to everyone’, he wrote, ‘but we have many friends in common, including Mr Chester Beatty and Mr Upham Pope, who have told me that you might be willing to allow me to see your pictures.’111 Gulbenkian was impressed with Clark’s appreciation of all his collections, not simply the paintings. As the international situation deteriorated in 1937, Gulbenkian reversed his policy of centralizing his collections in Paris, confiding in Clark a wish to ‘decentralize my works of Art’ until such time as he reached a final decision on where his collection would finally rest. In the meantime ‘I am most anxious they should have happy surroundings, and procure public enjoyment. No one will understand me better than yourself and for this I am very grateful.’112 In 1936 Gulbenkian lent twenty-nine paintings to the National Gallery, as well as twenty-six Egyptian statuettes and other artefacts to the British Museum. Renoir’s Mme Renoir on a Sofa was purchased by Gulbenkian on Clark’s advice and delivered to the National Gallery.113 They would remain there until 1939, when they were evacuated with the rest of the nation’s treasures. When the paintings returned to a bomb-damaged Trafalgar Square in 1945 Clark ensured that a room was devoted to the Gulbenkian loans, and tried to find a publisher willing to produce a catalogue he had agreed to write, to replace the rather thin work produced by the Gallery in 1937.114 Unfortunately Clark’s successor as director, Philip Hendy, proved unwilling to accord Gulbenkian the deference the latter had come to expect. For his part Gulbenkian (who never visited London after 1940) failed to appreciate the constraints of space, finance and manpower under which Hendy was labouring, and chose to interpret the failure to display all of his loans as evidence that London no longer appreciated them. As he wrote to Clark in May 1946, ‘my dear “children” should in no way remain in [sic] sufferance.’115 The chief curator of the National Gallery of Art in Washington, John Walker, seized on the opportunity to court a disaffected Gulbenkian. The Egyptian loans had remained in their crates at war’s end, again for want of space to display them within the bomb-damaged British Museum. In 1946, during a transatlantic crossing, Walker heard of this state of affairs from the Museum’s director, Sir John Forsdyke. Walker immediately made arrangements through the US embassy in Portugal to send information about the gallery (including a film) to Gulbenkian, now ensconced in the Hotel Aviz in Lisbon.116 It was intimated that Forsdyke had eagerly welcomed the suggestion that the collection be lent to Washington. David Lindsay, the Archbishop of Canterbury and other British Museum trustees managed to distance themselves sufficiently from Forsdyke to keep the collection for another year.117 When the US State Department sent him an official invitation to lend, however, Gulbenkian told the British that he felt honour-bound to oblige. Although he claimed to have been taken by surprise, the official invitation had been his own idea.118 In 1950 the paintings from Trafalgar Square joined the Egyptian collection at the National Gallery of Art, Washington. There they would remain until 1960, when they recrossed the Atlantic, this time to Lisbon. For both Clark and Walker the decision to reunite the collections in Lisbon represented the final rejection of plans to build Gulbenkian annexes to their respective galleries. Both had collaborated closely with Gulbenkian on these plans, and both had, at different times, believed Gulbenkian to have reached a firm decision in favour of their institution. In 1938 Clark had reported to his board in London that arrangements with Gulbenkian were sufficiently advanced ‘to make the possibility of a bequest almost a certainty.’ Gulbenkian would leave his collection and his enormous oil fortune to the National Gallery, the income from which would pay for the annexe, acquisitions and gallery publications.119 In 1951 Walker wrote to his director, David Finley, that ‘if [Gulbenkian] does not decide on another solution I think there is a will leaving the pictures and Egyptian pieces to us!’120 Both Clark and Walker secured their government’s approval of plans to build Gulbenkian annexes on public land next to their galleries, on sites where the Sainsbury Wing and East Wing now stand. Right up until the end, Gulbenkian kept both schemes on the back burner. The story of Gulbenkian’s testamentary plans is complex.121 Although the collector was genuinely unsure where to leave his collection, he was also spinning out discussions in hopes of favourable treatment by the tax authorities. The similarities with Gulbenkian’s friend Chester Beatty are striking. Like Beatty, Gulbenkian reviled ‘socialism’, felt himself ill-used by Forsdyke and expected special treatment in exchange for the promise of future bequests. Between 1949 and 1953 Beatty moved his collections of paintings, manuscripts and Egyptian artefacts from London to Dublin, where he initially enjoyed special treatment from Éamon de Valera’s government. All too soon, however, the tax advantages (relative to British tax rates) were whittled away, while Beatty was disappointed by how few people visited the Chester Beatty Library. As Beatty’s curator had warned, the Library was (in its early years, at least), ‘an isolated, if decorative institution’.122 There is, surprisingly, no evidence that Beatty and Gulbenkian conferred in these years. Gulbenkian also spun out discussions in the hope of wearing down London and Washington’s reluctance to honour his insistence that his collection be exhibited together. ‘You know very well my friend,’ he wrote to Walker in 1952, ‘that I shall never agree to separate my children.’ Walker replied that Andrew Mellon had founded the National Gallery of Art with the specific aim of avoiding ‘en bloc’ hangs. Where there were enough works of a certain school from a specific collection, however, the Gallery did hang them together.123 In the case of the 1909 Mond Bequest the National Gallery in London had provided a benefactor with assurances, opening a dedicated Mond Room (funded by a grant from the estate of chemicals magnate Ludwig Mond) in 1928, only to change course subsequently.124 Concerned that Gulbenkian might be lured back to London, Walker ensured that this information reached Gulbenkian in Lisbon. Feeding Gulbenkian pieces of (occasionally tendentious) information calculated to put the British in a bad light was another aspect of Walker’s plans, which the Chief Justice of the United States (a Gallery trustee) was ‘emphatic . . . would not be cricket’.125 Gulbenkian duly challenged Lindsay (who was a trustee of both the British Museum and National Gallery) about Trafalgar Square’s policy on gifts. ‘The Gallery would no doubt much prefer a gift without conditions – as we all would’, Lindsay conceded, ‘but it might take a very different line if the proposed gift were an important one – indeed, I think it would be crazy if it did not do so!’ Gulbenkian reassured Lindsay that he had yet to decide in favour of Washington, London or anywhere else. Indeed, Francis Henry Taylor, director of the Metropolitan Museum in New York was coming over to Lisbon to discuss his own proposals for Gulbenkian’s legacy.126 Gulbenkian’s philanthropic plans touched on a number of questions that continue to confront private collectors and public institutions’ relations with each other. One question surrounded the etiquette and expectations surrounding loans. Before accepting Burne-Jones’s Mirror of Venus in 1924 Aitken (director of the Tate) had warned Ruck that ‘my Board rather hesitates to accept even good pictures on temporary loan, unless it has some assurance that there is some prospect of any ultimate gift on bequest, or unless the owner fixes a price at which he would give the Trustee’s [sic] the option of purchasing, if he should desire to sell.’127 When Gulbenkian requested the return of Flore, a sculpture by Jean-Baptiste Carpeaux which had been on loan to the Louvre for twenty-two years (1905–27), the curator of sculpture Carle Dreyfus reportedly ‘told Gulbenkian in plain language what he thought of him’, leading a ‘very red-faced’ Gulbenkian to donate a bronze by André Riccio ‘as compensation’.128 Walker’s superiors suspected Gulbenkian of using his loans to string Walker along, when (as Finley put it) ‘it simply seems as if Mr Gulbenkian wants to send them abroad just for security.’129 Although it may be churlish to do so, it is worth noting that by his loans to London and Washington Gulbenkian spared himself having to make his own arrangements to protect his treasures (as he had done in 1918), not only in 1939, but also in 1950, when many felt World War iii to be imminent.130 Thanks to their hopes of a bequest as well as to the genuine friendship they had each formed with Gulbenkian, Clark, Lindsay and Walker joined George Davey and Gulbenkian’s restorer, Angenitu Martinus (Martin) de Wild, as unpaid advisers and agents: scanning forthcoming sales, viewing prospective purchases and liaising with owners and dealers.131 As part of his search for works of the highest quality Gulbenkian was prepared to give these advisers considerable room for manoeuvre. They were asked to exercise their own judgement over price when bidding at auction, for example. These advisers also suggested works which were already in his collection, but which did not to ‘belong’. As we have seen, Gulbenkian had long sought works by Hoppner, but in 1948 he wrote to Clark noting that ‘mutual friends’ (these included Lindsay) had counselled him to dispose of Miss Beresford, as ‘Hoppner has no place in my collection’.132 Gulbenkian was struck by the contrast between the prices he paid in 1943 at the New York sale of J. P. Morgan Jr.’s collection, for works which had reached much higher prices when acquired in London a couple of decades earlier: works such as the aforementioned Miss Beresford, Nattier’s Madame Delaporte and Thomas Lawrence’s Lady Coningham. The Duveen bubble (Joe had died in 1939) was clearly deflating. ‘All these fluctuations’, Gulbenkian noted, ‘are very interesting for an economist.’133 There was little sense, however, that shifting values had weakened Gulbenkian’s affection for eighteenth-century painting. While in wartime exile in Lisbon, Gulbenkian sought to acquire works from another refugee, Henri de Rothschild. As Rothschild noted, Gulbenkian had frequently visited his collection of Guardis, Chardins and other eighteenth-century French masters (largely inherited from Henri’s grandmother, Charlotte) when it was installed at La Muette, his chateau on the western edge of Paris. In 1943 it was not clear if the Chardins survived; sent to England at the start of the war, many had apparently been destroyed by enemy action in 1941–2.134 Owing to wartime currency controls, Gulbenkian had difficulties in paying for purchases – difficulties he may have hidden from Rothschild, who ended up feeling that his trust had been abused. Rothschild particularly resented discovering that these purchases were being made by Gulbenkian himself, not (as the latter had claimed), ‘on behalf of the National Gallery’.135 Gulbenkian nonetheless acquired several pieces, including a large Maurice Quentin de La Tour pastel (Louis Duval de l’Épinoy) which he remembered from the Doucet Sale, ‘an exceedingly fine portrait’, if ‘a little “maniéré”.’ He was also pleased to acquire ‘a really magnificent antique ewer, of a kind of hard-stone resembling lapis-lazuli, mounted in pure gold by Gouthière’, for which ‘in the old days’ Gulbenkian had offered £2,500.136 Gulbenkian also sought masterpieces by artists in whom he had never before expressed much interest. When Clark visited New York in 1946, Gulbenkian asked him to inspect Paul Gauguin’s Ia Orana Maria at Sam A. Lewisohn’s, from whom he had earlier (1943) acquired Manet’s Soap Bubbles.137 Davey had sold this, ‘the finest Gauguin in the world’, to Sam’s father Alfred in 1918, and had urged Gulbenkian to buy it.138 As for Chinchón, Gulbenkian had never shown much interest in Goya or Spanish art generally. He first made contact with the Duchess of Sueca regarding Chinchón (which her husband, the duke, owned jointly with several relations) in early 1942, when he found himself passing through Madrid. Further negotiations continued off and on through Alfred W. Barth, vice-president of Chase National Bank, and by 1946 Gulbenkian felt his chances of success were ‘fifty-fifty’.139 A price of 4 million pesetas was agreed in 1950, and Gulbenkian began thinking about how to export the work. He wanted the portrait transported to Washington under the seal of the American embassy (the same way he proposed to export his art from London and Paris), a way around the French Ministry of Foreign Affairs, which refused to issue an export licence.140 When Walker hesitated, Gulbenkian’s letters took a chillier tone. ‘You will quite appreciate, my friend, that if you leave everything to me to tackle, it will not be easy’, he wrote; ‘Surely when the State Department wrote to me, I could not expect that elementary moral support would be missing.’141 It is perhaps fortunate for Ibero-American relations that the duke reconsidered, then toyed with the Prado’s offer of 3.5 million pesetas, then decided to hold on to the portrait (it entered the Prado in 2000).142 À la recherche du Corot perdu, 1949–1955 Alongside Clark, Davey, Gutekunst, Lindsay and Walker there is a less-well-known figure who knew Gulbenkian’s collection better than them all, Gulbenkian included: Marcelle Chanet. Chanet was hired in 1927 to arrange the collections in their new Parisian home and served as de-facto conservateur up until Gulbenkian’s death and beyond. Chanet was one of the few members of Gulbenkian’s staff to join him on one of his tours in the 1930s, and the insights into his character revealed in her unpublished memoir were probably gleaned from discussions on board the Narcissus and other private yachts. Gulbenkian never consulted her regarding acquisitions or testamentary plans. Chanet was probably hired for her secretarial skills; there is no evidence that she had any prior experience of the art world or any relevant training. Her memoir nonetheless makes it clear that she considered 51 Avenue d’Iéna ‘a temple worthy of his idols’. This temple’s ‘double desertion’ by Gulbenkian and his art was, in her view, a national misfortune for France.143 Chanet’s account suggests that Gulbenkian’s relationship to the building and its contents changed during the war. Having followed the French government on its flight from Paris to Vichy in 1940, Gulbenkian did not see Paris again until 1949, by which time he was eighty. ‘He did not display the slightest satisfaction at seeing his treasures again’, Chanet recalled; ‘He seemed not to recognize the paintings which had been returned to their usual places.’ Instead ‘he searched for an imaginary Corot, Le Village Endormi, which he could see clearly in his mind’s eye.’ Despite ordering Chanet to search a catalogue raisonné of Corot’s works in vain for this painting, three years later he was still writing from Lisbon that ‘he expected to find it in the house’ on his next visit to Paris.144 Gulbenkian’s last acquisition was a Lépine, Rue Saint Vincent à Montmartre, acquired through Davey in June 1953 for 700,500 francs. Davey wrote reassuringly to Gulbenkian that even his colleagues at Knoedler’s had no idea Gulbenkian had bought it, believing rather that it had been purchased for stock.145 Although Gulbenkian told Davey he was still interested in the Earl of Radnor’s Velázquez (Juan de Pareja), Prince Liechtenstein’s Quentin Matsys (Portrait of a Canon) and other works, his health was now fragile and his interest in his collection fading.146 He had outlived his wife Nevarte and many of his associates from the world of banking and oil, as well as the art world. His memorial service, held on 20 July 1956 at the Armenian church he endowed in South Kensington, St Sarkis, was very thinly attended. ‘No one at all from the oil world’ bothered to attend, noted one member of his London staff. Apart from the immediate family, executors and staff ‘only [Vere] Pilkington of Sotheby’s and D[a]vey of Knoedler’s were there.’147 Far from being his harem, Gulbenkian’s paintings became his children. Although Gulbenkian used this metaphor himself, he would probably not have welcomed the analogy as drawn by Chanet: With the exception of the nine years when they were reunited in the hôtel on Avenue d’Iéna (1927 to 1936), C. S. Gulbenkian’s collections shared to some extent the fate of his own family: a great solicitude . . . but from a distance; correspondence by letter, cable and telephone over thousands of miles; specialist practitioners ordered hither and yon, but hardly ever a celebration spent with all the family together, and not a gallery without its deplorable gaps. These treasures, dispersed across two continents, are the fruit of more than half a century of research, intrigues, sometimes struggles. Each represents a conquest this tenacious collector had dreamt about for years before the right occasion presented itself.148 There is an almost conspiratorial aspect to the pursuit of great masterpieces which a ‘buccaneer’ like Gulbenkian relished, and which was particularly evident in his Hermitage purchases. Researching provenance, evaluating the condition and attribution of works, gathering intelligence from multiple sources and studying market fluctuations all drew on the same skills which had made Gulbenkian such renowned and envied figure on the stock market and within the oil industry. Though individual pieces could indeed represent ‘conquests’, Gulbenkian’s distaste for publicity ensured that they never became trophies of celebrity. This was evident from his desire that Clark produce ‘a scholarly catalogue, not a bumptious one like the American similar [sic] works, but very fine, sober and dignified.’149 ‘I take a deep interest in the study of art’, Gulbenkian wrote in 1928, ‘and whenever I have time I study the periodicals and endeavour to educate myself and derive benefit.’150 Although he never educated himself to appreciate Cézanne or what he termed ‘the ultra-modern school,’ his taste did become more catholic with time.151 Keeping up with the latest literature on art, scanning sales catalogues and liaising with a bevy of agents and dealers made considerable calls on Gulbenkian’s time. Collecting, in short, was not an activity fitted into Gulbenkian’s leisure hours – which is just as well, as he maintained a punishing work routine throughout his life, and never retired. Other aspects of Gulbenkian’s collecting persona are also worth emphasizing, notably his interest in gardens and landscape architecture, to which he gave full rein at Les Enclos, his Normandy estate near Deauville (now the Parc Calouste Gulbenkian). An eye for colour and respect for craft were partly an inheritance from his amira ancestors. Although these moments of contemplation have left few traces, the contemplation of his treasures enabled an otherwise restless business titan to snatch moments of restorative calm. As he confided in his son Nubar (who showed no interest in collecting anything, except wives), art provided ‘consolation and relief’ whenever one found oneself faced with disappointments in life.152 In the final assessment, forming a collection appealed to Gulbenkian’s desire for a legacy. This motivation only became more important as it became clear that he had failed to groom Nubar as a fitting successor. Gulbenkian’s loans to great public collections began a complex series of parallel negotiations with curators, directors and trustees: while he accepted the latter’s guidance on which artists did and did not belong in a collection of great masterpieces, Gulbenkian also fought to protect the integrity of his collections as a monument of his own taste. The story suggests several ways in which collectors could, to use Walker’s rather lurid terminology, be ‘predators’ as well as ‘prey’.153 Supplementary information The following appendices are provided at Journal of the History of Collections online: Appendix 1, Calouste Gulbenkian’s Desiderata lists, 1928–30, indicating items from the collections of the State Hermitage Museum and their present whereabouts (where known); Appendix 2, transcription of a memo from Calouste Gulbenkian to Georgy Piatakoff, 17 July 1930, in which Gulbenkian comments on Soviet art sales policy; Appendix 3, Principal acquisitions, a list of the fifty most important paintings from the Calouste Gulbenkian Museum, arranged chronologically by year of acquisition. Acknowledgements The author would like to thank the Calouste Gulbenkian Foundation, Lisbon, who funded the Calouste Gulbenkian Biography Project, as well as the staff of the Calouste Gulbenkian Museum and other scholars who offered comments on drafts: Mafalda Melo de Aguiar, Penelope Curtis, João Carvalho Dias, Peter Fuhring, Ozan Ozavci, Barbara Pezzini and Thomas Stammers. The views expressed here are those of the author, who bears sole responsibility for any errors. Footnotes 1 Vitry had been ‘dazzled’ by his visit in early 1937, according to his colleague Carle Dreyfus. Dreyfus to Gulbenkian, 7 May 1937. Calouste Gulbenkian Foundation Archives, Lisbon (hereafter cgf), mcg01862. 2 [Georges Salles], ‘Note sur l’affaire Gulbenkian’, in Georges Salles to Jacques Jaujard (Directeur Général des Arts et des Lettres), 10 February 1956. Archives Nationales, Peyrefitte. f12/8603. 3 Stephen Duffy, ‘French eighteenth-century painting in England and the opening of the Wallace Collection’, in Christoph Vogtherr, Monica Preti and Guillaume Faroult (eds), Delicious Decadence: The rediscovery of French eighteenth-century painting in the nineteenth century (Farnham, 2014), pp. 141–57 (153). When some of the Wallace’s French furniture was regilded in 1933, Gulbenkian complained, to the suppressed fury of its curators. See the National Archives, London, ar1/340. 4 Receipt from Libraire H. Floury, cgf, mcg01365. 5 For these and other ‘collection museums’ see Anne Higonnet, A Museum of One’s Own: Private collecting, public gift (New York, 2009). Unfortunately Higonnet does not include the Gulbenkian Museum in her study. 6 Shelley M. Bennett, ‘The formation of Henry E. Huntington’s collection of British paintings’, in Robyn Asleson, British Paintings at the Huntington (London, 2001), pp. 1–15. 7 As Anne Distel has noted, Doucet hung Monets alongside his eighteenth-century paintings, and his 1912 sale should not be seen as a case of a collector disposing of one collection to begin a ‘new’ impressionist one. Anne Distel, ‘Jacques Doucet et l’Impressionisme’, in Chantal Georgel (ed.), Jacques Doucet: collectionneur et mécène (Paris, 2016), pp. 98–111 (103). 8 ‘My Intelligence Service in New York informed me that you had a Holbein on sale which, I believe, came from Vienna. You never told me about it?’ Gulbenkian to George Davey, 11 March 1948. cgf, mcg02910. 9 E.S.G. Robinson, ‘Calouste Sarkis Gulbenkian as a collector of Greek coins’, in E.S.G. Robinson, A Catalogue of the Calouste Gulbenkian Collection of Greek Coins (Lisbon, 1971), pp. 13–16 (15). 10 Marcelle Chanet, ‘C. S. Gulbenkian’. cgf mcg 03615. 11 Although it contains some interesting anecdotes, José de Azeredo Perdigão’s Calouste Gulbenkian, Collector (Lisbon, 1969), is superficial and hagiographic in its treatment. João Carvalho Dias was the first scholar to move beyond such well-worn anecdotes and draw on the extensive archival record for Gulbenkian’s collecting. João Carvalho Dias et al., Calouste Gulbenkian and English Taste (Lisbon, 2015); João Carvalho Dias, ‘Calouste S. Gulbenkian. O colecionador que preferiu Portugal, 1942–55’, Artis 2 (2014), pp. 99–107. 12 Edward Fowles, Memories of Duveen Brothers (London, 1976), p. 174. 13 In his speech at the opening of the Gulbenkian Museum in 1969 Kenneth Clark claimed that Gulbenkian ‘once told me that he was tempted to burn them all, like Sardanapalus, on a funeral pyre.’ Tate Britain Archive, Clark Papers, 88188.8.131.521-2. 14 For context see Jonathan Conlin, ‘The amiras and the Ottoman Empire, 1880–1923: the case of the Gulbenkians’, Turcica 48 (2017), pp. 219–44. 15 Calouste S. Gulbenkian, Transcaucasia and the Apcheron Peninsula: Travel memories [orig. La Transcaucasie et la Péninsule d’Apchéron: Souvenirs de voyage (Paris, 1891), trans. Caroline Beamish (Lisbon, 2011). 16 For correspondence between Kevorkian and the Gulbenkians (Calouste and Karnig) between 1896 and 1898, see cgf, ldn00017. 17 cgf, mcg00611. 18 The Chaplin and the Greuze cannot be identified. The Henner was one of two acquired for £385 in 1902. Agnew’s receipt, March 1902. cgf, mcg00611. It may be the same as the Portland Art Museum’s Reclining Nymph (inv. no. 43.8.19). 19 cgf, ldn00041. Gulbenkian presumably gave the work away. It should not be confused with another ‘Drouais’, Du Barry and her page Zamore, which he acquired from Gimpel in January 1900 for £150, and which remains in the collection. 20 For a marked-up catalogue, see cgf, mcg00616. The Reynolds bore the title The Snake in the Grass, an alternative title for Nymph and Cupid (Tate), well-known from prints by J. R. Smith. The Gulbenkian copy is listed under ‘Related Sales’ in the entry in Mannings’s catalogue raisonné, albeit with a garbled date (1809 in lieu of 1899) and name (‘Gulbenstein’). David Mannings, Sir Joshua Reynolds: A complete catalogue of his paintings (London, 2000), cat. 2125. In July 1901 Gulbenkian acquired another ‘Reynolds’ (not in Mannings), Mr Duheney in a blue coat seated holding a snuff box, through Agnew’s for £49 at Christie’s 29 July 1901 (Lot 140). The whereabouts of this work are unknown. cgf mcg00610. 21 Agnew’s 7604 (Islands near Venice), bt. T. Bonar 9 May 1896 for £199, sold to Gulbenkian for £420, 26 March 1901. National Gallery Archives, London. Agnew’s Archive (hereafter nga). nga27/1/1/9, f. 7. In his ledger Gulbenkian gives no date and records the price as £400, but notes it was acquired from Agnew’s. cgf, mcg01235. 22 Gulbenkian to Fox, 21 January 1902. cgf, ldn01480, fol. 36. 23 In December 1900 Gulbenkian paid Agnew £200 for a pair of ‘proofs in the first state’ (The Lock and The Cornfield). cgf mcg00607. Sixteen entries appear under ‘r’ (all Reynolds and Romney) in Agnew’s Print Book for 1898–1903. nga, ng27/1/3/1, fols 205, 207–210 (stock numbers 6229, 6290, 6299, 6324, 6328, 6336–8, 6377–9, 6381–2, 6384, 6394). 24 Landscape: Home Field, Dedham (Agnew’s 9711, bt. T. H. Ward, £600), sold to Gulbenkian 20 February 1901 for £1,600; Flatford Mill (Agnew’s 9579, bt. J. Orrock, £250), sold to Gulbenkian 19 December 1900 for £420. nga, nga27/1/1/9, fols 71 and 78. Receipts at cgf, mcg00608. Chesil Beach (Weymouth Bay) and In Dedham Vale (Agnew’s 2537 and 2538, bt. Christie’s 21 May 1908, Lots 11 and 14) immediately sold to Gulbenkian for £66 and £44. nga, nga27/1/1/10, fol. 91. Agnew’s receipt, cgf, mcg00632. In each case Gulbenkian has written ‘returned’ or ‘Échangé’ over these receipts, and their present whereabouts are unknown. João Castel-Branco Pereira was clearly mistaken when he asserted that, other than Stratford Mill, Gulbenkian was not interested in Constable. Calouste Gulbenkian Museum, The Collector and his Taste: Calouste S. Gulbenkian (Lisbon, 2006), p. 29. 25 Gulbenkian to Clark, 14 July 1946 (quote). cgf, mcg02472. For Young Waltonians see Gulbenkian to Clark, 3 July 1946. cgf, mcg02472. Gulbenkian also sought to acquire the Ashton Salisbury Cathedral. See cgf, mcg02436. Gerald Agnew to Lord Ashton of Hyde, 28 November 1924. nga, nga27/11/1. 26 For the bookseller H. Glaisher’s receipts for 1899–1901 see cgf, mcg01414. For purchases of books written by Alfred Whitman, see cgf, mcg01415 and mcg01419. 27 Sheila O’Connell, ‘William second Baron Cheylesmore (1843–1902) and the taste for mezzotints’, in Antony Griffiths (ed.), Landmarks in Print Collecting: Connoisseurs and donors at the British Museum since 1753 (London, 1996), pp. 134–57 (141). 28 Colnaghi receipt, cgf, mcg00805. Otto Gutekunst to Gulbenkian, 23 December 1920 (quote). Waddesdon Archive, Windmill Hill (hereafter wa) col1/4/13. A proof had reached £1,218 at the 1901 Blyth sale. For another example of a proof of Mrs Carnac, see Griffiths, op. cit. (note 27), cat. 58. 29 Gutekunst to Gulbenkian, 19 July 1922. wa, col1/4/13, fol. 282. 30 Gulbenkian to Colnaghi, 20 November 1947. cgf, mcg02914. 31 ‘N. N.’, ‘The glory and the shame of mezzotints’, Brush and Pencil 16 no. 5 (1905), pp. 151–9 (154, 157). 32 Gulbenkian to Davey, 1 November 1945. cgf, mcg02906. 33 Portrait of a Lady in White Dress (Agnew’s 9134, bt. Robinson Sale for £530), sold to Gulbenkian 16 May 1900 for £1,035. nga27/1/1/9, fol. 46. Agnew’s receipt, cgf, mcg00604. Its whereabouts are unknown, but it may be that listed in a 1909 catalogue raisonné as belonging to Gulbenkian. William McKay and William Roberts, John Hoppner, R.A. (London, 1909), p. 288. 34 Gulbenkian to Charles Romer Williams, 1 June 1914 (quote). cgf, ldn00066. 35 In 1932 he acquired thirteen gouaches then on show at the Musée Guimet by the Iranian Sarkis Katchadourian (1887–1947) from the artist for 40,000 francs, reconstructions of the frescoes of Ali Qapu, the Shah Abbas I palace in Isfahan, gouaches. Five were later given away, and there is no evidence any were ever hung. A fellow Armenian who, as conseiller financier to the Iranian embassy in Paris, was charged with helping arrange exhibitions for Persian artists new to the city, Gulbenkian probably bought the works out of charity. Sarkis Katchadourian, invoice, 27 April 1932. cgf, mcg00495. 36 Gulbenkian to Poynter, 4 and 31 May 1898. cgf, ldn00895, fols 37 and 38. The Skirt Dance was also known as The Ionian Dance or The Greek Dance. See Bonhams New Bond Street, ‘19th-century paintings’, 10 July 2013, Lot 114. 37 Poynter to Gulbenkian, 4 May 1898. cgf, ldn00895, fol. 38. 38 J. W. Waterhouse to Gulbenkian, 31 May 1898. cgf, ldn00895, fol. 39. 39 Gulbenkian acquired three Chaplins from the Michel Pelletier Sale in 1922, two through Graat et Madoulé for 19,039 francs. Graat’s comments on the sale catalogue hint at Chaplin’s titillating appeal, dismissing one work (Lot 45, Indolence) as unsatisfactory on account of the mass of silk drapery making it impossible to make out the figure’s bust. Graat to Gulbenkian, 30 May 1922. cgf, ldn00461. The third Chaplin, L’Étoile du matin (Calouste Gulbenkian Museum), was presumably bought in by Georges Petit, who then sold it to Gulbenkian for 21,000 francs. cgf, mcg00371. One of the three Chaplins acquired at the Pelletier sale (L’Allegorie de la peinture) was later given by Gulbenkian to a close business associate, Moïse Adjemoff. ‘Liste complete des tableaux & dessins à 38 Hyde Park Gardens’, cgf, mcg01178. 40 Gulbenkian to Davey, 13 September 1942. cgf, mcg02431. 41 Gulbenkian to Davey, 18 September 1942. cgf, mcg02431. 42 The Mirror of Venus was acquired through Arthur Ruck for £5,000, and immediately lent to the Tate. The Tate’s director, Charles Aitken, claimed that Joseph Duveen had mooted buying and presenting it, and hence was delighted to receive the loan. Ruck receipt, 29 September 1924. Charles Aitken to Gulbenkian, 10 December 1924. cgf, mcg01067. 43 Joseph Duveen to Gulbenkian, 19 August 1915. cgf, ldn0090. 44 Gulbenkian to Joseph Duveen, 27 January 1916. cgf, ldn00101. 45 Joseph Duveen to Gulbenkian, 13 March 1920 (quote); Ernest Duveen to Gulbenkian, 22 March 1920. cgf, ldn00164. 46 The earliest transaction with Duveen seems to have been a purchase in 1901 of Sèvres porcelain, although it should be noted that the Gulbenkian Foundation’s archives have little correspondence pre-dating 1900. Louis Duveen to Gulbenkian, 6 December 1901. cgf, ldn00002. 47 Edward Fowles to Joseph Duveen, 15 October 1931. Cited in Nuno Vassallo e Silva, ‘The great opportunities. The Russian acquisitions’, in Calouste Gulbenkian Museum, op. cit. (note 24), pp. 48–57 (54). 48 The whereabouts of the van der Neer and the van der Heyden are unknown. The latter was sent to Mensing & Fils (Amsterdam) in 1931, but was not sold. cgf, mcg02825. 49 Herbert G. Gosling to Gutekunst, 19 August 1924 (‘Gulby’); Gosling to Nevarte Gulbenkian, 23 April 1926 (‘millionaires’). wa, col1/4/14, fols 63 and 311. 50 Gutekunst to Gulbenkian, 8 February 1921. cgf, ldn00412. The work was acquired by Gulbenkian’s closest business associate, Henri Deterding of Royal Dutch-Shell, who presented it to the Mauritshuis. When other Six collection works came on the market in 1928, Gulbenkian consulted Ant. Mensing. Mensing to Gulbenkian, 8 September 1928. cgf, mcg02825. 51 Christie’s, 26 June 1914 (Lot 82). Marked-up catalogue, cgf, mcg00655. Gutekunst to Gulbenkian, 22 July 1926 (quote). cgf, ldn00636. Alex Kidson, George Romney: A complete catalogue of his paintings (London, 2015), cats. 277 and 347. 52 Colnaghi to Gulbenkian, 5 August 1924. wa, col1/4/14. Kidson, op. cit. (note 51), cat. 1483. In 1924 the Tankerville family owned another Romney, Newton and the Prism (ibid., cat. 1784), but it seems unlikely Gulbenkian would have been interested in this work. 53 Demotte also offered his services as cataloguer, modestly predicting that ‘in a few years I will be the greatest editor in the world.’ Demotte to Gulbenkian, 8 June (catalogue) and 5 August (agency) 1921. cgf, ldn00412. 54 For bankruptcy, see Edinburgh Gazette [copied from the London Gazette], 14607 (10 December 1929), p. 1544. 55 Ruck to Gulbenkian, 27 October 1924 (Apollo). cgf, ldn00553. Meryle Secrest, Duveen: A life in art (New York, 2004), p. 125. 56 Arthur Ruck to Gulbenkian, 27 November 1918. cgf, ldn00137. 57 Ruck to Gulbenkian, 5 January (commission) and 3 February (Berwick) 1921. cgf, ldn00432. 58 Cited in Sir Howard Frank to CSG, 9 March 1916. cgf, ldn00103. 59 Arthur Ruck to Gulbenkian, 26 November and 3 December 1920. cgf, ldn00148. 60 Colnaghi inventory, 1 January 1926. Duveen inventory, 3 May 1927. cgf, mcg00170. 61 ‘As you are aware, Ruck is bankrupt, or something has gone wrong with his affairs, and therefore we cannot effect the payment direct to him.’ Gulbenkian memo, 11 March 1932. Gulbenkian paid a $400 commission on the $8,000 purchase (ex-collection H. C. Coleman). cgf, mcg01245. 62 Camille Benoît to Gulbenkian, 18 December 1903. cgf, ldn00002. 63 Gulbenkian to Charles Carstairs, 19 February 1918. cgf, ldn01447, fol. 33. 64 Bt. Christie’s, 26 June 1919 (Lot 27) and 9 July 1920 (Lot 40). They appear in the stockbook as a742 and a991. wa, coc3/1/2, fols 60, 80. 65 Rosenberg to Gulbenkian, 20 April 1920. cgf, ldn00182. 66 Rosenberg to Gulbenkian, 20 February 1918. cgf, ldn00137. 67 Madoulé to Gulbenkian, 23 May 1922. cgf, ldn00461. 68 Fowles, op. cit. (note 12), p. 172. 69 Gulbenkian to Lockett Agnew, 19 January 1916. cgf, ldn00066. Marginalia on a letter in Agnew’s archive refers to this as ‘the infamous sale by CRW to Gulbenkian!’ Gulbenkian to Charles Romer Williams, 11 July 1914. nga27/23/4/16. For Williams, who joined Agnew’s in 1905, but retired early, in 1921, see nga27/23/4/15. 70 Ruck to Gulbenkian, 3 September 1923. cgf, ldn00513. For a similar episode, see Ruck to Gulbenkian, 9 and 12 December 1921. cgf, ldn00432. 71 The Marriage of the Adriatic is now catalogued as The Grand Canal: Scene - A Street in Venice. Asleson, op. cit. (note 6), cat. 101. At the time of writing Caterina Hoogshaet’s foreign buyer has been refused an export permit. 72 Georges Wildenstein to Gulbenkian, 2 June 1920. cgf, ldno186. 73 Ruck to Gulbenkian, 8 October 1924. cgf, ldn00553. 74 Duveen (London Office) to Gulbenkian, 10 May 1921. cgf, mcg00136. 75 Pierre Soulaine, ‘Elles en ont toutes’, Figaro [19 Apr. 1920?], clipping enclosed in Soulas to Nubar Gulbenkian, 20 April 1920. cgf, ldn00182. 76 Gulbenkian to Joseph Duveen, 8 January 1920. cgf, ldn01451, fol. 13. 77 Gutekunst to Gulbenkian, 8 February 1921. cgf, ldn00412. Alec Martin to Gulbenkian, 8 April 1922. cgf, ldn00456. 78 Gutekunst to Gulbenkian, 10 April 1922. cgf, ldn00456. 79 Martin Conway, The Art Treasures of Russia (London, 1925). Conway urged the National Gallery (London) to buy, and proposed a return visit to the USSR in 1931. See C. J. Holmes to Robert Witt, 3 October 1924. National Gallery (London) Archives, ng26/122; National Archives, London. fo371/15621. 80 See the marginalia in Gulbenkian’s copy of Pierre P. Weiner, Les Chefs-d’oeuvre de la Galerie de Tableaux de l’Ermitage à Petrograd (Munich, 1923). cgf, Art Library ms 6 Res. Piatakoff sent out other, unspecified catalogues. Piatakoff to Gulbenkian, 30 June 1928. cgf, mcg01682. For Aucoc’s advice, see Aucoc to Gulbenkian (with memo and list), 30 October 1928. cgf, mcg01134. 81 Troïnitsky to Mikoyan, 11 October 1928. Russian State Archive of Economy, Moscow, file 5240, opis 9, delo 243, doc. 10. See also the report by D. A. Schmidt, 1 April 1929. The Hermitage Archives, St. Petersburg, file 1, opis 5, chast 2, 859/72, doc. 212. 82 Elena Solomacha, ‘Verkäufe aus der Eremitage’, in Waltraud Bayer (ed.), Verkaufte Kultur: Die sowjetischen Kunst- und Antiquitätenexporte, 1919–1938 (Oxford, 2001), p. 51. 83 cgf, mcg01136 (‘Documents officiels 2ième Achat) and mcg01138 (‘Troisième Achat’). 84 Tatyana V. Sapozhnikova cited in Natalya Semyonova and Nicolas V. Iljine (eds), Selling Russian Treasures: The Soviet trade in nationalized art, 1917–38 (New York, 2013), p. 139. 85 Gulbenkian to A. Chadanian, 26 June 1930. cgf, mcg01139. 86 cgf, mcg01143 (‘Quatrième Achat’). For the options see Wildenstein to Gulbenkian, 2 September 1930 and ‘Memo de Chad. sur compte en suspens avec Wildenstein’, 6 November 1930. cgf, mcg01145. 87 Ministère des Affaires Étrangères to Mirza Hussein Khan Ala, 5 June 1930. cgf, mcg01138. 88 V. V. Veeder, ‘The Tetiuhe mining concession 1924–32’, in Liber Amicorum Claude Reymond: Autour de l’arbitrage (Paris, 2004), pp. 325–42. 89 David Kynaston, The City of London, vol. iii:Illusions of Gold, 1914–45 (London, 1999), p. 339. 90 Chester Beatty to Gulbenkian, 20 February 1937. cgf, prs04037. 91 Chicherin memo, 3 November 1927. Foreign Policy Archives of the Russian Federation, Moscow. File 05, opis 79, folder 29, delo 5, doc. 7. 92 Gulbenkian, ‘Memorandum sur le pétrole’, 20 March 1929. cgf, mcg02429. For discussion of his proposals see Chicherin to Politburo, 30 May 1928; undated Politburo memo; Goldin to Khinchuk, 11 June 1928. Russian State Archive of Economy, Moscow, file 5240, opis 18, delo 903, docs. 214, 207 and 192. 93 Gulbenkian to Piatakoff, 27 September (‘insult’) and 19 December 1928. cgf, mcg02429. Gulbenkian, ‘Memorandum objets d’art’, 6 December 1930 (‘unfriendly’). cgf, mcg01147. Piatakoff to Mikoyan, 2 January 1929. Russian State Archive of Economy, Moscow, file 5240, opis 18, delo 2738, doc. 68. 94 Birenczweig to Khinchuk 25 October 1929. Russian State Archive of Economy, Moscow, file 5240, opis 18, delo 2740, doc. 40. 95 Piatakoff was sent on ‘holiday’ in August 1928 and placed under investigation by the feared Rabkrin (Workers’ and Peasants’ Inspectorate). Unknown to Piatakoff, 15 June 1929. Russian State Archive of Economy, Moscow, file 5240, opis 18, delo 2739, doc. 35. 96 See cgf, ldn01095 and ldn01154. 97 See cgf, ldn01094; Veeder, ‘Tetiuhe’, pp. 337-42; V. V. Veeder, ‘International arbitration: Anglo-US mining concessions in Soviet Russia (1920–1925)’, in Elizabeth Bastida, Thomas Waelde and Janeth Waren- Fernandez (eds), International and Comparative Mineral Law and Policy (The Hague, 2005), pp. 99--126. 98 Solomacha, op. cit. (note 82), p. 56. 99 Although he was also objecting to these agents dealing with rival would-be purchasers. Gulbenkian memo for Piatakoff, 31 July 1930. cgf, mcg02892. 100 Gulbenkian memo, 25 June 1929. cgf, mcg02429. 101 Unidentified correspondent to Piatakoff, 15 June 1929. Russian State Archive of Economy, Moscow, file 5240, opis 18, delo 2739, doc. 35. Both Birenczweig and N. G. Tumanoff (both Paris-based Russian trade representatives) met Gulbenkian in the months before this letter and may have written this letter, but it could have been penned by someone else entirely. 102 Gulbenkian to Piatakoff, 27 June 1930. cgf, ldn01030. 103 Gulbenkian to Piatakoff, 17 July 1930. cgf, ldn01030. 104 The Presentation in the Temple (ex-Northwick collection) was bought in 1921 from Duveen for £820. Duveen receipt, cgf, mcg00138. 105 Gulbenkian to F. O. Trog, 17 December 1950. cgf, mcg02332. 106 Solomacha, op. cit. (note 82); Semyonova and Iljine, op. cit. (note 84). 107 Gulbenkian acquired the painting (now Metropolitan Museum of Art) from Leon Mantacheff, the son of a former business associate, the Armenian oil magnate Alexander Mantacheff. The pair agreed that Gulbenkian would take the painting in partial consideration of Gulbenkian paying off a 75,000-franc loan owed to Mantacheff’s London bank. The ‘price’ was therefore 37,500 francs. Gulbenkian to Leon Mantacheff, 27 February 1931. cgf, mcg00873. In 1948 Gulbenkian gave it to Robert Guiterman of the New York bank Kuhn Loeb, who had been advising him on plans for a US-based charitable trust. Guiterman sold it in 1957. Kevork Essayan to Chanet, 21 November 1957. cgf, mcg02661. 108 Gutekunst cited in Jeremy Howard, ‘A masterly Old Master dealer of the Gilded Age: Otto Gutekunst and Colnaghi’, in J. Howard (ed.), Colnaghi: The history (London, 2010), pp. 12–19 (14). 109 Chanet, op. cit. (note 10), p. 76. As Higonnet observes with regard to other ‘collection museums’, the ‘domestic’ look was just an ‘effect’. Higonnet, op. cit. (note 5), p. xiv. 110 Draft codicils dated Cannes, Dec. 1924 and 1 Jan. 1925. cgf, csca00226/d-z/008. The 1916 will then in force is at cscs00226/ d-z/009. 111 Clark to Gulbenkian, 5 December 1935. cgf, mcg02644. 112 Kenneth Clark, Another Part of the Wood: A self-portrait (London, 1974), p. 229; Gulbenkian to Clark, 10 August 1937. Tata Britain, 88184.108.40.206a. For the loans, see cgf, mcg02905. 113 Board Minutes, 9 March 1937. National Gallery Archive, London. 114 Clark to Gulbenkian, 14 December 1945 (hang). cgf, Archive, Clark Papers. Gulbenkian to Clark, 17 May 1946. cgf,mcg02471. 115 Clark to Gulbenkian, 13 March 1946; Gulbenkian to Clark, 12 May 1946 (quote); Clark to Lindsay, 25 June 1946. ta, 88220.127.116.11c. Hendy also made the mistake of making slighting comments about a portrait Gulbenkian gave to Lisbon’s Museu Nacional de Arte Antiga in 1949, Joshua Reynolds’s Major-General William Keppel (mnaa 1880). Several National Gallery trustees attempted to have Hendy dismissed in 1952, citing Gulbenkian’s removal of his collection among other episodes as proof that Hendy was (in T.S.R. Boase’s words) ‘maladroit in all his relationships’. Minutes of trustee meeting with Sir Edward Bridges, 10 January 1952. National Archives, London. t273/43. 116 Walker to Theodore A. Xanthaky, 23 December 1946. National Gallery Archives, Washington (hereafter ngw), rg28, 8–13 117 Lindsay to Gulbenkian, 13 December 1947. cgf, mcg02450. 118 As noted by Gulbenkian’s confidant at the US Embassy in Lisbon, ‘The old gentleman does not wish to hurt the feelings of his friends in London and therefore confidentially suggests that a dressed-up invitation sent officially through the State Department be extended to him by your gallery.’ Xanthaky to Walker, 27 February 1947. ngw, rg28, File 8–13. 119 Trustee minutes, 11 January 1938. National Gallery Archive, London. 120 Walker to David Finley, 25 August 1951. ngw, rg28, File 8-6 121 Jonathan Conlin, ‘Calouste Gulbenkian and his Foundation’, in David Cesarani and Peter Mandler (eds), Great Philanthropists: Wealth and charity in the modern world, 1815–1945 (London, 2017). 122 Brian P. Kennedy, Alfred Chester Beatty and Ireland 1950–1968: A study in cultural politics (Dun Laoghaire, 1988), pp. 58 (quote), 88, 103. Right up until his death in 1968 Beatty’s commitment to Ireland wavered, depending on which of his circle was in the ascendant at any one time. 123 Gulbenkian to Walker, 2 January 1952; Walker to Gulbenkian, 18 February 1952. ngw, rg28, 7–13. 124 Jonathan Conlin, The Nation’s Mantelpiece: A history of the National Gallery (London, 2006), p. 132; Dennis Wardleworth, ‘The “friendly” battle for the Mond Bequest’, British Art Journal 4 no. 3 (2003), pp. 87–93. 125 Walker claimed to have heard rumours that the British would not allow Gulbenkian’s collection to leave the United Kingdom; elsewhere he referred to ‘the ominous possibility of partial or complete confiscatory legislation’ being passed in Britain. Walker to Gulbenkian, 22 December 1947 and 15 March 1948. ngw, rg28, 8–14. Chief Justice Fred Vinson cited in Finley to Walker, 28 June 1949. ngw, rg28, 8–6. 126 Lindsay hinted that he suspected Walker of feeding this information to Gulbenkian. David Lindsay to Gulbenkian, 3 February 1953; Gulbenkian to Lindsay, 12 February 1953. cgf, mcg 02473. There is tantalizing evidence that Gulbenkian considered lending his paintings to the Metropolitan in 1942. See Clark to [Francis Henry Taylor], 4 December 1942. cgf, ldn02103. The file does not contain a response. 127 Aitken to Ruck, n.d., copy enclosed in Ruck to Gulbenkian, 15 October 1924. cgf, ldn00553. As it happens, the Tate board agreed to accept the loan without conditions. 128 The source of this anecdote is Fowles, op. cit. (note 12), p. 80. It should be noted, however, that the bronze (Marsyas and a Maenad, bt. André Desvouges in 1918 for 20,000 francs) was given later than the date given by Fowles (1935 rather than 1927), and elicited fulsome thanks from Dreyfus. See cgf, mcg00100. Dreyfus’s other letters to Gulbenkian, preserved among the latter’s papers in Lisbon, show no trace of strained relations. 129 Memorandum of conversation Walker, David Finley and Chester Dale, 28 June 1949. ngw, rg28 Box 8-6. 130 In 1918 (when Paris was under German bombardment) Gulbenkian had sent six crates to Biarritz, depositing other items in the vaults of dealers and banks. See cgf, ldn00128. In 1939 moveable items remaining at 51 Avenue d’Iéna were evacuated to a friend’s estate in the Charente using an old Citroën (two trips). They returned to Paris in February 1941. Other items were moved to the basement, where most remained until 1954. Chanet, op. cit. (note 10), pp. 130–2. 131 For Martin de Wild, see Esther van Duijn and Mireille te Marvelde, ‘The art of conservation vii: Hopman and De Wild. The historical importance of two Dutch families of restorers’, Burlington Magazine 159 (October 2016), pp. 812–23. My thanks to Peter Fuhring for this reference. 132 Gulbenkian to Clark, 13 December 1948. cgf, mcg02472. For Clark’s view of the Hoppner, see Clark to Gulbenkian, 28 December 1945. cgf, mcg02471. For Lindsay’s, see Lindsay to Gulbenkian, 16 March 1948. cgf, mcg02473. 133 The $60,000 paid by Gulbenkian for Lady Coningham represented a quarter of what the seller had paid, he recounted. Gulbenkian to Davey, 18 July 1943. cgf, mcg02433. Gulbenkian to Clark, 14 July 1946 (‘economist’). mcg02472. Joseph Duveen had sold the painting with two others to Edward T. Stotesbury in 1913. When asked to value Lady Coningham in 1938, Duveen’s gave a figure of $250,000. Kenneth Garlick, Sir Thomas Lawrence: A complete catalogue of the oil paintings (Oxford, 1989), cat. 206. 134 See Harry W. Paul, ‘Collecting Chardins: Charlotte and Henri de Rothschild’, Rothschild Archive Annual Review (2005), pp. 21–6. For the fate of the paintings in storage see A. Cooper to Henri de Rothschild, 2 March 1944. cgf, Caixa csg15, rot147. 135 Henri de Rothschild to José de Azeredo Perdigão, n.d. [after 2 March 1944]. cgf, Caixa csg15. 136 Gulbenkian to Clark, 22 April (La Tour) and 22 May (ewer) 1943. cgf, mcg02897. In 1932 Gulbenkian had approached Henri de Rothschild through the Berlin dealer Hans Stiebel, but Rothschild was uninterested in selling anything. I am grateful to Peter Fuhring for this information. 137 The work was left to the Metropolitan Museum at the death of Sam Lewisohn in 1951. Gulbenkian to Clark, 7 June 1946. cgf, mcg02471. 138 Davey to Gulbenkian, 26 August 1943. cgf, mcg02450. 139 Gulbenkian to Clark, 16 June 1946. cgf, mcg02471. 140 Victor M. Oswald to Gulbenkian, 27 March 1950. cgf, mcg02457. 141 Gulbenkian to Walker, 17 March 1950. cgf, mcg02457. 142 Gulbenkian agreed to Barth continuing to negotiate, however, albeit ‘very slowly, in true Spanish fashion, more as a deal between friends than as a business proposition.’ Barth to Gulbenkian, 14 July 1950. cgf, mcg02457. 143 Chanet, op. cit. (note 10), p. 139. 144 Ibid., pp. 135–6. 145 Davey to Gulbenkian, 13 June 1953. cgf, mcg02437. Given how few Gulbenkian purchases are recorded in the Knoedler stockbooks – a search of the sales books from 1892 to 1953 reveals a mere six paintings; the Lépine is not among them – one wonders if Davey adopted the same practice in other cases. Getty Research Institute, Los Angeles. Knoedler Archive, Sale Books 7–17. 146 Gulbenkian to Davey, 4 November 1952 and 9 February 1953. cgf, mcg02436. 147 N. M. Ekserdjian, diary, 20 July 1956. Private collection. 148 Chanet, op. cit. (note 10), p. 140. 149 Gulbenkian to Clark, 17 May 1946. cgf, mcg02471. 150 Gulbenkian to Gutekunst, 10 August 1928. cgf, ldn00749. 151 ‘You know that everything interests me except the ultra-modern school.’ Gulbenkian to Davey, 18 September 1942. cgf, mcg02431. 152 Calouste Gulbenkian to Nubar Gulbenkian, 12 June 1926. cgf, nsg00019. 153 John Walker, Self-Portrait with Donors (Boston, ), pp. 235–7. © The Author(s) 2017. Published by Oxford University Press. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
Journal of the History of Collections – Oxford University Press
Published: Oct 13, 2017
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera