Thwaytes v Sotheby's

The case in summary

In the collection of the Kimbell Art Museum in Fort Worth, Texas there is a painting called The Cardsharps (I Bari in Italian) by the Italian Baroque artist Michelangelo Merisi da Caravaggio, known as Caravaggio. The painting depicts three men grouped round a table, playing cards. On the left side of the painting is a young man (‘the dupe’) dressed in a sumptuous purple velvet doublet, his eyes downcast looking at the two cards he holds in his hand. Slightly behind him and to his left, facing the viewer, is an older bearded man (‘the old sharp’) and opposite the dupe, to the left hand of the old sharp is a young man (‘the young sharp’). The old and young sharps are in league against the dupe. The old sharp is covertly glancing down at the dupe’s cards and signalling to the young sharp with the fingers of his right hand what cards the dupe holds. The young sharp leans forward eagerly towards the dupe, his left hand outstretched and resting on the richly patterned rug covering the table. The young sharp’s right hand is reaching behind his back where, tucked into the belt of his breeches, two additional cards are concealed, one of which he is removing. I shall refer to this painting in Texas as the Kimbell Cardsharps. It is accepted by the parties to this dispute and by Caravaggio scholars that the Kimbell Cardsharps is an autograph painting by the hand of Caravaggio.

The painting which is the subject of this dispute is another painting of the same scene, with the same three characters in broadly the same positions. This painting (‘the Painting’) used to be owned by the Claimant in these proceedings, Mr Thwaytes. In 2006 Mr Thwaytes entrusted the Painting to the Defendant, the well-known auction house (“Sotheby’s”). It was sold by Sotheby’s on 5 December 2006 for £42,000 plus buyer’s commission. The purchaser was Ms Orietta Benocci Adam but it is accepted that she bought the Painting on behalf, at least in part, of her close friend Sir Denis Mahon.

Sir Denis was himself a lifelong Caravaggio scholar of great renown and had been an experienced collector of Baroque paintings for many decades. After acquiring it at the Sotheby’s auction, Sir Denis carried out extensive investigations into the Painting including having it cleaned and restored. In November 2007 Sir Denis announced to the world that the Painting was an autograph replica of the Kimbell Cardsharps painted by Caravaggio himself. He was supported in this attribution by an Italian scholar Professor Mina Gregori. Mr Thwaytes now brings this action against Sotheby’s in negligence and for breach of contract. He alleges that the Sotheby’s experts failed adequately to research the Painting and failed to notice certain features of it that should have indicated to them that it had ‘Caravaggio potential’, that is to say that it might actually be by Caravaggio rather than a copy. He does not assert that the Painting is by Caravaggio. But he alleges that if Sotheby’s had performed their duties towards him properly, he could either have sold the Painting for much more money or he would have decided not to sell the Painting and would now own a work of art of much greater value than he received on its sale.

Caravaggio and the Cardsharps

Caravaggio was born in Milan in 1571 and arrived in Rome to work as a painter probably in 1592 - 93. Relatively few works by Caravaggio – only about 60 paintings – are known to exist. Although Caravaggio trained in an art studio he did not himself teach or organise his own studio of pupils and followers. Caravaggio painted the Kimbell Cardsharps in about 1594 so it counts as one of his early works; a genre painting depicting every day life rather than a scene with a religious or mythological subject. Its original owner was the artist’s first important patron, Cardinal Francesco Maria del Monte. The Cardsharps was an immediate success. It was later bought by Cardinal Barberini and was in his family’s collection for some time. In 1772 an engraving was made of the image by the Italian engraver Giovanni Volpato (‘the Volpato engraving’). There is some dispute between the experts in this case as to whether the Volpato engraving is copied from the Kimbell Cardsharps or from the Painting.

The autograph original passed through various hands over the centuries. It was sold in the 1890s and left Italy and it was then ‘lost’ for about 80 years. There was a photograph taken of it in about 1890 by a Parisian firm Braun, Clément & Compagnie. This was known to be a photograph of the autograph work and was the image against which the claims to authenticity of any contending painting fell to be judged. In 1987 the Kimbell Cardsharps was discovered at the Zurich Institut für Kunstwissenschaft where it had been deposited for restoration and analysis. Professor Gregori was one of those who saw it by chance and she realised that it must be the missing original. The Kimbell Art Museum acquired the painting through a New York art dealer. The cleaning and restoration of the Kimbell Cardsharps was undertaken in 1987 in New York by a team at the Metropolitan Museum of Art. The head curator of the Metropolitan Museum is the art historian Dr Keith Christiansen who is another well-known scholar of Italian Baroque art. In the January 1988 edition of the Burlington Magazine (a leading publication in the art history world) two important articles appeared. I shall refer to these together as the ‘Burlington Cardsharps Publication’. The first was by Sir Denis Mahon called Fresh Light on Caravaggio’s Earliest Period: His ‘Cardsharps’ Recovered. He recounts that he first saw the painting cleaned but before restoration: ‘It did not take very long – sceptical though I have always been on this subject – to convince me that this was indeed the autograph painting by the youthful Caravaggio himself which had figured in Braun’s photograph of a century ago…’. The two features that Sir Denis picked out in the Kimbell Cardsharps as particularly convincing of the artistic merits of the painting are the delicate handling of the feather in the hat of the young sharp and the ‘spectacular execution of his quasi-Savoldesque breeches’. The article included a picture of the Volpato engraving and of various copies of the work. Sir Denis also recounts that when the Kimbell Cardsharps was relined during restoration, a stamp was found in the bottom left corner of the original canvas, showing that the painting had been in Cardinal del Monte’s collection. The article contains infra-red reflectogram images of the work, including one showing that the position of the young sharp’s right hand had been altered during the course of composition.

As an Appendix to Sir Denis’ article there was a technical report from Dr Keith Christiansen. He described the condition of the painting as ‘exceptionally fine’ although there were small areas of wear. In that appendix Dr Christiansen refers to two aspects of the Kimbell Cardsharps that are important for our purposes. The first is that before restoration, the painting had an added piece of canvas running along the length of the top of the canvas, about 14 cm high. This strip was removed during restoration and no one currently knows its whereabouts. The second element Dr Christiansen points out is in relation to a black area by the right elbow of the dupe. I discuss the significance of this ‘black mass’ later.

Copies of Caravaggio’s works and of the Cardsharps composition

It is important to draw a distinction between ‘replicas’ or ‘copies’ where two paintings are virtually identical; ‘variants’ where the two paintings show the same scene but with significant modifications; and ‘repetitions’ of a subject where the two paintings deal with the same incident from the Bible, or a religious or mythological figure (such as Bacchus) but in a different way. It is accepted that Caravaggio painted repetitions of the same subject. For example The Supper at Emmaus in the National Gallery in London and the painting of the same name in Brera, Milan depict the same event where Christ appears to the two disciples after the resurrection. But the figures and the composition are very different and one is not intended to reproduce the other. It is also accepted that Caravaggio painted variants such as the two versions of The Fortune Teller painting, one in the Louvre, Paris and one in the Musei Capitolini in Rome. Those are both accepted as autograph paintings and show a young dandy having his palm read (and his gold ring surreptitiously removed) by an attractive peasant girl. Though the figures are similar and are dressed in the same clothing, they are clearly not intended to be identical.

What is more contentious in the art world is whether Caravaggio ever painted replicas of his own work, at least after the very early stages of his career. There is no doubt, however, that the instant popularity of the composition of the Cardsharps led to the making of high quality copies by other hands shortly after Caravaggio completed the work and over the centuries thereafter. There are several dozen copies known to exist. Sotheby’s annexed to its Defence a print out from Artnet which records paintings sold at auction worldwide. About 30 versions of this composition other than the Painting are listed as having been offered for auction between 1988 and 2012, over half of them by either Christie’s or Sotheby’s. They appear of varying quality and sold for a wide range of prices. Indeed during the first week of the trial of this action, two copies of The Cardsharps were sold at auction in London, one at Bonhams for £1,250 and one sold at Christie’s as ‘After’ Caravaggio’ for £10,000 (over an estimate of £2,000 - £3,000).

I have already briefly described the composition of the Cardsharps. The Kimbell Cardsharps in its current form is about 131 cm wide by 94 cm high and the Painting is very slightly larger at 131 cm wide by 104 cm tall. There are some passages in the composition which are important in this case in addition to the three figures:

Sotheby’s

Sotheby’s is an auction house of long standing, international high repute. Its Old Master Painting department (‘the OMP Department’) is highly successful and is considered perhaps the best in the world. The OMP department holds three Old Master Painting sales in London per year as well as sales in New York and Paris, selling about 600 works each year. In 2006 Sotheby’s had two auction rooms, at New Bond Street and at Olympia. The Olympia auction room was less prestigious and generally sold less valuable paintings although some more valuable, decorative works such as the Painting were known to do particularly well there. That showroom closed in 2007.

For every auction sale a catalogue is produced describing each painting to be sold. Every entry indicates the certainty with which Sotheby’s is prepared to attribute it to a particular artist. The catalogue entry may describe a painting in the following ways:

Some technical terms

The creation of a painting

A painting is made up of a number of different layers of paint. Usually the artist will first paint a priming layer covering the whole of the bare canvas. There will then be a layer called the ‘ground’ which will also cover the whole canvas although different colour grounds may be used over different parts of the canvas. The artist may then sketch out the basic shapes of the picture that he intends to paint. This may be referred to as the abbozzo layer. Finally the artist will start painting the figures and other items that make up the composition.

In many Caravaggio paintings it is apparent that he painted from live models and constructed the image, broadly speaking, from the back forwards so that figures further from the viewer were painted first and the closer figures afterwards. There are two ways an artist can do this. The first is to paint one item overlapping another item. For example, in the Kimbell Cardsharps it is clear that the dice holder was painted over a fully painted tric trac board because one can now clearly see the triangular point of the board through the dice holder; the paint of the dice holder having become more transparent over time. The other way is to paint the later figure ‘en reserve’. This means that the artist painted the background figure first but left a suitably shaped gap in which he would later paint another figure. In that case the later figure does not overlap the background figure but is painted on the same layer as it. An example of this in both the Kimbell Cardsharps and the Painting is the face of the young sharp. He stands in front of the old sharp’s left shoulder and his face is painted en reserve of the old sharp’s black cloak rather than being painted over the top of a fully painted cloak.

When a figure is painted en reserve the artist may not paint the later figure right to the boundary of the background figure. For example, in the Kimbell Cardsharps and in the Painting, there is a small space running along the contour of the young sharp’s profile and the edge of the old sharp’s cloak through which a lower level of the painting, the ground, can be seen. This is known as a bordo a risparmio. Caravaggio left this reserve strip in the Kimbell Cardsharps to form part of the final image which the viewer can see as a double profile.

When an artist is composing the image in paint on the canvas he may change his mind about the positioning of a figure and paint over what he has already painted to alter the final image. These changes are called pentimenti (or pentimento, singular). Sometimes pentimenti can be seen with the naked eye, for example, when the upper paint layers have become more transparent with age. Other pentimenti areonly visible through a process that allows one to look below the paint surface, using, for example, x-ray or infra-red imaging. Pentimenti suggest that the painter refined and altered the composition as they worked, and, for this reason, they are often cited as evidence that a painting is an original composition (i.e. not a copy after a known composition). If the artist has simply copied an existing image, one would not expect to see a major pentimento, for example with one of the figures facing in a different direction or an arm bent instead of straight.

Technical analysis of a painting

Various methods of examining a painting have featured in this case. Everyone is agreed that the examination of a painting in a strong natural light is an important first step in assessing a painting. When the Sotheby’s experts examined the Painting they also wiped it with cotton wool soaked in white spirit. This is a common method of examining old paintings where the paint surface is flat and where the details are obscured by a layer of varnish or dirt. The effect is to cut through the dirt and reveal a much clearer image – rather like putting a pebble in water. The effect was demonstrated to me during the short adjournment of the third day of the trial when Sotheby’s brought to court a 15^th century painting of a Madonna and Child (painted on a panel rather than on canvas). Wiping with white spirit does indeed make it easy to see the detailed paintwork although the effect lasts only until the spirit evaporates and the surface returns to a dry more opaque state.

A painting can also be viewed using ultra violet light shone from a small torch. This can reveal changes near the surface of the painting and is used particularly to reveal retouchings or restorations on the painting which have been made in the last 80 – 100 years. These show up clearly under ultra violet light as purple patches on the surface of the painting. Taking x-rays of a painting is an important method of looking below the surface image. X-rays pass easily through material with a low atomic weight (x-ray transparent material) and such material will look dark on the x-ray image. Materials which have a high atomic weight will block the rays and appear white in the image. Lead white is one of the most common pigments used by Italian artists in the early 17^th century and is highly opaque to x-ray producing white areas on the image. X-rays can therefore show a pentimento because the image they reveal may appear very different from the surface image. Infra-red imagingreveals carbon material and hence any drawing with a pencil or charcoal, especially if it is on a light background. There was some dispute between the parties as to how far infra-red analysis is used when examining Italian Baroque art where the artists typically painted on a dark ground so that any underdrawing in the abbozzo layer would not show up. My understanding is that infra-red analysis is more commonly used for Northern European paintings from the 15th and 16th centuries because those artists made extensive use of carbon black underdrawing on a white background and that shows up very well in an infra-red image of the painting. However there are clearly instances when infra-red analysis has been done of Caravaggio works and indeed there were infra-red images taken of the Kimbell Cardsharps when it was restored in the Metropolitan Museum.

Other techniques for analysing paintings include the taking of samples of paint and cross-sections of the work. Since it is accepted by Sotheby’s that the Painting is roughly contemporary with Caravaggio, this was not a case where analysis of the pigments or other materials used was likely to assist in attribution. It is not suggested that any materials used by Caravaggio were unique to him.

The factual witnesses in the case

There were five factual witnesses on behalf of Sotheby’s, two of whom still worked for Sotheby’s at the time of the trial.

Mr Alexander Bell is the Joint International Head and Co-Chairman Worldwide of Sotheby’s OMP Department based in London. He joined Sotheby’s in 1986 and has worked in the OMP department for nearly 27 years, becoming head of the department in London in September 1994. Mr Bell necessarily has to have a good general knowledge across the wide range of Old Master Paintings artists. His stated area of interest is Spanish and Italian art of the 16^th and 17^th centuries but he accepted that he is not an academic and that those working in auction houses have to be generalists rather than specialists. His evidence was that though he had not been involved in the sale of any painting where there had been a serious issue about whether it was an autograph work by Caravaggio, he had been involved in the sale of a number of contemporary copies of Caravaggio works – ‘such works’ he said ‘pretty regularly turn up’. Mr Bell had seen the Kimbell Cardsharps at The Genius of Rome exhibition in London in 2001. Mr Thomas Baring currently owns his own fine art dealership but at the relevant time he was working for Sotheby’s in the OMP Department and was responsible for overseeing the cataloguing of works allocated to Olympia sales. He prepared most of the Olympia catalogue entries himself and was also involved in cataloguing paintings sold at New Bond Street. Mr Matthew Barton was employed by Sotheby’s in its valuations department first in Billingshurst in West Sussex and later in Olympia. His work involved a great deal of travelling round his allocated area visiting the homes of clients to look at items that the client wanted to be valued either for insurance purposes or for sale. His evidence was that he carried out about 1000 valuations every year most involving client visits. Ms Letizia Treves was a senior expert in Sotheby’s’ OMP Department in London, specialising in Italian paintings. She joined the Department in 1996 and was always based in New Bond Street. Her main focus was to research and respond to queries on paintings by Italian artists, whether in the context of the New Bond Street auctions, insurance valuations or general enquiries. She would also look at other paintings being catalogued or researched by other specialists in the Department. Ms Treves left Sotheby’s in April 2013 to work as a curator of Italian and Spanish Paintings 1600-1800 at the National Gallery in London.

There was also a witness statement from Clarissa Post served by Sotheby’s under a hearsay notice producing an email sent to her by Dr Keith Christiansen on 28 May 2009. I will refer to this email later.

Part way through the trial, Mr George Gordon of Sotheby’s was called to give evidence in relation to some documents that were disclosed late by Sotheby’s’ New York office. I refer to his evidence in relation to the question whether if Sotheby’s had decided to consult an external expert to assist with the attribution of the Painting, they would have consulted Sir Denis Mahon and/or Professor Gregori.

Mr Thwaytes was the primary factual witness in support of his case. He also served a witness statement from Professor Gregori who lives in Italy and is now 90 years old. Her statement deals with her assessment of the Painting.

I found all five witnesses who gave oral evidence to be honest and straightforward and doing their best to assist the court. There was little of the Sotheby’s factual evidence which was challenged but Mr Baring and Mr Barton were cross-examined about some conversations they had with Mr Thwaytes where their recollection differed from his. Some of the differences do not seem to me material other than as casting doubt on the general accuracy of the witness’s other evidence. In assessing their evidence I bear in mind that for the Sotheby’s witnesses, the Painting and its sale was not particularly noteworthy until the controversy over its attribution arose at the end of 2007. The Painting was just one of a large number of paintings that they assessed and sold and their recollections of events were, not surprisingly, hazy in some details. For Mr Thwaytes of course, the sale was an important if not unique event in his life and I accept that for that reason generally his recollection is more likely to be accurate than that of the Sotheby’s’ witnesses. However, where Mr Thwaytes’ evidence involves an assertion that one of the Sotheby’s witnesses deliberately told him something that was not true, I find that difficult to accept. They had no motive for lying to him and I do not consider that any of them had the inclination to be anything other than truthful. As appears from the narrative below, I find that there are other explanations for why the parties drew different conclusions from what was said in a particular discussion. I recognise also that I must take care when assessing Mr Thwaytes’s evidence to consider whether it is coloured by hindsight and his understandable upset about what has happened.

The expert witnesses

The expert evidence in this case fell into three categories; (i) art historical evidence and connoisseurship; (ii) technical evidence about the Painting and the Kimbell Cardsharps and (iii) evidence of auctioneering practice and fine art valuation.

The Claimant called Dr Roberta Lapucci to deal with art history and connoisseurship and Ms Helen Glanville to deal with more technical evidence. Dr Roberta Lapucci has a PhD from the University of Rome, her thesis being on Caravaggio’s Technique: materials and methods. Since 1986 Dr Lapucci has taught post-graduate courses in conservation and artistic techniques as well as Baroque Art History at notable universities. She is currently the Head of the Department of Art and Archaeology at Studio Art Centers International, American University programme located in Florence. She has also worked as a private restorer of paintings. She states that in the last 35 years she has personally inspected numerous autograph Caravaggio paintings and numerous paintings proposed to be by Caravaggio. She is a much published author of articles for exhibition catalogues, magazines and books. One of Dr Lapucci’s supervisors for her PhD was Professor Gregori and in 1991 - 1992 they co-curated an important exhibition of Caravaggio’s work in Florence and Rome. She describes that exhibition as having been revolutionary because it included life-size illuminated panels containing the x-ray mosaics of 10 paintings of Caravaggio and explaining various image diagnostic techniques. She describes this exhibition as marking a turning point in demonstrating the importance of Caravaggio’s technique in the chronological and attributive analysis of his paintings. She goes on to say that she has developed as a scholar independent of Professor Gregori with whose attributions she sometimes agrees and sometimes disagrees.

Ms Helen Glanville is a conservator and technical art historian and research associate at the Hamilton Kerr Institute at Cambridge University. She trained as a conservator of easel paintings at the Courtauld Institute of Art. She has taught conservation and restoration at various universities in Europe and has worked as a freelance conservator/restorer for major museums including the Louvre. She states that she has written thousands of condition reports on paintings from every period, in particular preparing condition reports on all the works that were exhibited in the Royal Academy exhibition The Genius of Rome in 2001.

The Claimant’s expert on auction house practice was Mr Guy Sainty. Mr Sainty has been an art dealer since December 1975. He has sold many paintings to major museums all over the world. As an art dealer he says he has constantly interacted with auction houses primarily as a buyer but also occasionally as a consignor and as an adviser to private clients buying and selling at auction. He is thus familiar with the practices of art auctioneers particularly Sotheby’s and Christie’s. His evidence is that art dealers and major auction houses follow the same procedure when dealing with important works of art, in particular obtaining opinions from leading authorities on the artists they are offering for sale. His report and his oral evidence were strongly criticised by Sotheby’s as being of no probative value or weight at all. It was pointed out that he had never worked for an auction house and had therefore never had to research or catalogue an Old Master Painting for auction. I do not accept that Mr Sainty was not qualified to give evidence about auctioneering practice. There are very close links between art dealers and the auction houses since dealers are the main customers of the auction houses both as consignors of art works and as purchasers of paintings at auction. I also accept the point made by the Claimant that it is difficult to find an auction-house expert with relevant experience who is not involved with either Sotheby’s or Christie’s.

The Defendant’s main expert witness was Professor Richard Spear. Professor Spear’s career has been devoted to the study of Baroque art. He has a PhD in art history from Princeton University and taught for many years at Oberlin College in Ohio. During his tenure as a Professor of Art History at Oberlin, he was also for a time Director of Oberlin’s Art Museum and trustee and president of the Intermuseum Conversation Association. Since 1998 he has worked in Washington DC where he is Distinguished Visiting and Affiliate Research Professor at University of Maryland, College Park. In his report he refers to his fifty years of engagement with Caravaggio, and to having studied every one of Caravaggio’s paintings first hand, most of them many times. He has written and lectured widely on Caravaggio all over the world. He also describes his professional connection with conservation of paintings.

Sotheby’s also relied on the reports of Professor Dianne Modestini on more technical aspects of the evidence. Professor Modestini is a paintings restorer and conservator specialising in Italian Old Master paintings. She worked as Assistant Paintings Conservator at the Metropolitan Museum in New York from 1974 to 1987. She left there shortly before the Kimbell Cardsharps was restored at the Met but she has maintained close links with her colleagues in that department. She now works at the Conservation Center of New York University where she is Institute Research Professor. Sotheby’s expert witness on auctioneering practice and valuation was Ms RachelKaminsky. Ms Kaminsky is now a private art dealer working in New York. She started working for Christie’s in 1983 and rose to become head of the Old Master Paintings department there, a position she held for five years. She left Christie’s in 1994 and has since then worked for various galleries and art dealers. She founded her own dealership in 2006. She states that during her tenure at Christie's, they were shown in the original or via photographs more than 1,000 Old Master Paintings per annum for review, sale at auction and appraisal. They sold approximately 500 to 700 paintings per year, ranging in value from around $3,000 to $32 million.

Each of these experts was clearly extremely knowledgeable and dedicated to their work; to the study of Italian Baroque art in general and to the work of Caravaggio in particular. I regard them all as fully qualified to give the expert evidence set out in their reports. Each of them was subject to criticism from the opposing side, picking up mistakes in their reports which needed to be corrected or challenging alleged exaggeration or overstatement either in their evidence before the court or in earlier articles and monographs. During the course of the trial I was taken to a range of scholarly articles and correspondence written by these witnesses or other Caravaggio experts. From this two points emerge. First it is clear that an art historian may express his or her current view with considerable certainty based on what may appear to a lawyer to be scant available evidence. A greater freedom of surmise and speculation may also be legitimate in an article or treatise than is generally appropriate for a witness statement. This does not rule out another equally qualified expert expressing a different view with equal certainty based on the same scant evidence. The second point is that the role of technical analysis in the attribution of paintings to artists is still a matter of some debate. Although most scholars would conclude that technical analysis can establish that a painting is not from a particular period or not by a particular artist, its value in establishing a positive dating or attribution is less widely recognised. Some art historians and connoisseurs express concern about over-reliance on or misuse of scientific data about a painting, whereas some regard scientific evidence as more reliable than the traditional connoisseur’s ‘eye’. This latter point is important here because one of the issues in this case is whether more weight should be placed on traditional connoisseurship skills of assessing the work by visual inspection (albeit assisted with technical analysis) or whether reliable attribution is better arrived at by reliance on technical analysis of the work. Both Dr Lapucci and Ms Glanville naturally tend to the latter camp since their primary expertise is in the technical aspects of art historianship. Although they have seen and examined very many Italian Baroque paintings over the course of their careers, they are less used to relying on their ‘eye’ to assess the quality of a work of art. Their expertise is in undertaking technical analysis of paintings and interpreting those images and they not surprisingly therefore place greater weight on scientific analysis than on connoisseurship. Professor Spear, I find, is more of a traditional connoisseur who has examined x-ray and infra-red images of paintings many times but who still regards the visual examination of the surface of the work as paramount in assessing its potential. On questions of connoisseurship and the visual assessment of quality, I therefore will place greater weight on Professor Spear’s evidence than on that of Dr Lapucci or Ms Glanville.

The Painting before it was consigned to Sotheby’s

Mr Thwaytes inherited the Painting from his father’s cousin Surgeon Captain William Glossop Thwaytes who died in June 1965. Surgeon Captain Thwaytes was a collector of art who lived in a small country estate known as Holesfoot in Cumbria. In 1947 Surgeon Captain Thwaytes bought a different painting, called The Musicians, from a dealer in Kendal. That painting depicts four boys in Classical costume playing musical instruments and singing. In 1951 this painting was identified by two art historians as an original work by Caravaggio and in 1952 The Musicians or ‘Musica’ was purchased from Surgeon Captain Thwaytes by the Metropolitan Museum of Art in New York for a substantial sum. Sir Denis Mahon was closely involved in the ‘publication’ of The Musicians that is to say, he carried out the extensive research necessary to write the monograph that set out the proof that the painting was an autograph work. He published an article in the Burlington Magazine in January 1952 in which he described how he had been shown a tiny photograph of a picture from the collection of ‘a private amateur in an English country house’. He had concluded that the canvas had potential and he described how careful restoration ‘brought to light from beneath dirt and extensive old repaints an early work of Caravaggio which has every claim to be identified as the lost Musica painted for Cardinal del Monte.’

Returning to the Painting which is the subject of this dispute, little is known about its provenance before 1962. In February 1962 Surgeon Captain Thwaytes bought it for £140 at an auction of Old Masters Paintings held at Sotheby’s premises in New Bond Street. At that sale the Painting was described in the auction catalogue as ‘Caravaggio (After), The Cardplayers’. Mr Thwaytes’s evidence is that Surgeon Captain Thwaytes considered the Painting to be ‘very fine’ and thought that it could be by the hand of Caravaggio. In his will, Surgeon Captain Thwaytes left his house in Cumbria with its contents on trust for Mr Thwaytes. As well as the Painting there were two other versions of the Cardsharps, one of which apparently included a fourth figure in the image. No one has suggested that those other two paintings were by Caravaggio. In addition, Mr Thwaytes’ father had acquired his own copy of the Cardsharps in the early 1970s.

Mr Thwaytes first sought a valuation of the Painting in 1989/1990 when someone from Christie’s came to the house to value some of the contents of Holesfoot with a view to selling them. The value placed on the Painting at that time was £3000 - £4000. Mr Thwaytes was first in touch with Sotheby’s about the Painting in 2002 when he sought an insurance valuation. Matthew Barton came to Holesfoot accompanied by Judith Heelis, the local representative of Sotheby’s who was a friend of Mr Thwaytes. The value placed on the Painting then was £9,000, slightly higher than the value of another Cardsharps in the house.

In 2006 Mr Thwaytes was considering selling items from the collection at Holesfoot to help pay for school fees. He contacted Sotheby’s and arranged a meeting at Holesfoot on 4 April 2006. Both Mr Thwaytes and Mr Barton agree that the valuation was discussed at this meeting and that they also discussed different ways of researching the Painting in order to find out if it is by Caravaggio or not. It is agreed that x-rays were discussed. Mr Thwaytes says that he was aware that analysis of x-rays had been important in the discovery of The Musicians so he raised this with Mr Barton. Mr Barton told him that x-rays were not normally done as this involved sending the painting out of the building. There is a dispute over whether Mr Barton told Mr Thwaytes that infra-red testing would be done by Sotheby’s on the Painting. Mr Thwaytes asserts that Mr Barton told him that infra-red testing would be done almost as a matter of course. Mr Barton denies he would have said any such thing because he would not have known what was involved in infra red imaging and he now knows that it is only rarely done by auction houses. It is clear from subsequent events that from this point on, Mr Thwaytes had the idea fixed in his mind that Sotheby’s would subject the Painting to infra-red analysis. I accept his evidence that at the time he thought that infra-red inspection of the Painting would involve an expert simply looking at the Painting through a lens like a marksman looks through an infra-red rifle night scope when shooting in darkness. He did not appreciate that it would involve the production of images like the x-ray plates. I accept his evidence that although he had read the Burlington Cardsharps Publication about the discovery of the Kimbell Cardsharps, he did not have in his mind that the photographs included in the articles were of infra-red images rather than x-rays. However, no possible reason has been suggested as to why Mr Barton would have told him that infra-red testing would be done as a matter of course when it is certainly not true. I find that what happened was that Mr Barton told Mr Thwaytes that the Painting would be looked at carefully by Sotheby’s experts and that they would look at it under ultra-violet light. That is something that is done as a matter of course – and was done on the Painting - using a little torch that shines ultra violet rather than ordinary light. It is not something that usually generates an image but it is important as it shows up recent restorations to the surface of the Painting. What Mr Thwaytes describes Mr Barton having told him is not true about infra-red analysis but it does accurately describe ultra violet light analysis.

I therefore find that there was a misunderstanding between Mr Barton and Mr Thwaytes whereby Mr Barton told Mr Thwaytes about the likely use of ultra-violet light to examine the Painting and Mr Thwaytes came away from the conversation thinking that they had been talking about infra-red imaging.

The Painting at Sotheby’s

On 10 July 2006 the Painting along with two other items were collected from Holesfoot and taken to Sotheby’s’ premises. On 11 July 2006, Mr Thwaytes wrote to Mr Barton enclosing a copy of his driving licence and saying:

"As far as the Cardsharps is concerned this is going to be researched and studied with both infra-red and possibly x-rayed. I hope I have presumed correctly, this hopefully will confirm it is in fact the original!?"

In these proceedings, Mr Thwaytes relies on this letter to Mr Barton as an instruction to Sotheby’s to carry out infra-red imaging on the Painting. I do not accept that. I find that Mr Barton would have explained to Mr Thwaytes that he would not be responsible for the examination of the Painting once it got to Sotheby’s since he was not qualified to assess it. If Mr Thwaytes had really intended to give an instruction to Sotheby’s he would have got in touch with the OMP Department once the Painting had left Holesfoot.

The next step taken with regard to the Painting was at the ‘divide’ on 4 September 2006. This is an internal Sotheby’s procedure whereby the Painting was allocated to Tom Baring as the person who would be in charge of researching and cataloguing it for eventual sale, if indeed the Painting was going to be sold. Mr Baring spoke to Mr Thwaytes by telephone soon after the divide. Mr Thwaytes told Mr Baring that the Painting had been owned by Surgeon Captain Thwaytes who had owned The Musicians. Mr Thwaytes made clear that he thought the Painting might be an original and that he wanted it to be thoroughly researched. Mr Baring’s evidence is that it is extremely common for consignors and their families to have strong emotional beliefs in the origin and quality of the works they hold. This does not affect his approach to assessing the work which is considered on its own merits.

Mr Baring’s first task was to examine the Painting himself, take its measurements and form a preliminary view about it in order to produce the text that might form the entry in the auction catalogue if the Painting were to be put up for sale. He examined the Painting using white spirit, a torch and ultra violet light. His evidence was that he found a large clear photograph of the Kimbell Cardsharps in the Department library in the Caravaggio catalogue raisoné written by Dr John T Spike (‘the Spike book’). He recalls noting that the Painting appeared to reproduce the image of the Kimbell Cardsharps almost exactly. Mr Baring drafted a condition report and some initial details about the Painting for the catalogue. This first draft of the catalogue entry described the Painting as ‘Follower of Michelangelo Merisi da Caravaggio – The cardsharps’. It referred to the original now in the Kimbell Art Museum as being among the artist’s most popular and influential compositions. It described the card game that is depicted in the image. The catalogue entry at that point did not include any information about provenance even though Mr Baring had had a discussion with Mr Thwaytes about Surgeon Captain Thwaytes and the discovery of The Musicians by this time.

The Painting was considered at Sotheby’s New Bond Street premises at a picture meeting on 6 September 2006 (‘the Picture Meeting’). This was attended by Mr Baring, Mr Bell, George Gordon (at that time a senior specialist in the OMP Department in New Bond Street) and Sian Keene (the Sotheby’s catalogue coordinator at the time). Mr Bell could not remember how many pictures had been examined at that meeting. The Painting would have been put on an easel for them to look at and they would have had available only the information included at that stage in the catalogue entry. A can of white spirit and a supply of cotton wool is kept in the OMP basement and it is used extensively during picture meetings to examine the paintings. Mr Bell accepted that the image revealed by white spirit was not as good as the image of a painting that has been cleaned, but he said it gave ‘a very strong indication of what the picture would become’. Those present also inspected the Painting with ordinary torches and with torches that shine ultra violet light. It is not clear whether there was a photograph of the Kimbell Cardsharps at the Picture Meeting. It is accepted that the people attending did not consult a copy of the Volpato engraving or the Burlington Cardsharps Publication.

Mr Bell said that he concluded that the figures in the Painting looked ‘flat and dead’ and were ‘unconvincing’. Mr Bell could not remember whether he had noticed particular features of the Painting which were pointed out to him in cross-examination and which I discuss later. At the Picture Meeting, the Sotheby’s experts formed the view that the Painting was a 17^th century copy of the Kimbell Cardsharps painted by a ‘follower’ of Caravaggio.

Mr Baring telephoned Mr Thwaytes after the Picture Meeting and told him that Sotheby’s’ opinion was that the Painting was a copy and was not by Caravaggio. He said, correctly, that this had been the unanimous view of the specialists who had seen the Painting. Mr Thwaytes asked Mr Baring if infra-red or x-ray analysis had been carried out and Mr Baring told him that it had not. Mr Thwaytes said that he wanted x-rays to be carried out even though Mr Baring told him he would be ‘wasting his money’. Mr Baring’s evidence is that when Mr Thwaytes said he wanted x-rays and infra-red analysis of the Painting, Mr Baring suggested that they have x-rays first and then if they show anything interesting, they could move on to infra-red images. Mr Baring’s evidence on this point was criticised on the basis that it was implausible that he should have such a clear recollection of saying this when his recollection of most other important events in the case is very limited. I consider that if Mr Baring had said that, it would have been a reasonable thing to say but I doubt that he did actually say that to Mr Thwaytes. If he had, that would have disabused Mr Thwaytes of his misunderstanding about the nature of infra-red analysis and Mr Thwaytes’ evidence is that his misunderstanding about that continued after this conversation. However, I do not accept that Mr Thwaytes said anything that amounted to an instruction to Mr Baring to carry out infra-red analysis as well as x-ray analysis. If he had done so, there is no reason why Mr Baring should not have complied with this request, just as he complied with the request to take x-rays, even if he privately thought it was a waste of time and money.

Subsequently x-ray images were taken of the Painting by Dr Nicholas Eastaugh and delivered to Sotheby’s on 9 October 2014. The images take the form of 12 plates which are then joined together in an x-ray mosaic of the whole image. Mr Bell saw the x-ray plates of the Painting. His evidence is that he did not see anything in these x-rays that raised any question in his mind about the attribution of the Painting. He did not recall having noticed pentimenti of any kind in the x-rays.

The x-ray mosaic was emailed to Mr Thwaytes by Mr Baring on 12 October 2006. After the email had come through on Mr Thwaytes’ computer he had a phone conversation with Mr Baring for about 15 minutes. In that conversation Mr Baring told Mr Thwaytes that the x-rays had not shown anything that caused the people at Sotheby’s to change their view that the Painting was a 17th century copy. There is a dispute between Mr Baring and Mr Thwaytes as to what was said in this conversation about infra-red images. Mr Thwaytes says that before the phone conversation, whilst the large file comprising the x-ray mosaic was downloading onto this computer, he made some notes about things to discuss with Mr Baring. His handwritten notes list four items, ‘cleaning’; ‘x-ray and infra red’; ‘reserve’; and ‘date’. He says that in the conversation he ran through these items with Mr Baring. As regards cleaning, he asked whether it would help if the Painting was cleaned. Mr Baring replied that paintings do better at auction if they are not cleaned. Mr Thwaytes’s evidence is that he asked whether x-rays and infra-red had been done and Mr Baring confirmed that they had. He also recalls Mr Baring saying that there were no underdrawings visible in the Painting. I do not accept that Mr Baring told Mr Thwaytes that infra-red images had been taken of the Painting. That would have been a deliberate lie and would have been very odd since Mr Baring would have realised (and had no reason to know that Mr Thwaytes did not realise) that if infra-red images had been taken then Mr Thwaytes would have wanted to see the images and would expect to reimburse Sotheby’s for the costs of the analysis. I find that Mr Baring told Mr Thwaytes that everything that could reasonably have been done to investigate the Painting had been done. Mr Thwaytes interpreted that, incorrectly, as meaning that infra-red analysis had been done because he was still labouring under the misapprehension that Mr Barton had told him that that is done as a matter of course.

During this conversation it is accepted that Mr Thwaytes and Mr Baring agreed that the Painting would be placed in Sotheby’s Old Masters Paintings auction due to take place at Olympia in London on 5 December 2006. A written sale agreement dated 9 November 2006 was entered into between the parties for the sale of the Painting.

Sotheby’s prepared a catalogue entry for the auction. The catalogue defines the term ‘follower’ as meaning a work by a painter working in the artist’s style, contemporary or nearly contemporary, but not necessarily his pupil. The catalogue included a double page spread for the Painting with a colour illustration on one side and a description of the Painting on the other. The catalogue described the Painting as FOLLOWER OF MICHELANGELO MERISI DA CARAVAGGIO THE CARDSHARPS. The entry says that it is a 17^th century copy after the Kimbell original and again describes the card game being played. It contains an additional note about the provenance of the Painting:

‘Surgeon Captain W.G. Thwaytes was a very keen and important collector of compositions by Caravaggio, and indeed sold Caravaggio’s original of The Musicians to the Metropolitan Museum of Art, New York.

PROVENANCE

Surgeon Captain W.G. Thwaytes, Maulds Meaburn, Penrith and thence by descent.’

The estimate given was £20,000 – 30,000.

The sale of the Painting took place during ‘Old Master pictures week’ which draws wealthy collectors to London from all over the world. The Olympia sale on 5 December 2006 had 223 lots on offer. The Painting was given a favourable placing in the auction as the first painting to be auctioned at the start of the afternoon session. Before the auction takes place, there is a sale exhibition at the Olympia show room so that potential buyers and others interested in Old Masters can come to view the paintings and discuss them with the Sotheby’s experts. The Painting was hung in a prominent place in the exhibition.

Mr Baring who was in charge of the sale noticed that the Painting was attracting a significant amount of attention. So much so that on 4 December 2006 he contacted the Sotheby’s experts at New Bond Street and asked them to come and take a further look at the Painting. Mr Baring was asked a number of times in cross-examination what it was that had caused him to call New Bond Street at such a busy time and ask for the specialists to come to look at the Painting. He could not remember and indeed none of the Sotheby’s witnesses was able to cast any light on what prompted this meeting. Mr Baring rejected the suggestion put to him that it was because he had spotted a well-known dealer taking an interest in the Painting. He said that it was common for dealers to come and look at paintings during Old Masters Week so this would not have struck him as out of the ordinary.

At Mr Baring’s request, Mr Bell took a taxi across from New Bond Street to Olympia with two colleagues, Letizia Treves and Christopher Apostle. Letizia Treves was at that time working in the OMP Department in London and Mr Apostle was Head of the OMP Department in New York, visiting London for the sale. When they arrived at Olympia the Painting was taken down from the wall and placed in a side room for them to take a further look at it (‘the Olympia Meeting’). Mr Bell’s evidence was that they looked at the Painting for about 20 minutes and discussed it in detail. They used white spirit again to examine the Painting though he was not sure whether they had brought along an image of the Kimbell Cardsharps. It was put to him that if they had spent such a long time looking at the Painting, this must mean that they were not sure about the attribution. He rejected that suggestion:

“I don't think that is a fair way of putting it. I think the answer ‐the evidence to us was clear, but we kept on pushing ourselves: can we be making a mistake, and it's the question that we always ask at picture meetings: could this be better than we think it is? Could this be - it's almost like someone asking the question so that you don't ever slip into a frame of mind that something isn't right. From our point of view, obviously, if something is right, it is much more beneficial because we make our money when we sell things and we earn our commission and our revenue is greater the higher of the price. So the possibility of discovering a Caravaggio, which would have been potentially worth much more money, would have been an extremely attractive prospect for us. … I think that looking at the picture very carefully, wiping it over with white spirit which, as you have witnessed today, especially on a picture with a coarse new canvas, evaporates very quickly, you need to keep on doing, to examine all areas of it, to look at the painting in great detail and to have a discussion amongst ourselves, testing each other: what about this area, what about that area, it doesn't surprise me the length of time it took place.”

Mr Bell was using the word ‘right’ there in the sense that art experts use it to mean an autograph work rather than a copy or counterfeit work. At the end of the meeting the Sotheby’s experts remained of the unanimous view that the Painting was a copy and that there was no reason to change the catalogue entry.

Mr Thwaytes rang Mr Baring at some point during 4 December to ask whether there had been any particular interest in the Painting. Mr Baring could not recall whether that conversation took place before or after whatever it was that prompted him to ask Mr Bell and the others to come to see the Painting again. In any event, Mr Baring did not tell Mr Thwaytes either that there had been an unexpected level of interest in the Painting or that the Olympia Meeting had taken place.

The Painting was put up for sale on the afternoon of 5 December 2006. There is no record of how brisk the bidding was. The only record apart from the winning bid is of the underbid of £40,000. The underbidder was a consortium of prestigious art dealers (‘the Consortium’).The successful bidder was Ms Orietta Benocci Adam, a close friend of Sir Denis Mahon and the Painting was knocked down to her at a hammer price of £42,000. Of this Mr Thwaytes received, after commission, testing, shipping costs and other expenses the sum of £34,468.24.

On 11 December 2006, after the Painting had been sold, Mr Thwaytes emailed Mr Baring to thank him and asked him to confirm in writing that the Painting was not a Caravaggio saying ‘this is for family reasons and my file’. Mr Baring spoke to Mr Thwaytes on the phone and confirmed this.

The Painting after the sale

Sir Denis Mahon was initially a specialist in the artist Guercino but he later became the principal reference in the English speaking world for Caravaggio after writing several articles during an exhibition in Milan in 1951. There is no doubt that Sir Denis was greatly revered by everyone in the art world and that he was a great scholar and connoisseur. As I have already described, Sir Denis was involved in the publication of The Musicians which had been owned by Surgeon Captain Thwaytes and he had also researched and written the Burlington Cardsharps Publication publishing the Kimbell Cardsharps as the lost autograph work. He was a very wealthy man and had an encyclopaedic knowledge of the works of the artists in whom he took a particular interest. Mr Bell described Sir Denis’ reputation in the art world in the following terms:

“Sir Denis was an enormously important figure, not only as a champion for Baroque Italian paintings, but also because he had a very large collection of Italian paintings, many of which were on loan to institutions throughout the United Kingdom. He has been very passionate about the idea that museums should never charge entry fees, and there was always the potential threat that should the museum which were the current beneficiaries of his loan ever charge entry fees, those pictures would be taken away again. People admired, I think, the stance that he took on that. People admired the fact that he had been a champion for Italian Baroque paintings when they had been rather out of fashion and had been able to assemble quite a collection of them, and I think they treated him with great respect and deference for that.”

The following work was carried out on the Painting at Sir Denis’ instigation:

The charges for all this work were as follows (not including VAT): Dr Eastaugh charged about £2,500 for examining the Painting and preparing his report; RMS Shepherd Associates charged about £18,800 for cleaning and restoring the Painting, scientific analysis including x-rays and writing their report; Davide Bussolari charged €1000 for taking x-rays, infra-red images and a high resolution image of the cleaned Painting before it was restored; Simon Bobak charged about £6,400 for relining the Painting.

At a party to celebrate his 97th birthday in November 2007 Sir Denis announced that the Painting was an autograph replica painted by Caravaggio himself. This announcement received widespread publicity in the media and quickly came to the attention of Mr Thwaytes. Mr Thwaytes’ evidence on this was as follows:

“On 12 December 2007 a close friend of mine, Alasdair Darroch, who had frequently visited Holesfoot and was familiar with the Painting left a telephone message for me at home, which my wife Deborah told me was something about a painting. I telephoned him back and he said something like 'You know your painting, it's in the Telegraph'.

He was referring to an article entitled 'Caravaggio worth £50m" discovered”’ dated 12 December 2007, …. The article stated that:

Sir Denis Mahon ... bought the painting for £50,400 at an auction at Sotheby's last December. [. . .] Sir Denis, who has authenticated three other Caravaggios, decided that the painting was an early work by the Renaissance master himself, and dated it to 1595.'

The article stated that the Painting 'may be worth up to £50 million' and was to go on display in Trapani, Sicily. I then did some further research on the internet and located a further article suggesting Mina Gregori and Maurizio Marini supported Sir Denis Mahon's attribution.

Words cannot really do my emotions justice but I was in utter disbelief and absolutely horrified to see that the Painting was now being proclaimed to the world as an original Caravaggio, little more than a year after the auction. I thought by asking Sotheby's to properly research the Painting, and by asking them repeatedly if they were sure that it was a copy, that I had done everything that I could in my position. I felt extremely let down and very angry that Sotheby's had apparently not done their job properly.”

Not surprisingly there were subsequent discussions between Mr Thwaytes and people at Sotheby’s during which Sotheby’s maintained their view that the Painting is a copy and is not by Caravaggio. On 19 December 2007 Mr Bell wrote to Mr Thwaytes expressing surprise that Sir Denis was now of the view that the Painting is a genuine work by Caravaggio. He said:

“As you know, we studied the painting here carefully over a period of 3 ½ months before the catalogue went to press and arranged for X-rays to be made of it. We remain confident in our opinion, from our research and from the study of these X-rays, that the painting is not by Caravaggio but is an anonymous contemporary copy. Clearly the market (i.e. dealers, museum curators and private collectors who received the catalogue and viewed the sale) took the same view as the price realised was in line with prices for good contemporary copies of pictures of this type.

I can, of course, understand your concern that Sir Denis Mahon and also, it seems, two Italian art historians, Mina Gregori and Maurizio Marini, are reported to believe that the painting is by Caravaggio. It is worth pointing out, however, that an attribution to Caravaggio proposed by any or all of these scholars will not automatically be accepted by the wider art historical community or by the market and we think it most unlikely in this case that their view of the picture’s authorship will be accepted.”

Following Sir Denis’ announcement that the Painting was an autograph work, the Painting was deposited at the Ashmolean Museum in Oxford albeit it was never displayed there. In 2008 the Painting was taken to Italy and exhibited there as an autograph work by Caravaggio. It was shown first at the Museo Pepoli in Trapani as part of an exhibition Caravaggio: L’Immagine del divino and then in the town of Cento and then at the Musei San Domenico at Forlì. The catalogue for the Trapani exhibition contained a monograph on the Painting by Professor Gregori. For the Forlì exhibition a catalogue was also produced (‘the Forlì catalogue’) containing extended essays promoting the autograph status of the Painting including a reprint of the article by Professor Gregori. Her article was called ‘A Further Original of Caravaggio’s Cardsharps’. She referred to the discovery of the Kimbell Cardsharps and to the possibility of further autograph replicas concealed among the many copies of the work. She states in the article that ‘studies have revealed’ that during the early years of financial hardship prior to Cardinal del Monte’s patronage ‘it was common practice for Caravaggio to make copies of his own works “to sell”’. She also notes that letters from Giulio Mancini in 1615 referred to the commissioning of a number of copies to be made from Cardinal del Monte’s personal collection for 15 scudi each, including a copy of a ‘card game’. She was however ‘inclined to rule out’ that the newly found Cardsharps ever belonged to Cardinal del Monte. She believes it is more fruitful to follow another course of enquiry starting with an examination of the work itself ‘in whose authenticity, just like Sir Denis Mahon, I firmly believe’.

Professor Gregori goes on in the article to describe a large number of passages in the Painting which strike her as of high quality. The article also considers the x-rays and infra-red images of the Painting. Professor Gregori highlights the changes that are revealed to the positioning of the dupe’s head, the detail of his face and the palm of his right hand. She outlines her hypothesis that the Painting is in fact the first version of the Cardsharps that Caravaggio painted, surmising that the pentimenti to the young sharp’s right hand in the Kimbell Cardsharps shows the master trying to modify the hand in the Kimbell Cardsharps but then returning to the position that he had used in the Painting. She refers to the 14 cm strip that had been added to the Kimbell Cardsharps and the corresponding space above the old sharp’s head in the Painting. She surmises that when the strip was added to the Kimbell Cardsharps, probably after the death of Cardinal Barberini, ‘the earlier, authoritative version with more space at the top of the canvas was still held in some important collection (either as a replica or as the early original)’. There are various other passages in the Painting which she describes as being different from and as superior to the corresponding passages in the Kimbell Cardsharps.

The third essay in the Forlì catalogue is by the late Maurizio Marini and is called ‘Three cardplayers by Caravaggio, a darkened and neglected painting, from Rome to London in 1769’. It is fair to say that none of the experts in the case invited me to place reliance on the opinion of Mr Marini or to conclude that this would have affected the value of the Painting.

The Forlì catalogue also contained a technical article written by Davide Bussolari (who had cleaned and relined the Painting) and a short technical article by Thomas Schneider. Professor Gregori’s article in the Forlì catalogue is significant not only because she is the principal Caravaggio expert who has unequivocally supported Sir Denis’ attribution of the Painting to Caravaggio but because as I have described, she provided a short witness statement in support of the Claimant’s case in these proceedings. In the witness statement she states that her judgement of paintings is based on her ‘eye’, her experience and connoisseurship and that it is not influenced by other scholars. She notes in her witness statement that she was not asked by Sotheby’s to express a view on the Painting and goes on to say:

“8 … The quality of the Mahon 'Cardsharps' was as good as the Kimbell 'Cardsharps', which I also saw when it was discovered, before it was cleaned.

9.

I saw the painting from the Mahon collection for the first time when it was being restored in the Robert Shepherd and Associates studio in Wimbledon. At that time, it had been cleaned but not restored. I noticed the quality of the painting immediately, but I have not discussed it here, because I have talked about it in detail in the Forlì and Trapani catalogues.

10.

I would however like to point out in particular that, as the painting had been cleaned, I could see that the right eye of the old sharp was visible under the hat of the young innocent. For me, this was definite confirmation that it could not be a copy.

11.

Since I saw the painting, it has been restored, but I maintain that it has not been done well.”

Professor Gregori’s evidence was treated for the purposes of the trial as incorporating by reference her analysis of the Painting in the Forlì catalogue.

The Painting is now on display at the Museum of the Order of St John in Clerkenwell in London. It is apparently insured for £10 million. One evening during the course of the trial, arrangements were made by the parties for me to visit the Museum to see the Painting there. Since there has been some criticism of the restoration from the Claimant’s expert witnesses asserting that the restoration has made it look more like a copy, I have not based any assessment of the evidence on what I saw - beyond recognising that it is a strikingly attractive and charming work of art.

Mr Thwaytes put his case on the basis of both breach of contract and negligence. It was accepted by both parties that the test to be applied was the same for both causes of action. I note here that although the statements of case pleaded reliance on various clauses in the deposit receipt signed by Mr Thwaytes when he consigned the Painting to Sotheby’s in July 2006 and in the sale contract he signed later, no issues about the interpretation or validity of these clauses were raised at the hearing.

There was, however, a dispute about the scope of the standard of care owed by Sotheby’s to Mr Thwaytes. Sotheby’s submitted that only the ‘normal’ standard of care should be considered in this case and that there was nothing that happened here to impose any greater duty on Sotheby’s than it would owe to anyone who consigned a work of art for sale. Mr Thwaytes relied on some additional points which he says affect the scope of the duty owed by Sotheby’s to him because they put Sotheby’s ‘on special inquiry as to the quality and importance of the Painting’. The first is that Mr Thwaytes asked for the Painting to be researched and, he says, he consigned it initially to Sotheby’s for research without intending that they should sell it. He relied especially on the 11 July 2006 letter as giving instructions for the Painting to be researched and studied. In my judgment there is no basis for concluding that where a work is consigned to an auction house for research and assessment rather than for sale, that imposes on the auction house a duty to examine the work more carefully than they need to if it is consigned to them for sale. Such a finding would be illogical. An auction house is in the business of selling works of art; according to Ms Kaminsky, when they carry out valuations for insurance and probate purposes they will often charge for that service if there is no prospect of earning commission on the sale. It would be unfair to impose a more onerous obligation on an auction house to spend time and resources on investigating a painting which the consignor had not yet committed to sell compared with a work where they know they will earn some commission on the sale. Mr Thwaytes accepts that he did not tell Sotheby’s when he consigned the Painting to them that he had no intention at that stage of selling it. I reject the submission that Sotheby’s duty was more onerous either because Mr Thwaytes asked them expressly to research the Painting or because he had not yet decided to sell it by the time he sent it to them.

The second factor relied on as putting Sotheby’s on notice of the Painting’s potential is that the Painting belonged to Surgeon Captain Thwaytes and hence came from the same collection as The Musicians which had been acquired by him as a copy and had turned out to be an autograph work. This seems to me a non-sequitur. Surgeon Captain Thwaytes owned very many paintings of low value in his collection including two other versions of the Cardsharps. There was no evidence to suggest that Surgeon Captain Thwaytes had a particularly good ‘eye’ or that he had bought either The Musicians or the Painting believing that he had identified an autograph work (bearing in mind of course that he acquired the Painting before the Kimbell Cardsharps had been identified as the lost original). He certainly did not take any steps to authenticate the Painting himself. Although Mr Thwaytes referred to his uncle as having ‘discovered’ The Musicians it is clear from Sir Denis’s 1952 Burlington Magazine article that it was David Carritt and Benedict Nicholson who spotted the painting’s potential and tenaciously pursued its authentication. Sotheby’s included the reference to Surgeon Captain Thwaytes in the catalogue for the 5 December 2006 sale as a point of interest and in fact the evidence suggests that this connection did pique the interest of Sir Denis Mahon in the Painting. There was no reason for Sotheby’s to think that because Surgeon Captain Thwaytes had owned The Musicians, any other paintings in his collection were worthy of special attention.

The third is that it was believed by the Thwaytes family that the Painting was by the hand of Caravaggio and Mr Thwaytes made this clear to Mr Baring. I do not accept that this can affect the duty. I accept Mr Baring’s evidence that many consignors have a strong emotional belief in the ‘rightness’ of their paintings and that an auction house must approach each painting on its merits regardless of the state of knowledge or expertise of the consignor.

I therefore hold that the duty undertaken by Sotheby’s when the Painting was consigned to them is the duty that arises generally when a painting is consigned to a leading international auction house and that there are no special features in this case to extend that duty or make it more onerous.

As to what the general duty on an auction house is, this was considered by the Court of Appeal in Luxmoore-May and Another v Messenger May Baverstock [1990] 1 WLR 1009. That case was brought against a firm of fine art auctioneers outside London who failed to spot that two small paintings of foxhounds might in fact be the work of the celebrated painter of animals George Stubbs. The paintings, which had been very dirty and overpainted when assessed by the auctioneer, were given a reserve price of £40 for the pair and sold for £840. They were subsequently sold as being by Stubbs for £88,000. The Court of Appeal allowed the auctioneers’ appeal against a finding of liability. In that case the two pictures had initially been consigned by the claimants to the auctioneer ‘for research’. Slade LJ held that this term had no standard, recognised meaning but that in the context of that case the duty of the auction house was:

“to express a considered opinion as to the sale value of the foxhound pictures, and for this purpose to take further appropriate advice.”

The Court went on to consider what was the standard of skill and care which the plaintiff had the right to expect of the auction house in the discharge of their duties. Each member of the Court of Appeal emphasised that the defendant in that case was a provincial auction house and not a leading London house: see Mann LJ at page 1028F and the comments of Sir David Croom-Johnson at page 1029H-1030A. In the leading judgment Slade LJ referred to the analogy with the distinction in the medical world between general practitioners and specialists. He cited the judgment of Lord President Clyde in Hunter v Hanley 1955 S.L.T. 213, 217 where the Lord President said:

“In the realm of diagnosis and treatment there is ample scope for genuine difference of opinion and one man is clearly not negligent merely because his conclusion differs from that of other professional men… The true test for establishing negligence in diagnosis or treatment on the part of a doctor is whether he has been proved to be guilty of such failure as no doctor of ordinary skill would be guilty of it if acting with ordinary care…”

Slade LJ regarded the defendants in that case as akin to ‘general practitioners’ rather than ‘specialists’ and held that the standard of skill and care required of them was to be judged only by reference to what may be expected of a general practitioner. He also warned against assessing the defendant’s behaviour with the benefit of hindsight and set out an important rider:

“The valuation of pictures of which the artist is unknown, pre-eminently involves an exercise of opinion and judgment, most particularly in deciding whether an attribution to any particular artist should be made. Since it is not an exact science, the judgment in the very nature of things may be fallible, and may turn out to be wrong. Accordingly, provided that the valuer has done his job honestly and with due diligence, I think that the court should be cautious before convicting him of professional negligence merely because he has failed to be the first to spot a “sleeper” or the potentiality of a “sleeper”: … ”

The trial judge had held that the person who assessed the foxhound pictures on behalf of the auctioneers had been negligent because no competent valuer could have fixed on a low valuation without need for further investigation. This was because it was the duty of a general practitioner to guard against his own want of specialist knowledge and to exercise proper caution in arriving confidently at his own conclusion. He must know his own limitations. The judge posed the question ‘whether there was enough about these foxhounds to make it unreasonable for a competent valuer to be sure he was right when in fact he was so dramatically wrong’ and he held that there was. The Court of Appeal overturned the judge’s findings noting that although by the time of the trial the ‘Stubbs potential’ of the pictures was obvious and undeniable, in the case of a ‘sleeper’ it is all too easy for the court or anyone else to be wise after the event. Slade LJ listed six elements of the evidence that convinced him that the judge had demanded too high a standard of skill on the part of the auctioneer, even though there was an impressive list of people who had recognised some quality in the paintings. These elements included the large numbers of horse and dog paintings by insignificant artists that a valuer is likely to come across; the fact that the paintings were not themselves of the high quality one would expect from Stubbs and that only two bidders were prepared to bid for the paintings at auction – no one else present thought bidding was worthwhile.

In applying the test set out in Luxmoore-May I must of course take into account that here the defendant is a leading auction house not a provincial one and that it must be held to the higher standard that the Court of Appeal rejected in that case. In what ways must that higher standard of skill and care owed by a leading auction house be manifested? First, I consider that those who consign their works to a leading auction house can expect that the painting will be assessed by highly qualified people - qualified in terms of their knowledge of art history; their familiarity with the styles and oeuvres of different artists; and in terms of their connoisseur’s ‘eye’. In contrast, the valuer used by the defendant in Luxmoore-May had no formal fine art qualifications but that, the Court held, did not prevent him from being fully competent to assess the paintings in the defendant’s storeroom before the auction. Further the specialists at a leading auction house will have ready access to the opinions and services of art historians at the highest levels of scholarship around the world. I doubt that the valuer in Luxmoore-May would have been in touch regularly with Sir Denis or Dr Christiansen to ask their opinions about works of art in the same way that the Sotheby’s’ employees clearly are.

Secondly, a leading auction house must give the work consigned to it a proper examination devoting enough time to it to arrive at a firm view where that is possible. Again, this would contrast with the position Luxmoore-May where it appears from the evidence that the valuer gave the foxhounds a rather cursory examination as two among 50 paintings that he examined on his visit to the storeroom. It was not suggested that this of itself was negligent for that defendant but is far from what would be expected of Sotheby’s. Thirdly, I consider that it would be much more difficult for a leading auction house to rely on the poor condition of a painting as a reason for failing to notice its potential – one of the factors that the Court of Appeal did find militated against a finding of negligence in Luxmoore-May.

However, much of what the Court of Appeal said in Luxmoore-May is still relevant here in particular about the nature of the task of attribution, the need to avoid hindsight, the prevalence of copies of the Cardsharps and the absence of bidders prepared to take the price up above £42,000 at the auction. I also accept that the principle that an art expert must know his or her own limitations and when to bring in an expert would apply as much to Sotheby’s as it does to a provincial auction house albeit, of course, that the bar for where that threshold is crossed is set at a much higher level in Sotheby’s’ case.

There is one additional submission made by Mr Legge QC appearing for Mr Thwaytes that I accept. That arises from the difficulty of determining the prevailing standard of conduct when there are only two generally accepted auction houses of this stature at least as regards Old Masters, namely Sotheby’s and Christie’s. Mr Legge referred to the case of Edward Wong Finance Co Ltd v Johnson Stokes & Master (A Firm) [1984] 1 AC 297 where the claimant had suffered loss because of the manner in which the conveyance of a mortgaged property was carried out. The defendants’ case was that they had followed the normal and customary conveyancing practice current in Hong Kong. The Privy Council nevertheless restored the finding of the first instance judge that the solicitors had acted negligently. Although they accepted that the Hong Kong practice had obvious advantages for both solicitors and clients they held that it involved a foreseeable risk as operated in that particular case. Mr Legge drew from this the principle that if the accepted practice among professional people in a particular matter falls into bad habits and creates unnecessary and easily avoidable risks for the client, then following that practice may still be regarded as negligent. Whether that principle can in fact be drawn from the case is not entirely clear to me but I accept the proposition that merely because Christie’s and Sotheby’s can be shown to act in a particular way does not automatically mean that that way is not negligent. There must be a back stop consideration of the need to protect the interests of the client.

The allegations that Sotheby’s acted negligently can be summarised as follows. First it is alleged that Sotheby’s was wrong in its general approach to the Painting, namely to assess it solely in terms of its artistic quality. Secondly it is alleged that Sotheby’s failed to notice certain features of the Painting which should have alerted them to its Caravaggio potential. These features should have prompted them to undertake further technical analysis and to seek the views of external scholars. Thirdly it is alleged that Sotheby’s was negligent in failing to notify Mr Thwaytes of the Olympia Meeting (and of whatever it was that prompted the Olympia Meeting).

Sotheby’s’ general approach to assessing the Painting

Sotheby’s accept that the specialists who examined the Painting at the Picture Meeting and at the Olympia Meeting assessed the Painting by applying their connoisseurs’ eye to a consideration of its quality. A number of the witnesses tried to describe what is meant by the connoisseurs’ eye. Mr Bell said:

“Our main consideration in assessing a painting is quality. In the case of a painting suggested to be a copy of a work by a known artist, we will consider whether the painting being viewed is of the quality expected of a painting by that artist. The ability to determine quality is gained by experience in the profession, from looking at all sorts of pictures from the low quality end of the spectrum right up to works by the greatest artists. From that, one develops an 'eye' for quality. It is not something that I can reduce to words easily and, if I were to do so, it would be misleading as it would then appear to be a mechanical exercise of looking at various aspects of a painting, which is definitely not the case. On the contrary, it is necessary to take into account all aspects of a painting together to determine whether overall it is painted with the skill, finesse and energy that might be expected of the particular artist under consideration. In the case of an artist like Caravaggio, this will involve consideration of, for instance, the anatomy of the figures and whether this is convincingly rendered or looks awkward in any way, how the figures relate to each other spatially and how convincing the artist's use of light and shade is in creating a powerful image.”

Mr Bell accepted that if he reaches a conclusion about the quality of the painting, that determines what further work is carried out.

Ms Kaminsky also described what happens when a connoisseur first looks at a painting. She describes the process of attribution as comprising an intuitive component, which involves the application of connoisseurship, and a scholarly and methodical component. She described the intuitive component as follows:

“The intuitive component is what happens during the first few seconds that an expert stands in front of a painting. Almost instantaneously — in the blink of an eye — the brain processes an enormous amount of information, expertise, knowledge and years of experience to arrive at a hypothesis or series of hypotheses about a painting. These may relate to the attribution, subject, value or other aspects of the painting. It is difficult to explain how this process happens but, astonishingly, these split-second reactions are very often accurate.”

It is accepted by Mr Thwaytes and by his expert Mr Sainty that for the vast majority of works of art that come into a leading auction house, the specialists can perfectly properly assess the work themselves. But with this particular Painting, he contends, they ought to have recognised their own limitations in terms of experience and expertise and have sought the opinion of Caravaggio scholars. There are various aspects of the Painting which Mr Thwaytes says should have indicated to Sotheby’s’ specialists that they needed to seek expert assistance with assessing the Painting.

The first such feature is that the Painting was clearly a contemporary version of an early Caravaggio work and that Caravaggio’s early work was variable in quality. Further, no accepted work by Caravaggio has been sold at auction in recent decades so the Sotheby’s experts are not used to handling Caravaggio works. They could not have developed the necessary ‘eye’ for his work.

As to the variability in quality, Mr Bell was cross-examined about articles in which notable scholars have criticised certain passages in Caravaggio’s early works. For example Professor Spike describes the version of TheLute Player now in the Metropolitan Museum as a work which is ‘competent but never excites in any particular’. In an article called ‘On Some Aspects of Caravaggio and His Times’ in 1953 Sir Denis Mahon referred to the fact that Caravaggio had more or less ignored the conventional studio training and that ‘Nothing was easier than to point out features in his paintings which could be taken to indicate that he had not mastered “the tricks of the trade”’. Further, Mr Legge noted that Mr Bell had said that in looking at a potential Caravaggio he would focus particularly on the anatomy of the figures and how they relate to each other spatially. He put to Mr Bell that Sir Denis had written about the difficulties which Caravaggio had in the coordination and articulation of the human form on occasions; problems involving distance and perspective.

Mr Bell’s response to these points was that although he recognised that Caravaggio’s technical ability might be variable, this did not detract from the impact of Caravaggio’s early work. Various accepted works by Caravaggio were then put to Mr Bell as illustrating infelicities, in particular the lack of accurate perspective in some instances. One was the comb on the table in the Detroit Magdalene which I consider later. Another was the shoulder of the Borghese Ailing Bacchus. Mr Bell did not accept that there was anything wrong with this shoulder but rather thought that it was beautifully modelled giving a sense of its volume and form. Both Mr Bell and Professor Spear accepted that assessment of quality is subjective and that scholars of Caravaggio differed in their views of the quality of some works. But they did not accept that this devalued the usefulness of quality as a means of assessing the Caravaggio potential of a work. Mr Bell’s evidence, with which I agree, is that any technical shortcomings in Caravaggio’s work in no way diminish the overwhelming impression that one is looking at a masterpiece of composition and craftsmanship when one looks at Caravaggio’s paintings of this period. A good example is one that was put to Mr Bell, namely the fact that the hands of the figure with outstretched arms on the right side of the Supper At Emmaus in the National Gallery are out of perspective and that the foreshortening is not correctly done. Mr Bell’s response was that that did not affect the visual impact of the painting which he described as ‘absolutely stunning’ and ‘extraordinary’. He said that a passage in a painting, such as a hand, can be very convincing and powerful even if it is not anatomically correct or in perfect perspective. The same point was made by Professor Spear when he was asked about the variable quality of Caravaggio’s accepted works. He accepted that there were anatomical mistakes in his early work but went on to refer to Caravaggio’s:

“… uncanny ability to represent natural forms in light and the glistening surface or the nature of fruit, the what I think of as the thingness of things, he doesn't slip, and that's where the connoisseur sees the difference.”

I also accept the point made by Ms Kaminsky that the quality of the Painting does not have to be assessed in a vacuum – it can be compared with the quality of the Kimbell Cardsharps. If there are passages in the Painting which are markedly inferior to the quality of the corresponding passage in the Kimbell Cardsharps then that is unlikely to be due to an inability on the part of the young Caravaggio to capture the item in question, given that it is not suggested that the Painting was made substantially earlier in time than the Kimbell painting. If Caravaggio is able to paint a very realistic ear or pewter plate or lace cuff in the Kimbell Cardsharps there is no reason why he should not be able to paint an equally good ear, plate or cuff in a replica.

In the light of this evidence, I reject the submission that the variability in quality of Caravaggio’s early work makes it inappropriate for the Sotheby’s specialists to undertake the consideration of the Painting themselves on the basis of its quality or that they should immediately have sought external advice.

Mr Thwaytes’ second reason why Sotheby’s should not have tackled the assessment of the Painting themselves is that Caravaggio raises particular problems as regards attribution. Mr Sainty’s report described how there have been many disagreements in the past about whether a particular work was by Caravaggio or not; that eminent scholars have disagreed with each other and that scholars have changed their minds about a particular painting over time. Ms Kaminsky accepted that there are many Caravaggio attributions that are controversial and that an auction house specialist would be expected to know this – Mr Bell’s evidence was that he was aware of this. Allied with this aspect of Caravaggio scholarship is the fact that the question whether Caravaggio ever painted replicas of his own works is also hotly debated. A minority of scholars adhere to the view that he did paint more than one version of the identical composition. But there are some scholars who do not accept that any of the proposed replicas are really autograph. Mr Bell’s evidence was that he was aware of these academic controversies but that he did not regard Caravaggio as more difficult to attribute than other artists such as Velazquez, Rubens, van Dyck or Titian. Professor Spear also said that Caravaggio was not particularly difficult. He referred to another Baroque artist Guido Reni who is difficult because he ran a studio where pupils painted copies of his works, some of which were retouched by the master. Caravaggio did not have a studio so there is no problem with these different degrees of autograph status.

Although Mr Sainty regards these difficulties of attribution as distinguishing the Painting from the general run of artwork assessed by Sotheby’s, he did not say that in his experience auction houses are unwilling or regard themselves as unable to assess period copies of Caravaggio’s works. There is no evidence before me therefore that period copies of early Caravaggio works are treated in a different way by leading auction houses from how they deal with period copies of any other artist’s works. On the contrary there is plenty of evidence that copies of Caravaggio and other Old Masters are assessed and sold by auction houses all the time. I do not consider that this is an instance of the leading auction houses falling into ‘bad habits’ and exposing consignors to unnecessary risk. I find that the fact that academic opinion about the attribution of many Caravaggio paintings has differed in the past and the extent of the scholarly debate over whether Caravaggio ever painted replicas of his own works did not mean that Sotheby’s experts ought to have refrained from coming to their own assessment of the quality of the Painting.

The third feature Mr Thwaytes relies on is that the Painting was dirty – sometimes people’s views as to whether it is autograph can change once the painting has been cleaned and restored. Although the Painting was examined with white spirit both at the Picture Meeting and at the Olympia Meeting, it was accepted by the Sotheby’s witnesses that this does not show every detail that would be disclosed by cleaning.

In my view, it would be impractical to hold that Sotheby’s experts were not competent to assess a painting because it was covered in discoloured varnish. That is something they do hundreds of times a year. Certainly if a painting is very dark and overpainted then it may be impossible to assess it. But that was not the case with the Painting, a point made by Professor Gregori in her witness statement where she says that the varnish of the Painting before cleaning had not been so darkened or oxidised as to obscure its quality.

Finally, Mr Thwaytes relies on the interest shown in the Painting by the Consortium which bid against Ms Benocci Adam at the sale but stopped at £40,000. Mr Sainty said that these dealers were very unlikely to be interested in a painting that was nothing more than a decorative period copy. They would not have wasted their time bidding on something unless they thought it had a chance of being right. As to why they did not bid further than £40,000 he said that the dealers would have realised that to establish the Painting as an autograph work was going to involve a great deal of time and money and that this was a speculative bid. Ms Kaminsky said that she found the conduct of the Consortium ‘inexplicable’. She did not reject the possibility that they would be interested in buying the Painting as a highly decorative and potentially profitable period copy. She said that even very high-end dealers have such ‘bread and butter’ sales for particular clients. She did not accept that the cost of researching the Painting would have discouraged them from bidding more than £40,000 given the very high value of an autograph Caravaggio. She said that if these dealers really thought that the Painting might be by Caravaggio she would have expected them to ‘put some backbone behind it’.

In my judgment it is very difficult to speculate what prompted the dealers first to join in the Consortium to bid but then to stop bidding at £40,000. There is no evidence that anyone at Sotheby’s had seen someone from one of the dealers examining the Painting. Mr Baring denied that this was what prompted him to call the Olympia Meeting. The notification of the Consortium was only received by Sotheby’s very late on the day of the sale. I do not consider that the interest of the Consortium of dealers should have caused Sotheby’s to conclude that those dealers thought that the Painting might be an autograph work and hence to question their own assessment of the Painting.

My conclusion is therefore that Sotheby’s were entitled to rely on their expertise and connoisseurship and to approach the question whether the Painting is an autograph early Caravaggio or a copy by considering first and foremost its quality.

Was Sotheby’s’ assessment of the poor quality of the Painting unreasonable?

Sotheby’s’ defence to this action is and has always been that the quality of the Painting is obviously inferior to anything that Caravaggio would have produced. Sotheby’s accepted that the treatise in the Forlì catalogue represented Professor Gregori’s honest opinion on the Painting and that it had not been motivated by any financial interest in the Painting or in the outcome of the debate over attribution. Her opinion of the quality of the Painting is adopted by Dr Lapucci in her report although Dr Lapucci does not do much more than repeat Professor Gregori’s view and say she agrees with it. Professor Spear says nonetheless that both their views are not reasonably held.

Mr Thwaytes submitted that the very fact that eminent experts can disagree so starkly over the quality of the Painting must show that it is a borderline case and that of itself means that Sotheby’s must have acted negligently in dismissing the Painting on the basis of poor quality alone. I do not consider that that is a fair way to approach the evidence and it was an approach rejected by the Court in Luxmoore-May. A similar point was rejected by the House of Lords in Bolitho v City and Hackney Health Authority [1997] UKHL 46. There the question was whether the negligence of the doctor in failing to attend the child had caused the subsequent death of the child. This in turn depended on whether the doctor ought to have intubated the little boy if she had attended to him. Although there was truthful evidence from an eminent specialist that he would not have intubated the child in those circumstances, the House of Lords held that the judge was not effectively constrained by that evidence to hold that failure to intubate would not have been a further act of negligence. Lord Browne-Wilkinson held that the court is not bound to hold that a defendant doctor escapes liability for negligent treatment or diagnosis just because he leads evidence from a number of medical experts who are genuinely of opinion that the defendant's treatment or diagnosis accorded with sound medical practice. The court has to be satisfied that the exponents of the body of opinion relied upon can demonstrate that such opinion has a logical basis. In my judgment it would be wrong to hold that the mere fact that Professor Gregori and Dr Lapucci regard the Painting as showing elements of high quality is enough to establish that the Painting should have passed the first hurdle of quality assessment at the Picture Meeting. It is my task to consider their evidence and come to a conclusion whether Sotheby’s was negligent in that no reasonable leading auction house would have concluded on the basis of quality that the Painting could not be by Caravaggio.

There were many passages of the Painting that were praised by Professor Gregori and Dr Lapucci but criticised as of inferior quality by the Sotheby’s witnesses. Here I discuss those which appeared to me the clearest. I bear in mind Buckley J’s warning in Drake v Thos. Agnew & Sons Ltd [2002] EWHC 294 (QB)about substituting my own assessment of quality for that of the experts. However, it seems to me that the task is inescapable here, given the issues in this case. Further, since the quality of Caravaggio’s work lies in its ability to convey to the viewer a naturalistic and convincing depiction of items or people, a lay person may be more justified in forming a view as to quality than he or she can of an artist who paints in a more abstract or impressionistic style.

The dice holder In the Kimbell Cardsharps the painting of the little dice holder sitting on the tric trac board is completely convincing in terms of its solidity, perspective and its placement on the board. In contrast, Professor Spear described the dice holder as ‘the most blatantly weak passage’ in the Painting because the copyist failed to keep it in proper perspective. I agree that the dice holder’s opening and the base in the Painting are awkwardly rendered as the top half does not seem to belong to the bottom half and the shape of the circular opening at the mouth of the cylinder does not tally with what one would expect to see. Professor Gregori in the Forlì catalogue attributes the different shape of the dice holder in the Painting to the suggestion that the Painting is a view of the scene from a higher vantage point - this ‘allows us to see its inside and shows a larger opening than the one of the same object in the Texan version’. I do not accept this explanation. As Professor Spear says, the dice holder in the Painting is ‘out of kilter within its own perspective system’ and this is inconsistent with Caravaggio’s skill. I do not see why Caravaggio, being clearly capable of painting a perfect dice holder in the Kimbell Cardsharps, would paint one that was so much less than perfect in a replica work.

Mr Legge put to the Sotheby’s’ witnesses that they were wrong to regard the dice holder as inferior simply because it was painted with incorrect perspective. In particular he compared it with the ivory comb that is lying on the table in the Caravaggio painting called The Magdalene in Detroit. It is true that the ivory comb is painted as if viewed from above whereas the little white bowl next to it on the table is painted as if viewed from side on. In that sense the perspective of the two items is awkward because it does not match. But there is nothing wrong with the depiction of either the white bowl or the ivory comb in itself. The problem with the dice holder in the Painting is of a different order.

The feather in the young sharp’s hat Professor Spear refers to the ‘magical evocation of a feather’s featheriness’ in the ostrich plume extending from the young sharp’s cap. The Amended Particulars of Claim refer to the delicacy of the highlights of the ostrich feather in the Painting and Professor Gregori refers in the Forlì catalogue to ‘the tufts of the feathers bristling up from the younger cardsharp’s hat and gently swaying against the light coloured background’. To my mind, if one is looking at whether the depiction of the feather in the Painting is as convincing a representation of the softness and fluffiness of an ostrich plume as the feather in the Kimbell Cardsharps then it is clear to me that it is not. The feather in the Painting has a shininess that is inappropriate because it suggests a waxiness that ostrich feathers do not have. The artist of the Painting has not captured the barbs of the feather extending over the hat. I accept Professor Spear’s assessment that the depiction of the feather in the Kimbell Cardsharps is greatly superior to that in the Painting.

The clothing In the Forlì catalogue Professor Gregori points out aspects of how the clothing is rendered in the Painting which she says show details that ‘are all on the same level of execution as the corresponding details in the Fort Worth painting’. She refers in particular to ‘the brush strokes running at speed’ over the soft, plum coloured garments of the dupe and the golden glow of the young sharp’s clothes. Again Professor Spear took issue with this. He pointed out that the artist of the Painting had not taken the same pains to convey the nature of the fabric as Caravaggio had done in the Kimbell Cardsharps. During his oral evidence he focused particularly on the striped sleeve of the old sharp:

“If you, my Lady, look at the Kimbell picture, you see that a bit of the black stripes defines the contour of the sleeve, and especially if you look at all the yellow stripes, you will see the Caravaggio, with great labour, in the lower layer painted many, many, many diagonals to give that yellow an extraordinary richness of surface and texture of textile. Do you see what I mean?

MRS JUSTICE ROSE: You mean although they are very vertical stripes, he filled them in by painting almost horizontally?

A. But they go from lower left to upper right, from 7 o'clock to 2 o'clock, and there are those striations all over that sleeve to give it the richness of texture. If you come back and look at the [Painting] sleeve, there is nothing there, it's plain, flat, yellow. And that's the difference between these two images: the other is a piece of fabric that's been run through too hot of an iron and it's lost all of its definition. And it's not because of damage, it's because the artist took a shortcut and didn't bother with what Caravaggio bothered with. I think the same is true of that contour at the outer edge, which is fuzzy. If you want to say sfumato to give it a fancy Leonardo word, well and good. But I think the artist here just didn't bother much with this sleeve.”

What he says is clearly right. Similarly I accept Mr Bell’s and Professor Spear’s assessment of the realism of the muslin folds that protrude through the slits in the sleeves of the young sharp’s doublet. They are much more convincing of the softness of the cloth in the Kimbell Cardsharps than they are in the Painting.

The handling of light Professor Gregori refers in the Forlì catalogue to ‘the subtle fall of light’ visible on the subjects’ hands and faces. She refers to the fine and very subtly defined slithers of light outlining the edges of items such as the playing cards and the pewter plate. There does not appear to me anything particular convincing about the highlights. The most testing passage for conveying light and shadow is in the lace cuffs of the dupe’s sleeves. This is very well done in the Kimbell Cardsharps whereas the lace in the Painting is schematic and stiff looking.

There were many other passages in the Painting that were criticised by Mr Bell and Professor Spear – the dupe’s right ear, the weave of the carpet covering the table, the inside edge of the pewter plate and the gold stripes on the young sharp’s breeches. Having considered all these in comparison with the passages of the Painting that are particularly praised by Professor Gregori and Dr Lapucci I am firmly of the view that Sotheby’s were entitled to come to the view that the quality of the Painting was not sufficiently high to merit further investigation.

Sotheby’s’ failure to notice features of the Painting (visible at the Picture Meeting) which indicate that it is not a copy

The second main allegation of negligence against Sotheby’s is that Mr Bell and his colleagues failed to spot features about the Painting that should have alerted them to its Caravaggio potential. These features fall into two classes – features which are ‘characteristic of Caravaggio’ (‘Caravaggio features’) on the one hand and features which are characteristic of a painting not being a copy (‘non-copy features’). Before considering the individual features there are some preliminary points to consider.

As regards Caravaggio features, of course when one is considering whether a copy of a well-known work is by Caravaggio or not, there is no point relying on features of the composition of the image as being typical of Caravaggio; that is what the copyist has tried to reproduce. So Caravaggio features here are features to do with the construction of the Painting and the techniques used. The difficulty however with identifying a particular technique with being characteristic of Caravaggio – at least as regards a period copy – is that it is accepted that there is very little research into copies that enables one to say that Caravaggio produced his paintings in a particular way which none of his contemporaries used. Mr Bell made the point about the lack of money to fund research into copies of Old Masters:

“… an important point to make, is that very often study of artists like Caravaggio are made with paintings that are thought to be by Caravaggio. So [infra red analysis] is done on a body of work which is broadly accepted. There is not a view -- what doesn't happen is a group of a hundred pictures, some of which are by Caravaggio, the majority of which aren't, are studied in the same way; because it may well be that an awful lot of pictures which are not by Caravaggio exhibit very similar characteristics to paintings that are thought to be by Caravaggio. It's where the money is. You know, you don't get a grant to study fifty ‘not-Caravaggios’ in order to determine what techniques – what differentiates them from the five pictures which are by Caravaggio.”

One must be very cautious therefore, when assessing a contemporary copy, about what features are or are not significant.

So far as non-copy features are concerned, it was accepted by Professor Spear that it is possible to identify features in a painting which are ‘characteristic of it not being a copy’. Three kinds of non-copy features were discussed in the course of the trial:

A number of passages in the Painting were proposed as indicating that it is not a straight-forward copy.

The additional area above the old sharp’s hat Mr Baring looked at the photograph of the Kimbell Cardsharps in the Spike book when he initially examined the Painting. Mr Bell could not remember whether he also looked at the Spike book at the Picture Meeting but I will assume that he did or that he should have done. From a comparison of the photograph and the Painting it is clear that there is more space above the old sharp’s hat in the Painting than there is in the Kimbell Cardsharps. Should this have alerted Sotheby’s that this was not a straightforward copy because the artist of the Painting had clearly used some artistic creativity in varying the proportions of the image rather than slavishly copying what he saw? Mr Bell was asked whether this difference in size would have excited further interest. His response was as follows:

“I think we do find copies of not identical dimensions to originals, and what I'm always conscious to look out for in cases like that, to indicate whether the picture that you think is a copy might be something more interesting than that, is in the area which doesn't exist in the original, is there something of interest? Is there something that's well painted, something that shows the artistic mind at work? Here you have an upper strip which is simply bland and pretty undifferentiated. So even if I had known that, there is nothing in that upper band that would have made me think immediately: gosh, there's something special going on here.”

There is an additional point here in that Mr Thwaytes asserts that Mr Bell and his colleagues should have considered not only the Spike picture of the Kimbell Cardsharps but also the Burlington Cardsharps Publication by Sir Denis and Keith Christiansen. This, as I have set out, refers to an additional strip at the top of the Kimbell Cardsharps having been removed during restoration. Dr Christiansen’s appendix states that there is no way of establishing precisely when the addition was made but he notes that it was included in the Volpato engraving. The article contains a photograph of the Kimbell Cardsharps before restoration showing a substantial area above the feather in the old sharp’s hat. Sir Denis’ article also says that it ‘seems impossible to pin point the period when the enlargement occurred’.

There was a great deal of technical evidence from the experts at the trial as to whether the Painting could have been copied from the Kimbell Cardsharps after the strip was added. The evidence was that the 14 cm strip was glued onto the main canvas rather than being sewn on – a fact not mentioned in the Burlington Cardsharps Publication. Professor Modestini accepted that if a 17th century painter wanted to enlarge his composition he would have sewn the addition on rather than glued it. The use of glue therefore indicates that by the time the strip was added, the canvas had become too brittle to sew. This in turn points to the addition of the strip taking place when the Kimbell Cardsharps was already over 100 years old because that is the period after which canvas of this kind becomes brittle. Since Sotheby’s accept (in cataloguing the Painting as by a Follower of Caravaggio) that the Painting was contemporary or near contemporary, it cannot be the case, it was submitted, that the Painting is a copy of the Kimbell Cardsharps with the added strip.

On this point I must keep in mind what conclusions might reasonably have been drawn at the Picture Meeting or at the Olympia Meeting about the difference in size between the images in the two paintings. I find that if the Sotheby’s specialists had noticed the additional space in the Painting above the old sharp’s hat as compared with the Kimbell Cardsharps, they would reasonably have concluded, on reading the Burlington Cardsharps Publication, that the Painting had been made after the strip had been added. There is nothing that they could have seen at that point that would have alerted them to the puzzle about the gluing of the strip and the dating of the Painting.

On that basis I do not accept that the additional space in the Painting was something that would have been regarded as a non-copy feature by a reasonable auction house.

The black mass Another difference between the Painting and the Kimbell Cardsharps is the presence of a black mass behind the right elbow of the dupe that is not there in the Kimbell Cardsharps. Mr Bell acknowledged that if he had compared the Painting closely with the photograph of the Kimbell Cardsharps in the Spike book, he would have spotted that there is a black mass in the Painting but not in the autograph work. He denied that this ‘mushy black area’ as he called it would have excited his interest because ‘it doesn’t compositionally do anything’. Ms Glanville’s evidence was that she would have been ‘intrigued’ by the existence of the black mass and by the fact that it appeared in the Volpato engraving but not in the Kimbell Cardsharps. Her evidence is that when she examined the Painting she concluded that the black mass was in the abbozzo layer. This was based on the fact that whereas other dark areas of the Painting were very abraded, this area was not; suggesting to her that it had been protected over the centuries by being covered by layers of paint and varnish. She relied on this as indicating that it was an element of creativity introduced by the maker of the Painting and not just a reproduction of an element in the work being copied.

On this point I accept Mr Bell’s evidence that a compositional change which appears to have no artistic merit of itself and does not seem to have any particular purpose in the composition is not something that amounts to a non-copy feature. Further the same point arises here as I have discussed in relation to the additional strip. If the Sotheby’s experts had read the Burlington Cardsharps Publication as Mr Thwaytes says they should, they would have noted that Dr Christiansen refers in the Appendix to this:

“Prior to cleaning there was a blackish area that ran over the carpeted table top in a meaningless fashion. The pigment here was old but certainly not original and it has been removed. However, in both the Volpato engraving and in an old copy .. there appears a well defined form suggestive of a cape folded over a chair. Whether this was an alteration introduced by Caravaggio or whether it was added later cannot be established definitively with the evidence at hand. Microscopic examination showed that the background grey had been used by Caravaggio to tidy up and slightly reduce the contour of the sleeve. At one point this grey had been broken through by an early restorer, probably in an attempt to remove the 'cape', and the penumbral black area was substituted (contradicting the actual source of light from the left). It is conceivable that at the same time the 'folded cape' was removed the candle scorch marks visible on the back wall of the copy were also reduced, but the matter necessarily remains somewhat speculative.”

Again, if the Sotheby’s experts had been curious about the existence of the black mass in the Painting, they would reasonably have concluded on reading Dr Christiansen’s account that at some stage of the life of the Kimbell Cardsharps, possibly during the lifetime of Caravaggio himself, it contained a black mass of some kind and this explains why the Painting, the Volpato engraving and indeed many other copies of the work also have something dark at the dupe’s elbow. I do not consider that Sotheby’s could reasonably have discovered that the black mass in the Painting is in the abbozzo layer, if indeed it is. I therefore conclude that the black mass was not something that should have been regarded as a non-copy feature by Sotheby’s.

Difference in view point or expression Both Ms Glanville and Dr Lapucci said that the scene depicted in the Painting was painted from a different view point – by someone looking at the scene from higher up - than the view point of the painter of the Kimbell Cardsharps. They illustrated this by overlaying the composition of the Kimbell Cardsharps on top of the Painting and comparing the outlines of the figures and items. This showed, they said, that Caravaggio improved the composition of the scene in the Painting from what he had painted in the Kimbell Cardsharps by shifting the pewter plate, the dagger and the left leg of the young sharp slightly in a way which increases the apparent spatial relationship between the figures to give the composition more depth. Both Ms Glanville and Dr Lapucci rejected the suggestion that any shift in view point was simply the result of slippage of the pieces of paper used to trace the image from the Kimbell Cardsharps to the Painting (a method of copying commonly used at the time). This could not account for the fact that within the same small area of the composition, the shifts between the images were not all in the same direction.

Dr Lapucci made a different point about the moment which is depicted. She believes that the Painting is not in fact intended to be a copy of the Kimbell Cardsharps at all but is a different composition by Caravaggio of a slightly later moment in time. She reads the stances of the bodies of the three figures and the expressions on their faces as showing that the Painting depicts a different ‘psychological moment’ when the cheating had been discovered:

“In terms of the poses, the Kimbell Cardsharps seems to be temporally first: the two cardsharps are relaxed (as if they had not yet been discovered) and the innocent youth seems to hold back his body with dynamic tension, an elastic pose that springs back, and recoils, as if to prevent the cards in his hands being seen ... Caravaggio has portrayed two different consequent moments or instances that happen in a few seconds: the Kimbell Cardsharps depicting the moment before the deceit, and the painting depicting the moment after the deceit has been unveiled.”

As another example of a Baroque artist painting the same scene at two different moments, Dr Lapucci referred to Gentileschi’s two versions of Judith Beheading Holofernes where it is clear that in one version Judith is further advanced with the task in hand than she is in the other.

On this point, I find that these supposed differences in the composition are illusory. It is much more likely that any small differences are due to the failure of the copyist to reproduce the image accurately. I do not accept that the spatial appearance of the Painting is better than the Kimbell Cardsharps. I do not accept that it should have struck anyone – however expert - on examining the Painting that it was not intended to be an accurate copy of the Kimbell Cardsharps. I do not accept Dr Lapucci’s evidence that it would be unreasonable for an expert looking at the Painting to consider that it was intended to be a copy of the scene depicted in the Kimbell Cardsharps. I therefore find that these differences do not exist or at least that they have no significance as features that should have alerted Sotheby’s to the Caravaggio potential of the Painting.

Different light source and palette Ms Glanville pointed out that the scene is lit differently in the Painting compared with the Kimbell Cardsharps, in particular that in the Painting there is an aura of light area around the outline of the dupe’s right side that does not appear in the Kimbell Cardsharps. Dr Lapucci agreed and also considers that the colour palette used by the artist was different. Mr Bell accepted that the light source appeared different in the two paintings but he denied that this had any significance. As regards the suggested change in palette he said:

“I wouldn't necessarily expect the palette of every copy to be identical to the original for the reasons in the way I have, I think, started to explain, which is it depends a little bit when the original was painted, whether the pigments are the same, whether the painting has aged in the same way because of where it has been kept, whether those pigments have degraded at the same rate and the change of colour has happened in the same way. So there are many reasons why a picture might not look identical, even if at the outset the two did.”

I struggle to see the difference in palette that Ms Glanville and Dr Lapucci have identified. Certainly if the young sharp’s left sleeve had been scarlet rather than turquoise that would have been a striking difference. But Mr Bell’s evidence accords with common sense and I have not seen any expert evidence to gainsay it.

The looped bow Dr Lapucci and Ms Glanville point to the difference in the bow at the young sharp’s right elbow. They state that the bow in the Painting is straight whereas in the Kimbell Cardsharps the bow shows a definite loop. It emerged during the course of the evidence that the loop in the bow may not have been included by Caravaggio in the Kimbell Cardsharps but may have been added by a later restorer painting over the original tacking edge of the canvas. In any event, this seems to me to be such a small detail that it cannot reasonably be regarded as a non-copy feature.

Decoration on the dice holder Finally, Mr Bell’s evidence was that they would have looked at the Painting with ultra-violet light at the Picture Meeting. When one shines UV light onto the dice holder in the Painting, some dots and lines of paler paint become apparent. This is not the case with the Kimbell Cardsharps. These dots and lines are described in the Amended Particulars of Claim as ‘decorative detail’. Both Ms Glanville and Dr Lapucci regarded this as a significant creative difference between the two works. They rejected the suggestion that these were just accidental or random droplets of paint because the paint is confined to the side of the dice holder that appears lit and not spread over the whole dice holder. Dr Lapucci described the decoration as characteristic of or at least compatible with Caravaggio’s work. Mr Bell did not recall seeing this detail when he examined the Painting.

I have not been convinced by what I have seen that there is really any decorative detail on the dice holder. Although I accept the point that the dots and lines appear to be confined to the lit area of the object, they do not appear to form any logical pattern or shape. If the artist really intended to change the dice holder from the plain object in the Kimbell Cardsharps to a decorated object he would probably have done so with rather more conviction than is evidenced by these ambiguous additions. I certainly would not accept that no reasonable auction house could have failed to spot these additions or to have accorded them significance.

In summary, my conclusion as regards all the elements that it is alleged should have been visible to the Sotheby’s experts at the Picture Meeting is that none of them either individually or cumulatively was sufficiently important to have triggered in their minds the possibility that these were non-copy features. As regards the two largest differences, namely the additional space above the old sharp’s hat and the black mass at the dupe’s elbow, these were not of any artistic merit in themselves, they are features found in many other copies of this work and if Mr Bell had consulted the Burlington Cardsharps Publication he would reasonably have concluded that the Painting had been made at a time when the Kimbell Cardsharps had been enlarged and contained the black mass as described in Dr Christiansen’s Appendix. The other differences relied on are either non-existent or so minor as to be entirely consistent with a copyist either failing accurately to capture the image or making a few very minor changes of his own.

In my judgment there is nothing disclosed on visual examination which should have counteracted Sotheby’s view that the Painting was of poorer quality than the Kimbell Cardsharps and did not therefore have Caravaggio potential.

Sotheby’s’ failure to compare the Painting with the Volpato engraving

The Volpato engraving bears the inscription Michelangelo da Caravaggio pinxit Johannes Volpato Sculpsit Romae 1772. It also states that it was taken from a picture in the collection of the Palazzo Barberini in Rome. This was one of the images through which the composition was known whilst the original was considered lost – that is before the Kimbell Cardsharps was discovered and published as the autograph original. It is alleged that Mr Baring and those attending the Picture Meeting should have looked at the Burlington Cardsharps Publication and that if they had done so they would have seen the photograph of the Volpato engraving. Such a comparison would, it is alleged, have made them realise that there were strong similarities between the image in the Volpato engraving and the Painting such that they should have concluded that the Volpato engraving may have been copied from the Painting and not from the Kimbell Cardsharps as had been previously been assumed.

The suggestion that the Volpato engraving is taken from the Painting and not from the Kimbell Cardsharps is in my view extremely speculative. In the Burlington Cardsharps Publication Sir Denis describes the stamp found on the back of the canvas of the Kimbell Cardsharps showing that it was once in the collection of Cardinal del Monte who then sold it to the Barberini collection. The Volpato engraving states that it is taken from a picture in the Barberini collection. Even if it is accepted that Caravaggio’s patrons may have had replicas and copies in their collections as well as autograph unique works, there is no evidence at all to suggest that the Painting was ever in the Barberini collection.

I am prepared to assume for present purposes that Mr Bell and his colleagues should have looked at the Burlington Cardsharps Publication or otherwise have been aware of the Volpato engraving at the time they examined the Painting. As to the passages in the Volpato engraving which are said to be closer to the Painting than to the Kimbell Cardsharps my conclusions are as follows:

As a counterweight to the points made by Mr Thwaytes on the Volpato engraving, Sotheby’s point out what is clearly true, namely that the depiction of the light in the Volpato engraving is closer to the depiction of light in the Kimbell Cardsharps than in the Painting – there is no lit area behind the dupe’s back.

Having looked carefully at all three images and considered the evidence of Dr Lapucci and Ms Glanville I have concluded that there is nothing in the Volpato engraving or the Burlington Cardsharps Publication that ought to have alerted Sotheby’s to the idea that the engraving was taken from the Painting and not from the Kimbell Cardsharps.

Sotheby’s’ alleged failure to examine the x-rays properly

Sotheby’s accept that once they had undertaken to obtain x-rays for Mr Thwaytes, they were under a duty to examine the x-rays competently. I note that the x-rays of the Kimbell Cardsharps are very obscure because Caravaggio seems to have used an x-ray opaque material in the priming or ground of the work so that there is very little that can be seen of the figures in the composition.

The first challenge to Sotheby’s’ handling of the x-rays was the allegation that Mr Bell and his colleagues should not have attempted to interpret the x-rays at all themselves but should have sent them out for expert consideration. This was particularly the case, it was submitted, given that Mr Thwaytes had stressed to Mr Barton and Mr Baring that he wanted to be absolutely sure that the Painting was not by Caravaggio. The evidence was that Mr Bell was used to looking at x-rays of works although of course this was not as frequent occurrence for him as it was for Dr Lapucci and Ms Glanville. His evidence on this point was that:

“I am certainly not a specialist and I certainly wouldn't have the degree of skill to interpret an x-ray in the way that someone who specialised in this area alone would have done. What I have learnt over the years from talking to scholars and conservators is to look out for telltale signs on an x-ray which might be indicative of a creative process going into the production of that particular work of art.”

It was put to him that when Sotheby’s took x-rays of the painting St John at the Well (discussed later) they sent them for analysis to an external expert. He explained that there had been particular reasons for that at the time that did not pertain to the Painting. He recognised that his competence would go as far as seeing whether there were significant changes or significant creative process going into the production of the work.

There was no evidence before me that auction houses or art dealers regard themselves as not competent to form a preliminary view as to whether x-rays of copies of Baroque paintings in general or of Caravaggio’s works in particular show something worthy of further investigation. I therefore reject the suggestion that Mr Bell or his colleagues should automatically have sent the x-rays out for external consideration because they were not competent to assess them themselves.

As to what Sotheby’s did do with the x-rays, Mr Thwaytes criticised them for not having compared the x-rays alongside the Painting itself. Mr Baring’s written evidence was that he did remember doing this. Mr Bell could not remember. The significance of this was, principally, that if they had done so this would have enabled them to notice some pentimenti that were not obvious from the x-rays themselves, in particular the length of the ribbon dangling from the left elbow of the young sharp. I consider this point later.

Mr Bell was also criticised on the basis that his evidence was that he had limited his consideration of the x-rays to looking for major pentimenti. This would have been negligent, it was submitted, because it is well known that many of Caravaggio’s early works do not contain major pentimenti. I do not regard it as a fair reading of Mr Bell’s evidence to say that he dismissed the x-rays just because they did not show major pentimenti. Certainly he said that the presence of a major pentimento would have been inconsistent with the Painting being a straight-forward copy of the Kimbell Cardsharps. But he recognised that the converse did not hold true – the absence of major pentimenti did not rule out the Caravaggio potential of the Painting. He was also looking for something rather more difficult to define:

“I think if I had seen an indication of a creative process involved in the x-ray, with changes, significant changes, that would have rung an alarm bell, it wouldn't necessarily have provided an answer, but there would have been a question: why is the x-ray so free and showing indications of change when the surface of the picture is so static and copy-like? What happened is that the surface of the picture appeared to us to be static and copy-like and nothing that I was able to see in the x-ray led me to believe that we were dealing with anything that exhibited features other than you would expect to find in an x-ray of a copy.”

In my judgment he was right to conclude that the absence of major pentimenti in the x-rays meant that in this regard the x-rays contained nothing to change the view he had reached after visual inspection of the Painting – the x-rays did not move the debate forward. What he also noticed about the x-rays and what is striking even for a lay person is that the x-ray mosaic of the Painting overall looks like a black and white photograph of the image. There is little difference between the x-ray image and the surface image of the Painting. This Mr Bell says, and I accept, is typical of x-rays of copies because the artist has painted only what is necessary to reproduce the surface of the composition. Again I accept that the x-rays images of some early Caravaggio works may have a similar quality. But the conclusion that Mr Baring and Mr Bell reached on examining the x-rays was, in my judgment, absolutely right. The x-rays did not prove that the Painting was or was not by Caravaggio or was or was not a copy. They simply did not provide any additional information that should have caused the Sotheby’s specialists to reconsider the view they had formed from their visual inspection of the work.

Ms Glanville and Dr Lapucci’s reports listed various passages in the x-ray that they say should have sounded alarm bells in the minds of the Sotheby’s specialists. I discuss below the passages that I consider are the most significant.

Pentimenti Although there was no major compositional change, Dr Lapucci draws attention to some alterations in the composition of the Painting namely some repositioning of the dupe’s right hand holding the cards, the expansion of the outline of the right side of his hat and a clump of hair at the back of his neck. There is also the dangling ribbon from the young sharp’s left elbow. On the surface of the Painting, as on the Kimbell Cardsharps, the ribbon dangles down so that it obscures part of the sides of the stack of cards placed face down on the table between the young sharp and the old sharp. The stack of cards is painted in x-ray opaque material whereas the ribbon is x-ray transparent. From the x-rays one can see that the painter of the Painting has initially painted the ribbon slightly shorter; then painted the stack of cards around it. The painter must then have extended the end of the ribbon over the edge of the stack of cards. This extension appears on the surface of the work but not in the x-ray because it is transparent to x-ray and so does not block out the relevant passage in the x-ray opaque stack of cards as it does on the surface image. Is this pentimento significant? I do not see that it is. It is entirely consistent with a copyist initially painting the ribbon too short, then later realising that the ribbon in the original image was longer and so he extended the ribbon in the Painting to make it look more like the ribbon in the Kimbell Cardsharps. I do not consider that this pentimento should have alerted Sotheby’s to the existence of some creative mind at work in the composition of the Painting.

The handling of the paint It was said by Mr Thwaytes that various passages in the Painting were revealed by the x-rays as showing ways of handling the paint that were either characteristic of Caravaggio or at least not methods that the copyist would use. These included a broad stroke in white lead picking out the outline of the dupe’s left arm against the background of the old sharp’s doublet and the highlights on the cuffs of the dupe’s sleeves. As regards the former, there was a disagreement between the experts as to the purpose of this white line. Dr Lapucci and Ms Glanville regarded this as a subtle technique building up the composition in terms of light and shade prior to turning it into colours. Professor Spear’s explanation was more prosaic: that it was simply the copyist drawing the contour of the sleeve in a mechanical or pedantic way and then filling in the sleeve with colour. I have considered each of the instances relied on and concluded that Professor Spear’s explanation is equally if not more likely to be the case than the Claimant’s experts’ explanation. It is impossible to say that these small features should have alerted Sotheby’s to the existence of non-copy features in the Painting.

More generally it was not, as I understood it, suggested that the x-rays showed the kind of freedom or energy of painting that Mr Bell says he was looking for. There was said to be a contrast between the ‘vigorous brushstrokes’ painting the background and more delicate work painting the feather. But this seems to me explicable on the basis that anyone painting a large homogenous area may well do so more vigorously than a delicate passage in the work. I do not regard that as significant.

Different painting techniques Ms Glanville points out that the x-rays show that the painter of the Painting used different techniques in different passages of the work. Thus, the dagger at the side of the young sharp was painted en reserve of the carpet on the table and the carpet colour was then painted around the pommel of the dagger. In the Kimbell Cardsharps it is accepted that the dagger was painted over the top of the young sharp’s breeches. There are other areas of the Painting where the items overlap, for example the cards in the dupe’s hand and the young sharp’s doublet has been painted over the top of the old sharp’s left hand. Two other passages were contrasted namely the fact that the dupe’s hair is painted in the same layer of paint as the flesh of his forehead whereas the old sharp’s beard is painted on top of a layer in which his chin is painted. As regards this last point, Dr Lapucci said:

“If you are copyist you know that the beard arrives here so if you have a tracing you follow the shape of the beard and now it is abraded and transparent, but in the past it was probably more strong and intense so you could stop – a copyist does not lose time in doing the entire face and then … waiting for it to dry and again going on top with another colour generally. A copyist tries to get the copy done quickly because it is only worth 15 scudi”

I found this point unconvincing. The old sharp’s beard in the Kimbell Cardsharps is a rather scrubby ‘five o’clock shadow’ through which his chin needs to be seen. I do not see why Dr Lapucci should assume that the beard of the old sharp in the Painting was originally much more intense and hence why it would not have made sense even for a copyist to paint the chin first and sprinkle the stubble over it later. More generally, there is no reason to suppose that a copyist would not paint with slightly different techniques when depicting different items. I regard Dr Lapucci’s suggestion that the fact that the stripes of the old sharp’s right sleeve overlap a few millimetres with the left sleeve of the dupe indicates to her that the Painting was made from live models to be very speculative.

My conclusion on the examination of the x-rays is that there is nothing here that should have alerted Sotheby’s to any Caravaggio potential of the Painting.

Sotheby’s’ analysis of the infra-red images

I have already rejected the submission that Mr Thwaytes’s letter of 11 July 2006 amounted to an instruction to Sotheby’s to carry out infra-red testing. I do not consider that any other instruction was given by Mr Thwaytes in relation to infra-red testing or that Sotheby’s ought to have offered Mr Thwaytes the opportunity to have infra red analysis carried out, even though Mr Thwaytes stressed that he wanted to be sure of the Painting’s status. I accept that Mr Thwaytes was prepared to pay for infra-red testing just as he was prepared to pay for x-rays. But Mr Baring would not have been justified in encouraging Mr Thwaytes to spend his money on further tests, merely because Mr Thwaytes would have been happy to pay for it. An auction house’s task is to advise consignors as to how much research into a painting is justified by the likely results. Mr Baring had no reason to suppose that infra-red analysis would reveal anything of note about the Painting.

At the trial we had the benefit of looking at the infra-red images that had been made on the instruction of Sir Denis after he acquired the Painting. Various passages were put to the Sotheby’s witnesses as being non-copy features. As previously, I consider below those where I thought that Mr Thwaytes’s case was the strongest.

Eye under the hat In her treatise in the Forlì catalogue, Professor Gregori says this: (emphasis added)

“On closer inspection of the London version of The Cardsharps … the radiography seems to confirm to me .. that the dark brim of the dupe’s hat, which obscures half of the older cardsharp’s face (a rather brutal stroke of realism), in fact covers the latter’s face painted in its entirety, complete with details. This particular detail of execution, also found in the Fort Worth version, excludes the possibility that the newly found London version may have been painted by another artist and brings us straight back into the Lombard painter’s work…”

This is the one point she also confirms in her witness statement in these proceedings. In the witness statement she seems to suggest that she could see the right eye of the old sharp on visual examination of the cleaned Painting although in the Forlì catalogue she refers to this being confirmed in the x-ray images. Dr Lapucci’s evidence in her witness statement was that the eye ‘is very clear’ in the infra red images and can also be seen in the x-ray. In cross examination her evidence seemed more equivocal as to whether she could see the eye, or some other modelling of the old sharp’s face including his cheek and part of his forehead and whether there was a ‘basic sketching’ of the face or a face fully painted.

I accept that if the old sharp’s eye was painted under the dupe’s hat, this would be a significant indication that the Painting was not a copy. But I am sure that there is no eye under the hat. There is nothing apparent on the x-rays or the infra red images to indicate anything other than the painted folds of the velvet of the dupe’s hat. There are some darker areas and shadows in the images but these do not seem to me to form anything that looks like an eye or part of a face. I also do not accept that Professor Gregori saw anything material through the thin paint before the Painting was restored. It is more likely that she saw some of the brownish pink ground to which Mr Bobak refers in his 2007 condition report.

Drawing lines on the dupe’s fingers and facial features The infra-red images show some drawing lines particularly round the hands of the dupe and around the facial features of some of the characters. The lines on the dupe’s hands are notable because they do not seem to outline the feature but rather to shade in the tips of the fingers that are going to be in shadow in the final image. There was some debate between the experts as to whether infra-red analysis would have shown these lines in 2006 because shortly after that date advances were made in the wavelengths that could be used in this analysis and more drawing could become apparent. For present purposes I will assume that if Sotheby’s had taken infra-red images before the sale of the Painting they would have seen what we saw in the images presented at trial. Professor Spear says that he is not aware of any other instance in Caravaggio where there is this shading of features. He regards this as a clear indication that the Painting is not by Caravaggio. Dr Lapucci on the other hand says that the markings are ‘characteristic of Caravaggio’. On this point it does not seem to me unlikely that a copyist would make these kinds of marks when trying to replicate the image he sees. There seems no reason to conclude that these are Caravaggio features rather than something that any competent painter might have done. Even if there are similar drawing marks in other works by Caravaggio the existence of these marks is not a strong indication that the Painting is an autograph work.

The profile of the young sharp As I described when dealing with technical terms, the face of the young sharp is painted within the reserve of the cloak of the old sharp. Moreover there is a bordo a risparmio between the edge of his profile and the cloak. It is not clear whether this would have been capable of being seen by someone copying the Kimbell Cardsharps shortly after it had been finished but it has become more apparent over time and is visible on the infra-red image. Mr Thwaytes argues that this is a non-copy feature because a copyist tends to reproduce only the surface of the work rather than using a sophisticated and subtle effect like this. Mr Bell did not recall having spotted this feature when he examined the Painting or the x-rays. He denied that if he had seen this, it would have ‘rung an alarm bell’ given the other visual information they were obtaining from the Painting itself.

This point was also put to Professor Spear, in particular referring him to an article he wrote in 1998 called ‘What is an original?’ in a publication called ‘The Italians in Australia’ published by the University of Melbourne. In that article Professor Spear compared an autograph work by Guido Reni and a copy of that work which he described as ‘far from Reni’s personal manner’:

“It has a slick, superficial aspect, and I mean superficial both literally and metaphorically, because everything appears to be right there on the surface, as is typical of so many copies, which often are made that way. That is, instead of being built up as originals are, they frequently rely on shortcuts that replicate only the final, surface layer of their models and hence end up looking like decals.”

Despite this, Professor Spear rejected the suggestion that this use of the ground as part of the final image in the Painting was a non-copy feature:

“Q. And you would accept that that shows an artistic intelligence at work?

A. No.

Q. And why would you say it doesn't?

A. As I have said throughout my report, my Lady, copyists, and especially diligent ones, copy what they see, including effects of ground. I could ask you, if you wished, to go to the current beautiful Constable exhibition at the V&A which I did as a break over the weekend. … Look at John Constable's very detailed copy of Ruisdael's Landscape with Windmills. In the Ruisdael, especially in the lower left area, you see the ground coming through, which was surely intended as part of Ruisdael's painting. Look at Constable, and what did he do? He left the ground showing through in those same passages. A good copyist who wants to replicate what he is looking at will do things like that: let you see what's in the original, including ground coming through. Now, why the Mahon painter decided to basically cover up that area, I don't know.”

Professor Spear admitted that he could not point to any particular example of a copyist of the Cardsharps having made use of the ground in this way because there are very few x-rays and infra-red images of copies. But he did not rule out that a copyist would do this. On this point I accept Professor Spear’s evidence. The passage quoted in his article was not intended as an absolute statement. Copyists were sometimes accomplished artists in their own right (even if not of the calibre of Constable) and were trained in the same techniques as their more famous contemporaries.

My conclusion on the analysis of the infra-red images is therefore that even if Sotheby’s had arranged for these images to be produced before the sale, there is nothing in those images that would have alerted them to any Caravaggio potential of the Painting.

Was Sotheby’s negligent in not informing Mr Thwaytes about the Olympia Meeting?

It was alleged that Sotheby’s ought to have told Mr Thwaytes that something had happened to prompt Mr Baring to ask the three experts to come in a taxi to Olympia to reassess the Painting on 4 December 2006. It is accepted that Mr Thwaytes and Mr Baring spoke on 4 December 2006 and Mr Thwaytes asked Mr Baring whether there had been any interest in the Painting. Mr Baring replied that there had been a bit of interest ‘but nothing untoward’. It was not possible to ascertain whether this conversation took place before or after the occurrence of whatever triggered the Olympia Meeting. Mr Thwaytes argued that if the conversation with him took place after the Olympia Meeting then what Mr Baring said was seriously misleading. If it took place before the trigger then it was submitted that it was wrong of Mr Baring to fail to correct the impression that he had given Mr Thwaytes – he should have telephoned Mr Thwaytes to tell him about the Olympia Meeting and let him decide whether to withdraw the Painting from sale.

It is submitted that the effect of this failure was that the Painting was sold without any further attempt to procure academic support. There were examples cited in evidence of paintings being withdrawn from sale at the last minute because of interest shown. Mr Baring gave evidence of a painting having been ‘pulled’ from a sale at Bonhams in Oxford because he was seen looking at it. The painting turned out to be by Velazquez.

I do not accept that Mr Baring was at fault here. I accept Ms Kaminsky’s evidence that it would be unlikely that a consignor would be informed about repeat viewings made by the experts unless they resulted in significant change in the status of the painting. She said:

“During this phase in the auction cycle, all of the department experts will be focused on meeting with and speaking to potential buyers, in essence to try to 'sell' the paintings. Every auction has several hundred paintings and as such many consignors. It would be unreasonable, in fact, impossible, for the experts to update each client about the interest shown in their lot(s), unless they had important news. This would distract from the most pressing and time-sensitive task at this stage in the auction process, which is beneficial to both Sotheby's and the consignors.”

As I understood her evidence, ‘interest’ in a painting means more than just a lot of people looking at it. Interest can be gauged in a number of ways such as people requesting a condition report or registering to make a telephone bid for the work. She would not regard the fact that a number of reputable dealers are looking closely at the work as any sign that there might be a mistake about the cataloguing of a work. The interest shown may subsequently be reflected in a large number of bids at the sale or it may not. As regards the Painting, the evidence is that there was only one request for a condition report before the auction and that did not come from someone likely to pay a large sum for the work.

I do not consider that the approach that Sotheby’s took was unreasonable or fell below what can be expected of them. I do not consider that the attitude described by Ms Kaminsky as being adopted by the auction houses is a ‘bad habit’ which operates against the overall interests of the clients. Further I consider that the suggestion that Mr Thwaytes would have withdrawn the Painting from sale if he had been told what had happened at the Olympia Meeting to be implausible and affected by the benefit of hindsight. On the contrary, the fact that two ‘fresh pairs of eyes’ in the form of Ms Treves and Mr Apostle had also formed the confident view that the Painting was correctly catalogued as a copy should have reassured Mr Thwaytes that Sotheby’s were taking care properly to assess the Painting.

Overall conclusion on negligence

In the light of the conclusions set out above, I find that Sotheby’s were not negligent in their assessment of the Painting:

In case I am wrong on the issue of negligence I now turn to the question of causation. What would have happened if either (a) Sotheby’s had concluded that they should not assess the Painting themselves but should call for expert assistance or (b) they had spotted Caravaggio features or non-copy features in the Painting which caused them to think that the Painting had Caravaggio potential.

The key question so far as this part of the case is concerned is which experts Sotheby’s would have consulted. Would they have gone to Sir Denis Mahon and Professor Gregori for their opinions and hence found out before the sale that those two experts thought that the Painting was an autograph replica? It was not asserted that it would have been negligent of Sotheby’s not to go to Sir Denis and Professor Gregori. What Mr Thwaytes says is that the evidence shows that in fact they would have done so.

It appears to be common ground that the decision which experts to consult would be taken by Mr Bell. It is also common ground that there is no single ultimate authoritative voice on the attribution of Caravaggio as there is with some artists. Ms Kaminsky’s evidence was that that the ability to navigate the difficult waters of seeking scholarly views on attribution is an important skill for an auction house senior specialist to have. It also appears that scholars are generally willing to give their views when consulted by Sotheby’s, often on the basis of looking at photographs of a work. Sometimes the scholar is not, however, willing to have their view disclosed to the public and indeed, some who work for museums or other institutions are constrained from allowing their view on a work to be made public.

The principal scholar whom Mr Bell says he would have invited to look at the Painting would have been Keith Christiansen at the Metropolitan Museum of Art, New York. Other potential consultees proposed by Sotheby’s were Professor Sebastian Schütze (Professor of Art History at the University of Vienna and author of Caravaggio: the complete works, 2009) or Professor Sybille Ebert-Schifferer (the current Director of the Bibliotheca Hertziana-Max Planck Institute for Art History in Rome and author of Caravaggio: the Artist and his Work, Los Angeles, 2012). Other experts were referred to by Mr Thwaytes in the Amended Particulars of Claim as potential consultees, namely Ulrich Birkmaier, Antonio Paolucci and Daniele Benati (who wrote an article in the Forlì catalogue). Ms Kaminsky’s evidence was that she had not heard of these individuals and hence she would not consider them as likely consultees.

The evidence also established the following facts as regards consultation of outside experts. Sotheby’s tend to rely on the view taken by their own specialists rather than deferring to the outside experts who are consulted. If Sotheby’s are confident that a painting is right then they will catalogue the painting accordingly though they will refer to contrary views expressed by others. Similarly if they are convinced that a painting is not right they will not catalogue it more optimistically unless the positive views they receive cause them to change their minds. Mr Bell was also clear that Sotheby’s will take however long is necessary to build academic support for a painting if they think it is right. He referred to a painting by Vermeer that they were convinced was right but it took 11 years for them to gain sufficient support to move the cataloguing from ‘attributed to Vermeer’ to ‘by Vermeer’.

It was put to Mr Bell that he would have gone to Sir Denis and Professor Gregori first for their views because they had a reputation for taking an ‘expansionist’ view of Caravaggio’s oeuvre (that is of being more willing than some other scholars to accept that proposed works are by Caravaggio). It was suggested that once one reputable scholar has supported the attribution, it might be easier to get other scholars on board and that was a good reason to seek the views of an expansionist scholar first. Mr Bell denied this. He said rather that one would go to the scholar with the best reputation in the scholarly community. If that person backed the picture then others were more likely to support it also. I accept Mr Bell’s evidence on this point. There is no point in an auction house gathering scholarly views unless those views are likely to convince potential buyers to bid a large amount of money for the painting.

As to whether he would have consulted Sir Denis, Mr Bell’s evidence was very firm that he would not. As I have said, Mr Bell – and all the other witnesses in the case – expressed the highest regard and respect for Sir Denis’s lifelong devotion to studying and promoting the arts. But Mr Bell said that in 2006 Sir Denis was already 96 years old and in his opinion and in the opinion of many in the art world, Sir Denis’s ‘eye’ was no longer reliable so far as attribution of Caravaggio was concerned. Mr Bell referred in particular to what had happened in 1998 with a painting called Saint John at the Well. The painting came to Sotheby’s with the potential to be a late Caravaggio. The question was whether it was a hitherto ‘lost’ full-length picture of which there were copies around but also a possible autograph smaller painting just of the head and shoulders of the figure. Sotheby’s sent transparencies of the painting to Professor Gregori because she had published an article stating that the smaller head and shoulders painting was an autograph work. Photographs were also shown to Sir Denis. Both Professor Gregori and Sir Denis were emphatic in their view that the painting was not by Caravaggio. Other scholars also expressed the same view. The painting was then cleaned and sold to a third party as ‘circle of Caravaggio’. Subsequently Professor Gregori and Sir Denis saw the painting in its cleaned state and changed their minds. They both stated emphatically that they now did believe that the painting was the lost work by Caravaggio. There is a contemporary file note for Sotheby’s prepared by Mr Bell recording this incident, from which his irritation at the turn of events is clear. He notes that Sir Denis did not seem to recall that he had previously given a negative opinion or to know that Professor Gregori had also previously given a detailed negative assessment of the painting. As I understand it, the painting of St John at the Well has not been sold since so it is not known whether anyone would be prepared to pay for it the price that a Caravaggio would command on the strength of Sir Denis’ and Professor Gregori’s changed view. Mr Bell’s evidence was that the attribution to Caravaggio is not widely accepted by scholars, though it appears it may be supported by Nicola Spinosa as well as by Professor Gregori and Sir Denis. Mr Bell was asked why this incident caused him to change his mind about the reliability of Sir Denis. It was put to him in particular that scholars may well change their view of a painting once it has been cleaned because there is more information available. The following exchange took place:

“Q. Perhaps we can get onto that in relation to this painting, but you say you thought after this that Sir Denis was unreliable, I think those are the words you used. Can you just share with the court what particularly did you think was unreliable about his process of assessment?

A. There was a general view in the art world, I would go as far as ‐‐

Q. I am talking about you personally, Mr Bell?

A.

My experience with this is that his dismissal [of St John at the Well] had been so categorical at the outset, and his opinion had changed, and he didn't really seem to have an awfully good recollection of what was a very clear initial analysis of the photographs of the painting. I suppose my opinion was formed against a backdrop of the general art historical community's feeling that Sir Denis' views were becoming less reliable generally. In a way, if I could give you an analogy, he was almost like a great wine expert who still knew, you know, where every single vineyard was, he still knew when the best vintages were, but his ability to distinguish one glass of wine from another had been severely impaired. So he had all the information, but his critical judgment to distinguish one work of art from another, one picture being original, one picture being a copy, had, in my view, and the view of everyone else I was aware of, become severely impaired.”

Ms Treves also said in her oral evidence that by 2006 she believed from her own experience of showing Sir Denis pictures that his ability to distinguish one hand from another and his attributional abilities had become impaired by then.

However, there were several examples put to the Sotheby’s witnesses where they did consult Sir Denis Mahon about possible attributions of Baroque paintings after the St John at the Well incident and at around the time they were considering the Painting. There are a number of catalogue entries which express Sotheby’s’ gratitude to Sir Denis for expressing his support for an attribution based on his examination of the painting; Ms Treves herself wrote to Sir Denis in April 2003 asking for his advice on an ‘enormous painting’ that he had seen at their premises. In relation to each of these the Sotheby’s witnesses put forward a particular reason why Sir Denis had been consulted over that particular painting either because it was by an artist (Guercino, Salvator Rosa, Pietro Testa or Giovanni Sirani) for whom Sir Denis was the undoubted repository of expertise or because of some earlier connection he had with the work or the artist. They denied that these instances showed that they were still relying on Sir Denis’ expertise. Part way through the trial some additional documents were disclosed from the Sotheby’s New York office in relation to the sale in New York in January 2005 of a copy of Caravaggio’s work The Supper at Emmaus. The painting was part of the Berkman estate and had been bought as an autograph Caravaggio in 1958. Sotheby’s were sure it was in fact a copy and not autograph. George Gordon seems from the contemporaneous documents to have advised that both Sir Denis and Professor Gregori should be consulted as to the attribution of the painting, in part at least “just to cover our backs”.

On this point, I find on the balance of probabilities that Sotheby’s would have consulted Sir Denis Mahon if they had considered that the Painting had Caravaggio potential. Not only was he available in London to look at the Painting first hand but he also had a strong connection with this work because of his publication of both the Kimbell Cardsharps and of the Musicians from the same collection. That is just the kind of connection which in other instances caused Sotheby’s to consult him. Although Sir Denis had caused an upset in 1998 with his apparent volte face about St John at the Well it is clear from later catalogue entries and correspondence that Sotheby’s continued to consult him, whatever doubts they had about his reliability. They realised that potential buyers of Italian Baroque paintings would be interested to know what Sir Denis thought about a particular work. This extended beyond lesser known artists where there was a smaller pool of expertise available to Caravaggio where, although there was a much larger pool of experts available, Sir Denis still enjoyed a considerable reputation in London because of his lifelong scholarly dedication to the artist.

It is likely, since he would have given his honest opinion, that Sir Denis would have given the same positive opinion of the Painting if consulted then as he did once he had bought it. I therefore find that Sotheby’s would have had at least one positive attribution for the Painting if they had consulted outside scholars.

I find that the position of Professor Gregori is different and I am not satisfied that Sotheby’s would have consulted her if they had spotted Caravaggio potential in the Painting. Although there is evidence of Sotheby’s asking for her opinion on works after the St John at the Well incident and she had some connection with the discovery of the Kimbell Cardsharps in Zurich, there is also plenty of evidence that the Sotheby’s experts did not place much weight on her expertises for Caravaggio’s works. For example, she had supported a painting called The Sacrifice of Isaac which had been bought by a collector as a Caravaggio. The painting had been extensively exhibited as being by Caravaggio during the 1990s. Mr Bell’s evidence was that the art world realised that the painting was in fact by Bartolomeo Cavarozzi and that although it was a beautiful work it was worth a fraction of what it would have been worth if it had been by Caravaggio and hence a fraction of what the collector had paid for it. The collector had been unreceptive to challenges to attribution of the work and it was only after her death that the painting was sold by Sotheby’s with the correct attribution to Cavarozzi in July 2014. There were other examples given of Sotheby’s rejecting attributions supported by Professor Gregori.

My further finding is that if Sotheby’s had received a positive opinion from Sir Denis they would have sought to garner support from other experts on Caravaggio but they would have been disappointed. I am satisfied that they would also have consulted Dr Christiansen and that he would have given a firm contrary view that the Painting was a copy and not a very good quality copy. In the email introduced into evidence by Ms Post’s witness statement, Dr Christiansen said this:

“Clarissa,

I was sorry to hear about the ongoing discussion regarding the version of the Cardsharps owned by Denis Mahon. I saw this picture in Forli, where it was exhibited for a time at the Cagnacci exhibition. As much as I admire the scholarship and connoisseurship of Sir Denis and his enormous contribution to Caravaggio studies, I very regretfully cannot agree with his idea that this is a work by the artist. It seemed to me an obvious later copy -- and not of particularly outstanding quality (to be truthful). Currently, a number of scholars have embraced the view that Caravaggio made "trial versions" for his paintings as well as replicas. So far as I am aware, there is no documentary evidence for this and no reason to believe it part of his practice. I have yet to be shown a single case that convinced me .. On the other hand, there is abundant evidence that Caravaggio's paintings were copied - and copied very well - at an early date and that these copies were later inventoried as originals, which creates a sometimes baffling situation for the sorting out process. Personally, I believe that the over-riding criterion must be quality, and I just don't find the requisite quality in the work in question.

I've discussed the picture with a number of colleagues and have yet to find one who goes along with Sir Denis.

With all best wishes,

Keith”

I reject the suggestion that Dr Christiansen’s view as expressed in his email to Ms Post was distorted because he only saw the Painting in its restored state. Ms Glanville was taken in cross-examination in detail to the passages of the Painting that had been restored by R M S Shepherd (as manifested under ultra violet light). I accept that the restoration may have hardened some of the edges of the figures to make them appear flatter. But the passages of the Painting that are relied on by Sotheby’s as showing the poor quality of the Painting (the feather, dice holder etc) have not been greatly restored.

Similarly if Sotheby’s had gone further in seeking views of other experts, I find that they would have received many more negative views than positive. I reject the suggestion that the negative views expressed by various Caravaggio scholars for the purposes of these proceedings were the result of some arm twisting by Professor Spear. From what I have seen in this case of art historianship, the scholars do not hesitate to disagree with each other in forthright terms without generating any apparent ill feeling. I regard the opinions given by the experts who have expressed a view on the Painting as their genuinely held views based on the application of their skilled connoisseurship to consideration of the Painting. I find that the counterfactual world is therefore one where Sotheby’s:

I also find that this state of affairs, given the evidence of the Sotheby’s’ witnesses, would not have been enough for Sotheby’s to be prepared to catalogue the Painting as being “by Caravaggio” or even as being “attributed to Caravaggio”. They would still have proposed to Mr Thwaytes that the Painting be auctioned as by a Follower of Caravaggio, albeit that the catalogue entry may have mentioned the positive view expressed by Sir Denis. Even if Professor Gregori’s endorsement had been acquired somehow by Mr Thwaytes I find that would not have persuaded either Sotheby’s or any of the other experts to change their negative view of the Painting. I also consider that if Sir Denis had endorsed the Painting, Professor Gregori’s additional endorsement would not have materially affected the outcome that I now describe as likely.

Mr Thwaytes’s pleaded case is that if he had been in that counterfactual world, the Painting would have been sold either by auction at Sotheby’s or by private treaty with the benefit of a description that reflected the scope of the academic support that existed for the Painting. He then pleads that the quantum of his loss is the difference between the value of the Painting being sold at auction or by private treaty with that description and the amount that the Painting in fact realised at auction. The question now is how to arrive at the value of the Painting with the endorsement of Sir Denis (and perhaps Professor Gregori). Any attempt at arriving at a valuation is necessarily speculative. But I bear in mind the comment in McGregor on Damages (19^th ed.) at paragraph 10-002 that the fact that an assessment is difficult because of the nature of the damage is no reason for awarding no damages or merely nominal damages. My attention was also drawn to the statement of Blackburne J in Zabihi v Janzemini [2009] EWCA Civ 851 that the court must do its best on such evidence as it feels able to accept to place some kind of value on the loss suffered. The present case differs from the Luxmoore-May case because no one has tried to sell the Painting since Sir Denis’ death so it is not known how the art world would react to the views that have been expressed. Even in Luxmoore-May the court was careful to make clear that the first instance judge’s apparent assumption that the quantum was the difference in the sale achieved by the claimant and the £88,000 ultimately received at auction may have been too simplistic.

Mr Sainty’s evidence gave a number of examples of paintings that were of doubtful status but all sold with catalogue entries which set out the conflicting views of scholars. Most if not all of these instances were where a painting had been sold as ‘attributed to’ the artist rather than as by a follower of the artist. He said that the prices achieved by those paintings reflected their controversial status because they would have been sold for substantially more if their attribution had been more generally accepted. He gave as an example a painting called Temptation of Saint Anthony (or Saint Anthony Tormented by Demons) which was sold with one expert attributing it to Michelangelo for under £1 million and then resold after further research had strengthened the work’s claim to be by the hand of Michelangelo for more than $6 million. Although almost all acknowledged Michelangelo scholars have rejected the attribution, Mr Sainty says that this shows that ‘even with a work that could only very tentatively be identified with a great artist, it is possible to catalogue it in such a way as to build excitement around the work to achieve a good result’. His evidence is that the Kimbell Cardsharps is worth about £55 million and that although there would have to be a substantial discount for the negative views, he considers that £11 million is a realistic estimate of what the Painting would have fetched if it had been sold with the opinion of all those scholars who currently accept the work fully detailed in a well-prepared catalogue.

Ms Kaminsky’s evidence was that sometimes even the attribution by a well respected scholar can be rebuffed by the market. She gives the example of the offer for sale by Sotheby’s in January 2013 in New York of a Still Life with Fruit on a Stone Ledge as by ‘Roman School, circa 1605-1610’ with an estimate of $2-3 million. The painting belonged to a notable Caravaggio scholar who had acquired it ten years earlier when it was sold as by ‘Follower of Caravaggio’. He had then built a case for its authenticity and published it as by Caravaggio on a number of occasions. The work had also been exhibited in several prominent museums as being by Caravaggio. However it went unsold at auction because the market was not convinced. She also thought that the Painting might well have fetched less than £42,000 if it had been sold on the basis that only Sir Denis Mahon and Professor Gregori supported the Painting and other Caravaggio scholars were firmly of the view that it was no more than a reasonable contemporary copy. The ‘sense of mystery’ would have gone out of the work and any purchaser interested in acquiring the Painting on a speculative hunch that it might be autograph would realise that they would have an uphill struggle in trying to turn around the strong negative views expressed.

We cannot know how much more than £42,000 Sir Denis would have been prepared to bid for the Painting on 5 December 2006 if the Consortium had stayed the course; to achieve £11 million at auction one needs at least two people prepared to bid up to those heights, not just one. Although Mr Sainty suggested that an art dealer may have access to clients who do not bid at auction, there was no convincing evidence before me that the Painting would have sold for more by private treaty than it could fetch at auction. The parties agreed that quantum should not be based on the possibility that Mr Thwaytes would have consigned the Painting to an unscrupulous dealer who would suppress negative scholarly views and sell the Painting for a vast sum to a wealthy but naïve art collector.

On balance my conclusion is that the Painting probably would have made slightly more at auction or by private treaty if it had been sold with a catalogue entry detailing the positive and negative attributions of respectable scholars but not a great deal more. I find that the views of those scholars who have expressed a negative view of the Painting would have carried much more weight in 2006 than the views of Sir Denis Mahon and Professor Gregori.

Given my findings on liability, I will not speculate further about a precise amount. I note for completeness that the question whether the quantum of Mr Thwaytes’s loss would in fact be the difference between what the Painting would have fetched at auction in the counterfactual world that I have posited and the price he in fact received is very far from clear. Sotheby’s argued that the Painting is not by Caravaggio and that the duty of Sotheby’s to Mr Thwaytes does not extend to a duty to obtain an inflated price for the Painting on the basis of scholarly opinions that are in fact wrong. As Mr Onslow QC put it, Sotheby’s would dispute the contention that it is responsible for Mr Thwaytes not benefiting from the currency of a wrong opinion about the Painting. I will leave that interesting legal point to be decided in a case where it properly arises for determination.

In the light of my conclusions I dismiss Mr Thwaytes’ claim.

Finally I wish to record my gratitude for the exemplary way in which this fascinating case was presented at trial by Mr Legge QC and Mr Bruce for Mr Thwaytes and Mr Onslow QC and Mr Edwards for Sotheby’s, and for the huge amount of work put into the preparation of the case by the legal teams and by the expert witnesses.