Qatar Investment & Projects Development Holding Co & Anor v John Eskenazi Limited & Anor

INDEX

This claim concerns seven objects (“the objects”) which were sold during 2014 and 2015 by the First Defendant (“JEL”), a specialist London antiquities dealer. The Second Defendant (“Mr Eskenazi”) is a director of JEL. It was common ground that JEL, including through Mr Eskenazi, is one of the most highly respected dealers and experts in the fields of Indian, Gandharan, Himalayan and Southeast Asian works of art in the world.

The purchaser was the First Claimant (“QIPCO”), a company incorporated under the laws of Qatar. The Second Claimant (“Sheikh Hamad” or “the Sheikh”) is the Chief Executive Officer of QIPCO and a senior member of the Qatari royal family. There was at one stage an issue as to whether QIPCO or Sheikh Hamad was the purchaser. The sale discussions were indeed between Sheikh Hamad and Mr Eskenazi, but the invoicing evidenced sales from JEL to QIPCO and, at the end of the trial, it was no longer in dispute that QIPCO was the purchaser.

QIPCO claims US$4,990,000, being the purchase price of the objects, and its case is that each of them is a modern forgery, not an ancient object. In consequence, it alleges various causes of action which can be summarised as follows. In respect of all objects, they advance claims for breach of contract, misrepresentation and in the tort of negligence. The contractual claim is advanced on the basis that the authenticity of the objects was contractually promised by JEL. However, they also advance alternative claims on the basis that JEL did not have reasonable grounds for believing that the objects were authentic. In respect of one object, QIPCO goes further and alleges fraud on the part of JEL and Mr Eskenazi personally. This claim concerns the most expensive object, the Hari Hara. QIPCO contends that JEL and Mr Eskenazi knew that this statue was not authentic, or sold it not caring whether it was authentic or not.

The Hari Hara statue is different to the other 6 objects in a number of respects. It comes from South East Asia, and it represents a combination of Hindu deities. The other objects came, or purported to come, from a different part of Asia, broadly speaking from an area which encompassed Afghanistan and North Western Pakistan. The invoices described various of these objects as coming from Gandhara or Greater Gandhara or Bactria: these areas are illustrated on a map in Section F below.

The relationship between Mr Eskenazi and Sheikh Hamad began in early 2014 when Sheikh Hamad visited JEL’s gallery in Maida Vale, London, with his cousin, Sheikh Saoud bin Muhammed Al Thani (“Sheikh Saoud”), who was a keen collector and a long-established client of JEL. Sheikh Hamad himself is (and was) a prolific and renowned collector of antiquities and works of art (his/QIPCO’s collection is known as the Al Thani collection), and his acquisitions have been the subject of numerous museum exhibitions, including at the Victoria & Albert Museum, London, and the Metropolitan Museum of Art, New York. Sheikh Hamad’s London residence, Dudley House, is a showcase for his many acquisitions. The late Queen Elizabeth II is reported to have visited and commented that it made Buckingham Palace look rather dull.

Following that first introduction, Sheikh Hamad was thereafter a reasonably regular visitor to the gallery and to the Eskenazis’ nearby house; and in turn Sheikh Hamad invited Mr Eskenazi and his wife to a number of events held by the Al Thani collection and to Dudley House.

Between April 2014 and October 2015, QIPCO purchased the objects from JEL for a total of US$4,990,000. Each of the objects was purchased following oral discussion and agreement between Sheikh Hamad and Mr Eskenazi. No written contracts were concluded. However, each sale was evidenced by an invoice from JEL to QIPCO. In their opening submissions, the Claimants said (in the context of the then dispute as to the identity of the buyer) that the best evidence of the contracts were the invoices.

Each invoice contained the following words following a brief description of the relevant object:

“I declare that to the best of my knowledge and belief the item detailed on this invoice is antique and therefore over one hundred years of age.”

The objects were as follows, dealing with the chronological sequence of sales.

Head of a Goddess. This was described by JEL in the invoice as originating from the Greater Gandhara region in the late Bactrian period 2^nd/3^rd century CE. It was described as marble with garnet inlay. The sale was concluded in a telephone conversation between Sheikh Hamad and Mr Eskenazi in late March 2014. The date of the invoice is 3 April 2014. The price was US$400,000.

Head of Dionysus. This was described as originating from Bactria in circa 2^nd century CE. Its material was marble with gemstone inlays (garnets, dyed quartz). The sale was concluded in a conversation between Sheikh Hamad and Mr Eskenazi at the gallery, in November 2014. The date of the invoice is 25 November 2014. The price was US$1,275,000, which made it the second most expensive piece.

Serpent Bracelet. This was described by JEL as originating from Afghanistan in circa 1^st century BCE to 1^st century CE. Its material was gold with turquoise and garnet inlay. The sale was concluded in a conversation between Sheikh Hamad and Mr Eskenazi at the gallery, again in November 2014. It was invoiced together with the Head of Dionysus on 25 November 2014, at a price of US$125,000. In their closing submissions, the Defendants accepted that this piece was inauthentic.

Hari Hara. This statue was described in the invoice as originating from the Kingdom of Zhenla – present day Vietnam – in the late 7^th century CE. Its material was sandstone. The sale was concluded in or following a conversation between Sheikh Hamad and Mr Eskenazi at Dudley House on around 23 March 2015 and in any event at latest by 23 April 2015. The date of the invoice was 29 April 2015. The price was US$2,200,000. This is the most expensive piece.

Head of a Bodhisattva. This was described in the invoice as originating from Gandhara in the 4^th century CE. Its material was schist (which is a type of stone, with some similarity to slate). The sale was concluded in a conversation between Sheikh Hamad and Mr Eskenazi at the gallery on around 13 October 2015. The date of invoice was 20 October 2015. The price was US$730,000, making it the third most expensive item.

The Visit of Mayadevi to the Emaciated Bodhisattva (“the Frieze”). This was described in the invoice as originating from Gandhara, Swabi in the early 3^rd century CE. The material was, again, schist. The sale was also concluded in the conversation between Sheikh Hamad and Mr Eskenazi at the gallery on around 13 October 2015. It was invoiced in the same invoice as the Head of a Bodhisattva. The price was US$120,000.

Head of a Krodha Vighnakarta (“the Krodha”). This was described in the invoice as originating from Greater Gandhara in the 5^th/6^th century CE. Its material was unfired clay with traces of coloured pigments. The sale was concluded in the same conversation as the previous two items, and invoiced in the same invoice. The price was US$140,000.

The evidence at trial was given in the course of 8 days in July 2022. The parties then prepared and exchanged written submissions, and these were followed by oral closing arguments in October 2022.

The principal focus of the evidence was on the question of whether the pieces were, as the Claimants contended, inauthentic and the related but separate question of whether Mr Eskenazi had reasonable grounds to consider them authentic. These issues gave rise to the exploration of the circumstances in which each of the pieces had been acquired by Mr Eskenazi on behalf of JEL. In the case of some objects, such as the Hari Hara where there was an allegation of fraud, this involved extensive cross-examination. The circumstances of acquisition, and provenance of an object, are of potential relevance to questions of authenticity.

In addition to this factual evidence, there was substantial expert evidence in the disciplines of art history and materials science. The nature and significance of this expert evidence is described in Section D below.

The evidence also covered the discussions and dealings between Sheikh Hamad and Mr Eskenazi. There were a variety of factual disputes between them as to what had happened, and what Sheikh Hamad knew or had been told. However, I did not think that the factual disputes were of any great significance to the central issues which require resolution.

There was in fact a good deal of common ground between the parties, as recorded in the “List of Common Ground and Issues” which the parties prepared. The following was therefore common ground or not in dispute. There had indeed been sales of the relevant objects as evidenced by the invoices. Prior to the sale, Sheikh Hamad had viewed and discussed the objects with Mr Eskenazi, and Mr Eskenazi had identified the age and origin of the objects as later set out in the invoices and as summarised above. There was no suggestion that Mr Eskenazi had expressed any doubts as to authenticity to Sheikh Hamad. On the contrary, Mr Eskenazi willingly accepted in his evidence that he had been positive in saying that the pieces were authentic.

It was also common ground that it was an implied term of the contracts of sale between QIPCO and JEL (as evidenced by the invoices) that JEL honestly and reasonably held the opinion that the objects were of ancient origin. In addition, it was agreed that JEL through Mr Eskenazi represented that it held the honest and reasonable belief that the objects were of the age and origin so identified.

It was also common ground that, subject to a qualification concerning paragraph 19 of the List of Issues, Sheikh Hamad relied upon Mr Eskenazi’s opinion that the objects were of ancient origin as being honestly and reasonably held. I do not, however, need to analyse the nature of that qualification. This is because I have no doubt that Sheikh Hamad did rely upon Mr Eskenazi’s opinion that the objects were of ancient origin as being honestly and reasonably held. Sheikh Hamad had no expertise himself in antiquities of the relevant period or geographical area. Prior to the death of his cousin, Sheikh Saoud, he had the benefit of the latter’s views. However, it was Mr Eskenazi who was, and who carried on business on the basis that he was, knowledgeable about the objects that he was selling. I have no doubt that, at the time, Sheikh Hamad trusted Mr Eskenazi and the information that he was being given.

In the following section, where I deal with the chronology in more detail, I therefore deal relatively briefly with the evidence as to the dealings between Sheikh Hamad and Mr Eskenazi. I deal in more detail with the circumstances in which Mr Eskenazi for JEL came to acquire the objects, and I return to the evidence on that issue in later sections.

In this section, I describe the facts concerning the sales to QIPCO, the circumstances in which the objects had been acquired by JEL prior to sale, and the events leading to the dispute. The oral evidence on these topics was given, principally, by Sheikh Hamad and Mr Eskenazi. I shall start by setting out my general views on these witnesses.

I considered that Sheikh Hamad, by and large, gave his evidence fairly and to the best of his recollection. Some parts of his statement had, however, clearly been over-lawyered: at one stage in cross examination he had to ask Mr Green KC what was meant by the term “unequivocal assurance”, which was an expression used in his witness statement. He also gave some unimpressive answers when asked about WhatsApp exchanges concerning the decision not to give Mr Eskenazi certain reports which had been prepared by Professor Ahuja.

That said, I think that Sheikh Hamad did give a generally reliable account of the flavour of his meetings with Mr Eskenazi, even though (unsurprisingly) it was apparent that he did not remember the detail. In a number of passages in his evidence, he gave answers to the effect that he believed that Mr Eskenazi was giving an honest and reasonable opinion. At one stage, he put the matter into his own words, when it was put to him that he knew that no antiquities dealer could or would ever give an absolute assurance of certainty:

“Absolute assurance, there is not – it is not a science – what can I say? It is not a science – it is not a science by itself, but it’s to the best of their knowledge and experience over a wide – long time of years, and knowing the provenance, knowing where did they buy it from, how much they paid for it, who handled, they can understand much better than.”

I thought that this was an honest and indeed sensible answer, albeit that it was obviously not helpful to the case that QIPCO received a contractual assurance of authenticity.

Having heard Sheikh Hamad’s evidence as a whole, I was left in no doubt that, during the period when the objects were being purchased, Mr Eskenazi was always enthusiastic about the pieces, and clearly conveyed the impression that they were authentic. Sheikh Hamad said in evidence that: “my trust with him was beyond doubt”. I fully accept that evidence.

Mr Eskenazi gave evidence for just over 2 days. He was a witness who did try to answer, directly, all of the questions which he was asked. Many of his answers were very frank and honest. It was put to him that he was always very emphatic about the genuineness and authenticity of the pieces that he sold. He gave a straight “yes” to that question. I considered, however, that I needed to treat his evidence with a degree of caution, unless it was supported by contemporary documents or the inherent probabilities. There were a number of related reasons for this.

First, the cross-examination revealed a number of areas where Mr Eskenazi had not conducted his business with integrity. As will appear from the more detailed discussion in this judgment, there were a number of important documents which were backdated or which otherwise did not accurately record the true position. Examples are as follows.

A provenance letter provided by Mr Mark Hubbard, in respect of the Head of Goddess, was backdated. The letter bears a date of 15 September 2014, but was clearly not created on that date.

When a question arose concerning the authenticity of the Head of a Bodhisattva, Mr Eskenazi wrote to Sheikh Hamad on 12 November 2015 stating that, as part of his due diligence into the work, he had commissioned a scientific test. This was not true. There was a scientific test which had been commissioned by the person who had sold the piece to JEL, but Mr Eskenazi had not commissioned it himself nor sent a sample for analysis.

On 8 December 2015, Mr Eskenazi asked his restorer, Neil Perry-Smith, for a report on the Head of Goddess, and told him broadly what he wanted him to deal with. This was, as Mr Eskenazi said in evidence, for the purpose of providing evidence to the Sheikh that he had had the object examined. The report was, however, backdated to 9 November 2015.

When there was pre-action correspondence about the Hari Hara, the Claimants were provided with a provenance letter signed by Mr Paul Jewell. This letter (dated 6 February 2013) stated that Mr Jewell had purchased the piece in Singapore about 22 years previously, and that since 1998 it had been in his house in London. Whilst this letter does not appear to have been backdated, it was false: the piece had been purchased by Mr Jewell at the behest of Mr Eskenazi (and acting on his, or his company’s behalf) in February 2013. As will appear hereafter, the Hari Hara was, if genuine, a quite exceptional piece: Mr Eskenazi described it as one of the most extraordinary Southeast Asian sculptures that he had ever had and was “by far the best of the different pieces from Vietnam I have ever seen or handled”. Albeit that the pre-action correspondence was some years later, I find it very difficult to believe that Mr Eskenazi had forgotten the circumstances of the acquisition (i.e. that the piece had been purchased on his behalf by Mr Jewell in Vietnam a few years earlier) and forgotten that it had not been in Mr Jewell’s house in Muswell Hill since 1998.

Secondly, serious question-marks as to Mr Eskenazi’s integrity were raised by other documents that came into existence in connection with the Hari Hara. There was considerable documentary evidence and cross-examination which concerned the amounts paid for the Hari Hara in reality, in contrast with the figures utilised for the purposes of JEL’s accounts. As discussed in Section L below, it is difficult to be precise as to exactly what was paid originally, for the Hari Hara, since the documentary evidence on that issue is incomplete. Various documents were, however, later produced in order to reflect sales involving Mrs Eskenazi and a company called Sunny Country Ltd (“SCL”). The story is described in more detail below and in Section L. The Claimants submitted, in their closing submissions, that the true position was that there had been no genuine sale by Mrs Eskenazi to SCL, nor any genuine sale by SCL to JEL. Rather, documents dated February 2015 were fabricated to create the impression that JEL had paid SCL USD 1.8 million for the Hari Hara, thus making a profit of only USD 400,000 which was included in JEL’s accounts. The evidence before me strongly suggests that there indeed was false accounting for tax purposes; i.e. to reduce JEL’s tax liability which would otherwise have needed to reflect the lion’s share of the receipt of USD 2.2 million from QIPCO for the Hari Hara. Mr Eskenazi was one of the directors of JEL, and I do not accept that the accounting was unknown to him because it was dealt with by his wife and an accountant.

Thirdly, in relation to the acquisition of the Hari Hara, the Defendants’ solicitors refused to provide disclosure of the documents of Mr Jewell, on the basis that there was no agency relationship between him and the Defendants. It is in my view difficult to see how the Defendants could legitimately have taken that position. The contemporaneous documents, described in more detail in Section L, show Mr Jewell in effect taking Mr Eskenazi’s instructions in connection with a possible bid for the Hari Hara. Furthermore, Mr Eskenazi in his oral evidence accepted, at least at one stage, that there was an agency relationship.

Fourth, as discussed in section B2 below, I cannot accept the evidence of Mr Eskenazi concerning the discussions with Sheikh Hamad relating to the Serpent Bracelet.

Fifth, it was clear from Mr Eskenazi’s evidence as a whole that his recollection of events, which of course had taken place, some years ago, was not at all good.

The only other factual witness who gave evidence at trial Dr Amin Jaffer, who is now the senior curator of the Al Thani collection. There was very little in his statement upon which he was challenged in his brief cross-examination, or indeed that was relied upon by either party in their closings.

Against this background, I turn first to the facts concerning the sales that took place, and my fact findings in relation to any relevant disputed issues concerning the sale process.

There was relatively little in dispute in relation to the meetings or discussions between Sheikh Hamad and Mr Eskenazi which led to the various sales. The main disagreement concerned the circumstances of the sale of the Serpent Bracelet. Despite the over-lawyering of his witness statement, I considered that Sheikh Hamad’s evidence in relation to the meetings, including that concerning the sale of the Serpent Bracelet, was broadly consistent with the probabilities and contemporaneous documents. The following fact findings are therefore based principally upon his evidence.

Sheikh Hamad was first introduced to Mr. Eskenazi in 2014 by his late cousin, Sheikh Saoud. At that time, as Mr Eskenazi would have appreciated, Sheikh Hamad was a novice in the fields pertaining to the objects which were later acquired. There were no objects of Gandharan origin in QIPCO’s collection, and Sheikh Hamad had no specialist knowledge or expertise in Gandharan or Bactrian art. Indeed, at a later stage, Mr Eskenazi gave him some books on the subject. Sheikh Hamad liked the aesthetics of this kind of art, with (as he described it) a resemblance to “Greek art but with an Asian twist”. Sheikh Hamad became interested in expanding QIPCO’s collection accordingly.

Sheikh Hamad visited Mr Eskenazi’s gallery for the first time on 16 January 2014, with Sheikh Saoud, to see his latest exhibition. There were a number of sculptures and art works on display. Upon the first meeting, Mr Eskenazi told him that he was one of the world’s biggest experts in Gandharan art, and he showed him books and images of pieces that he had sold to important museums and collectors around the world. He also showed his extensive collection of Gandharan and Bactrian books in his library and generally praised the pieces in his gallery.

On this first visit, Mr Eskenazi was keen to show the marble Head of Goddess with garnet inlays. He described it as an extraordinary piece from the Greater Gandhara region, dating to the 2^nd to 3^rd century CE. He said that this was one of the most exceptional and unique Gandharan pieces that he had seen in such good condition, and he explained the historical context of the work within the Gandharan sculptural tradition. Sheikh Hamad told Mr Eskenazi that he did not know much about Gandharan art in general, and asked him to explain the field to him. Mr Eskenazi explained the context of the piece. The details of the discussion do not matter, but it is likely that the points made were to some extent included in the research note on the piece that JEL subsequently sent to the Sheikh.

Sheikh Saoud recommended that Sheikh Hamad should acquire the Head of Goddess, so as not to lose it. Sheikh Hamad was cross-examined on the basis, in substance, that the Head of Goddess and later acquisitions were based solely on the recommendations of Sheikh Saoud. This is not the case. Sheikh Hamad was no doubt encouraged by Sheikh Saoud’s recommendation that he buy the Head of Goddess. But this recommendation had followed Mr Eskenazi praising the piece, explaining its historical context, and describing its importance, and this – coming from a person who presented himself as, and indeed was, an expert in the field – was obviously very influential in Sheikh Hamad’s decision.

I was not persuaded that Sheikh Hamad had directly asked Mr Eskenazi whether the piece was authentic. This is because Sheikh Hamad was dealing with a well-established and reputable dealer, and because the whole context of Mr Eskenazi’s explanation of and praise for the piece was that they were dealing with a piece that was, or at least was firmly believed by Mr Eskenazi to be, a genuine antiquity. I doubt whether Sheikh Hamad would have directly asked the question. However, Mr Eskenazi would have left Sheikh Hamad in no doubt that the piece was very firmly understood by Mr Eskenazi to be a genuine antiquity: as already mentioned, Mr Eskenazi accepted that he was always very emphatic about the genuineness and authenticity of the pieces that he sold.

I accept that, as Sheikh Hamad said, at no point during the discussion did Mr Eskenazi express any doubt or reservation as to the authenticity of the object. I do not, however, accept that this meant that, as Sheikh Hamad said, “Mr Eskenazi was providing an unequivocal assurance that the work was as he described it”. Sheikh Hamad’s evidence in cross-examination revealed that he did not understand the concept of an “unequivocal assurance”. Sheikh Hamad did not ask for an unequivocal assurance, either at the meeting or subsequently in writing. He also understood, in the passage quoted above, that – in relation to the authenticity of antiquities – “it is not a science”. He went on to say that a mistake could be made in any field. Accordingly, I do not consider that Mr Eskenazi was giving, or that Sheikh Hamad understood him to be giving, a categorised or unequivocal assurance of authenticity. What he was giving was a very firm and unequivocal opinion as to the genuineness of the piece. I discuss in Section C below the legal effect of this conclusion.

Sheikh Hamad’s evidence was that he asked about the provenance of the piece. I think that this is likely to have happened: it would have been a natural question for Sheikh Hamad, who was very knowledgeable about art generally, to have asked. Mr Eskenazi explained that the piece had been in a European collection for a significant time, but that he could not give him details of its previous ownership for reasons of discretion.

Subsequent to the meeting, on 29 January 2014, Mr Eskenazi sent to Mr John Baker a research note that JEL had prepared. This research note, as with similar research notes later sent, was prepared by Jane Thurston-Hoskins, who worked as a researcher for JEL. She had a BA degree in art and archaeology, and had worked for JEL for around 20 years, having previously worked for Spink. Mr Baker was Sheikh Hamad’s private secretary at the time. Sheikh Hamad said, and I accept, that the research note very much reflected what Mr Eskenazi had discussed at the gallery. It described the piece as Gandharan and dating from the 2^nd/ 3^rd century, and it contained no hint of uncertainty about that attribution to that time nor any doubt or reservation about it. Sheikh Hamad said, and again I accept, that he trusted Mr Eskenazi and relied on his description of the work, and that he had no reason to doubt it.

On around 3 April 2014, Sheikh Hamad confirmed to Mr Eskenazi that he wished to acquire the piece. The invoice was to be made out in the name of QIPCO, and there is no longer any dispute that QIPCO was the purchaser. On the following day, Sheikh Hamad received the invoice dated 3 April 2014. Mr Eskenazi’s covering letter described the piece as an extraordinary work, and he offered to bring it to Dudley House personally.

Sheikh Hamad’s evidence was that if he had known that the piece might not properly be attributed as a Gandharan work of the 2^nd/3^rd century, or that it (or any of the pieces that he bought on behalf of QIPCO) may not be authentic, he would not have purchased it on behalf of QIPCO. I accept that evidence both in relation to that piece, and the pieces subsequently purchased.

In around October 2014, Sheikh Saoud told Sheikh Hamad about a newly acquired head sculpture, which Sheikh Saoud considered amazing. They visited the gallery in late October, and saw the Head of Dionysus prominently displayed. Mr Eskenazi praised the work as an absolute masterpiece in Gandharan art, and hence the initial asking price of over USD 2 million for the object. Mr Eskenazi also explained the geographical, cultural and art historical context of the piece. Sheikh Hamad asked about provenance, and he received a similar answer to that given in relation to the Head of Goddess: it had been in a private collection, but the need for discretion prevented Mr Eskenazi giving any details.

Sheikh Hamad’s evidence as to how the Serpent Bracelet came be purchased was as follows. Whilst he was in the main exhibition area of the JEL’s gallery, he also noticed the gold Serpent Bracelet piece. He said that Mr Eskenazi told him that the piece was of Bactrian origin, and was dated between approximately 1^st century BCE and 1^st century CE. This was how the piece was later described on the invoice. Contrary to Mr Eskenazi’s evidence, Sheikh Hamad said that Mr Eskenazi did not express any doubt or reservation as to its authenticity or attribution. Mr Eskenazi did not say that he had not yet submitted the bracelet for testing or for expert analysis, or make any statement that he lacked expertise in the field of Bactrian jewellery. Sheikh Hamad was interested in an important piece of antique jewellery, since QIPCO had a collection of jewellery albeit not Bactrian pieces. Sheikh Hamad told Mr Eskenazi that he was interested in acquiring the Serpent Bracelet.

Subsequently, he received a research paper on the Head of Dionysus, but not on the Serpent Bracelet. He then visited the gallery again on 22 November 2014, in order to see both pieces again. He was very sad at this time, because Sheikh Saoud had died on 9 November 2014. However, Sheikh Hamad recalled how much Sheikh Saoud had loved the Head of Dionysus. At the gallery, he discussed with Mr Eskenazi his growing passion for collecting important works from the ancient world through QIPCO, and his desire for QIPCO to loan important works to pre-eminent cultural institutions around the world, and in particular the Metropolitan Museum in New York. During that visit, Mr Eskenazi described the Head of Dionysus as a very important work, and one which demonstrated the position of the cult of Dionysus in the region at around the 2^nd century CE. He claimed it was a masterpiece, a world treasure, and one that would be perfectly suited for exhibition in major museums around the world. Sheikh Hamad also looked again at the Serpent Bracelet briefly, and agreed the purchase price for both pieces. Shortly after the visit in November 2014, QIPCO was invoiced for the two pieces, reflecting the sale that had been agreed during that visit. In December 2014, Mr Eskenazi sent Sheikh Hamad provenance documentation and an Art Loss Register certificate in relation to the Head of Dionysus. The provenance document comprised a letter from a Mr Mark Hubbard, and this is discussed in a later section. The Art Loss Register certificate stated that a search had been carried out of the database of the Art Loss Register and that the piece had not been registered as stolen or missing.

The only significant dispute about the circumstances relating to these acquisitions – indeed the only significant factual dispute relating to the various meetings between the two principals – concerned the Serpent Bracelet. Mr Eskenazi gave an account in which he sought to distance himself from any suggestion that he made clear statements as to the origin of the piece, and indeed disputed that the Serpent Bracelet was on display at all. His evidence was that he was not an expert in such pieces (“it wasn’t my field”), and that he explained this clearly to Sheikh Hamad. The piece was not in the gallery, but was in his office and Sheikh Hamad happened to see it. In fact, he happened to see it when Mr Eskenazi was not even at the gallery. He said that when the possible acquisition was discussed, he explained to the Sheikh that he had not yet had it tested or examined, and that he would give a refund if the Sheikh had the piece tested by Dr Ogden and it turned out not to be genuine. He was therefore, as he said in cross-examination, “ready to accept a scientific test”.

I consider that Sheikh Hamad’s account of the background to the sale of the Serpent Bracelet was accurate, and I have no hesitation in preferring it to the account of Mr Eskenazi. There is nothing in the documentary evidence which suggests that this sale was made on any different basis to any other sale. The invoice was in similar format to the other invoices. The invoice contained no qualification to the description of the object, and neither the invoice nor any other document suggests that the sale was in effect conditional on the result of later scientific testing.

I also consider it far more probable that the Serpent Bracelet would actually have been on display in Mr Eskenazi’s gallery, rather than that it was in his office. Mr Eskenazi’s evidence was that it was on his desk because he wanted to try to look at it repeatedly in order to try to understand it more. This was in my view implausible. If Mr Eskenazi had no expertise in this area, then it is difficult to see that there was anything to be gained by Mr Eskenazi giving it further study in his office. If he wanted to find out more about the piece, and in particular its authenticity, he could have obtained a view from an expert. There was plenty of time for Mr Eskenazi to have obtained a scientific report on the piece: it had been acquired in February 2014 and was sold to the Sheikh in November 2014. Mr Eskenazi said that he was busy with other things, and that the possibility of scientific testing had slipped his mind.

In my view, Mr Eskenazi’s evidence was an improbable account of what happened in relation to the Serpent Bracelet, and it was aimed at minimising possible criticism of his approach in selling this object. Mr Eskenazi’s acknowledgment that he lacked expertise in this area gives rise to the question of how he could properly make representations as to the age of the piece. This is not the only difficulty. A gold object such as the Serpent Bracelet can be subjected to a test for cadmium, and this test is more or less conclusive (as discussed in Section D below) as to whether it is an ancient object. Here, the cadmium test eventually carried out by the Sheikh indicated that the bracelet was not ancient, giving rise to the question: why (particularly bearing in mind Mr Eskenazi’s lack of expertise) Mr Eskenazi had not had the object tested beforehand? Mr Eskenazi’s response to these difficulties was that he did not express a clear view as to the antiquity of the piece (“I never expressed any doubt on anything, except for the gold bracelet”), that he told the Sheikh that he was not an expert, and offered a refund if scientific tests proved inauthenticity. In my view, this account, designed to put Mr Eskenazi in the best possible light, is unreliable. I also observe that even though the Defendants now accept that the Serpent Bracelet is inauthentic, and despite Mr Eskenazi’s evidence that a refund would be given if the piece was shown to be inauthentic, QIPCO’s entitlement to a refund is nevertheless (as I understand it) still disputed and no refund has been made.

The next piece acquired was the Hari Hara. Sheikh Hamad saw this for the first time on a visit to Mr Eskenazi’s home, where it had a prominent position. He recognised it as a sculpture that Sheikh Saoud had previously admired. Mr Eskenazi described it as a magnificent piece of top museum quality, which was why he kept it in his home. It was, he said, from late 7^th century Cambodia, and he explained in detail the cultural and art historical context of the work. He said that whilst the piece was held in his family’s personal art collection, he would be willing to sell it and would later confirm the price. Sheikh Hamad asked for a further discussion at a follow-up meeting.

There was then a further discussion, briefly, in March at Dudley House. Mr Eskenazi again praised the piece as a magnificent piece. They agreed on a price of USD 2.2 million, which was negotiated down from USD 2.4 million. At some stage in the process, Sheikh Hamad received JEL’s research note on the object.

In the course of the summer, Sheikh Hamad was told that Mr Eskenazi had acquired a major Gandharan masterpiece. On 1 October 2015 Sheikh Hamad attended the gallery with another dealer, Mr Noriyoshi Horiuchi. He saw the schist Head of a Bodhisattva and discussed it with Mr Eskenazi and Ms Thurston-Hoskins (JEL’s researcher). Mr Eskenazi showed him the schist Head of a Bodhisattva and described it as dating from the late Gandharan period, 4^th to 5^th century. Mr Horiuchi looked closely at the piece and questioned Mr Eskenazi as to whether it was an authentic piece, in response to which Mr Eskenazi and Ms Thurston-Hoskins spent a considerable amount of time explaining why they had no doubt that the piece was authentic, and that Mr Eskenazi was entirely confident that the work was a very important Gandharan work. During the meeting, Ms Thurston-Hoskins went to another room in the gallery in order to bring books and papers, to which Mr Eskenazi then referred. At this meeting, or at a follow up meeting, Mr Eskenazi said that he had shown the work to Carlos Picon of the Metropolitan Museum New York. Mr Picon had congratulated him and praised the work as a fantastic Gandharan statue. Mr Eskenazi said that this was one of the finest heads he had handled in his many years as a specialist dealer. There was, he said, a lot of interest in the object, which was prominently displayed across all of JEL’s promotional material for London’s Asian Art fair.

On 13 October 2015 Sheikh Hamad, accompanied by Mr Baker, met Mr Eskenazi again at the gallery. Mr Eskenazi continued to praise the Head of a Bodhisattva very highly and again confirmed its authenticity. He also mentioned that a number of museum curators had seen the piece and praised it. Sheikh Hamad’s focus was on the Head of a Bodhisattva, but there was also discussion of the two other pieces: the Frieze and the Krodha. Sheikh Hamad said that he would like to purchase all three pieces, which would compliment the other pieces that he had recently acquired.

In the following weeks, Mr Eskenazi sent invoices for the three objects to Mr Baker, and confirmed that his detailed research papers would shortly follow.

Mr Eskenazi and Sheikh Hamad dispute what, if anything, was said in relation to provenance when agreeing the sale of these objects. Mr Eskenazi says that Sheikh Hamad did not ask him about their provenance and indicated that he was not concerned with it. By contrast, Sheikh Hamad says that provenance was important to him. The documentary evidential position is as follows.

Head of Goddess. No provenance documentation was provided at the time of sale (March/April 2014). Sheikh Hamad first requested such documentation in March 2016, around two years after the sale. He then repeated this request (notwithstanding that it had been provided) a few weeks later on the basis that an American museum which was interested in having it on loan required this information.

Head of Dionysus. As discussed above, a statement by Mr Hubbard was provided to Sheikh Hamad on 19 December 2014. It appears that this was because, at the time, Sheikh Hamad wanted to display the Head of Dionysus in a museum, ideally the Metropolitan Museum of Art, New York. No objections were raised as to the documentation provided. Sheikh Hamad then again requested provenance documentation in March 2016, around eighteen months after the sale, for the American museum which was interested in having it on loan.

Serpent Bracelet. No provenance documentation was provided at the time of sale (November 2014). Sheikh Hamad did not request provenance documentation.

Hari Hara: No provenance documentation (for this most expensive of the objects) was provided at the time of sale (April 2015). Sheikh Hamad did not request provenance documentation. Subsequently, in the context of pre-action correspondence, Sheikh Hamad was provided with Mr Jewell’s letter dated 6 February 2013.

Frieze, Head of a Bodhisattva, Krodha. No provenance documentation was provided for these objects at the time of sale (October 2015). Sheikh Hamad first requested provenance documentation in March 2016, around six months after the sale again in the context of a possible loan to the American museum.

This summary of the documentary position indicates that, at the time of the sales, Sheikh Hamad did not ask for or generally receive provenance documentation. In my view, the likely factual position is that (as Sheikh Hamad’s evidence indicates) there was a brief discussion about the provenance of the first two marble pieces that were purchased, and JEL provided documentation in relation to the Head of Dionysus. Thereafter, I am not satisfied that there was any specific discussion of the provenance of each piece. This reflects the fact that Sheikh Hamad, as matters developed, trusted Mr Eskenazi, probably thinking that if he needed provenance documentation (for example if it was required by a museum) he could ask for it.

The first suggestion made by Sheikh Hamad to Mr Eskenazi that any of the objects was not authentic was in early November 2015, when he telephoned Mr Eskenazi and suggested that the Head of a Bodhisattva was not authentic. Mr Eskenazi addressed those concerns formally in a letter of 12 November 2015 including referring to scientific reports from 2014 by an Italian conservation science laboratory, Adamantio, which endorsed its authenticity.

In early 2016, Sheikh Hamad asked his consultant Dr Jaffer to identify a scholar who could write a series of reports about all the objects except the Serpent Bracelet. Dr Jaffer suggested Professor Naman Ahuja, an expert in the field of Gandharan and Bactrian antiquities. Around this time, Sheikh Hamad also requested provenance documentation for five of the objects, which Mr Eskenazi duly provided to the extent that there was such information. Dr Jaffer considered that the provenance information he received was in line with what he would expect, at least in the context of a sale to a private (rather than a museum) purchaser. However, discussion with the museums potentially interested in showing the objects came to nothing.

Professor Ahuja provided his reports on what he described as “the six wonderful pieces in the collection of Sheikh Hamad” in December 2016. He suggested that two of the objects, the Head of Dionysus and the Krodha, had:

“tremendous potential for further study…I would love to find out what adhesive has been used to affix the jewelled eyes and gemstones in the jewellery, and if there has been any conservation to secure them? Have the pieces ever been X-rayed?”

I discuss Professor Ahuja’s involvement and reports in greater detail in Section L below.

Dr Jaffer followed up with Professor Ahuja about his suggestions for x-ray. The name that was put forward was AnalyzeArt, a company owned by Dr Anna Bennett of Conservation and Technical Services (CTS). At Dr Bennett’s suggestion, her initial examination of the objects (all but the Serpent Bracelet) was purely visual under the microscope. Her first set of reports (the “First CTS Reports”) were provided on 11 August 2017; and, on the basis of only a visual inspection, she concluded that each object was a forgery.

Sheikh Hamad also commissioned another company, CIRAM (run by a Dr Olivier Bobin) to test the objects. The CIRAM Reports were received piecemeal throughout October 2017, with the final reports, on the Head of Goddess and the Krodha, received in December 2017.

On 22 November 2017, Mr Eskenazi and Sheikh Hamad met at Dudley House. Sheikh Hamad told Mr Eskenazi the objects were “problematic” and said that he wanted JEL to buy them back, but Mr Eskenazi required proof of inauthenticity. Sheikh Hamad then asked Dr Jaffer to gather together the reports. His exchange with Dr Jaffer was as follows:

“Sheikh Hamad [SH]: I had a long session with Johnny ESKENAZI He is asking to read all the papers that relates to the scientific tests on the pieces. Plus the scholars who has said is wrong etc can you please gather all reports for me and send them to me please. Urgent.

Amin Jaffer [AJ]: I will have this prepared and sent to him urgently.

SH: NO DON’T SEND TO HIM SEND TO ME AND I WILL FORWARD IT AND I DON’T WANT HIM TO SEE YOU ENGAGED IN IT. WHO HAS WRITTEN ABOUT THE PIECES?

AJ: I will have the files sent to You directly. Naman Ahuja wrote about the pieces as an art historian. Anna Bennett and Ciram have tested them.

SH: Ahuja what did he say? I think he was saying good No?

AJ: Ahuja was impressed by the Pieces. However, he raised questions about the excellent condition of the pieces, for example the survival of garnets in tact [sic] on the female head. He felt that such a survival was remarkable

SH: So his report we can’t face Johnny with. Don’t you think?

AJ: Yes I agree. The main reports that contest the authenticity are those of Anna Bennett and Ciram. These indicate the use of modern materials and tools in the production of the works.”

On 20 December 2017 JEL was given the first CTS Reports and the CIRAM Reports, except that the CIRAM Report on the Krodha (which was carried out by Dr Bobin and his colleague Dr Armel Bouvier) was not provided until early February 2018, and the CIRAM report on the Serpent Bracelet was not provided to JEL until the letter of claim was served on it in May 2019. Professor Ahuja’s reports were not provided to JEL, at that time, and were not provided until disclosure in the action.

By early 2018, Sheikh Hamad was demanding that JEL refund the price and take back the objects. On 10 January 2018, Mr Baker emailed Mr Eskenazi stating: “HH asked me if I’ve received a refund from you? I know the peace’s [sic] are ready for collection once moneys are received. Can you let me know the date I can be expecting refund?”. Mr Eskenazi asked to be put in contact with Sheikh Hamad’s legal advisers to “proceed in a rational and legal framework” to “see if there is a way to address His Highness’s concerns”.

On 27 February 2018, Mr Eskenazi wrote to Sheikh Hamad asking to meet to discuss the “current issues together privately”, and to present JEL’s comments on the reports, including those of a conservation scientist, Dr Wim Lustenhouwer whom JEL had asked to consider them.

It is not necessary to describe the subsequent course of events. It is sufficient to say that the parties were unable to reach agreement on the dispute which had arisen, and in due course the present litigation was commenced. This has been a lengthy process, partly as a result of an adjournment of the trial originally scheduled for 2021. However, I need not lengthen this judgment by describing the course of the litigation.

I have hitherto discussed the dealings between Sheikh Hamad and Mr Eskenazi. An important part of the factual background, relevant to the issues relating to authenticity and whether there were reasonable grounds for JEL to describe the pieces as antiquities, concerns the circumstances in which each piece was acquired. This was explored in some detail in the evidence at trial, in particular in relation to the Hari Hara, where fraud is alleged. I consider the detail of that evidence, where relevant, when discussing each piece. This section therefore introduces the factual background concerning each acquisition. I do so in the chronological sequence of acquisition.

This piece was acquired by Mr Eskenazi with the assistance of Mr Paul Jewell, who was in Vietnam in late 2012 and early 2013. Mr Eskenazi agreed in cross-examination that Mr Jewell was acting as his or JEL’s agent in acquiring the piece. An unsatisfactory part of the litigation history is that this proposition was emphatically disputed by JEL’s solicitors in correspondence, and there has never been any disclosure from Mr Jewell.

The negotiations for the acquisition took place between Mr Jewell and an unidentified individual in Vietnam. The evidence suggests that the seller was a dealer. Mr Eskenazi’s evidence at trial was that he was told that the piece had been sold off by a Buddhist monastery in order to raise funds. The monastery was not identified, and Mr Jewell’s dealings were with the dealer not the monastery.

There are a handful of e-mails passing between Mr Eskenazi and Mr Jewell in late 2012/ early 2013, leading to a sale in around February/ March 2013. It is clear that the Hari Hara was at that stage in 5 separate pieces. Mr Eskenazi agreed to buy it sight unseen. There is no evidence, certainly no documentary evidence, that he was sent any photographs of it prior to sale.

The price paid to the dealer for the piece is unclear. There is no genuine invoice issued by the dealer. JEL’s disclosure contains what is accepted to be a fraudulent invoice and packing list in the sum of USD 575 purportedly issued by the Hien Minh gallery, Hanoi. On any view, this was not the price paid for the piece. The Hari Hara, still in pieces, was shipped from Vietnam initially to Hong Kong, and then to England. If JEL’s case as to authenticity is accepted, there can be little doubt that the export would have been prohibited by Vietnamese law: generally speaking, countries such as Vietnam do not permit their genuine antiquities to be exported. Again, if JEL’s case as to authenticity is accepted, the shipment documents were fraudulent in describing the Hari Hara as a garden ornament.

In due course, the Hari Hara arrived in England, where it went directly to a restorer, Neil Perry-Smith, with whom Mr Eskenazi had a business relationship. He carried out some work on the statue, including its reassembly. The Hari Hara then went to JEL’s storage on or around 22 July 2013, and from there to Mr Eskenazi’s home on or around 16 October 2013.

Apart from the uncertainty as to the amount paid to the dealer, it is unclear how much Mr Jewell was paid for his efforts. Mr Eskenazi’s case was that there was a down-payment of USD 85,000, with a further fee to be paid to Mr Jewell once the Hari Hara had been restored and Mr Eskenazi had inspected it. He said that a further substantial fee of USD 200,000 was then made, although there is no contemporaneous evidence which records it.

The disclosed documents contain a series of documents which cannot in my view be taken as authentic, either in terms of evidencing a genuine transaction or as being correctly dated. In summary, these documents purport to show that the Hari Hara was purchased from Mr Jewell by Lapo N.A. Fine Arts Limited (“Lapo”) a Canadian company, in April 2013. Lapo is a company which is, in practical terms, controlled by Mr Eskenazi and/or his wife. However, a purchase from Mr Jewell is not consistent with Mr Eskenazi’s evidence that Mr Jewell was acting as his agent.

The documents then record that the piece was subsequently acquired by Sunny Country Limited (“SCL”), a Hong Kong company, from Mrs Eskenazi. This sale is purportedly evidenced by a sale agreement and an invoice each dated 13 February 2015 for USD 1.5m. SCL then sold it to JEL, purportedly for USD 1.8m. These and related documents, strongly suggest a scheme to understate JEL’s profits and thereby reduce tax that would otherwise be payable. I have already referred to this in the context of my assessment of Mr Eskenazi as a witness, and I return to some of the detail in Section L below.

JEL acquired the Head of Goddess and the Frieze in 2013. These objects were acquired from a Hong Kong company, incorporated in October 2012, Far Eastern Antiquities Ltd (“FEAL”). This was a $ 1 company, without substantial capital. Mr Eskenazi said that the company was the corporate vehicle of Mr Vigorelli, an Italian gallery owner that he had known for many years.

The background to the purchase of the Head of Goddess, according to Mr Eskenazi’s evidence, was that he had first seen this piece some years prior to 2013. It was in the collection of Mark Hubbard. He had known Mr Hubbard for over 35 years, and had acquired several pieces from his collection over the years. Mr Hubbard had inherited the collection from his father, who acquired many pieces on business trips in the Orient from the late 1970s to mid 1990s. In September or October 2013, Mr Hubbard telephoned Mr Eskenazi to offer the Head of Goddess for sale. He was selling the pieces through FEAL, and Mr Eskenazi went to Mr Vigorelli’s gallery in Milan to inspect it. He agreed to buy it following that inspection.

There is no contemporaneous documentary support for this account, except for the invoice dated 18 October 2013 from FEAL to JEL for the sale of the Head of Goddess for USD 75,000. Some years later, in March 2016, Mr Hubbard provided a provenance letter in respect of the Head of Goddess, stating that it had indeed been in his father’s collection since the early 1990s.

As regards Mr Eskenazi’s purchase of the Frieze, Mr Eskenazi’s evidence was that this too was purchased from FEAL in October 2013. He said that Mr Vigorelli brought the piece to JEL’s gallery where Mr Eskenazi inspected it and agreed to buy it. Mr Vigorelli explained that he was selling the piece on behalf of Corinna Mazza. Mr Vigorelli provided him with a declaration made by Mrs Mazza in Milan in November 2012, which confirmed that the piece had been in her and her husband’s collection since the mid-1970s, when they purchased it at an antiques exhibition. Shortly after purchasing it, he said that he showed the Frieze to Mr. Bowles, who informed him that the stand on which the Frieze was mounted was the work of Frank Thomas, a base maker and restorer who worked in London in the 1970s and 1980s.

There is an invoice dated 31 October 2013 from FEAL to JEL for the Frieze. JEL’s stock book confirms that both the Frieze and Head of Goddess were purchased at that time.

JEL acquired the Serpent Bracelet in February 2014 from FEAL. In his witness statement, Mr Eskenazi described Mr Vigorelli as an expert in jewellery, and that he trusted his knowledge of gold in particular. There is an invoice from FEAL to JEL dated 7 February 2014 for US$100,000, and an entry on the master stock book of the same date.

All of these objects were therefore acquired prior to the first sale to QIPCO – the sale of the Head of Goddess in April 2014.

In October 2014, JEL purchased the Head of Dionysus and Krodha at the same time. Mr Eskenazi’s evidence was that he was contacted by Mr Hubbard in around September 2014 to say that he was selling the Head of Dionysus through FEAL. Mr Eskenazi later inspected the work at Mr Vigorelli’s Milan gallery, and agreed to buy it. At the same time, Mr Eskenazi received the written declaration from Mr Hubbard confirming that the piece had become part of his father’s collection during the early 1980s.

Mr Eskanazi’s statement does not explain the background to the purchase of the Krodha from FEAL. He says only that he purchased the Work from FEAL at the same time as the Head of Dionysus. This is confirmed by the invoicing for both objects in October 2014 and the master stock book entries showing the acquisition of the objects. The prices paid for the objects were EUR 110,000 (Dionysus) and EUR 17,000 (Krodha).

In respect of all the objects hitherto described, Mr Eskenazi did not carry out any Art Loss Register search at the time of acquisition. The first such search was carried out on the Head of Dionysus on 3 December 2014 after he had sold it to QIPCO. A search on the Krodha (and the Head of a Bodhisattva) was carried out on 8 April 2016, when he was being pressed for provenance by QIPCO.

This object was acquired from Mr Marco Pilati in August 2015. Mr Eskenazi had known Mr Pilati for 40 years. The sale is confirmed by an invoice for the purchase price of EUR 250,000 and a stock book entry.

Mr Eskenazi also says that he received documents, at that time, from Mr Pilati relating to this object, namely an affidavit from a Dutch collector Elisabeth Coeburgh and a report from Dr Nicola of Adamantio. The affidavit said that it had been in her family’s possession since 1980. Adamantio’s report excluded the presence of modern organic materials in the sample he had been provided with.

Various aspects of the circumstances of the acquisitions were relied upon by the Claimants in the context of their case on authenticity, absence of reasonable grounds for the believing that the objects were genuine antiquities. Generally speaking, however, the Claimants’ case did not involve a significant challenge to Mr Eskenazi’s factual account, as summarised above, of how the objects came to be purchased from the various sellers. However, I did not understand the Claimants necessarily to accept that account. For example, in relation to the marble heads, Mr Stewart KC drew attention to the fact that they were apparently being offered for sale before JEL had bought them from FEAL. He submitted that it was unknown as to what exactly was going on in relation to the objects. However, the main thrust of the Claimants’ case was to submit that the circumstances of acquisition, even on Mr Eskenazi’s case, were such as to negate authenticity or reasonable grounds for believing in authenticity.

For reasons given, I am reluctant to place reliance upon evidence of Mr Eskenazi which is not corroborated by contemporaneous document or the inherent probabilities. I shall proceed on the basis that Mr Eskenazi’s account of the acquisitions, as summarised above, is broadly accurate. When dealing with the various objects, I will explain any relevant areas where I have not accepted his evidence concerning the acquisitions, or where I am not satisfied that I can make a fact finding on the balance of probabilities.

The Claimants advance causes of action for breach of contract, misrepresentation and in tort for breach of a duty of care. There was substantial common ground between the parties, and the principal dispute was as to extent of any contractual promises provided on sale.

Each object was described, as previously set out, as being of ancient origin in the invoices. These descriptions reflected what Sheikh Hamad was told in the discussions prior to sale and in the JEL research papers that he was given. It was common ground that by describing the objects as he did, Mr Eskenazi (for and on behalf of JEL) expressed a belief that the objects were of the age and origin so identified. It was also common ground that JEL through Mr Eskenazi represented that it held the honest and reasonable belief that the objects were of the age and origin so identified.

As previously discussed, it was also common ground (subject to one qualification) that Sheikh Hamad relied upon Mr Eskenazi’s opinion that the objects were of ancient origin as being honestly and reasonably held. My fact finding (see Section B above) is that Sheikh Hamad did indeed rely upon Mr Eskenazi’s opinion that the objects were of ancient origin as described in the invoices and prior discussions.

It follows the claims for misrepresentation will succeed if, as the Claimants contend in relation to all the objects, Mr Eskenazi did not hold the reasonable belief that the objects were indeed ancient. There was no dispute that the various causes of action for misrepresentation, if successful, would lead to rescission of the contracts of sale and the return of the monies paid.

In relation to the Hari Hara, the Claimants went further and alleged that Mr Eskenazi’s representation was false because he did not honestly believe that the Hari Hara was ancient. Accordingly, the Claimants in that respect advance a claim in deceit (i.e. fraudulent misrepresentation). The essential difference between the claim in deceit, and the other claims for misrepresentation, is that the Claimants must prove (for the deceit claim) that Mr Eskenazi knew that his representation, as to ancient origin, was false or had no belief in its truth. Nothing short of proof of fraud will do. In order to prove fraud, it must be shown that a false statement has been made (1) knowingly, (2) without belief in its truth, or (3) recklessly, careless whether it be true of false. It is not necessary that the maker of the statement was ‘dishonest’ as that word is used in the criminal case. What is required is dishonest knowledge, in the sense of an absence of belief in truth. This ingredient of dishonesty (in that sense) must not be watered down into something akin to negligence, however gross: see Vald. Nielsen Holdings and Ors v Baldorino [2019] EWHC 1926 (Comm) at [145] – [148].

The Claimants did not need to allege fraud in order to succeed against JEL. A claim in fraud requires them to cross a higher hurdle than a claim in misrepresentation based upon an allegation that Mr Eskenazi did not reasonably believe that the objects were of ancient origin. Where a party alleges a lack of a reasonable belief in a representation of an opinion expressed, the issue in effect is whether there was negligence in the formation of the opinion: see Avrora v Christie, Mason & Woods Ltd [2012] PNLR 35 at [131]. If the negligence hurdle is met, then the claim will succeed irrespective of the honesty or otherwise of the individual making it. If the case does not cross the negligence hurdle, then it is improbable to suppose that it might nevertheless cross a fraud hurdle.

The claim in fraud, if successful, does however have one important practical consequence: if successful, it would enable a claim to be made against Mr Eskenazi personally. The Defendants did not dispute this consequence, but strenuously disputed the case that there was fraud in relation to the Hari Hara. For their part, the Claimants did not make a claim against Mr Eskenazi personally, except in relation to the Hari Hara.

It was also common ground that it was an implied term of the contracts of sale that JEL honestly and reasonably held the opinion that the objects were of ancient origin. The factual enquiry, as to whether this term was breached, is the same as that involved in deciding whether there was a misrepresentation. For breach of contract, reliance does not have to be proved, but in any event reliance was established on the facts of this case.

A very important issue between the parties was whether or not this implied term was the only relevant contractual term concerning the ancient origin of the objects. The Claimants contended, in substance, that JEL gave a contractual promise that each object was indeed of ancient origin. If correct, this would mean that the question of Mr Eskenazi’s belief (reasonable or otherwise) in their ancient origin would not matter; a claim for breach would succeed if the Claimants could prove that each object was not ancient. The Defendants submitted that the only relevant standard was that of an honest and reasonable opinion or belief. I address this issue in Section C2 below, but in summary I accept the Defendants’ case on this issue.

It was common ground that JEL owed QIPCO a duty of care in tort in respect of the object.

At the start of the trial, there appeared to be an issue as to the scope of the duty, although it was not very easy to identify what the issue was. The Defendants contended that the duty was only as wide as the contractual duty, and that therefore it added nothing to the contractual claim. The Claimants suggested that it was wider. In their written opening, this wider duty was expressed as being “a common law duty to exercise the reasonable skill and care to be expected of one of the world’s leading dealers and experts in Indian, Gandharan, Himalayan and South-East Asian antiquities and works of art in describing or opining on the age and origin of each of the Works”.

By the end of the trial, however, any issue had really evaporated, and I did not understand Mr Green substantially to dispute the Claimants’ formulation. In his closing submissions, Mr Green said that Mr Eskenazi was indeed a world-leading expert in this area, and that therefore he met the criteria described in paragraph [76] of the judgment of Rose J in Thwaytes v Sotheby’s [2015] EWHC 36 (Ch). In that paragraph, Rose J referred to the difference between the standards to be applied to a provincial auction house, and the higher standard of skill and care owed by a leading auction house.

That said, I think that the Defendants were also correct to say that the case based on a duty of care in tort did not add materially to the contractual case. Leaving aside the argument on fraud, each cause of action requires the consideration of reasonableness. In misrepresentation, the question is whether there was a misrepresentation because Mr Eskenazi’s belief in the ancient origin of the goods was not reasonably held. In contract, the question is whether Mr Eskenazi reasonably held the opinion that the objects were of ancient origin. In tort, the question is whether he exercised reasonable care in describing or opining on the age and origin of the objects. In relation to all three causes of action, the question of reasonableness of the opinions expressed must be considered in the context of a dealer which held itself out as one of the world’s leading experts in the relevant fields. In assessing reasonableness in each of these contexts, the essential factual inquiry is the same. Indeed, in their closing submissions, each party treated the factual inquiry into reasonableness as raising the same questions. I shall return to aspects of “reasonableness” later in this section.

The Claimants accepted that they bore the burden of proving a misrepresentation, or breach of contract, or breach of their duty of care. There was at one stage a suggestion that the burden of proof was reversed in relation to the Claimants’ cause of action based upon the Misrepresentation Act 1967, section 2 (1). However, I accept the Defendants’ submission that where the relevant statement is as to the representor’s reasonable grounds of belief, the Claimants must prove the absence of reasonable grounds in order to prove the falsity of the statement made. If the representor does not have reasonable grounds, then falsity is proven and there is no point in further enquiring whether the representor has discharged a burden of proving that it had reasonable grounds for the false statement made. In any event, since I have had ample evidence to make up my mind on the facts, the incidence of the burden of proof does not really matter.

As previously described, an important legal issue is whether there was a contractual promise that each object was in fact genuine, or whether the relevant standard is that of an honest and reasonable opinion/ belief. The Claimants argue that the attribution of the objects as ancient was an express term of the contract, and that JEL is therefore liable for damages for breach. This case was itself advanced on a number of bases: that an unconditional assurance of genuineness was given in the discussions that preceded the relevant sales; that there was an express term to that effect; that there was a breach of the term implied by the Sale of Goods Act 1979 Section 13 (1) that the goods sold would comply with their description, and also of the term implied by Section 14 (2) that they would be of satisfactory quality.

It seemed to me that each of these ways of putting the case ultimately depended upon whether there was a contractual term which went beyond the (agreed) implied term that JEL honestly and reasonably held the opinion that the objects were of ancient origin.

That issue has been considered in a number of cases where a purchaser of a work of art or a classic car has alleged that the relevant sale was a sale by description, and that there is a breach of description if the work of art or classic car is not genuine. The law is fairly and helpfully summarised in Benjamin’s: Sale of Goods 11^th edition paragraph 11-011:

“And sales of works of art may also be held not to be by description where the seller does not guarantee the attribution of the work. In Harlingdon and Leinster Enterprises Ltd v Christopher Hull Fine Art Ltd, for example, a sale of a picture between dealers was held not to be by description where it was clear that the buyer relied on his own judgment and that the seller disclaimed knowledge as to the supposed artist, despite the fact that there were attributions in early negotiations, in an old auction catalogue to which reference had been made, and in an invoice issued after the sale. Similarly, in Drake v Thomas Agnew & Sons Ltd, it was held that an art dealer’s attribution of a painting to van Dyck was an expression of opinion that did not turn the sale into one by description. According to Buckley J., the statutory implied term as to description does not come into effect merely because “some descriptive words were used or written”, but only “if the proper conclusion from all the evidence is that the parties intended the description to be a term of the contract”.”

It is not necessary for me to analyse the judgments of the Court of Appeal in Harlingdon and Leinster Enterprises Ltd v Christopher Hull Fine Art Ltd [1991] 1 QB 564, which was a case involving the attribution of a painting to a particular artist. That task is not altogether easy: there was a dissenting judgment, and the two judges in the majority each gave somewhat different judgments. However, it is a task which was performed by Buckley J in Drake v Thomas Agnew [2002] EWHC 294 (QB), and his approach seems to have met the approval of the Court of Appeal in Brewer v Mann [2012] EWCA Civ 246. Both parties referred to the judgment of Buckley J in their submissions, and the question was not whether I should follow it, but rather how the principles there articulated should be applied on the facts of the present case.

In Drake, the question was whether a painting was by Van Dyck. The judge held that it was not, but nevertheless that there was no breach of contract. In paragraph [24], he referred to the words of Lord Kenyon in Jendwine v Slade (1797) 2 Esp. 571:

“It was impossible to make this the case of a warranty; the pictures were the work of artists some centuries back, and there being no way of tracing the picture itself, it could only be a matter of opinion whether the picture in question was the work of the artist whose name it bore, or not. What then does the catalogue import? That, in the opinion of the seller, the picture is the work of the artist whose name he has affixed to it.”

In paragraph [25], which was referred to by both parties, he said:

“In general mere expressions of opinion or belief are not contractual; without more they do not become terms of any subsequent contract. Clearly, one party may be so confident in his opinion, for example, as to the authenticity or origin of an object or painting that he is prepared to contract on that basis. He may have good commercial reasons for doing so. But in such cases an objective assessment of all the circumstances must point to that conclusion. The conclusion must be that the common intention of the parties was that the content of the opinion or belief was to become a term of the contract. The obvious and sensible way to achieve that result is to say so; but the courts are often called upon to resolve cases in which the parties have not so clearly expressed their intention and although it may be tempting, it is not always just to conclude that they did not have the necessary intent simply because they did not express it.”

In relation to the question whether there was a sale by description, Buckley J said at [26]:

“… a sale cannot be “by description” unless the parties intend the description to be a term of the contract. It is only then that the implied condition that the goods must correspond with the description arises. With respect, it seems to me that some of the dissenting judgments in the cases cited to me give insufficient weight to that basic point and proceed from the premise that because some descriptive words were used or written, the statutory implied term comes into effect. It only does so if the proper conclusion from all the evidence is that the parties intended the description to be a term of the contract. That makes good sense if one bears in mind the serious consequences that flow from a breach of the implied term, when the statute makes it a condition that the goods should correspond with the description”.

Buckley J then went on to consider all the circumstances of the case, and his conclusion was that there was no term of the contract that the relevant painting was by van Dyck and it was not a sale by description.

As the Court of Appeal said in Brewer v Mann [2012] EWCA Civ 246 para [280], Buckley J’s approach adopts the test of Slade LJ in Harlingdon at 583H/584B:

“the fact that a description has been attributed to the goods, either during the course of the negotiations or even in the contract (if written) itself, does not necessarily and by itself render the contract one for “sale by description.” If the court is to hold that a contract is “for the sale of goods by description,” it must be able to impute to the parties (quite apart from section 13(1) of the Sale of Goods Act 1979) a common intention that it shall be a term of the contract that the goods will correspond with the description.”

Brewer v Mann concerned a classic Bentley car, and the issue arose as to whether the goods had been bailed or hired “by description”. The judgment of the court was given by Rix LJ. It is clear from paragraphs [272] to [274] that where a description embodies an opinion, that description does not necessarily become a term of the contract, and also that reliance by the buyer is not in itself sufficient to make it a term. In particular, Rix LJ said:

“[272] In our judgment, if matters are considered purely in terms of reliance, which was the primary way in which Mr Brant put his case under this heading, we do not think that Fortis can avoid the conclusion that this was prima facie a bailment by description, so that, having failed to appeal against the judge’s view of its terms as unfair, Fortis is unable to rely on them as negativing reliance by Mrs Brewer. Indeed, Mr Brant barely pressed that point of reliance on Fortis’s terms. He rather relied on some citation of treatises and jurisprudence. However, subject to one important consideration, we do not consider that such citation assists him.

[273] That one important consideration relates to the alternative way in which Mr Brant’s point can be put, namely that, to the extent that the description of the car can be viewed as a matter of opinion, rather than as intended to become a term of the contract, then the proper inference to draw is that, irrespective of any reliance on the part of Mrs Brewer, there could be no bailment by description.

[274] That point can, we think, be considered in the following way. An opinion, for instance as to the attribution of a painting to a particular artist, may plainly be relied on by a person, particularly a potential buyer, to whom that opinion is expressed, but the essence of an opinion is that, although if given negligently, and a fortiori if given dishonestly, it may give rise to a breach of duty owed to the buyer on which the buyer may sue, nevertheless it would not generally give rise to an expectation that it would become a term of any contract into which the buyer entered in reliance on it. For a discussion of the situation in which, for instance, an auction house may be liable in breach of duty for an opinion as to the description of an antique, see the well-known case of the pair of Louis XV urns, Thompson v. Christie Manson & Woods Ltd [2005] EWCA Civ 555, [2005] PNLR 38. We will revert below to the issue of whether, correspondingly, the description of the car contained in the hire purchase contract is better regarded as an opinion rather than a warranty.”

At [282] – [283], Rix LJ returned to the question of whether a description is better regarded as an expression of opinion recorded in, rather than as a promise made under, the contract. He quoted the “wise words” of Nourse LJ in Harlingdon as to the nature of the art market (at 577/8):

“All this is a matter of common knowledge amongst dealers in the art market, and, I would expect, amongst all but the most inexperienced or naïve of collectors. It means that almost any attribution to a recognised artist, especially of a picture whose provenance is unknown, may be arguable. In sales by auction, where the seller does not know who the buyer will be, the completeness with which the artist’s name is stated in the catalogue, e.g. “Peter Paul Rubens,” “P.P. Rubens” or “Rubens” signifies in a descending scale the degree of confidence with which the attribution is made. Nowadays an auctioneer’s conditions of sale usually, perhaps invariably, so declare and, further, that any description is an opinion only.

…

For my part, being confident that that principle [i.e. that dealers deal with each other on a caveat emptor basis] would receive general acceptance amongst dealers, I would say that the astuteness of lawyers ought to be directed towards facilitating, rather than impeding, the efficient working of the market. The court ought to be exceedingly wary in giving a seller’s attribution any contractual effect. To put it in lawyer’s language, the potential arguability of almost any attribution, being part of the common experience of the contracting parties, is part of the factual background against which the effect, if any, of an attribution must be judged.”

In the end, the Court of Appeal ordered a retrial, and did not decide the question of whether or not the description of the car as a Bentley Speed Six was or was not intended as a term of the contract. It is clear from paragraph [285] that this depended on all the circumstances of the case, including the discussions between buyer and seller and the expert evidence.

The Claimants relied upon various matters in support of its argument that the attributions were express statements of fact and terms of the contract of sale. The attributions and assurances were emphatically delivered, and they were of cardinal importance to QIPCO in its decision to buy the objects. There was only a short interval between the statements being made and the objects being bought. There was a vast disparity of knowledge between Mr Eskenazi and Sheikh Hamad. Reliance was a very significant factor, and the Sheikh relied upon Mr Eskenazi’s knowledge and expertise, and his attributions and assurances, when he made the acquisition decisions. Mr Stewart also referred to the Sheikh’s evidence to the effect that he understood that he was receiving unequivocal assurances of authenticity. He therefore submitted that the court should conclude, objectively, that the attributions were contractual promises that JEL’s attribution was true.

I do not accept this argument, and I consider that the points made by the Defendants were far more persuasive.

In Harlingdon, Nourse LJ said that the court should be exceedingly wary before concluding that a seller’s attribution be given contractual effect. Whilst that statement was made in the context of a case where one dealer had sold to another, and indeed where the seller indicated a lack of knowledge concerning the relevant artist and reliance by the buyer was not established, I think that it remains an appropriate starting point in a case such as the present. This too is the approach of Buckley J in paragraph [25] of Drake: in general, mere expressions of opinion or belief do not become terms of any subsequent contract. More is required if that is to happen.

In the present case, there is nothing which supports an argument that it was the parties’ common intention that JEL’s attribution be given contractual effect. These sales involved objects which were between 1,000 and around 2,000 years old. It would be very surprising for a dealer, particularly an experienced dealer, to give a guarantee of authenticity in respect of objects of that age. If this result had been intended, then I would have expected the parties to have said so expressly. This is not what happened.

I consider that Sheikh Hamad’s evidence recognised the difficulty in a dealer giving an absolute assurance of authenticity of the sort of antiquities which he was purchasing. It was put to him in cross-examination that he appreciated that “the best a dealer could, and would, do was give you an honest and reasonable opinion”. His answer was: “Yes, after doing all their homework and checking the provenance of course”. I consider that this answer did fairly reflect Sheikh Hamad’s understanding at the time. Later on in his evidence, Sheikh Hamad gave his answer (quoted in Section B above) acknowledging that it is “not a science by itself, but it’s to the best of their knowledge and experience”. These answers are reflected in the fact that his lawyers, Pinsent Masons, were heavily involved in drafting detailed sale and purchase agreements in respect of various purchases that were made from other parties. None of those agreements contained a contractual promise of authenticity. This is no doubt because, notwithstanding the potential strength of Sheikh Hamad’s negotiating position, a dealer could not reasonably be asked to give an absolute assurance of authenticity.

In my view, the Defendants’ argument that there was no common intention that the seller’s attribution be given contractual effect is put beyond any doubt by the declaration at the bottom of each invoice. The invoices were, as the Claimants’ submitted in their opening (albeit in a different context), the best evidence of the contract that the parties made. The only reasonable interpretation of those words is that no guarantee of antiquity was being given. Mr Eskenazi was, rather, declaring that the description in the invoice was correct to the best of his knowledge and belief. These words also confirm that, in describing the objects, Mr Eskenazi was indeed giving an opinion, albeit the opinion of a highly experienced person. But the descriptions given remained statements of opinion: they were neither statements of fact nor contractual promises.

I do not accept the Claimants’ argument based upon alleged unequivocal assurances given by Mr Eskenazi to Sheikh Hamad during the course of the discussions. I have no doubt that, as indeed Mr Eskenazi accepted, he was emphatic in expressing his views as to the antiquity of the pieces. The emphatic or unqualified nature of the statement is relevant when it comes to considering whether there were reasonable grounds for the statement made. As discussed below, there must be reasonable grounds for a statement expressed in emphatic or unqualified terms. The emphatic nature of the statements made does not, however, convert a strong statement of opinion into a contractual promise. As Jack J said in Thomson v Christie Manson & Woods [2004] PNLR 42, para [188]: “Where [an opinion] is expressed in absolute terms it remains nonetheless an expression of opinion, which may later be shown to be doubtful or wrong”. Mr Eskenazi was never asked, either orally or in writing, to give a contractual assurance of antiquity, and I do not accept that the strong opinions that he expressed are to be regarded as unequivocal assurances – an expression which Sheikh Hamad used in his witness statement, but did not understand.

It will be apparent from earlier parts of this judgment that I accept that Sheikh Hamad relied upon what he was told. However, as Brewer makes clear, reliance is not determinative on the question which I am considering, namely whether the description (expressed in the opinion) is a term of the contract.

Accordingly, I accept the Defendants’ case that the descriptions given in the invoices, and expressed during the course of prior discussions, did not become terms of the contracts. Accordingly, I reach the same result in the present case as did the Court of Appeal (by a majority) in Harlingdon and Buckley J in Drake. The relevant contractual standard is therefore honest and reasonable belief or opinion.

This conclusion disposes of the Claimants’ arguments based on breach of the terms implied by the Sale of Goods Act 1979. Indeed, I did not understand the Claimants to contend that those claims could succeed unless the court were to conclude that the description in the invoices (and prior discussions) became terms of the contract. If they did not become terms of the contract, and the relevant contractual obligation was honest and reasonable belief, then there was no sale by description. Equally, if the relevant contractual obligation was honest and reasonable belief in authenticity, the Claimants cannot contend that the goods were not of satisfactory quality because they did not meet a higher standard of actually being authentic.

As indicated above, it nevertheless remains relevant that the opinions expressed were unqualified. In Thomson v Christie Manson & Woods, Jack J (at first instance) described at [188] how an opinion may be expressed in absolute terms or be more guarded, using words such as “probably” or “possibly”. In the present case, Mr Eskenazi’s opinions as to the antiquity of the objects were always expressed in emphatic or unqualified terms. Reasonableness must therefore be judged in this context. The question, therefore, applied to the present case is whether it was unreasonable for Mr Eskenazi to have reached his emphatic or unqualified opinion: see Thomson at [191].

In Thomson, the judge held that Christie’s had reached an opinion “without qualification” as to the dating of a pair of vases, which an auctioneer of their standard could reasonably reach. He nevertheless held Christie’s to be in breach of duty on a related point, and this aspect of his decision was reversed by the Court of Appeal.

However, it is clear from the judgment of the Court of Appeal that reasonableness must be assessed in the context of the fact that the expression of opinion was unqualified. Thus, Mr Sumption QC, for Christie’s accepted on appeal [see [35]) that if Christie’s had, or ought to have had, any real doubt as to the authenticity, and in particular the dating of the vases described in their catalogue, it was their general obligation to qualify the catalogue description appropriately. If they did not have such doubt, there was by definition no relevant qualification which they were obliged to express to Ms Thomson. This submission was accepted by the Court of Appeal at [38]:

“In the present case, there either were, or were not, in 1994 facts which should have led Christie’s to be cautious about their dating of the vases. If there were such facts, they should (as they accept) have expressed them in the catalogue but certainly to Ms Thomson. If there were not, there is no basis on which Mr Cooney [of Christie’s] should have qualified his advice”.

The Court of Appeal returned to this point in paragraphs [155] – [159], where a distinction was drawn between real as opposed to fanciful doubts. In short, if there is a real (as opposed to a fanciful doubt) an opinion will need to be qualified by appropriate words which reflect this. I return to this point in Section L below.

I was referred to a number of authorities in which the question has arisen as to whether an “art professional” (to use the expression in Jackson & Powell on Professional Liability 9^th edition, chapter 22) has fallen below relevant standards. This issue has principally arisen in the context of claims against auction houses in relation to works of fine art, in particular where the issue is whether a painting is properly to be attributed to a particular artist. The parties cited Luxmoore-May v Messanger May Baverstock [1990] 1 WLR 1009 (CA); De Balkany v Christie Manson & Woods [1995] Lexis Citation 3745 (Morison J); Drake v Thos Agnew [2002] EWHC 294 (QB) (Buckley J); Thomson v Christie, Manson & Woods Ltd [2004] PNLR 42 (Jack J) reversed in part [2005] EWCA Civ 555 (CA); Coleridge v Sotheby’s [2012] EWHC 370 (Ch) (HHJ Pelling QC); Avrora v Christie, Manson & Woods Ltd [2012] EWHC 2198 (Ch) (Newey J); Thwaytes v Sotheby’s [2015] EWHC 36 (Ch) (Rose J). The authorities are discussed in a valuable article by Isabel Paintin: The Art of Connoisseurship Through Judicial Eyes: The Law of Negligence and Fine Art Attribution Art Antiquity and Law, Volume XX, Issue 2.

None of these cases concern an antique dealer, such as JEL, selling objects on its own behalf and where the allegation is misrepresentation or breach of the implied terms of the contracts of sale with the purchaser. Indeed, one issue raised by the Defendants concerns the absence of expert evidence, called by the Claimants, as to the practice of antique dealers in antiquities. I address that issue in Section L below.

Although the various cases involve auction houses and (generally) attribution of a work to a particular artist, nevertheless they seem to me to provide a useful framework within which to consider the issues which arise in this case. The following points, which are potentially of significance, emerge from the cases.

First, when assessing the reasonableness issues which arise in this case, the question is whether it can be said that no reasonable leading specialist antique dealer would have concluded that these objects were ancient, and expressed an unqualified opinion to that effect: see e.g. Luxmoore-May at 1020D and 1023H; Thwaytes para [98], and Thomson (CA) para [158], where May LJ asked whether the unqualified opinion expressed was “professionally sustainable”. This reflects the fact that there may possibly, in a context such as the present, be a range of views that can reasonably be held.

Secondly, if the dealer had or should have had a real rather than a fanciful doubt, then it would not be appropriate to give an unqualified opinion: Thomson (CA) paras [155] – [158] and [167] (“In short, if Christie’s were, or should have been, less than completely confident about their description of the vases, they should have qualified their catalogue description of them”).

Thirdly, when considering the opinion expressed, it is important to avoid the benefit of hindsight. Compliance with the required standard is therefore to be judged “by reference to the actual circumstances confronting the practitioners at the material time, rather than with the benefit of hindsight”: Luxmoore-May at 1020C; Coleridge at [24].

Fourth, expert evidence in cases such as the present is usually very important. Where the issue involves whether a painting can properly be attributed to an artist, expert evidence of “connoisseurship” will invariably be tendered. This evidence concerns whether a particular work is consistent with an artist’s other work, and it involves consideration of the extent to which characteristics of the painting in issue are to be found in paintings or drawings that are known to be by the artist or, on the other hand, are atypical of the artist: see Avrora at [37]. In that case, Newey J described expert evidence of that kind as being of crucial importance. He referred to a well-known passage in Drake, where Buckley J said that the judge should not presume to have an expert’s “eye”. Buckley J said the following about evidence as to the attribution of a painting to an artist:

“Attribution of an Old Master can depend on various matters, including: provenance, historical research and the experienced eye of an expert, usually a trained art historian. In this case neither provenance nor history gives an answer or even very much help. The knowledge of van Dyck’s studio practice which art historians have acquired is certainly of some assistance, but in the end, both Sir Oliver and Mr. Agnew agreed the matter was to be resolved by ‘eye’. From listening to them both I understood that to mean rather more than just observation. Whilst it is vital to have keen observation it is also necessary to have knowledge of an artist’s methods and style and to be sufficiently familiar with his work to be able to recognise his artistic ‘handwriting’. Even that is not all. It involves also a sensitivity to such concepts as quality, emotion, mood and atmosphere. To an extent ‘eye’ can be developed but, like many other human attributes it is partly born in a man or woman. Were it otherwise there would be many more true experts. This is not a digression. It is rather important to my function in this case. A judge is not bound by expert opinion. A judge may presume to find that an expert’s final opinion is based on illogical or even irrational reasoning and reject it. But a judge should not himself assume an expertise he does not possess. Thus here, if the question had turned on analysis of historical data or inferences to be drawn from surviving documents, I would have been entitled, with such assistance from the experts as I had received, to have drawn my own conclusions; but it does not. It turns on ‘eye’. However I may regard my own taste or appreciation of things artistic, I must not presume to have an expert’s ‘eye’ for a van Dyck...”.

In the present case, I am not in fact concerned with “attribution” in the sense of the objects being attributed to a particular artist. Rather, I am concerned with whether the objects can reasonably be dated to a particular time period, here antiquity. It was in that sense that the parties used the word “attribution” in their submissions. Having heard the art historical expert evidence described in more detail in later sections, I am not persuaded that the assessment of whether a piece is or is not ancient is quite the same as the “eye” being discussed in Drake. The relevant inquiry is not as to the characteristics of a particular artist, but rather as to the features of comparable works which date from the same period. The process of comparing works nevertheless has many similarities, because the expert will need to be familiar with the comparators and their features. However, as will become apparent from the discussion in Section D and the later more detailed discussion, an art historian in the antiquities field is concerned too with other matters; for example archaeology (in particular what pieces are indisputably genuine, because they form part of the accepted archaeological record), and knowledge of relevant religions. Furthermore, whilst some of the authorities concerning attribution of paintings to an artist are to the effect that attribution is largely based on visual inspection, the art historical expert evidence in this case indicates that, in relation to antiquities, more is or may be required certainly in terms of knowledge of the historical or religious background and the archaeological record.

Fifth, it is apparent from Drake, and the article by Paintin, that it is not always easy for a party to be able to obtain the services of a well-qualified expert. They may be few on the ground, and those with relevant knowledge may work for museums which will not permit them to give evidence. In the present case, as will become apparent, the Defendants have not been able to call a reliable art historical expert to give oral evidence. Their expert on 5 of the objects was, as the Defendants accepted in their closing submission, unsatisfactory. Their expert on the Hari Hara was, apparently, too old and/or ill to give oral evidence at trial. It scarcely needs saying that this creates some difficulties for the Defendants, both on the question of whether the objects were authentic and also on the related question of whether their opinions were reasonable.

Sixth, judges in prior cases dealing with auction houses have identified, generally in fairly broad terms, the steps that auction houses should take in order to meet their standard of care. In the most recent case on this topic cited to me, Thwaytes, Rose J considered prior case-law, such as Luxmoore-May which concerned a provincial auction house. She identified a number of steps that a leading auction house should take in order to fulfil their duty in respect of works consigned to it. These were, in summary: (i) ensuring that the assessment of the painting is carried out 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; (ii) devoting sufficient time to a proper examination in order to form a firm view where possible; (iii) appreciating that it cannot necessarily rely on the poor condition of a painting as a reason for failing to spot its potential; and (iv) knowing his or her own limitations, and when to consult an outside expert.

These cases, concerning what should be done by a leading auction house, provide a degree of broad guidance, but they should obviously not be read as laying down a firm set of rules applicable to the circumstances of the present case. It seemed to me, as in different ways both parties submitted, that the question of whether Mr Eskenazi had reasonable grounds for the unqualified opinions which he expressed ultimately depends on all the facts of the present case and the circumstances of each object. The case law does not therefore provide a prescribed set of rules as to how any particular object should be approached.

This can be illustrated by considering the circumstances in which an auction house, or indeed dealer, should consult an outside expert. Rose J in Thwaytes refers to this possibility, but without trying to prescribe when this should be done. In her article, Paintin suggests that a leading international auction house might be expected to “consult external experts only when there were more than minimal doubts or differences of opinion internally as regards the correct attribution, or a significant gap in relevant internal expertise”.

This seems to me to be a sensible approach, but it may be that there are cases where an outside expert should be consulted in any event, depending for example on the ease and expense of doing so, the nature of the outside expert, and the potential value of the piece. For example, one of the objects in the present case is the Serpent Bracelet, and (and discussed below) there exists a relatively simple scientific test that can be performed on the gold, by a specialist, to see whether it contains cadmium. If so, then this would indicate that the item is not ancient. If a dealer is selling a gold item as an antiquity for a substantial sum, it may well be that they should have this test performed in any event. In the present case, again as discussed below, Paintin’s approach and indeed Thwaytes would also result in Mr Eskenazi having to consult an outside expert; since Mr Eskenazi’s evidence indicates that he lacked expertise in jewellery, and that he was uncertain as to whether it was ancient.

In other cases, the dealer’s expertise may be such that it is not reasonable to expect an outside expert to be consulted at the outset. However, if there are real doubts, or question-marks concerning the authenticity of an antiquity, a dealer who then proceeds to give an unqualified opinion as to authenticity may well be acting unreasonably. This is not in itself because, or necessarily because, there has been a failure to consult an outside expert. It is because a reasonable basis for the unqualified opinion is, in view of the real doubts or question-marks, lacking. If, however, an outside expert has been consulted, and has provided a positive opinion or scientific analysis, then this may be of considerable assistance to the dealer or auction house as supporting the opinion expressed.

Accordingly, everything depends on the circumstances, which may vary considerably. The judgment of Jack J in Thomson at [188] – [189] illustrates this. He declined to give a prescriptive answer to the question of “what needs to be done” whilst again providing some useful guidance as to how an auction house should approach objects such as the vases in that case.

“[188] It is the well-established practice of auction houses and the art world generally to date items such as the Houghton urns by visual examination in conjunction with such information as to the item as may be already available or obtainable by such research as is reasonably to be expected of the auction house in the circumstances. The examination must be as thorough as the circumstances reasonably require and it must be done by persons of appropriate qualifications and experience. It will not, cannot, and need not be as exhaustive as the examinations which have been carried out for the purposes of this trial: that is impractical. The outcome will be an expression of opinion as to date, which may be expressed in absolute terms such as here, or in more guarded terms, using words such as “probably” or “possibly”. Where it is expressed in absolute terms it remains nonetheless an expression of opinion, which may later be shown to be doubtful or wrong. That is the basis on which the market proceeds, and that is generally well-understood.

[189] It is difficult to define the degree of certainty which an auctioneer should have before he ascribes a date without qualification, and it is probably unnecessary for me to try to do so. It may be whether, having reviewed all factors, he finds he has a sufficient positive basis for the view he has formed combined with an absence of matters which raise a real rather than fanciful doubt by pointing another way.”

I return to some of these points in Section L below.

Finally, it is certainly logically possible for a court to reach the conclusion that an object is inauthentic, but that nevertheless an opinion that it was authentic had a reasonable basis.

In addition to the admitted representation that JEL had the honest and reasonable belief that the objects were of the stated age and origin, the Claimants submitted that there were two further implied representations, namely:

I am not persuaded that either of these representations adds anything to the representation which is admitted.

The first implied representation that it had reasonable grounds for its description or opinion in my view simply restates, in different words, the representation that has been admitted. JEL admit that it represented that it had an honest and reasonable belief. JEL therefore accepts that its opinion must be based on reasonable grounds – a point that is established by authorities such as Avrora at [131]. However, that paragraph of Avrora also indicates that it is possible to reformulate the admitted representation, so as to say that there was an implied representation by JEL that it believed that facts existed which reasonably justified the opinion expressed: see too SK Shipping Europe Plc v Capital VLCC 3 Corp [2020] EWHC 3448 (Comm) at [113 (iii)]. I cannot, however, see any difference of substance between the admitted representation, and the first implied representation for which the Claimants contend.

The second implied representation introduces unnecessary complexity, and in my view was not made. The question of whether JEL had carried out inspection, research and other work to establish reasonable grounds is simply a facet of the question which admittedly arises, namely did JEL have a reasonable belief? That necessarily involves looking at whether it had carried out sufficient work in order to justify the reasonableness of the opinion expressed. There is no need to say, and indeed no basis for saying, that there was an additional implied representation to that effect.

A central factual issue in the case is whether each of the 7 pieces is an authentic antiquity, or is a modern forgery. In their closing submissions, JEL accepted that the Serpent Bracelet was, on the balance of probabilities, not an authentic antiquity.

In deciding the question of whether each object is a genuine object as described to the Sheikh, the ordinary civil test of “balance of probabilities” applies. This involves looking at the evidence as a whole, and deciding whether the weight of evidence is such that the Claimant has discharged the burden.

The evidence adduced in relation to this issue fell broadly into three categories (factual evidence, art history, materials science) with a degree of overlap between them.

First, there was factual evidence as to the circumstances in which the objects were acquired, their ownership history or “provenance”, and how they had entered the art market. The Claimants submitted that this evidence was relevant to the enquiry into authenticity, and that the expert art historical and materials science evidence needed to be considered in conjunction with it.

For example, the Claimants submitted that a number of the pieces in issue (the Hari Hara, the marble Head of Dionysus and Head of Goddess, and the schist Head of Bodhisattva) were not only expensive, but also – if genuine – very significant and important pieces. It was surprising, therefore, that little or nothing was known about their ownership history and that they had never previously been photographed or discussed. They said, again by way of example, that the circumstances in which the Hari Hara came on to the market were suspicious, giving no confidence that the piece was authentic. It had been acquired from a dealer in Vietnam, with the assistance of Mr Jewell, a friend/client of Mr Eskenazi. The identity of the dealer was unknown to Mr Eskenazi at the time, and nothing is known about him even now. Mr Eskenazi at the time knew little or nothing about how the dealer had himself acquired the piece, or how it had come to light. Mr Eskenazi said in his oral evidence, for the first time, that he had been told by Mr Jewell that it had been owned by a Buddhist monastery, which was selling it in order to raise funds. However, the identity of the Buddhist monastery was unknown to Mr Eskenazi at the time, and still is. There is also (as the Claimants’ art historical expert, Mr Beguin, pointed out), something odd about the idea that the Hari Hara would be owned by a Buddhist monastery: it is a representation of a Hindu, not a Buddhist, deity.

Contrary to Mr Green’s submission, I consider that evidence of this kind is relevant to the issue of whether the pieces are authentic, as well as the related question of whether Mr Eskenazi had reasonable grounds for the unqualified opinion that he expressed that the pieces dated from antiquity. It is part of the overall evidential picture on both issues.

Indeed, there are, unsurprisingly, authorities that recognise the potential importance of provenance in the context of authenticity: see for example Thompson at [111] (Jack J) and Avrora at [38] (quoting Drake) and [65]. I accept, however, that the position on provenance in relation to antiquities must be viewed somewhat differently to, for example, a painting dating from recent centuries. It is possible, even for an antiquity dating back 2,000 years, for its origin and chain of ownership to be known and fully established: there may be a documented record of the piece being found in an archaeological excavation, with its subsequent ownership being known. I accept that the mere fact that that this does not exist is not in itself evidence of inauthenticity, not least because genuinely authentic objects may have been discovered in undocumented or illegal excavations, bearing in mind that (as in the present case) they may come from parts of the world which have been subject to war or the lack of stable government and society. However, this does not in my view mean that evidence as to the more recent history of an object, including its recent ownership and how it came to enter the art market, may not be relevant.

Secondly, there was the evidence from art historians in relation to the objects, in particular as to their style and iconography. The Claimants called four highly qualified and experienced experts on the objects: Mr Beguin in relation to the Hari Hara; Dr Luczanits in relation to the Head of Bodhisattva, the Frieze and the Krodha; Professor Lo Muzio in relation to the two marble pieces, the Head of Dionysus and Head of Goddess; and Dr Jack Ogden in relation to the Serpent Bracelet. In broad terms, these experts were able to give evidence as to the artistic and historic context of the period from which the pieces were alleged to date, and to draw conclusions as to authenticity based upon that context and what is known about objects dating from that time. It is clear from many cases that evidence of this kind is potentially very important.

For their part, and no doubt recognising the importance of such evidence, the Defendants adduced evidence from two witnesses on the issues of art history. They did not call any art history expert on the Serpent Bracelet. Dr Ulf Jager gave written and oral evidence as to all of the other objects, apart from the Hari Hara. The Defendants’ expert on the Hari Hara was Dr Roveda. He had produced a written report, but thereafter his participation in the expert steps ordered by the court was minimal or at least incomplete. He did participate in an expert discussion with Mr Beguin, but did not complete a joint report. He did not give evidence at trial, either in person or by video. The Defendants explained that he was not fit to do so: he is now 93, and had been diagnosed with cancer for whichhe had been treated in 2012. There was some medical evidence, dated 18 March 2022, in which his physician recommended that he should not travel and should avoid stressful conditions.

As would be expected, points made by Dr Jager or passages from his report were put to Dr Luczanits and Professor Lo Muzio in cross-examination. By and large, they disagreed with the points made. Dr Jager then gave oral evidence, and it became clear, fairly rapidly, that he had no real expertise in art history at all, and certainly none that was in any way comparable to or a match for the expertise and experience of Dr Luczanits and Professor Lo Muzio. In their closing submissions, and in the light of Dr Jager’s oral evidence, the Defendants took the realistic if not inevitable position that they could not place any significant reliance on Dr Jager’s evidence, describing that evidence as unsatisfactory.

This meant that the evidence of Dr Luczanits and Professor Lo Muzio was not countered or disputed by any reliable art historical expert called by the Defendants. I accept that this does not necessarily mean that I am bound to accept all of their evidence. The Defendants invited me to look critically at what they had said, and in particular at what they submitted was the very constricted approach to authenticity which they had taken (discussed further below). In his closing submissions, Mr Green made various points in support of the argument that there was no satisfactory art historical evidence on which to base a conclusion of inauthenticity, and that therefore the effective collapse of the Defendants’ expert art historical case on 5 of the objects (i.e. their inability to rely on Dr Jager) did not matter.

I do not accept this argument, which has an air of unreality about it. In a case of this kind, art historical evidence is obviously important, and one would expect a party to call an appropriately qualified expert if the evidence of the other’s sides experts was to be effectively challenged. Mr Stewart referred me to the evidence of the Defendants having spent or budgeted for very substantial sums of money on expert evidence, in support of the proposition that art history was plainly taken very seriously. But in any event, as will be apparent from my discussion of the detail of the pieces in later sections of this judgment, I considered that both Dr Luczanits and Professor Lo Muzio were very impressive witnesses. This was my impression when I listened to their responses to questions asked in cross examination, and it has been confirmed by my re-reading of the transcript of their evidence as well as their written reports. They both reached positions of distinction in their profession, and were highly qualified to give the art historical evidence which they gave: there was no suggestion to the contrary. They gave their evidence well, and clearly with a view to assisting the court. I see no basis on which I should take a different view on art history from the evidence given by these witnesses. I deal in more detail below with the argument that their approach, and that of Mr Beguin, was too constricted.

Notwithstanding the fact that the Defendants have placed no reliance in their closing on Dr Jager’s evidence, I sometimes refer to it in this judgment. This is because, at the time of cross-examination, the basic case with which the Claimants’ experts were dealing was the case advanced by Dr Jager. Their evidence must therefore be seen in that context.

The position in relation to the Hari Hara, and Dr Roveda, is somewhat different to the position in relation to Dr Jager, but it is scarcely more satisfactory from the Defendants’ perspective. Dr Roveda appeared on paper to be an appropriately qualified witness. However, he had not fully participated in the expert process and he did not give evidence before me. I therefore do not know how he would have responded to cross-examination, or indeed to some of the points made by Mr Beguin in his report and oral evidence. The Defendants’ decision to place no reliance on Dr Jager, notwithstanding extensive reference to his evidence in their skeleton argument for trial and cross-examination of the Claimants’ witnesses, shows the potential impact of effective cross-examination.

By contrast, and as discussed in more detail in Section E below, Mr Beguin was not only well-qualified, but he had enjoyed a longer and more distinguished career in the art history world than Dr Roveda. Mr Green submitted in closing that he was not an expert in Khmer art. But this was not a point that was put to him in cross-examination, and I am satisfied that he had appropriate expertise. As with the Claimants other witnesses (including Dr Ogden, whose evidence on the Serpent Bracelet need not be discussed in detail), he gave his evidence well, and with a view to assisting the court. As discussed in Section E below, I have taken into account the points made by Dr Roveda, but I prefer the evidence of Mr Beguin where they differ.

Faced with their inability to rely upon Dr Jager, and the fact that Dr Roveda did not give oral evidence, Mr Green sought in his oral closing to downplay the significance of the art historical evidence. He submitted that the art historical evidence was of less importance than the materials science evidence. The former involved subjective judgments in an area where different people could take different views. By contrast, materials science was more hard-edged. He referred to the Claimants’ pleading, which contained greater and earlier reference to their materials science case than to art history. He also referred extensively to reports of Professor Ahuja which had been provided to the Claimants when they began to look carefully at authenticity.

I do not accept this approach. The art historical evidence in other cases involving art professionals has been similarly subjective, but has nevertheless been treated as very important. Furthermore, it is not right to say that the entirety of the evidence of art history is subjective. Much of the evidence concerns the history of art during the relevant time, and in particular the nature of objects of undoubted authenticity which have been discovered. It also concerns the condition in which such objects have been found. In my view, this type of evidence is not really subjective, but it is potentially important. For example, as discussed below in the context of the two marble heads, the absence of marble and acroliths in the authentic historical record from the material time is significant.

Equally, I do not accept that the materials science evidence is completely hard-edged. Indeed, as the Claimants’ expert Dr Bennett said, there was much agreement between her and Dr Nicola as to what was shown by scientific test results, but there was nevertheless considerable disagreement between them as to how the evidence as a whole was to be interpreted.

Accordingly, I regard the art historical evidence as an important part of the picture. It has a degree of overlap with the materials science evidence. For example, the Claimants’ art historical experts’ views as to inauthenticity relied to some extent on the remarkably pristine condition in which the objects are. As Dr Luczanits said: “considering the condition of the object is a very important part of assessing the authenticity of an object”. The Claimants’ experts were able to say, with considerable force in my view, that these are not comparable to what would be expected if the pieces were genuine, given the condition in which authentic pieces are usually found. This is in my view relevant and indeed important evidence in relation to authenticity. A very similar point, from a materials science perspective, is made by Dr Bennett. Indeed, as discussed below, a repeated point which she made was the absence of the “weathering” which would be expected in ancient pieces. This was not her only point, but it was one on which she placed significant reliance.

Although it is necessary to bear in mind the evidence as a whole, there needs to be a starting point for my consideration of that evidence. I do not think that it matters where one starts, provided that the evidence as a whole is considered. In the detailed discussion of each object, I start with consideration of the art historical evidence. I then consider, where I consider it relevant, the factual evidence relating to provenance and acquisition, which in some respects is an aspect of evidence of art history. I then discuss the materials science evidence. For reasons which will become apparent, my conclusions on art history are relevant to my conclusions on the debate between the materials scientists’ evidence.

As far as Professor Ahuja is concerned, I do not consider that any significant weight can be attached to his reports on the issue of authenticity, still less anything which would outweigh the evidence of the Claimants’ experts to whom I have referred. Professor Ahuja has not been called as a witness by the Defendants. It would have been open to the Defendants to do so, notwithstanding that he had provided a report to the Claimants. There is no “property” in a witness, even an expert witness: see Harmony Shipping Co SA v Saudi Europe Line Ltd [1979] 1 WLR 1380. Moreover, there was evidence that Mr Eskenazi is a friend of Professor Ahuja, or at least that he has had some sort of professional dealings with him: Mr Eskenazi sent him some photographs in around April 2016 (at around the time that the Claimants were themselves in contact with Professor Ahuja), and Mr Eskenazi’s response in 2018 to the reports of CTS and CIRAM included a reference to what Professor Ahuja had told him about the Head of a Bodhisattva. There is nothing in the evidence to suggest that Professor Ahuja could not have been approached by the Defendants with a view to his acting as an expert, and certainly nothing that would justify any adverse inference being drawn against the Claimants for their decision to rely upon the evidence of experts other than Professor Ahuja.

But leaving those points aside, there are well-established procedures for the calling of expert evidence, including the need for reports to be in accordance with CPR Part 35. In the present case, the Defendants decided to call Dr Jager, but have ultimately been unable to place any reliance on his evidence. I have no Part 35 compliant expert evidence from Professor Ahuja. This is not simply a technical point on procedure. It is important, because Professor Ahuja was engaged at a very early stage by the Claimants, and in fact he recommended further testing. I have no idea what his views would now be in the light of the carefully considered views of the Claimants’ distinguished experts, including Dr Bennett, who was engaged after he had recommended her. He has not considered those views, or indeed been able to take into account the wider context (now known) in which the objects were acquired. It is reasonably possible that, having considered those views or the entirety of the evidence now available, he might agree with what the Claimants’ experts have said. Furthermore, even in the brief reports that he gave Dr Jaffer in December 2016, he did not express a clear view that the objects are authentic: see Section L below.

Accordingly, on the issue of authenticity, I attach no significant weight to Professor Ahuja’s views. In Section L, I consider the relevance of his views in the context of the “reasonable grounds” issue.

It is convenient at this stage, before introducing and discussing the evidence as to each object, to discuss a number of features of the evidence, as established by the expert evidence which I accept, and which are significant in the assessment of whether a particular piece has been established (to the requisite standard) to be a fake. This includes the question, described above, as to whether the approach of the Claimants’ art historical experts is too constricted.

First, it is clear that there is a significant industry, in or near the regions relevant to the objects that I am considering, involving the production of fake objects. Indeed, it was common ground (as recorded in the document also setting out the list of issues) that there “was a large and thriving industry producing inauthentic pieces of work which could not properly be compared to works with an established provenance and JEL was aware of forgeries in the market”. The expert evidence in the case provided further detail in relation to this large and thriving industry.

Some of the fake objects generated by this industry have no pretensions to authenticity: they may be sold, for example, as garden ornaments in the equivalent of garden centres. However, many objects are produced with a view to sale on the international art market, and are indeed designed to deceive. The craftsmen involved will often be very skilled. This skill will not only relate to the quality of the carving itself. A skilled forger will not simply copy an authentic piece, but will (as Dr Luczanits explained) copy different elements from different pieces.

The success of skilled forgers in infiltrating the market is illustrated by a well-known Japanese publication on Gandharan art: Kurita: Gandhara Art, Vol 1, the Buddha’s life Story” (1988); Kurita; Gandhara Art, Vol II, the World of Buddha (1990), with new and enlarged editions in 2003. This contains, as all relevant witnesses agreed, a mixture of authentic and fake objects. Many of the latter have found their way into private collections in Japan. The availability of books such as Kurita, with good illustrations of many pieces, itself serves to facilitate forgery; because numerous images are available from which different elements can be copied.

In relation to Gandharan art, Dr Luczanits was of the view that well over 50% of objects on the international market were forgeries, and indeed that no more than 10% of objects were genuine. He said that everybody who works in the field knew that there were thousands of fake objects on the market. He accepted that there was no empirical research data which proved the percentage, but he said that he had observed what was on the market in galleries, and he also followed auction house catalogues. He had seen what was offered in Pakistan or in India in different branches of Gandharan art. In his function as curator of the Rubin Museum of Art, he had visited different galleries in New York. He described demand in Japan resulting in accelerated production of new Gandharan works in Pakistan.

Whilst not offering a percentage, Professor Lo Muzio said (and I accept) that it is undeniable that during the last three or four decades, the international antiquary market has been literally flooded with sculptures, cult objects, jewellery, inscribed artifacts and other objects allegedly coming from Afghanistan or Northern Pakistan, and attributed to Gandhara, or Greater Gandhara or Bactria. For what it is worth, Dr Jager accepted that he was not in a position to disagree with their evidence that the market had been flooded with fakes.

Whilst it is not possible to identify with precision the percentage of Gandharan objects on the international market that are forgeries, I do not think that this matters. There is clearly a very substantial industry and a very significant problem in terms of the number of fakes on the international market.

The position in relation to Khmer art is similar. Mr Beguin said that in the field of Khmer arts, he had seen many forgeries. Forgeries abound, with the industry supported by an exponential market partly destined for the United States. The workshops of forgers and the fraudulent trade of fake Khmer sculptures has become better known as a result of the work of a number of academics, Martin Polkinghorne of the University of Flinders and Jim Sanborn. Their studies have identified and illustrated many fakes. These are manufactured in number, copying all eras and all iconography. The most skilled are copied from famous works and can be found in museums.

The flow of materials described by these witnesses generally has no provenance: in other words, the materials cannot be identified as having been unearthed in, for example, an officially sanctioned archaeological excavation. Such official archaeological excavations have been continuing in Pakistan, which is part of the relevant geographical area for present purposes. However, for much of this period Afghanistan has been war-torn, and there is no or no relevant accepted archaeological record: as Professor Lo Muzio said, the “most difficult situation is in Afghanistan, not in Pakistan”. The flow of materials of undocumented provenance will, as Professor Lo Muzio said, include genuine objects. However, he said that the number of “blatant fakes in private collections, antique shops and even in museums is impressive”. The result has been a “real plague affected Gandharan and Bactrian studies, with disheartening effects on all scholars who take this field of research seriously”.

Secondly, the skill of those producing fakes goes beyond the quality of the work of the craftsman who produces it. As Dr Bennett explained, the industry in the production of fakes will attempt to cover up the fact that it is a recent fake, for example by attempting to remove modern tool marks or by the application of artificial deposits or by various techniques aimed at simulating age. She gave some striking evidence about this, illustrated by photographs in the section of her report on the Hari Hara. She said that the chemical treatment of fake Khmer sandstones is well known in archaeological circles. In 1985-1988 she was able regularly to visit on a weekly basis some of the fake workshops in central Thailand, and allowed to witness their production. She saw that after manufacture, and sometimes after the sculptures had been intentionally broken, they were artificially patinated either by the application of chemicals to the surface or by burying them and applying chemicals to the soil over a prolonged period. The burial ground, which could contain several sculptures, was regularly tended by the application of chemical solutions, much as one would water a garden. Sculptures would remain, as she was told, for a period that varied between months and years. At the end of the patination process, deposits were sometimes induced on the stone surfaces by using old engine oil with iron filings to simulate deposits and discolouration. These fake sculptures were sent all over the world, and could be found in museums and private collections. Some of the items that she had seen being manufactured were offered for sale at auction. Khmer sculptures offered at Sotheby’s in 1989 were modern examples and were later withdrawn from sale.

Third, there does nonetheless exist a significant number of undoubtedly authentic works which can be used for the purposes of comparison with the objects under consideration in the present case. This is because, going back to the 19th century, there have been archaeological excavations whose results have been properly documented and recorded. These will result, for example, in a discovery being unearthed and photographed at the time. The art history experts referred to this as the accepted archaeological record. Professor Lo Muzio described the credited archaeological record of Gandharan sculptures as “very rich”, and that it therefore provided reliable criteria for confirming or disproving the alleged authenticity of artifacts of unknown provenance. Dr Luczanits referred to the existence of a “good body of Gandharan art”, which gives “a very good idea about Gandharan art and how it’s developed over time”.

Fourth, the excavations have, as Dr Luczanits explained, yielded some wonderfully preserved and exceptional works, albeit that such works are the exception (Dr Luczanits described them as the “extreme exception”) rather than the rule. Most excavated objects are severely damaged and often demonstrate clear signs of long exposure to weather or moisture. The art historical evidence was (unsurprisingly in my view) that it was therefore very unlikely for a roughly 1,800 year old object in any material to survive completely unscathed, let alone an object made of more sensitive material such as stucco or clay. Broadly speaking, this was the effect of Dr Bennett’s evidence as well; the absence of “weathering” was, as I have said, her repeated and usually most significant point. Dr Luczanits was cross-examined on the basis that his starting point was that a well-preserved purportedly Gandharan object was overwhelmingly likely to be a forgery. He accepted that this was so; because, as he said, this is “what the archaeological evidence tells us”.

Fifth, it would be wrong to assume that an object is definitely authentic simply because it appears in a museum collection. Some collections, such as the British Museum, are full of authentic objects. However, this is by no means the case with every museum. Indeed, on the evidence of the Claimants’ experts, even the most renowned museums in the world will have acquired and may display fake objects. This does not necessarily mean that the curator had acted with insufficient care in his decision to acquire an object. Some curators may well not be specialists in a particular narrow area, but will be acquiring objects across a broad range of periods and civilisations. Other curators may have taken the view that a particular object, even if an archaeological record was lacking, was sufficiently comparable to a piece of known authenticity that it could be treated as genuine. Other curators may perhaps have acted with less care, or may not have adopted a methodology which involved comparing a potential acquisition to objects in the historical record. Sometimes, a curator may reconsider prior views, with acquisitions being taken off display because of concerns as to authenticity: Dr Luczanits’ evidence was that the curator at the Singapore museum is likely to remove an object acquired from Mr Eskenazi. Ultimately, however, the present case is not concerned with the question of whether any particular piece in a museum is or is not genuine, let alone whether or not a curator in that museum acted with reasonable care in deciding to acquire or display it. I am concerned with the authenticity of the 7 objects in issue, and (as explained below) I consider that there is an appropriate art historical methodology for approaching that issue. This does not involve making a determination as to whether any other object, whether in a museum or a private collection, is or is not authentic.

Against this background, the approach of the Claimants’ art history experts was to base their opinions substantially if not exclusively on the accepted archaeological record. Professor Lo Muzio said that his approach was to discard from the range of reliable comparators all materials of undocumented provenance, and therefore to “restrict the search for relevant comparators in the accepted archaeological record”. He added that given that the number of accepted archaeological pieces was significant, the methodological condition which he set was “by no means limiting”. Mr Beguin took a similar approach. He said that an objective study must be based exclusively on archaeologically referenced works and thus must reject any piece – even a piece of exceptional quality – which was devoid of a scientific pedigree.

Dr Luczanits took a similar, but slightly more nuanced approach. He said in his report that he carefully avoided the use of market or museum objects of doubtful provenance, with the consequence that (in his view) the use of authentic works “reveals the language of Gandharan art”. That methodology involved, as he described it in his report, carrying out a “careful assessment of each object on the basis of a broad knowledge of those objects that were already excavated in the 19th or early 20th century and/or have an excavation provenance”.

It was clear from his oral evidence, however, that whilst his prime focus was on excavated objects, he did not think it appropriate to ignore all other objects. His “position as a scholar is that we have to keep in mind what is the evidence for each object that we use as evidence”. Accordingly, even if an object had not been excavated, his methodology might lead to the conclusion that the object was authentic, in which case it could be taken into account. Each object relied upon therefore needed to be considered on its merits. Thus, in the joint report, both Dr Luczanits and Dr Jager had agreed that “excavated objects and objects that occur in early photographic documentation are important reference objects to evaluate Gandharan art”. Dr Luczanits agreed in the joint report, however, that they are not the only objects that should be considered as reference; but “for him objects without provenance or early documentation are lesser value for evaluation”.

The Defendants argued that the approach of the Claimants’ experts was too narrow, and that in consequence I should be cautious before accepting and indeed should not accept their views. They submitted that the approach resulted in disregarding, for comparison purposes, many pieces which had been treated as authentic by scholars and indeed major museums. Since the burden of proving inauthenticity was on the Claimants, and since the approach of the Claimants’ experts had been unduly restrictive, the conclusions of their experts could not be accepted.

It seems to me that, in circumstances where the evidence indicates that the market has been flooded with inauthentic pieces, it is appropriate, at least primarily, to base conclusions by reference to pieces of undoubted authenticity. Those pieces provide a firm foundation for the inquiry. Otherwise, as Professor Lo Muzio explained, the ever-growing amount of potentially or patently spurious materials will infect the analysis and result in research which “has little or nothing to do with rigorous scholarly investigation”. In a market which is rife with fakes, it does seem to me that the best comparators, and the best source of knowledge of the development and features of the relevant art (whether Gandharan, Bactrian, or Khmer) are the pieces of undoubted authenticity whose provenance is known as a result of the archaeological record. Similarly, in Avrora at [37], Newey J referred to comparisons with paintings or drawings “known to be” by the artist in question. If the objects with which I am concerned are not comparable to pieces of undoubted authenticity, then that is in my view weighty evidence against their authenticity. It would in those circumstances be highly unlikely that the evidence of any other piece, with no archaeological record, would displace that weighty evidence.

Accordingly, I consider that the methodology of the Claimants’ experts is basically sound. Indeed, I remind myself that the Defendants have not called any well-qualified expert to give evidence which suggests that the approach of those experts does not accord with that taken by art historical experts in this field of antiquities.

The only possible qualification to the methodology of Professor Lo Muzio and Mr Beguin is that accepted by Dr Luczanits. As I have said, Dr Luczanits accepted that a piece without provenance or early documentation is “of lesser value for evaluation”. He was willing to accept, therefore, that such pieces could not simply be ignored, provided that they could be established – applying the sound methodology which he proposed, of proper comparison to authentic pieces – that they were themselves authentic.

I do not think that this qualification makes any practical difference, and it certainly does not lead to the conclusion that the evidence of Professor Lo Muzio and Mr Beguin is to be disregarded because of their methodological approach. It does not make any practical difference for at least two reasons.

First, at a theoretical level, it seemed to me that it would rarely, if at all, be necessary or helpful to look for possible comparator pieces outside the archaeological record. If the pieces with which I am concerned are comparable to pieces of known authenticity, then it is not necessary to look elsewhere. In that situation, a comparison with the authentic historical record is sufficient. If, however, the pieces with which I am concerned are not comparable to pieces of known authenticity, then it is difficult to see the value of looking at other pieces which have no provenance within the historical record. To do so would then give rise to the need to compare those “no provenance” pieces with pieces which are within the historical record. There was, however, nothing in the expert evidence which suggested that a piece – and in particular the pieces here – might (i) not be directly comparable to an authentic piece in the record but (ii) might nevertheless be indirectly comparable (i.e. on the basis that it could be compared to a “no provenance” piece, which was itself comparable to an authentic piece from the historical record). It seems to me that this is at best a theoretical possibility.

Secondly, when considering the pieces under consideration in this case in the context of Dr Luczanits’ possible qualification of the views of the other experts: I did not consider that any possible comparator “no provenance” pieces, outside the archaeological record, have been shown to be authentic by reference to that record. In their evidence, the Defendants did not set out to establish this. Indeed, in the case of the Gandharan/ Bactrian objects, the Defendants have not called any reliable art historical evidence at all on the authenticity of the objects in dispute, let alone on other possible comparator objects.

The importance of looking at pieces of known authenticity is reinforced by the evidence that there are many fakes in private collections. Objects which have no relevant archaeological record, and which are simply in private collections, have no or very little evidential weight, given that the existence of inauthentic objects in private collections is so widespread. The point is illustrated by the evidence, not in dispute, concerning the Kurita publication. This contains many photographs of objects from private collections, but the evidence is that many of the objects are indeed fakes. It would of course be possible, as Dr Luczanits’ evidence contemplates, for proper scholarly analysis to be carried out, by reference to objects of known authenticity, which demonstrated the likely authenticity of a particular object in Kurita or in some other private collection. However, there is nothing in the Defendants’ evidence which attempted this exercise in relation to any piece in a private collection. I also agree with a point made by Professor Lo Muzio that no evidential weight is to be attached to a comparison between an object in issue in these proceedings, and another object in the private collection of Mr Eskenazi. Professor Lo Muzio made the point, and I agree, that the “technique of drawing a positive confirmation using objects belong to one’s own collection … is a practice which should be rejected”.

It might be thought that the position is different in relation to pieces which are in museums, in particular well-regarded museums. However, I do not think that this is so. Such pieces will usually originate in private collections: they will have been given (or sold) to the museum by a private collector. The relevant curator at the museum, particularly a well-respected museum, may well have formed a positive view of its authenticity at the time of acquisition, and again at the time when they are actually put on display (and maintained on display). The fact that a well-respected museum has acquired and displayed a piece is, therefore, clearly evidence that the piece is regarded by a curator as authentic, even though it does not appear in the archaeological record. However, the mere fact that it has been acquired and displayed by a museum does not mean that a sound methodological approach has been applied when considering its authenticity. Nor does it lead to the conclusion that there is any flaw in the methodology which the Claimants’ experts have adopted.

In considering the relevance and any question of authenticity of objects in other private collections or museums, I do not attach any importance to the absence of published articles which attack the authenticity of such pieces. The absence of published articles does not mean that a piece, whether in a private collection or a museum, should be treated as authentic, despite forming no part of the archaeological record. As the Claimants’ experts explained, it is not usual or sensible for a scholar to write a piece which directly calls into question the authenticity of a piece owned by a museum or private individual. This could result in potentially expensive legal proceedings against the scholar, which he or she may well be unable to finance. Accordingly, the approach of individuals such as Professor Lo Muzio is to base their academic writing on those pieces which they consider to be authentic, and to avoid reference to pieces believed to be inauthentic. Accordingly, it is not fruitful to look for scholarly articles which directly call into the question the authenticity of a piece. However, the absence of scholarly discussion of a piece may well be an indication that the scholar does not consider it to be authentic.

The third area of evidence was from materials scientists, Dr Bennett for the Claimants and Dr Nicola for the Defendants. There was also materials science evidence from Dr Ogden and Dr Nicola in relation to the Serpent Bracelet, although in the end the Defendants conceded the inauthenticity of this object.

The evidence adduced on the other 6 objects was vast. Dr Bennett’s report, excluding appendices, ran to 141 pages. This included text as well as photographs and other visual material. Each page of text contained two columns of single-spaced text. Dr Nicola’s first report, prepared before he had physically inspected any of the objects, ran to 455 pages (again with text, photographs and other visual material). Following his inspection, he produced a further report of 92 pages. No joint expert reports were available for pre-reading prior to trial, since the experts remained in disagreement not only about the substantive issues but also about the content of the joint reports. Eventually, unsigned joint reports were made available during the trial, and each side treated those as embodying the views of each expert notwithstanding that they had not been signed off. There was a separate joint report on each object, with a further 74 pages of single-spaced text being produced.

Dr Bennett was then cross-examined for the better part of 2 days on Days 6 – 8 of the trial, and Dr Nicola for around ¾ of Day 8, which was the final day for evidence at the trial. (The parties had agreed that there would be an equal split of the available time at trial, and the time available for cross-examination therefore depended upon how much time had previously been taken in cross-examination). There were many issues, developed in the expert evidence, which were not explored in cross-examination. There are therefore many disagreements between the experts which I am unable to resolve. Nor do I consider it necessary to resolve them. To adapt the direction that is often given to juries in criminal cases: I do not need to resolve every issue which has been raised, but only those which are necessary to reach a conclusion on authenticity. Since the cross-examination did expose and examine the important areas of disagreement, I consider that I am in a position to reach a view on important points.

Despite the vast quantity of expert material on this topic, the scope of the debate on the 5 Gandharan/Bactrian objects was to some extent a narrow one. In broad summary, the Claimants’ case was that each of the objects showed a number of different signs of modern intervention, in particular in relation to deposits applied to the layer of the stone or modern tool marks. Perhaps most importantly, Dr Bennett relied upon the absence of evidence of the “weathering” that one would expect to see on ancient objects. Each of them was in remarkably good if not pristine condition.

The focus was on these issues because, as was common ground, it is not possible to draw a conclusion by reference to the age of the stone from which each object was made. Carbon dating techniques, which can be applied in the context of paintings for example, will not provide any useful information. Stone is rock, and the age of rock depends upon the geological time-frame in which it was formed, and this does not assist in saying when the stone was carved.

Dr Nicola accepted that there was evidence of modern working or modern materials on each of the 5 objects. He also accepted that, broadly speaking, there was little or no evidence of weathering. However, in his view this did not mean that the objects were modern. There were many reasonable explanations for the positive evidence of modern working and the absence of weathering. Prominent amongst these were the conditions in which the object may have been kept, and restoration work that may have been performed over the years. Accordingly, he considered that the matters relied upon by Dr Bennett were inconclusive on the question of authenticity. As he said in his final answer, concerning the stone objects other than the Hari Hara:

“I can say that it is very difficult to state that they are recent or they are ancient because there are too few elements that can give a conclusive point on this topic, and it is exactly what normally happens on many stone objects because it’s one of the most difficult materials to authenticate”.

Accordingly, whilst the Claimants advanced a positive case that the materials science evidence on these 5 objects established inauthenticity, the Defendants did not suggest that the evidence established authenticity. In relation to these 5 objects, the Defendants did not suggest that the materials science evidence could negate a conclusion of inauthenticity based upon the art historical evidence.

The position in relation to the Hari Hara was different. Here, Dr Nicola considered that there was positive evidence that demonstrated authenticity. That evidence concerned what he described as “chromatic variation” of the sandstone from which the Hari Hara was sculpted. He was referring here to some discolouration of a part of the stone, as shown on samples which he had taken from the sculpture. The examination of a stone sculpture by materials scientists involves removing very small parts of the sculpture for examination, usually from places which will not be apparent and will not affect the aesthetic appeal of the sculpture. Dr Nicola’s conclusion was that this provided evidence of natural weathering that had taken place over many years, and that therefore the sculpture was ancient and authentic. There was a substantial dispute between the experts as to whether the discolouration was evidence of natural weathering or, as Dr Bennett suggested, the application of a liquid to the stone.

I will address, to the extent necessary, the details of the debate between the experts in the sections which follow. However, I will here set out my more general conclusions in the light of their evidence including my views of the two experts.

Dr Anna Bennett was trained as a conservator and holds a first class BSc degree in Archaeological Conservation and Materials Science from the Institute of Archaeology, University of London. She then held a one-year fellowship at the J. Paul Getty Museum California. She was awarded her PhD in 1988 from University College London, and a post-Doctoral Research Fellowship at the University of London in 1988 – 1990. She is a principal of Conservation and Technical Services Ltd (referred to at the trial as “CTS”), a company which she founded in 1989 and which is affiliated with the Centre for the Scientific Investigation of Works of Art. She has undertaken a wide variety of conservation and analytical projects over the years, providing conservation and analytical services to numerous well-known museums and institutions such as the National Trust. It is not necessary to set out the details of the work that she has done, or her fairly extensive recent bibliography. There was no dispute that she was an experienced and very well-qualified materials scientist, who has been working in conservation and analysis for over 30 years.

I saw her cross-examined over a number of days. At the time, I formed the view that she was an impressive witness who was seeking to assist the court. By and large, she gave short and direct answers to the questions raised, often agreeing with the proposition that was being put. On occasions, her answers were lengthy, but they were still (generally) directed to the question which had been asked, and the length of her answers usually reflected either the nature of the question and/or the fact that she was seeking to explain scientific concepts which were not straightforward to a judge without a background in materials science. Her evidence was clearly expressed and usually well-explained.

The Defendants submitted in closing that she was an unsatisfactory expert witness, more inclined to argue her clients’ case than to assist the court, and that she was not a witness on whose evidence the court can safely rely. They said that her evidence was shown in cross-examination to be unsupported, speculative and in many cases positively misleading. She was driven to admit, reluctantly, that assertions in her court report had no evidence to support them, and that she did the same in reports which had been prepared prior to the proceedings. She abandoned arguments without identifying the fact that she had done so, and without admitting that they were false arguments. She advanced entirely novel and unsupported explanations. Her evidence was constantly shifting. Overall, no part of her evidence on inauthenticity could be relied upon. The Defendants supported these submissions by reference to 7 particular points drawn from the evidence that she had given at various stages.

I have considered each of the points raised, and I do not accept this submission. My reasons as expressed in the following paragraphs address certain technical issues relating to the objects themselves, and therefore need to be read in conjunction with my discussion of the materials science in later sections of this judgment.

It is right to say that Dr Bennett, on some occasions, accepted that what she had said in her report was incorrect or too emphatic. An example, given much prominence in the Defendants’ closing, concerned her acceptance in cross-examination that some chlorine was present in the deposit which had been applied to the Frieze. As it seemed to me, this was to her credit: where a good point was made, she was willing to acknowledge it.

Moreover, I did not see that this point, about chlorine in the deposit, materially affected her evidence as a whole as to the authenticity of the Frieze. She had some much more fundamental and significant points concerning inauthenticity, including those relating to the absence of weathering, and modern tool marks. These are discussed in Section I below. Whether or not chlorine was present in the deposit was not in any way crucial to her conclusion. Indeed, she had made it clear almost at the start of her cross examination that definitive conclusions could not be drawn from deposits:

“Well, you have to be very careful when you look at deposits because deposits can have occurred, you know, after … a genuine object was overcleaned or something. So, you know, an applied deposit is -- you note it, but it may not be the most paramount for the authenticity of an object, because the actual weathering of the object itself, that’s the most important in terms of authenticity”

This passage, in my view, is evidence from an expert who is indeed seeking to assist the court, by warning against drawing a conclusion simply from a deposit, and being careful not to overstate things. As she said, an artificial deposit applied to the surface of an object is “part of the story of the object”. It was apparent throughout her evidence that, as she said in that quoted passage, the actual weathering of the object was the most important point. The Defendants’ various criticisms of her evidence in their closing did not, as it seemed to me, focus on her evidence as to the lack of weathering. Indeed, in relation to all of the objects apart from the Hari Hara, it was essentially common ground that there was an absence of any significant evidence of weathering.

A related criticism of Dr Bennett was that she made points in her pre-action reports, but these points were not then made in what was referred to as her “court” report, and that she did not draw attention to the fact that she was no longer making that point. I do not think, however, that an expert is to be criticised for omitting, from a report prepared in accordance with CPR Part 35, points which she no longer considers to be important, or that she may have reflected upon and decided were not good points. To my mind, that is what an expert should do. Nor do I think that an expert is required to draw specific attention to points which are no longer pursued. Those points will be apparent to the other party, which can if necessary ask questions about them.

Mr Green submitted that, in relation to answers made in her lengthy cross-examination, Dr Bennett sometimes made points which had not been made in her previous reports or in her court report. When a witness is under cross-examination for the best part of 2 days, and is dealing with a large number of questions of which she has not received advance notice, it is almost inevitable that responsive answers may include information or statements which have not, in exactly that form, been contained in earlier reports. A trial, and in particular cross-examination, has a considerable element of fluidity, with particular points becoming the focus of attention and being explored and responded to. In the present case, I did not think that any of Dr Bennett’s answers relied upon in this context (for example, her evidence as to feldspars in the context of the Hari Hara) represented a significant departure from points which she had previously made.

A point on which great emphasis was placed concerned statements by Dr Bennett which were made in relation to the existence of ceramic particles in the Krodha. As with most if not all of the points made by the Defendants as part of their attack on Dr Bennett, this is a long way from being a central point on the issue of whether the Krodha is or is not authentic. Indeed, it is not a point of any importance at all, and the existence or otherwise of ceramic particles was not relied upon by Dr Bennett in her court report or in her evidence in the witness box. Her conclusion in her court report is that the work contains modern particles, such as modern brass, nickel chrome steel with a vanadium hardener, gold dust, and pure molybdenum. There is no mention of ceramic particles. It is clear from that report that ceramic particles played no part in her analysis. Moreover, it was ultimately common ground that there is indeed substantial evidence of modern materials in the Krodha. Dr Nicola in his oral evidence said that he was sure it was not 100% an original object, although he could not know exactly the number of percent that the object was original: it could in his view be 80 per cent original, or 50% or 30%. He described the object as being subject to a few deep restoration interventions, and he agreed that it was possible to see plastic protruding from the cheek of the object. It followed from these answers, and indeed Dr Nicola’s overall conclusion, that it was possible that the object was 0% original.

Nevertheless, there was extensive cross-examination of Dr Bennett on the issue of ceramic particles in the context of statements which she had made in the second CRS pre-action report that she had written, and also some correspondence (exhibited to her court report) with Oxford Authentication. It was suggested to her that she had made knowingly false statements in that pre-action report and/or in the correspondence. She was, I think, completely baffled by the suggestion and denied it emphatically. I consider that she dealt well with this line of questions, and I do not think that there is any substance in the suggestion that, in relation to ceramic particles or otherwise, she made deliberately false statements intending to deceive. Nor is there anything which leads me to conclude that I cannot rely on her evidence. In order to explain why this is so, it is necessary for me to give some explanation of the history of how the question of ceramic particles arose, and to refer to Dr Bennett’s evidence about what she had written. However, the evidence on this issue is best explained and understood in the context of the more detailed discussion of the Krodha, and it is therefore contained in Section K below.

Dr Bennett’s conclusions on various issues were criticised on the basis that they were “merely speculative opinion”. Reference was made to her evidence that a deposit had been applied to simulate a burial sediment. In my view, however, Dr Bennett was fully entitled and qualified to offer an opinion as to the likely reason for a relevant feature of an object. When one is dealing, as here, with allegedly fake objects, there will be no direct evidence, from someone who saw what happened, of the precise steps that the forger took. However, just as a pathologist may be able to draw conclusions from the state of a body at a post mortem examination, a qualified materials scientist may be able to draw conclusions from an object as to what is likely to have happened. Whether or not I accept that evidence, and Dr Bennett’s conclusions, is a different question. But her evidence cannot be disregarded because she sought to infer what is likely to have happened.

Mr Green also submitted that I should take into account the possibility of “confirmation bias”. His point was that Dr Bennett had previously written pre-action reports which concluded that the pieces were inauthentic. She was therefore likely to reach the same conclusion in her court report. I can certainly see that an expert who has previously expressed firm views in an earlier report may well adhere to those views in a later report. I do not see anything wrong with this, provided that the expert is willing to consider new information and points which are made. In that regard, I have already noted Dr Bennett’s willingness to acknowledge good points that were made to her in cross-examination, and good points made by Dr Nicola. I also bear in mind that if the expert had expressed very different views in an earlier report, the opposite point would be made, namely that the expert has been inconsistent. I can see the possible danger of confirmation bias, with an expert rationalising points so as to accord with a previously held view. Whether or not this happened seems to me to depend upon my evaluation of the evidence of Dr Bennett on the particular pieces. This is addressed in the sections below.

Accordingly, I do not accept that Dr Bennett is an unreliable witness. On the contrary, and as will be apparent from my detailed discussion of her evidence below, I consider that she was impressive, and indeed more impressive than Dr Nicola. I certainly have more confidence in her views than the contrary views expressed by Dr Nicola.

Dr Nicola is a conservation scientist with a background in chemical and materials sciences, graduating with a bachelors and masters degree in chemistry from the University of Turin in 2003. In 2019, he was awarded his PhD in Chemical and Materials Sciences, also from the University of Turin. In 2005, he co-founded his present company (of which he is now the sole owner) Adamantio Srl. This is an academic spin-off from the University of Turin, and includes a network of laboratories and professionals who work together in studying and preserving materials of artistic and historical interest and in developing innovative materials to safeguard the cultural environment. Between 2013 and 2016, he was a professor at the Fine Art Academy of Cuneo, and taught a conservation course including on the topics of chemical diagnostics and stone diagnostics. As well as teaching, he has written many book chapters and articles, spoken at conferences, and also worked for his family’s restoration company. He has worked on artworks owned by museums throughout Italy, and also on archaeological excavations. His report was prepared with the assistance, under his supervision, of a number of other people, including Dr Riccardo Frencia, an external consultant who was a geologist.

Dr Nicola was therefore clearly a well-qualified expert, and his reports were detailed and thorough. However, Dr Bennett had many years more experience than Dr Nicola, and in my view this is a relevant factor in considering the evidence of both witnesses. Furthermore, it was clear that in certain respects Dr Bennett had skills and experience which Dr Nicola lacked, specifically in the recognition and analysis of tool marks. Dr Nicola frankly acknowledged that, as a chemist, this was not an area of his expertise. He was also, very fairly, admiring of some of the work that Dr Bennett had performed, namely the taking of silicon casts in order better to identify the nature of tool marks.

In the present case, I have not only had the opportunity of comparing Dr Nicola with Dr Bennett, but also considering Dr Nicola’s evidence against that of Dr Jack Ogden in relation to the Serpent Bracelet. It was on that topic that Mr Stewart began his cross-examination in relation to the specific objects. In my view, Dr Nicola’s evidence on that issue was rather revealing and had features which appeared, in different ways, throughout his evidence. In their closing argument, the Defendants accepted (for the first time) that the Serpent Bracelet was not genuine, and (as Mr Green explained in his oral closing) this was because they recognised the force of the evidence of Dr Ogden. Dr Ogden was, as Mr Green’s submissions recognised, a most impressive witness.

The significant point to my mind, however, is not that Dr Ogden was more impressive than Dr Nicola. Rather, it is the nature of the evidence on the materials science concerning the bracelet that was given by Dr Nicola.

Dr Ogden’s evidence in relation to the bracelet concerned both art history (on which the Defendants had no expert at all) and materials science. Both aspects led him to the conclusion that the bracelet was not genuine. In relation to the materials science aspect, Dr Ogden relied upon various matters, including certain abrasion marks on the surface of the gold. Dr Ogden considered that it was highly unlikely that these marks were the result of recent cleaning or restoration. They appeared to be far more like marks made in the final stages of manufacture. In addition, the composition of the gold was significant. Dr Ogden’s referred to two analyses, carried out by two experts or expert organisations CIRAM and Pieter Meyers, which detected high levels of cadmium, at a level around 650 parts per million (or ppm). Both CIRAM and Pieter Meyers cited this level of cadmium as a sure sign of modern origin. CIRAM had said that it was “formal evidence of modernity”, and Meyers that it “provides convincing evidence that the gold was melted within the past 150 years”. Dr Ogden concurred with the views of those experts, explaining in his report that cadmium was first isolated as an element in the early 19th century, and by the early 1900s began to be found in gold fakes from about that period. There had been some attempt in the 1980’s to argue that cadmium was possibly in ancient gold, but this was firmly rebutted by the British Museum research laboratory. Dr Ogden’s conclusion was to agree totally with the conclusion of Meyers and CIRAM that the cadmium presence was convincing and formal evidence that the bracelet was not genuine.

In the joint report, Dr Nicola and Dr Ogden identified the levels of cadmium in the gold, which were in broadly similar ranges: 280 – 800 ppm (Dr Nicola), 650 ppm (Pieter Meyers), 315-882 ppm (CIRAM). They said that specialists debate the extent to which levels of cadmium as low as those found in the bracelet should be taken to indicate modern origin. Dr Nicola said that such minute traces in genuine objects could not be ruled out, because cadmium is a naturally occurring element. Dr Ogden agreed that it would be unwise to rule out completely the possibility of minute traces of cadmium in or on an ancient object, but was more sceptical that this was likely to occur in practice. Both experts agreed that, as far as they know, no instances of cadmium in gold objects of undoubted antiquity had been reported. But they noted that the very large majority of undoubtedly ancient objects so far analysed were analysed using only tools that could not detect cadmium at very low levels. Thus, very low traces of cadmium as found in the bracelet “may be possible in theory in genuine gold objects, but have not been detected so far in practice”. They went on to say that neither of the experts “could rule out the slight possibility that very low levels of cadmium might occur in an ancient gold object as the result of old or recent contamination”. Dr Ogden considered that there was far less likelihood of this than Dr Nicola.

Dr Nicola was cross-examined on the impact of the evidence of cadmium levels. It seemed to me to very obvious, not only from Dr Ogden’s evidence (supported by CIRAM and Pieter Meyers) but also from what was said in the joint report, that it was more likely than not (in my view, far more likely than not) that the cadmium levels found in the gold meant that the bracelet was not ancient. Cadmium had never been found in gold objects of undoubted antiquity, and the joint report had referred to the “slight possibility” that very low levels of cadmium might occur in ancient gold objects. However, Dr Nicola would not accept, in cross-examination, that it was more likely than not, based on the levels of cadmium, that the object was not ancient. When asked whether his view was that it was more likely than not that the Serpent Bracelet was of modern manufacture, he said in a series of answers that he had “no clue” whether it was ancient or modern. To my mind, that was a most unimpressive response, which could not really be squared with what was said in the joint report. He later referred to not being able to “rule out” that the object was ancient. It was then pointed out to him that there was a difference between not being able to rule something out, and forming a view on the basis of the evidence that existed. There was then the following question:

“Q. Given the evidence we have got, it is more likely than not that it is a modern forgery.

A.

No, no it is not more likely than not. I think it is almost in the 50 per cent. But if my customer came to me and asked me, “Oh, I have to buy this object or I don’t have to buy this object” I will answer, “So if you want to buy the object, this is your own risk. You know there is a warning there is some cadmium, but it could be possible it is for some reason present”.”

On the basis of what was said in the joint report, as well as the evidence of Dr Ogden, I have no hesitation in rejecting Dr Nicola’s evidence that he had no clue as to whether the piece was ancient or modern, that it was not more likely than not to be modern, and that it was “almost in the 50 per cent” as to whether it was ancient or modern. They were the answers of a witness who was trying to give nothing away which might damage his client’s position. A straightforward and obvious answer to the line of questions would have been for Dr Nicola to recognise that, on the basis of the evidence that cadmium started to be used in the 19th century, and had never been found in (genuine) ancient objects, it was indeed more likely than not that the bracelet was modern – even if one could not rule out the slight possibility that cadmium might be found in ancient objects.

Dr Nicola’s answers on this issue were illustrative, in my view, of recurring features of his evidence as a whole: his inclination to identify an area of uncertainty, leading to his conclusion that no view on the issue could be formed one way or the other; a tendency to identify greater difficulties in reaching a conclusion than really exist; and an unwillingness to recognise what is more likely than not to be the position.

Here, therefore, there was (I accept) an element of uncertainty as to whether cadmium might possibly be found in ancient objects, if different tests were conducted, notwithstanding that its use started in the 19th century and had never been found in such objects. However, it does not follow that no view could be formed one way or the other. Both from a scientific perspective (as confirmed by Dr Ogden’s evidence), and from the court’s perspective (where the issue is whether something is more likely than not), it is possible to reach a conclusion as to what is more likely on the evidence. The existence of uncertainty, and the fact that certain theories cannot absolutely be ruled out, does not mean that appropriate scientific or legal conclusions, as to likelihood, cannot be drawn. Here, it was clearly more likely than not, on the basis of the existing scientific knowledge, that the cadmium levels in the Serpent Bracelet meant that it was inauthentic.

These features of Dr Nicola’s evidence can also be illustrated by his evidence in relation to a number of the other objects, and here I give two examples.

First, there was his evidence concerning the marble heads. It was put to Dr Nicola that the heads were “remarkable in showing no visible signs of weathering whatsoever, but have clean features”. His responses were to the effect that there may have been a careful restoration and very favourable storage conditions could be a possible explanation. He said that, normally, calcareous stones are quite prone to be diminished in time, because it is a weaker stone than silicate stones. However:

“… if I am in a protected place where there is no wind, there is no very active weathering. I can’t rule out that an object can stay in a very sound state for a prolonged time if there is not the main cause of everything, that is water. So I think it is not impossible at all that an object stays for 2000 years in a place with – and stays without few signs of weathering”.

I address the evidence in more detail below, but it is sufficient to say that (consistent with the evidence of Dr Bennett) I consider that it is likely, indeed overwhelmingly likely, that the marble heads would show signs of weathering, and that it is at best a remote possibility that this would have been avoided because of favourable storage conditions over 2,000 years.

Secondly, and illustrative of Dr Nicola’s inclination to identify more difficulties than really existed, and to identify possible (and in my view theoretical) explanations, I refer to his evidence about one aspect of the Hari Hara. A relevant point, relied upon by Dr Bennett, was that there had been modern polishing of the Hari Hara, but also that this had occurred when the statue was in pieces. She could say this, in particular, because she had identified polish which had spilled over the break edges of the piece. She illustrated this with photographic evidence, and in my view her visual observation was clearly borne out by the photograph. Indeed, she was not cross-examined on the basis that this finding was wrong. (Its significance is a matter which I need not discuss here). In fact, in Dr Nicola’s second report, I understood him to accept that this was correct. I refer to the detail in Section E3 below.

However, when Dr Nicola came to be asked about this in cross-examination, and he was asked whether you could see that the polishing extended over the areas of the breaks in the statue, his response was that it was:

“Difficult to say because restorer – also if I well remember, also the last restoration that took place, the one by Neil, was a restoration that involved the connection of pieces, different pieces together”.

In fact, it was not difficult to say whether the polishing extended over the areas of the breaks in the statue: Dr Bennett’s evidence and photograph showed that it did. Furthermore, the suggestion that the polishing had been carried out by Neil (Perry-Smith) had no factual basis. The documentary evidence of the work he carried out did not refer to any such polishing, and there was no evidence that he had polished the statue at all, let alone polished it when it was in pieces.

One other aspect of Dr Nicola’s evidence in relation to the Serpent Bracelet is also relevant, and again illustrative of a recurring feature of his evidence. In his report, Dr Ogden had (as described above) certain abrasions which in his view were the result of modern manufacture, rather than recent cleaning and conservation. (Modern tool marks were also a feature of many of the other objects, and this was an important point relied upon by Dr Bennett). In the joint report, Dr Nicola said that the marks were “possibly the result of recent, inexpert cleaning or restoration”. (Broadly speaking, this was his response to all the points on modern tool marks made by Dr Bennett). In response, Dr Ogden said in the report that he “concurred that this was perhaps not impossible in theory, but did not agree that it could fully explain the marks on the bracelet”. His report had explained why he took this view, and I have no hesitation in accepting his evidence on this point in preference to that of Dr Nicola, who did not have real expertise in tool marks. As it seemed to me, Dr Nicola’s evidence, as to the possibility of the tool marks on the Serpent Bracelet being modern restoration, reflected an expert who was looking for possible explanations of clear evidence of modernity, and advancing a positive view on the point even though the possibility was only, as Dr Ogden said, “not impossible in theory”.

The same approach can be illustrated by his evidence about the Krodha. It was put to him that this object was a miracle. It had been “delivered to us” from almost 2,000 years ago in unfired clay and pretty well perfect condition. Dr Nicola accepted that this was “very strange”, but he went on to suggest that there had been extensive restoration. In my view, however, the strangeness of the survival of this object in itself gives rise to the likelihood that it is a modern forgery – a conclusion which is in my view put beyond doubt by the art historical evidence discussed in Section K below.

More generally, in relation to the absence of weathering, the modern tool marks and other aspects of modernity, my overall view is that Dr Nicola’s possible explanations were at best not impossible in theory, but were not likely – and indeed in most cases were most unlikely – on the balance of probabilities.

In summary, and as will be apparent from my discussion of the particular objects below, I accept that Dr Nicola is a very capable scientist, but I am sceptical about the key conclusions which he has reached. Consideration of the evidence of Dr Nicola in relation to the Serpent Bracelet, which is now accepted to be a modern forgery, is important when evaluating Dr Nicola’s evidence as a whole, and I have no hesitation in rejecting his conclusions for the reasons above. That leads me to a degree of caution in accepting his evidence generally. When I come to consider his evidence against that of Dr Bennett, there are in my view reasons (as introduced above) for exercising the same degree of caution. My inclination, by reason of the various matters described above, is to prefer the evidence of Dr Bennett to that of Dr Nicola on key issues. Ultimately, having considered the evidence on each object in detail, that is the view which I have formed.

This section has been addressing the issue of whether the Claimants have succeeded in establishing that the objects are inauthentic, and the evidence relevant to that issue.

There is obviously some overlap with issues which arise on the question of whether, if the pieces were inauthentic, the Claimants have established that Mr Eskenazi did not have reasonable grounds for his statements. That topic is addressed in Section L, after I have considered whether the Claimants have proved that the pieces were indeed inauthentic.

The Hari Hara is very different to the other objects in issue. It is unconnected with Hellenism, Buddhism, or the areas of Greater Gandhara or Bactria. Its religious origin is Hinduism, and its physical origin was Vietnam. It was referred to in the invoice as follows: “Kingdom of Zhenla – present day Vietnam. Late 7^th century. Sandstone. Height 119 cm. Width 65 cm”. It is indeed made of sandstone, and it is a substantial and imposing object. The issue in the case is whether it was a late 7^th century object, or a modern forgery.

It was common ground in the expert evidence that, if the piece was genuine, it was a very important object indeed. Dr Roveda described it as an outstanding original example of the Angkor Borei family of statutes belonging to the same cultural cycle of the Pre-Angkorian period, and as a major example of a statue from a fascinating region and time. Mr Beguin described it as rare and expensive, and “such a significant piece”.

The late 7^th century would place the piece in a time period prior to the rule of the Angkorian dynasty which sat at the heart of the Khmer empire. The period of the Angkorian dynasty lasted from the early 9^th century for approximately six centuries. Hence the piece, if 7^th century, would be a pre-Angkorian object.

Both experts considered the religious and historical background to the object, but it is not necessary to describe this in detail. It suffices to say that in the 7^th century, Hindu sculptures of great majesty and sobriety were produced in the area from the Mekong Delta (now in the south of modern-day Vietnam, which was then known as Cochinchina) to the centre of modern-day Cambodia. These sculptures mainly represented Vishnu and Hari-Hara. By the 8^th century, however, the stone sculptures were often of lesser quality.

The Hari Hara is a form which melds the two great gods of Hinduism into a single entity: Hari, referring to Vishnu, and Hara, which evokes Shiva. The sculptures of Hari Hara have the body of Vishnu on the (actual) left side in the aspect of Hari and that of Shiva on the (actual) right side in the aspect of Hara. In Hinduism, Hari is the protector and Shiva is the destroyer. Representations of the figure of Hari Hara are, in Khmer lands, among the oldest Hindu representations. The work of French, Cambodian and Vietnamese archaeologists has led to the discovery at various sites of several Hari Hara and Vishnu statues dating from the pre-Angkorian era. The expert evidence contained illustrations of many works which could be used for comparison purposes and which dated from around the time that the Hari Hara in issue was alleged to have been made.

The piece in question is standing upright and still, and (as its iconography requires) has one head and four arms. The first pair is broken at the level of the elbow and forearm. The rear right hand holds the trident, a key attribute of Shiva. The left hand holds the conch, one of the usual attributes of Vishnu. The head wears a royal tiara. An engraved sampot encircles the loins and upper thighs. A sampot is a sarong; i.e a length of cloth, wrapped around the waist, which is worn throughout Southeast Asia by men and women. A large support arch brings together the hands and back of the head to give them more strength.

The Claimants’ expert, Mr Beguin, acknowledged in his report that the iconography of the Hari Hara presented no anomalies. His evidence, that the piece lacked authenticity, relied upon stylistic anomalies in five aspects of the piece, coupled with a number of what he described as “further observations”. The anomalies concerned (i) the sampot (the sarong) and its treatment; (ii) the polish on the surface; (iii) the facial features; (iv) the whiskers (i.e. moustache); and (v) the navel. His observations concerned (i) the beauty whiskers and nipples; (ii) a comparison with the Vishnu in the Fine Arts Museum, Ho Chi Minh City; (iii) the absence of provenance. It was this combination of factors which led him to conclude that the piece was not authentic, although it is fair to say that he did not make a great deal of the first of these “observations”.

The Defendant’s expert, Mr Roveda, concluded in his written report that the Hari Hara was an outstanding original genuine 7^th century piece. The style of sculpture was either attributable to workshops located at Angkor Borei or workshops working in or influenced by the style emanating from Angkor Borei. The sculpture was in his view grounded within the appropriate comparators, which his report described in detail with the benefit of illustrations. He did not consider that the appearance and style of the sculpture were inconsistent with established comparators.

Mr Beguin had enjoyed a distinguished career which encompassed many aspects of Khmer and South East Asian art. He is in his mid-70’s, and studied art history and archaeology at the Sorbonne and the Louvre School. When preparing for his specialised master’s degree, he had followed the teachings on Khmer art by its master, Jean Boisselier, former director of the Angkor site and professor at the Sorbonne. (Mr Boisselier is quoted in JEL’s research paper on the Hari Hara.) His career included a lengthy period at the Guimet Museum in Paris, where for 23 years he was responsible for the collections of Nepal and Lamaist Buddhism. He was appointed Head Curator there in 1992. In 1994, he became director of the Cernuschi Museum, Paris Museum of Asian Arts, and held that position until 2011. In 1997, he was appointed General Curator of Heritage, which is the highest rank in the hierarchy of national museums. He had organised many exhibitions, both in Paris (for example in the Grand Palais) and internationally. He had an extensive bibliography, including numerous articles, educational brochures, exhibition catalogues and art books, both in his speciality and in neighbouring fields. Following his retirement in 2011, he had been an advisor on Asian arts. Although there was a suggestion in Dr Roveda’s report that Mr Beguin had “limited knowledge of the Pre-Angkorian style”, there was no suggestion in cross-examination that he lacked any expertise in Pre-Angkorian or Khmer art. Mr Beguin’s selected bibliography (in Appendix C to his report) included a number of publications concerning Khmer and other South-East Asian arts, and Khmer art had featured in courses which he had taught and in speeches given to cultural associations.

Mr Beguin’s written report was concise, with a focus on particular issues which he considered to be of importance. He was cross-examined for just over half a day. He answered in English where he could, but sometimes used the interpreter in order to explain certain matters. On occasion, his answers were a mixture of English and French, but it was not difficult to follow what he was saying. Overall, I considered Mr Beguin to be an impressive witness. He was clearly very knowledgeable about the issues that he was addressing, but also fair and keen to avoid overstatement. For example, when his written report made a general point as to an artistic feature, he was careful to identify any exceptions to that general point. Similarly, when discussing aspects of the sampot, Mr Beguin volunteered that he did not consider himself to be an expert in exactly how sampots were draped, and that he had relied to some extent on the work of Jean Boisselier. I refer to aspects of his reports and cross-examination in more detail below.

Dr Roveda is now in his 90’s and (as I was told) was not sufficiently well to give live evidence before me. He had apparently suffered from cancer some years ago and had been advised to avoid stress. Dr Roveda’s absence from the trial presented some difficulty for JEL in so far as they sought to persuade me to prefer his evidence to that of Mr Beguin. One difficulty was that I did not have the opportunity to see how Dr Roveda would have responded to questions, or would have dealt with some potentially powerful points raised by Mr Beguin. Another difficulty was that, as is inevitable, the case had developed after Dr Roveda had prepared his written report. That report addressed an earlier report which Mr Beguin prepared before he had actually seen the statue. Whilst Mr Beguin’s conclusion on authenticity in his first report was essentially the same as his conclusion in his report in these proceedings, there was some development in his thinking and the focus of the points being made. Since reports had not been exchanged sequentially, Dr Roveda’s written report did not address, at least in any detail, some of the important points that Mr Beguin made. Dr Roveda had not completed a joint report, despite the court order requiring him to do so.

Furthermore, the process of cross-examination will usually, and did in this case, involve some further elucidation of an expert’s (here Mr Beguin’s) views. If an opposing expert is not then called to give evidence, the court will not know how he would have dealt with the points which had been elucidated.

These difficulties for JEL might have been surmountable if, for example, Mr Beguin had been shown to be unqualified or poorly qualified to give evidence, or had been unimpressive when cross-examined, or if Dr Roveda was clearly a much better qualified expert than Mr Beguin, or if his written evidence was far more persuasive than that of Mr Beguin. However, in my view, none of these circumstances applied.

It seemed to me that Mr Beguin had enjoyed a more distinguished career in the relevant field than Dr Roveda. Dr Roveda had spent most of his professional career, from the 1950’s to the 1990’s, working in the oil industry, after obtaining a degree in geology. From the 1990’s onwards, however, he had developed a career as an art historian, and was awarded a PhD by the London School of Oriental and African Studies (SOAS) in “Narrative reliefs of SW and NW corner pavilions of Angkor Wat”. He had taught art history for a short while in Singapore, and then had moved to Cambodia to develop what he described as his final speciality in Khmer Art. He had published various books and papers, had presented the results of his research at various international congresses, and had also lectured at various academic institutions such as SOAS. I accept that, against this background, he was knowledgeable (as an art historian) about the matters which he discussed in his written evidence. However, I had no reason to think that he was more knowledgeable than Mr Beguin, and the respective career records of the two experts would indicate that Mr Beguin was well ahead of him. Mr Beguin said that he had not heard of Dr Roveda before these proceedings, although he had now seen some of Dr Roveda’s published work. Mr Green described Dr Roveda as one of the leading experts in the world in this area, but I did not think that this was established by the evidence. Mr Beguin was evidently surprised to see that one of Dr Roveda’s books appeared in a bibliography of one of Mr Beguin’s own books. This surprise appeared to me to be genuine, and I have no reason to disbelieve Mr Beguin’s evidence that his editor was responsible for the general bibliography in that particular edition of his book.

I will now consider each of the particular points made by Mr Beguin, together with any relevant written response from the Defendant’s expert, Dr Roveda. I leave aside the point on the beauty whiskers and nipples, where Mr Beguin accepted that his point did not support authenticity or inauthenticity.

The sampot in the present case was fully engraved. This contrasted with other (indisputably) authentic pieces where, generally, the sampot was sculpted in a slight relief design and was not fully engraved.

The first peculiarity identified by Mr Beguin concerned the design of the sampot. It encircled the thighs, one of the ends tightly winding its way up to the height of the genitals and dropping forward into two symmetrical pleats on each side of the crotch. Mr Beguin considered this to be whimsical. Divinities in the 7^th and 8^th centuries usually wore a long sampot, falling down to the calves, or short, down to the middle of the thighs. In his report, however, he acknowledged that there were exceptions in genuine pieces: he identified three such pieces, and discussed the differences between the sampots there and that on the Hari Hara in the present case.

In cross-examination, Mr Beguin said that there were no statues which had the same clothing as the present statue, and that whilst he considered the sampot here to be whimsical, he would accept “unusual” as an alternative description. He accepted that sampots were depicted in different lengths and in different ways: there were indeed a variety of sampots, and artistic fashion changed.

In his written report, Mr Beguin developed what seemed to me to be a related point or series of points, although in his oral evidence Mr Beguin considered that they were separate from his “whimsical” point. He said that in the case of a short sampot, a flap tightly encircles the crotch and is tightly held in place at the front by a knot, but often the fabric is long enough to fall to the front with one or perhaps two drops of the pleats. This was illustrated by various (indisputably) authentic statues of Hari Haras and a Vishnu in the Phnom Penh National Museum. However, during the period from which the JEL Hari Hara is alleged to have dated (the 7^th century), there was never a side apron laterally pressed on the left thigh.

Furthermore, on the present Hari Hara, there were two small pieces of floating fabric, falling symmetrically over the front of the thighs. Mr Beguin described this as being “against all plausibility”. It raised the question as to where these elements came from, and how they related to the long fabric that makes up the sampot. Those elements do not appear in the book produced by the distinguished French scholar, who was Mr Beguin’s “master” in Paris, Jean Boisselier. Mr Beguin said that these decorative details, apparently unique, were “impossible in the very logical and rigorous context of pre-Angkorian art”. Those elements belonged to a more recent period. At the beginning of the 10^th century, the fashion of tightly pleated garments, in the manner of Scottish kilts, appeared. The pleats at the waist then became very complex for both men and women.

In the course of cross-examination, Mr Beguin illustrated some of these points by pointing out features of the Hari Hara in the courtroom. He explained that the problem was not that the sampot was short: although sampots were usually long, there were some with short sampots. Where, however, there was a short sampot, there should be a pleat on the front, and it would be held by a belt. It was the equivalent of a fig leaf. But the Hari Hara here did not have that. There was, he said, no kind of cover, no kind of equivalent of a fig leaf, and you could not see what was happening on the sides. He said that the problem was that part of the (engraved) textile was between the two legs and it then went back down, and this was very strange and mixed up; because you could not see exactly what was going on. At the back of the statute, you could see that material going between the legs; but at the front, it disappears, and you could not see what was happening.

Mr Beguin accepted that he was not a specialist in the way that sampots were draped. But in his report, he had provided drawings from a book written by Jean Boisselier. The draping in those drawings was very different to the draping on the Hari Hara.

In his written report, Dr Roveda did not think that it was reasonable to criticise the sampot on the basis that it was whimsical. His principal point was that sampots can be long or short and that more frequently a short garment was worn. He said that, in summary, that the sampot can be depicted in different ways.

Dr Roveda did not address Mr Beguin’s interrelated points about the implausible floating fabric, the decorative detail which was impossible in the context of pre-Angkorian art, and the difficulty in understanding how the sampot worked. This was because these points were first developed in Mr Beguin’s report in these proceedings. He had not made them in the earlier report which he had written, and upon which Dr Roveda had commented in his report. Dr Roveda had not written a supplementary report, there was no joint statement, and he did not give oral evidence.

Dr Beguin’s evidence on this issue was, in my view, very fairly given. His written report drew attention to the different ways in which sampots were portrayed, and identified two statues (the Visnhu in the Ho Chi Minh City museum, and the Vishnu in the Phnom Penh National Museum) which had some affinity with the present statue. He also accepted that he was not an expert in the draping of sampots.

I accept his evidence that the appearance of the sampot in the present case is unusual, and does not appear in other authentic objects. Indeed, it did not seem to me that Dr Roveda, in his report, had specifically relied upon any comparable object as having a sampot similar to that in the present case.

Dr Roveda’s principal points were that sampots could be long or short (a proposition with which Mr Beguin agreed); that they were usually short (a proposition with which Mr Beguin disagreed, but which was not central to his argument); and that the sampot can be depicted in different ways (a proposition with which Mr Beguin again agreed). It did not seem to me that any of this really addressed the points that Mr Beguin made in his report in relation to the implausible floating fabric, the decorative detail which was impossible in the context of pre-Angkorian art, and the difficulty in understanding how the sampot worked. JEL’s inability to call Dr Roveda to give oral evidence, combined with the fact that Dr Roveda’s written report addressed an earlier report written by Mr Beguin, meant that I was not able to know how, or how effectively, he would have addressed these important points that were being made by Mr Beguin.

My overall view was that Mr Beguin had approached this issue, and indeed his evidence generally, with care and with a desire to avoid overstatement. I do not attach any real significance to the fact (put to him at different stages of his cross-examination) that additional points were made in Mr Beguin’s report served for the purposes of these proceedings, as compared to the initial report which he had prepared in 2019. At the time of the first report, Mr Beguin had worked from photographs, and had not actually seen the statue. It is also understandable why, as Mr Beguin thought about the issues carefully for the purposes of drafting his court report (which went through many drafts, as he told me), his thinking would develop.

As I have said, Mr Beguin gave his evidence well. He sought to answer all questions directly and fairly, and he himself volunteered the fact that he was not an expert in the draping of sampots. In the context of a case where Mr Beguin is clearly a well-qualified expert, where the Claimants were unable to test the evidence of Dr Roveda in cross-examination, and where Dr Roveda did not address some of Mr Beguin’s important points, I accept Mr Beguin’s evidence that there are various anomalies in relation to the sampot, as summarised above. If, however, these had been the only points relied upon, they might very well not have carried the day in relation to authenticity; not least because the evidence shows that sampots can be depicted in different ways. However, they are points which are to be weighed, as factors against authenticity, in the overall balance.

Mr Beguin identified another anomalous feature concerning the engraving of the sampot. One point was that the sampot was completely engraved, rather than being a slight relief design. However, I did not think that this was a particularly powerful point against authenticity, because (as Mr Beguin fairly accepted in his report), there is one (indisputably) authentic Hari Hara which has a completely engraved sampot. Mr Beguin also considered that there was an awkwardness to the engraving: the engraved features were not clear, and this contrasted disagreeably with the very clean look of the statue. Whilst this point had some force, because it was an unusual feature of the present sampot, I did not think that it added materially to the other anomalies previously discussed.

The polish on the surface

Mr Beguin’s written evidence was that there was a very attractive, exceptional polish over all of the statute, with reflections that could be described as metallic. This was an extravagant feature. Whilst some rare Indian schools preferred an extreme polish, this was not the case in Khmer lands in the pre-Angkorian period, where the polish was never as accentuated.

Furthermore, the polished feature would be incompatible with the imperfections inherent in an archaeological object that had been in the ground for a long time This appearance would be even more unusual if it was stored for more than 1,000 years in a place of worship. He contrasted this figure with others in various museums. The Hari Hara (Sambor Prei Kuk) in the Phnom Penh National Museum had a well-cared for polish that could be compared to the present Hari Hara, but had inevitable traces of wear and mutilation. Mr Beguin said that one possibility for the strange appearance of the present Hari Hara is that there had been an acid treatment, designed to make serious study of the surface difficult. He described the extreme polish of the piece as “dumbfounding”.

In cross-examination, Mr Beguin explained that the problem was not the existence of polish: it was perfectly possible for a statue to have been highly polished 1000 years ago. The problem was that there should be traces of ageing. If the sculpture was in the ground, there would be changing of colour due to acidity and underground water. There would also be ageing if it was kept in a shrine, out of the ground. He pointed out that in the (indisputably) authentic pieces which are in museums, there is evidence of both polish in some parts but also a mixture of parts which have been reconstituted, or missing, and corrosion.

Dr Roveda said that the polish is less usual, but not at all unheard of. “Extravagant” polish was, he said, found on notable museum works such as the Hari Hara in the Musee Guimet in Paris and that in the National Museum, Phnom Penh. The present Hari Hara was, he said, in “excellent condition”, and he could not see (from the photographs – he had not seen the statue itself) any signs of restoration other than the minimal signs where the sculpture was broken. He said that the polish may have been a thin wax applied by the previous owner, restorer or conservator, to achieve a sheen. But this was not known from his papers, and he did not think that this had any bearing on authenticity. Dr Roveda’s report did not really address, at least in any detail, Mr Beguin’s point as to the absence of the signs of ageing on the polished surface.

It seems to me that the points which Mr Beguin made – which are similar in many ways to the points made by Dr Luczanits and Professor Lo Muzio on the Gandharan objects – were important points against authenticity. The indisputably authentic statues in the Phnom Penh National Museum, and in the Musee Guimet, do show signs of polish. But they also, as Mr Beguin said, show signs of the ageing that might be expected on an object which is over 1,000 years old. The Hari Hara in the present case shows only the former (i.e. the polish), and in a remarkably uniform fashion. Mr Beguin was cross-examined on the basis that the polish was original and comparable to other statues. He disagreed, and I accept that he was right to do so. Indeed, the materials science evidence (discussed below) clearly established that the polish was not original, but had in fact been applied by a modern high-speed polishing wheel.

I therefore think that Dr Roveda was wrong to suggest that the polish on this statue was in any way comparable to that seen on the museum objects to which he referred. He was also wrong to say that he could not see any sign of restoration from the photographs other than minimal signs where the sculpture was broken. Mr Beguin was right when he said that there was exceptional or extreme polish, and that it could be seen on the photographs. This polish on the statue was, on the basis of the photographs, clearly not original. If the statue was authentic (as Dr Roveda contended), this was clearly a sign of restoration, contrary to Dr Roveda’s evidence. I take these matters into account in my assessment of Dr Roveda’s evidence, and my view that the evidence of Mr Beguin is to be preferred.

The question remains, however, whether this extreme polish is evidence of inauthenticity. Dr Roveda’s evidence acknowledged the possibility that a previous owner, restorer or conservationist had applied a thin wax layer to achieve sheen as they preferred that aesthetic. I accept that since this might have happened, the extreme polish is not conclusive evidence of inauthenticity. Nevertheless, in my view the polish is indeed a point, and a point of some weight, against authenticity. Once it is established that the polish is not original, then that demonstrates that there has been modern work on the statue. Whilst there is a theoretical possibility that an owner would prefer the aesthetic of a polished statue, it would be surprising if he would wish so significantly to alter the appearance of an ancient statue. The more likely explanation for the modern appearance is that the piece as a whole is modern. There is, of course, nothing in the evidence which suggests that a previous owner preferred this aesthetic. Apart from the dealer, the only suggestion of a previous owner is the Buddhist monastery.

Furthermore, I cannot see that the suggestion of polishing to improve the aesthetic of an ancient statue is at all likely, given what is known about the statue. If there was a desire to improve the aesthetic, then it would be highly surprising for the piece thereafter not to be preserved as a whole. However, when it was identified for sale, it was in 5 separate parts. There was also nothing in Mr Eskenazi’s evidence as to the provenance of the piece, sketchy as it was, to suggest that there was any previous owner, restorer or conservationist who had applied a thin layer of wax in order to improve the aesthetic. Furthermore, a significant aspect of Dr Bennett’s materials science evidence, discussed below, is that the polish was applied when the statue was still in pieces. She could tell this from the fact that there were some traces of polish on the break edges of the pieces. Accordingly, the “improve the aesthetic” suggestion makes no sense, because it would then be very odd for the statue to be left in pieces. If a previous owner was keen to improve the aesthetic by polishing, they would also improve it by having a statue which was in one piece.

Accordingly, the uniform polish found on this piece, without any signs of ageing, is indeed evidence of modernity and inauthenticity.

The facial features

In his written report, Mr Beguin said that although there was some variability in the craftsmanship of pre-Angkorian works, the face of the present Hari Hara could not be compared to any comparison pieces. This was because of its excessive geometry. This excessive geometry differed from both a subtle rendering of flesh inherited from India, and the severe majesty of developing Khmer art. Abrupt plane changes and sharp edges are to be found in metal statue works in the manner of certain stone statues from the Bakheng period of around 900 CE. This could be seen especially in the sharp eyebrows of the Hari Hara. This feature was unknown during the 7^th and 8^th century. Mr Beguin thought that the introduction of this feature into the piece represented a mistake by the maker in his attempts to replicate a particular period: this geometric, sharp style did not arrive until after the pre-Angkorian period, and it therefore raised questions as to the authenticity of the work.

In cross-examination, he confirmed that the excessive geometry of the face was more consistent with imagery from around 900 CE than with imagery from the late 7^th century. If there were 9^th/ 10^th century traits on a piece purportedly from the 7^th century, then there were problems with authenticity. It was suggested to him that the same traits could be seen on the Hari Hara in the Musee Guimet in Paris. Mr Beguin had worked as a curator at that museum for 23 years, including for the last 3 years as head curator. He disagreed with the idea that the face in the Guimet hari Hara was geometric, and he referred to various other differences between that statue and the Hari Hara in issue.

Dr Roveda did not agree with Mr Beguin’s assertion that the Hari Hara’s face was excessively geometric or that its features could not be compared with any of the comparison. He said that there were various examples in his illustrations which showed how closely comparable the features of the sculpture were. He did not identify any particular comparable sculpture for that purpose.

I accept Mr Beguin’s evidence on this issue. He has considerable expertise in this area, and he would clearly have been familiar with the Hari Hara in the Musee Guimet, where he worked for many years. This was the only allegedly comparable object on which he was cross-examined in relation to this topic, and I could well understand why he took the view that it was not comparable to the Musee Guimet piece. I accept too that Mr Beguin had what might be fairly described as a connoisseur’s eye for the point that he was considering. It may be that Dr Roveda has a similar eye, but I have not had the benefit of seeing him being cross-examined. There is in my view no reason to reject the evidence of Mr Beguin, who was cross-examined and who dealt fairly and convincingly with the points that were put to him.

Accordingly, I accept that the facial features are another factor pointing towards inauthenticity.

The moustache

Mr Beguin’s evidence in his written report was that a moustache was very rare in the pre-Angkorian era, although he did identify one Hari Hara (the Hari Hara of Prasat Andet) which had this feature. Moustaches did, however, become a fashion style starting from the early 9^th century. He accepted that this feature does not on its own demonstrate inauthenticity, but that it was another anomaly that was only rarely found in this era. It therefore reinforced his overall conclusion on inauthenticity.

Dr Roveda addressed this point in his report. He considered that the facial hair or “whiskers” that became more common in the 9^th century comprised finely trimmed beards. He too identified the Prasat Andet Hari Hara as having a fine moustache. His conclusion was that these examples showed that early Angkorian art did not follow a completely prescribed form, and that Dr Beguin was being overly precise.

It seemed to me that both in his report, and in the brief cross-examination on this issue, Dr Beguin was careful not to overstate the importance of this point. Both experts could only identify one other Hari Hara with a fine moustache, and it can therefore be said that this feature was unusual in a Hari Hara of this period. Mr Beguin’s point, ultimately, was that this should be considered in conjunction with the other unusual features which he had identified. On its own, my view is that this would clearly not be a sufficient point in favour of inauthenticity. I consider, however, that the rarity of moustaches is a small additional point in favour of inauthenticity.

The navel

A similar point can be made in relation to the navel, which has a shape similar to an inverted letter “W”. Both experts identified only one example of a similar navel. Most other statues, according to Mr Beguin, had a simple circular hollow. It seems to me that the rarity of navels of this shape in comparative pieces would not on its own be a sufficient point in favour of inauthenticity. Again, it is a small additional point.

Similarity to the Vishnu, Hao Ninh, of the Fine Arts Museum Ho Chi Minh City

Mr Beguin made three additional observations, of which two appear to me to be significant. The first concerned similarities between the Hari Hara and the Vishnu, Hao Ninh, at the Fine Arts Museum Ho Chi Minh City. This authentic piece shared a number of similarities with the Hari Hara, and I considered that these similarities were striking. In both pieces: the lower part and side supports are missing; there is a support arch in the upper part; there is engraved clothing, with the sampot being fully engraved rather than being a relief; there is a navel evoking an inverted “W”. A significant difference, however, is that although the Hao Ninh is neatly finished, it did not have the almost metallic sheen of the work. Mr Beguin considered that the sampot of the Hari Hara was so similar to the sampot of the Vishnu that it raised suspicion. It was very unlikely that one would see two pieces that were so similar in this period. That was because, during this period, each piece would have been crafted individually, as can be seen from the available corpus of work.

It seemed to me that this was a powerful point. Dr Roveda did not specifically address it in his report. I suspect that his response would have been a point made in his report: that it is a mistake to assume that a technically qualified and talented sculptor from the time only created one sculpture in his or her entire career, and that it is “perfectly possible that, having completed one master work, he or she was asked to create a second example for a different temple or group of patrons which may or may not have been of the same subject”. Whilst I accept that this is a possibility which cannot completely be discounted, I do not consider that it is a probability in the context of the statue that I am considering. It would in my view be a surprising coincidence that the two statues would both have similar parts missing. It is also, in my view, more probable, particularly bearing in mind the other features which I address in the section, that the Hao Ninh Vishnu served as the basis for a modern forgery.

Absence of provenance

Mr Beguin also considered that it was important to consider the provenance of any work whose authenticity is being considered. His point in summary was that it was strange (as he said, paradoxical) that a statue as rare and as expensive as this Hari Hara would be devoid of pedigree. The absence of any reference to its origin and history is surprising. If the work was genuine, it was reasonable to expect that it would be well documented.

Mr Beguin considered the possibility that the work may have been subject to clandestine archaeological excavation and then fraudulently exported. Indeed, as it seemed to me, the Defendants’ case was that the Hari Hara had been fraudulently exported; since the bill of lading from Vietnam had described it as a garden ornament worth USD 575.

However, Mr Beguin did not consider that this, or any alternative hypothesis, adequately explained the lack of documentation about provenance. For such a rare and important work, he would expect to see documentation such as scientific and academic publications, public auction catalogues, old photographs and any document pertaining to the status of any restoration work to the piece. Whilst a possible explanation was that the piece was genuine and excavated illegally, the other possibility was that it was a forgery. When considered in the light of the anomalies and various observations about the work, he was firmly of the view that it was a forgery. The absence of documentation supported that conclusion.

Again, it seemed to me that this was a powerful point against authenticity. It was not a point which was specifically addressed by Dr Roveda.

In summary, Mr Beguin considered that there were significant anomalies in the polish, the facial features and the shape of the garment. The unusual presence of the moustache made it a rarity, but not the only instance. No piece in the body of (indisputably) authentic work had so many unusual details at the same time. He thought that one could only be shocked that the same parts are missing on the Vishnu of Hao Ninh and this piece. The suggestion that the work had been preserved in a Buddhist temple was also unlikely: the Hari Hara belongs to a different religion (Hindu), and it would be strange for the piece to have been preserved as five more or less shapeless fragments. Accordingly, in his eyes the work was a forgery.

I consider that the views expressed by Mr Beguin were well-reasoned and persuasive. I accept his evidence in preference to that of Dr Roveda.

In his witness statement, Mr Eskenazi described the Hari Hara as one of the most extraordinary Southeast Asian sculptures that he had ever had. It was by far the best of the different pieces from Vietnam that he had ever seen or handled. To my mind, this reinforced Mr Beguin’s evidence as to the strangeness that such an important statue should be so completely devoid of pedigree.

The evidence indicates that almost nothing is known about the statue prior to the dealings between Mr Eskenazi and Mr Paul Jewell in late 2012. Mr Jewell was described by Mr Eskenazi as a collector and sometime dealer of Southeast Asian art who had spent most of his professional life in Southeast Asia. He understood him to have contacts with many local dealers. The contemporaneous documentary evidence indicates that, in late 2012 and early 2013, Mr Jewell was in discussion with someone in Vietnam to purchase the Hari Hara. Mr Jewell subsequently signed at least two documents (dated 6 February 2013 and 23 May 2019) which stated that he had bought the piece in Singapore many years earlier, that he had imported it into the UK in 1998, and that it had since been in his house in London. These statements were clearly untrue: the piece was purchased in early 2013.

The correspondence between Mr Eskenazi and Mr Jewell in late 2012/ early 2013 does not identify the person with whom Mr Jewell was in discussion at that time. It is reasonable to conclude that the Hari Hara was in Vietnam at that time: it was later sent, in March 2013, from Vietnam to Hong Kong with the invoice and bill of lading evidencing a sale and shipment of “Garden Stone Decoration”, and an invoice price of USD 575. The Defendants’ case is that these documents falsely represented what was being shipped, and its value; because the shipment was of a valuable antiquity worth far more than USD 575. The shipping documents identify the exporter as the Hien Minh Gallery, Hanoi Vietnam. There was no evidence, however, about this gallery. Indeed, in view of the fact that the shipping documents were (on the Defendants’ case) false, it must be questionable whether the exporter was correctly identified.

At all events, Mr Jewell was clearly in discussion with a seller, whoever it may have been, in late 2012. In his email of 11 October 2012, he told Mr Eskenazi that the price was creeping down and that it was now about “200”, which I take to mean USD 200,000. On 1 January 2013, Mr Jewell asked Mr Eskenazi whether he wanted him to resurrect discussions about the Hari Hara. He thought that it was “still there”, and he had made a few stories to “try to put them to sleep for a few months”. He indicated that with Tet (the Vietnamese New Year festival) coming up, the seller would be needing money. But Mr Jewell could not “start asking questions unless you really want to buy it”. Matters moved forward in early February, with Mr Jewell telling Mr Eskenazi:

“Now the body is here, I want to act quickly to stop the chance of it being shown more generally”.

This e-mail indicated that, prior to that time, the body of the Hari Hara had been moved, and that the statue was now more likely to be shown to other purchasers. It appears that the purchase of the statue was completed not long thereafter, possibly by 6 February 2013 and certainly by around 10 March 2013 which is the date of the bill of lading.

In cross-examination, Mr Eskenazi said that the piece had come from a Buddhist monastery. His evidence was that he did not know the name of the dealer, although the name did appear somewhere in the documentation. He said that, in his discussions with Mr Jewell, he had been told that it was with a dealer, and had come from a Buddhist monastery. The monastery was selling Hindu material in order to sustain the monastery. He had not included this information in his witness statement, because he had forgotten about it. However, it was well known that Buddhist monasteries had been selling mostly damaged Hindu gods for many years. He did not know which Buddhist monastery had sold the piece: there were many of them, and Mr Eskenazi had never been to Vietnam and he didn’t know the situation. It was, he said, all hearsay. But the piece was with the dealers’ family, and they wanted a lot of money for it despite its condition. There was no explanation of why the statue would be in 5 separate pieces: Mr Eskenazi said that he did not know, but that it could have been dropped.

This was really the entirety of the evidence as to the provenance of the piece. In my view, there is nothing here to suggest authenticity and (as Mr Beguin’s evidence indicated) much to indicate otherwise. There is no history to the piece and no identification of where it was found, despite its considerable importance (if genuine). As Mr Stewart submitted in closing, there were no photographs showing it standing in a Hindu temple, no record of it being excavated, nor even a record of it being in someone else’s collection 30 years before. As Mr Stewart said, it appeared by magic from an unknown dealer in Vietnam, with nothing behind it. There is also the oddity of a Hindu statue being owned by a Buddhist monastery, if indeed that was what Mr Eskenazi was told. Apart from his oral evidence in cross-examination, there is no documentary or other evidence that he was given this information, and it was not included in his witness statement, and there is also no other evidence (e.g. photographs taken of the object) that it was ever in a Buddhist monastery. There is also the oddity of the body of the piece apparently becoming separated from other parts, resulting in a degree of urgency (“Now the body is here...").

The Claimants also referred to various aspects of the way in which the piece was dealt with, and in particular: (i) that, if authentic, the piece had been illegally exported with false documentation, and (ii) the creation of false documents which were fabricated to create the impression, for tax purposes, that JEL had paid USD 1.8 million for the item when in fact it had not done so at all. Whilst those matters are relevant to the overall assessment of Mr Eskenazi’s credibility, including the question of whether he acted dishonestly, discussed in Section L below, I did not think that they were relevant to matters concerning the provenance of the piece. The information, or rather the striking lack of information, on that topic is as summarised above.

Nothing was known then, or is known now, about any restoration or conservation history of the piece prior to its purchase. Subsequently, when the statue was sent to England, it went directly to Neil Perry-Smith, the restorer with whom Mr Eskenazi dealt. He did reattach the broken sections and remodelled and toned in small areas of loss and light abrasion. There is nothing to suggest that he carried out an extensive restoration or that he was the person who applied the machine polish. (Dr Nicola said, in his first report, that he understood that the item was purchased in a polished state). There is no contemporaneous report or evidence of the work that Neil Perry-Smith did in 2013, but he did produce a brief report in 2018 after authenticity had been challenged. This states:

“Upon visual and microscopic examination the sculpture was determined to be in good condition.

A light manual clean under X 10 magnification was carried out to remove any loose surface debris.

Broken Sections of the sculpture were re-attached with non ferrous pins and polyester resin.

Small areas of loss were sympathetically re-modelled and toned in with fully water reversible earth pigments and polyvinyl acetate (see images).

Light abrasions and areas of surface loss were improved and toned in. The Sculpture was mounted on a patinated bronze mount.”

Accordingly, there is nothing to suggest that it was Neil Perry-Smith who applied the polish, the modern tool marks which are evident on the statue (as described below) or the patina (again described below) which was applied to the stone. This work was all done prior to the statue coming to England. Mr Stewart submitted that these matters all pointed to modern manufacture. It made no sense for someone to do heavy restoration on the statue (using modern tools on, for example, the engraving, applying the patina, using a machine polisher), but nevertheless to leave the statue in pieces. I agree.

Dr Bennett’s conclusions in her written report were in summary as follows. She had conducted a number of different examinations of the Hari Hara and particular features such as the tool marks. Her conclusion, based on a visual examination of the statue, examination of the work itself and samples under microscope, and of silicon rubber moulds taken of the tool marks on the statue, was that there was clear evidence that the statue was not ancient. She placed particular reliance on the absence of any natural weathering of the stone. She also considered that there was clear evidence of the use of hydrochloric acid: in her view, this had been used with the purpose, and certainly with the effect, of removing what would otherwise have been extensive evidence of modern tool marks. There was also evidence of modern machine polishing, and the application of an artificial modern deposit to the surface of the statue. The jagged shape of the walls and edges of the grooves of the decorative lines of the statute suggested that a modern high speed pendant drill had been used.

In her view, there was pitting visible over the entire sculpture, and this regular erosion/pitting was characteristic of the use of an acid etchant and a high-speed machine polisher. Examination of the petrographic samples and microprobe analyses confirmed the absence of any long-term natural weathering of the stone or of any biological activity at the surface, and this confirmed that constituents within the stone surface had been aggressively chemically dissolved, leading to the formation of pits with a shape characteristic of the use of an acid. She detected high levels of chlorine within the pits themselves, and chlorine in smaller amounts was detected throughout the flat machine polished areas surrounding each of the pits. The chlorine could not be explained by any natural weathering phenomenon, and in her view the high quantities of chlorine were derived from the use of hydrochloric acid with chloride salt residues trapped within the acid etched pits at the stone surface. Because the modern polishing could not access the pits, it was not able to polish away the chlorine rich chemically attacked layer of stone in these areas.

She considered that the statue was of modern manufacture, and that it had been broken prior to being chemically treated with hydrochloric acid in an attempt artificially to age the surface and to remove the modern tool marks. The work was then polished using a high-speed electric polishing wheel.

Electron microprobe analyses characterised the superficial artificial deposit at the stone surface as containing selenium and iron which could not derive from the weathering of the stone. The selenium within this deposit was likely derived from the use of a modern pigment, since selenium is rare in any other materials. Fragments of copper chloride and uncorroded copper within this layer further identified it as modern. The superficial artificial deposit was absent from the acid etched pits, which suggested that the selenium/ iron may be associated with the modern polishing in the final stages of manufacture.

At the heart of Dr Nicola’s conclusions, in both his first report and a second report produced after he had been able to see the statue (and other objects), was a critical point based on some discolouration which was visible, as shown by his samples, at or towards the surface of the stone. Dr Nicola described this discolouration as “chromatic variation”. He said that in his analysed samples, he observed a chromatic alteration of the stone, which changed in colour from grey-greenish in the interior to yellow/ red-brownish towards the surface. This phenomenon was commonly observable in natural weathering of this kind of stone and with similar sculptures of known authenticity from a slightly later period. The chromatic alteration was likely to have arisen from the oxidation of the iron present in the stone where it was closer to the surface as a result of the effect of external atmospheric conditions (for example, oxygen and water). In his opinion, this kind of weathering was “not easy” to simulate artificially.

Dr Nicola identified two possible mechanisms by which this natural weathering had occurred, although it was not possible conclusively to identify the cause of the phenomenon. It could be due to the process of oxidation described in a 36 page scholarly article by Wedekind and others: Damage phenomenon and petrophysical properties of sandstones at the Phnom Bakheng Temple (Angkor, Cambodia): first investigations and possible conservation measures. Environmental Earth Sciences 2018. 77 (21): pages 1 – 36. I shall refer to this as “Wedekind”. The mechanism involved was the transformation of chlorite and biotite, which are iron containing crystals within the sandstone. Dr Nicola also referred to the possibility that the stone had been exposed to humic acids, or of groundwater containing particular substances. In either case, there would be a process of weathering over a substantial period of time, and this explained the chromatic variation. Although Dr Nicola identified both possible causes, the primary focus of his evidence was on the process described by Wedekind. His report contained a photograph from the Wedekind article, alongside a photograph of one of Dr Nicola’s samples, and Dr Nicola’s conclusion was that they were similar because the processes described by Wedekind had operated in both cases.

Dr Nicola concluded that since the weathering profile of the samples showed a distinct chromatic alteration consistent with weathering over many centuries, the statue should be considered authentic.

This conclusion was the one area of the case where Dr Nicola’s evidence was that authenticity had positively been demonstrated. As discussed elsewhere, in relation to the other objects in issue in the litigation, Dr Nicola’s position was, generally speaking, that the evidence advanced did not prove inauthenticity, but was inconclusive. In relation to the Hari Hara itself, Dr Nicola’s evidence as to authenticity was based upon the chromatic variation, rather than upon any other distinct features of the evidence. He accepted that there was, in various respects, evidence of the statue having been subject to modern work. However, as with the other objects in issue in these proceedings, and as illustrated in the following paragraphs, he considered that the evidence of modern work was consistent with conservation and restoration of an ancient statue, and did not mean that there had been modern manufacture.

Thus, Dr Bennett had described an artificial deposit on the surface of the stone. Dr Nicola accepted that this was not a natural patina. The contact surface between the stone and the superficial layer was clear and not tightly bonded to the underlying stone. Dr Nicola accepted that this suggested that it was an “applied layer and not one naturally formed over a long period of time”. However, the presence of such deposits, and particular elements within them, could in Dr Nicola’s view be due to many common situations: for example, the presence of residue of soil or residue from past conservation treatments.

Dr Nicola did not detect significant amounts of chlorine in any of the samples that he analysed. In his view, however, even if there had been treatment with hydrochloric acid, its presence could not be taken as reliable evidence of modern production. The presence of high levels of chlorine could derive from common past conservation practices; for example, residues of products used for cleaning the surface.

In relation to weathering, Dr Nicola’s main and important point was that the chromatic alteration was the effect of weathering. He accepted that there was, on his samples, no direct signs of biological degradation or attack, which is a specific form of weathering that results from the interaction of organic material (such as lichens and mosses) with the stone. However, the absence of biological material or evident sign of biological degradation could not reliably demonstrate that an object was not ancient: for example, restoration or conservation treatments could have removed the presence of such organic material.

In relation to modern tools, the position at the time of Dr Nicola’s first report was that he had not inspected the statute, and his samples did not show any evident traces of modern tools. However, even if signs of modern tools were present, that would not in his view necessarily imply recent production. Such marks could well be due to past conservation treatments. In a case where, as here, the restoration history of the statue was unknown, any such signs could not reliably be taken to be an indicator of modern production.

Following his visual inspection in November 2021, Dr Nicola in his second report said that the statute had been “much more extensively restored than has previously been recognised between the parties or in the Claimants’ expert reports.” He said that there were mixed restored and unrestored zones across the item. His report reiterated his views as to discolouration being evidence of weathering. Insofar as there were points – for example the possible use of modern tools – that might suggest modern manufacture, Dr Nicola suggested that previous restorations or heavy conservation treatments could be responsible.

The views of both experts were then further set out in a lengthy (10 page, single spaced) joint report. Much of the joint report (and later the cross-examination and closing arguments) focused on the question of whether the discolouration or chromatic variation visible in Dr Nicola’s samples was the result of a natural weathering process. Dr Bennett was firmly of the view that it was not, with Dr Nicola taking a firm contrary position.

On other issues, there was a certain amount of common ground between them, but much remained in dispute. They agreed that the type of stone used for Khmer sculpture is still available today, and that further identification of the sandstone used was not critical to determining authenticity. They agreed that there was a thin superficial layer on the stone, but there was substantial disagreement both as to the nature and origin of this layer. They agreed that the stone surface was not entirely smooth, and that chlorine was used, but they disagreed on the amount present. There was disagreement as to the likely origin of this chlorine: Dr Bennett considered that her samples provided clear evidence of the use of hydrochloric acid, whereas Dr Nicola considered that it was likely derived naturally from crystallization of some of the most common salts contained in ground water or was due to contamination with sea salt from a marine or coastal environment.

There was a significant disagreement as to whether there was any evidence of natural weathering derived from long-term exposure to the environment or burial, with the experts expanding upon their views on the discolouration or chromatic variation issue described above. The experts were, however, agreed that there was no evidence of biological material or evident sign of biological degradation. Thus, there was no visible evidence of any microbiological colonisation by lichens, algae, fungi, bacteria or cyanobacteria which form biofilms on sandstone surfaces. There was also no evidence of any root marks, which are mineral deposits that accumulate along the paths of plant root surfaces during natural deposition. Dr Nicola reiterated his point, made in various contexts, that the absence of such material could not be taken as evidence that an object is not ancient; since its presence is dependent on the conservation history of a work, including any cleaning.

They were agreed that there was evidence of modern intervention to the stone surface, although they were not agreed as to the reason for, or the extent of, this modern working. Dr Bennett had identified the use of a modern pendant drill, and Dr Nicola did not dispute that this had been used. He suggested, however, that it had been used in areas of restoration/ recarving/ reinforcement. Dr Bennett also reiterated her evidence as to the use of a high speed polishing wheel. Dr Nicola identified at least one area where such a polishing wheel had possibly been used. He acknowledged that this was not a recommended tool in modern conservation practice, but said that it had been used widely in the recent past especially by untrained restorers.

Before examining the detail of the evidence concerning the chromatic variation issue, I consider it important to look at the other evidence relating to (in)authenticity, as part of the overall picture. If that other evidence pointed clearly towards inauthenticity, then the evidence as to chromatic variation would in my view need to be very clear and powerful in order to outweigh the other evidence. Equally, if the other evidence was more equivocal, then it would be more likely that inauthenticity could be disproved by the point made on chromatic variation. As appears below, I consider that the other evidence in the case does point clearly towards inauthenticity, and that there is nothing in the case on chromatic variation which outweighs that evidence. Indeed, I was ultimately not persuaded by Dr Nicola’s chromatic variation point at all, and I prefer the evidence of Dr Bennett. However, I begin with the other evidence in the case.

First, I have already concluded that the art historical evidence shows, on the balance of probabilities, that the Hari Hara is a modern forgery. As with the other pieces in issue in these proceedings, this conclusion provides the answer as to whether the evidence of modern working on the Hari Hara (in particular the applied deposit, the modern tool marks, and the polish applied by a modern high speed polisher) is the result of the process of manufacture, or restoration. I accept that my conclusion on the art historical evidence would need to be revisited if the case on chromatic variation (i.e. that this showed an ancient object) was clear and powerful. But in my view, that is far from the case here.

Secondly, as discussed above, there is a complete absence of evidence as to provenance or the history of what would otherwise be a very important piece. Again, this is evidence of inauthenticity, although I again accept that this would potentially be affected by the strength of the case on chromatic variation.

Thirdly, in relation to all of the objects, I reject the case, advanced by Dr Nicola, that the evidence of modern working can be satisfactorily explained by restoration or conservation. I reach the same conclusion in relation to the Hari Hara, and there is in my view a particularly strong reason for doing so here; namely the evidence that modern work, or at least an important aspect of it, was performed on a statue which was and remained in pieces.

Dr Bennett’s evidence was that the polishing had taken place when the statue was in pieces. She could see that this was so because the high-speed polishing machine had polished some small part of the break edges of the pieces. This was because, as she explained, it is not easy to control the machine. Her evidence in that regard was supported by a photograph which she had exhibited to her court report. She was also, clearly, a very observant and experienced expert, and I see no reason at all to doubt her observations as to the polish on the break edges. Indeed, she was not cross-examined on the basis that she was mistaken. Although Dr Nicola, when asked about this, said that it was difficult to tell; his written evidence seemed to me to support Dr Bennett. In his first report, Dr Nicola said that he understood that the item was “purchased in a polished state”. He was correct in his understanding, and I was not impressed by his tentative effort in cross-examination to suggest that the polishing may have been carried out by Neil Perry-Smith. Furthermore, in his second report, he said:

“There are very small indications of polish in some places at the intersection where the arm meets the break… In my view it is very likely that the polishing process which left these traces of polish took place after the break”.

In my view, this evidence makes it inherently unlikely that a genuine process of restoration or conservation was being applied to the Hari Hara. If the Hari Hara was being restored, then one would have expected the statue to be re-assembled, and that the polishing would take place as the final stage of the process. There is, in my view, no satisfactory or plausible explanation as to why a restoration should take place, with the individual pieces being polished, but the statue being allowed to remain in pieces. It is in my view a somewhat extraordinary restoration for a restorer to carry out the work that is evidenced on the statue – modern tools, the application of a deposit, and then a high speed polish – but nevertheless to leave the statue disassembled. In my view, it is inherently probable that, as Dr Bennett explained in her evidence based on what she had seen in South East Asia some years ago, modern forgers had made the statue, broken it, and then applied various techniques in order to give it the appearance of age and/or remove the signs of modern manufacture.

This conclusion is in my view reinforced by the fact that, as was undisputed, a restoration of an ancient statue should not in modern conservation practice involve the application of machine polishing at all. Dr Bennett was able to point to evidence that a modern high-speed polisher had indeed been used. Indeed, Dr Nicola did not ultimately dispute that this was, or at least might have been, the case. I accept that the statue was machine polished. This is in itself some evidence that what happened here was not the restoration of an ancient statue. Machine polishing of the type observed on the statue should not be used, as a restoration technique, on an ancient stone statue of this kind: it would remove and damage the original surface, which is in contradiction to the aims of conservation/ restoration practice. Whilst a possible explanation is that it reflected very poor conservation practice in the past, the far more probable explanation is that it was part of the modern manufacturing process that created this statue.

Fourth, it is in my view improbable that the only (alleged) sign of weathering on the Hari Hara, which was otherwise in very good condition is the chromatic variation or discolouration relied upon by Dr Nicola. The experts were agreed that, in relation to the Hari Hara, there was no visible evidence of any microbiological colonisation by lichens, algae, fungi, bacteria and cyanobacteria which form biofilms on sandstone surfaces. Nor were there any root marks.

Dr Bennett’s evidence was that one would expect to see very much more, if the statue was indeed ancient. That evidence was persuasive. I accept Dr Bennett’s obvious point that microbiological colonisation – for example by lichens, algae, or fungi bacteria – might be expected on an ancient sculpture which had weathered under natural conditions in a monsoon influenced tropical climate typical of mainland Southeast Asia. That point applies with particular force to Dr Nicola’s suggestion (albeit that he did not focus on this) that the discolouration may have occurred as a result of the entry into the stone of humic acids. Those are, as Dr Nicola explained, acids which are contained in the ground from decomposing material. It would therefore be very surprising if this had happened, but there was a complete absence of microbiological colonisation.

I do not consider that the absence of such evidence of these matters can be explained, with any degree of conviction, by conservation and restoration. As Dr Bennett explained: if a strong acid had been applied to stone in order to remove all biological material, then this could not have left the surface fully intact. Here, there were very sharp areas of decoration, and they would not have survived in this form if acid had been applied to remove all biological material. The acid would indiscriminately remove both biological material and the areas of decoration. Dr Bennett’s basic conclusion was that there was no biological material which needed to be removed, and that was because this was not an ancient statue. As she said in evidence, and I accept, there was here the original surface of the stone, and the stone was never encrusted in deposits which required removal but nevertheless left the surface of the stone basically unaltered.

Fifth, Dr Bennett made what seemed to me to be important points in relation to the use of modern tools. Dr Nicola acknowledged in his report, and his oral evidence, that this was not really his area of expertise. In contrast, Dr Bennett explained in her evidence that she had spent 40 years working in museums, and had spent many periods looking exactly at tool marks in order to trace ancient workshops, and to the extent possible, even to tell when the tool was sharpened. The study of tool marks was something that she had spent years doing. The use of silicon rubber moulds (which clearly impressed Dr Nicola) enabled an expert such as Dr Bennett to look at tool marks including tool marks used in ancient workshops.

In relation to the use of modern tools, Dr Bennett identified on the statue the tool signature of a modern pendant drill. This resulted in a systematic and regular pattern both within the groove and at the edges of the groove. She had confirmed this by taking silicon rubber impressions of the tool marks. The grooves on the headdress display the characteristic pattern of an electric pendant drill. There was also the characteristic tool signature of a pendant drill on the sampot, the eyes, and the third eye. In her view, because the whole work had been treated with acid, the tool marks were not as crisp as they would have been at the time of manufacture.

I accept that the use of modern tools is not conclusive, because it is theoretically possible that there was at some stage a restoration. However, there is (as discussed above) no evidence of any restoration prior to sale, and I have already referred to the improbability of the modern work on the statue being evidence of restoration. Dr Bennett’s view was that the tool marks reflected modern manufacture. She made the telling point that there was no original wear on the object: the wear that was on there was over the tool mark. This was not therefore a case of the object having been worn, and then the tool mark being introduced into a worn surface. She was also able to say that there was no evidence of any reworking of ancient tool marks. Had this happened, she would have seen evidence of the ancient marks, and this might have led to the conclusion that they were not sharp or well-preserved enough. However, there was, as she said, “absolutely no evidence of that”.

Against the background of these matters clearly indicating inauthenticity, I turn to the evidence that the discolouration or chromatic variation requires a different conclusion.

There was no dispute that, on Dr Nicola’s samples taken from the statue, there was some discolouration evident on the portion of the sample that represented the part that was at or close to the surface of the stone. Dr Nicola ascribed this to weathering, as evident on similar statues. Dr Bennett’s opinion was that the colour was likely to be the result of a liquid artificial surface treatment which had penetrated via cracks in the surface of the stone, albeit that she was not able to identify what liquid had been used. Having reviewed the written and oral evidence in detail, I am unable to draw the conclusion that the discolouration was the result of weathering. In my view, the evidence that this was the cause is not established on the balance of probabilities. On the contrary, I consider it more likely than not that this discolouration was indeed the result of liquid artificial surface treatment, as proposed by Dr Bennett. Whilst I accept the evidence of Dr Bennett generally on this area of the case, it seemed to me that the following points were significant.

First, Dr Nicola’s evidence on this issue primarily rests on the comparison which he sought to draw between the discolouration on his samples and the discolouration which can be seen in one of the photographs in the Wedekind article. However, the Wedekind article – which is a very scholarly piece work running to 36 pages – describes something which goes well beyond discolouration. The article is, as its title suggests, concerned with the need for conservation of the Phnom Bakheng Temple at Angkor, Cambodia. (It has nothing to do with authenticity). Dr Nicola correctly identified the fact that the Hari Hara sandstone is the same as the problematic sandstone at Phnom Bakheng, and hence the potential relevance of the Wedekind article. The need for conservation at the temple arose because there were natural processes of weathering underway which led to “deterioration and damage”. This was illustrated by photographs which gave examples of five different conditions leading to deterioration and damage. Dr Bennett’s evidence was that Wedekind was describing something which led to a friable crust and significant damage. This seems to me to be an important aspect of the problem that Wedekind was describing, and in respect of which conservation measures were required. The article refers in various places to problems leading to detachment. For example, the authors write in their conclusions:

“Previous studies suggest that hydric dilatation is the reason for contour scaling at the Angkor monuments. This study hypothesizes chemical weathering by the precipitation of iron and calcite at the surface combined with insolation-affected thermal dilatation as a probable cause for the contour scaling identified. The precipitation and crystallization of the case-hardening agents (iron and calcite) throughout the surface rind is a probable formation mechanism for the crust. Insolation and the increase of thermal dilatation due to the accumulation of calcite lead to extensional stresses and cracking within the crust zone. Hydric dilatation is high in the largely unweathered original PB sandstone behind the crust zone. Both the thermal expansion of the crust and the hydric dilatation of the stone material behind the crust behave in a contrary fashion. This causes extensional stresses and a detachment of the crust that leads to the weathering form of scaling.”

As Dr Bennett explained in her oral evidence, Wedekind had gone through an extensive analysis of what the material was, how the stone had altered, and how a crust had been formed. There had been a volume change (because the iron oxide produced by the weathering was bigger than the iron in the crystals), and therefore an alteration and the crust had become hard and brittle, with the outer dark brown crust tending to fall off. The article was very concerning in relation to conservation at Angkor. She summarised the position, accurately in my view, as follows:

“… his argument is that, because you’ve got this brown crust and because the iron has oxidised and you’ve got this increase in volume, the crust no longer adheres to the underlying stone and the volume change causes it to fall off the surface, and that’s what he’s concerned with in trying to conserve. That’s the basis of his article”.

However, in the case of the Hari Hara, there is evidence of discolouration on some of Dr Nicola’s samples (and indeed the places on the statue from which those samples were taken), but there is nothing to suggest that there is any friable crust, or parts of the statue that are becoming detached, or even that the statue is in need of some restoration, let alone a restoration equivalent to that being considered by Wedekind. On the contrary, the Hari Hara appeared to be in very good condition, as noted in 2018 by Neil Perry-Smith.

Secondly, the authors of the Wedekind article carefully analysed what was happening to the crystals which made up the sandstone. This sandstone includes various crystals including feldspars which make up approximately 30% of the total. Contrary to some of the closing submissions of the Defendants, Dr Bennett had identified the figure of 30% in her first report, and had discussed there the process of weathering. She said that these feldspar crystals were “particularly susceptible to hydrolysis, a form of chemical weathering by which water/ moisture interacts with the feldspar breaking it down to residual clays and silica”. The Wedekind article also discusses feldspars, and indeed other crystals, in some detail, with blown up photographs identifying different crystals. One photograph shows “altered feldspars”. The text of the article says that a large percentage of the feldspars (which have a “common cross-hatched pattern”) are weathered. The feldspars showed “varying degrees of alteration from fresh to moderately weathered ranging 20 – 50%”. The authors also describe how, under transmitted light, the “framework grains are highly broken (fractured) giving the grain fabric a shattered appearance”. This was illustrated by another photograph.

Dr Nicola’s report, and evidence, did not seem to me to demonstrate that the crystals in the Hari Hara had shown anything comparable to the process described, and illustrated, in Wedekind. Indeed, it became clear in cross-examination that the chromatic alteration theory, and the comparison to Wedekind, was something that Dr Nicola’s geologist colleague had identified. Dr Nicola had agreed with it, and adopted it. But Dr Nicola was unable, as he explained, to answer any detailed questions about the crystal structure shown in his photographs, because he was not himself a geologist. For example, he could not even identify a feldspar.

By contrast, Dr Bennett did identify, in Dr Nicola’s samples and photographs, where he was relying upon the colour change, examples of feldspars. Her identification seemed to me to be consistent with the description given in Wedekind. Her point was that the feldspars were intact and “there is absolutely no deterioration to them whatsoever”. I did not think that Dr Nicola provided a good answer to this point. In cross-examination, his evidence was that he could not identify, on his photographs, the distinct feldspar crystals: he was not comfortable doing so, because he was a chemist not a geologist.

His principal response, however, was that weathering was complex and that he did not feel comfortable to state that a mineral “is the one that should be gone first and then the other”. I did not find this a persuasive answer. The impact of weathering on the feldspars is clearly discussed in the Wedekind paper. Dr Nicola was drawing on that paper for the purposes of his point on discolouration, and was saying that the same phenomenon could be seen in the Hari Hara. There is no reason to think that the phenomenon of feldspar weathering, as discussed in the Wedekind paper, should be any different when it came to the Hari Hara. I considered Dr Bennett’s evidence on this more persuasive, particularly bearing in mind the fact that she had made, in her initial report, the point that feldspars were particularly susceptible to hydrolysis.

Dr Bennett had further points concerning the comparison with Wedekind that Dr Nicola had made, and the composition of the crystals within the sandstone. She said that Wedekind had talked very clearly about the iron particles, which would have to alter in order to produce the weathering on the surface. This was because the iron had to start from somewhere. However, Dr Bennett said that Dr Nicola’s samples showed that the biotite was completely unaltered. She also referred to Wedekind’s discussion of calcium dissolution and calcium reprecipitation, for which there was no evidence in Dr Nicola’s samples. It seemed to me that these were powerful points which reinforced my conclusion that Dr Nicola had not shown that the phenomenon described in Wedekind was the same as whatever had caused the discolouration shown in his samples. I do not accept the Defendants’ submission, in their closing argument, that this was not a reasoned analysis which met the argument that Dr Nicola had advanced.

Thirdly, Dr Bennett was insistent that the discolouration could not be seen on the samples that she had taken. There were some scan images (at Figures 4.6 – 4.7) of sections of the samples in her report. It did not seem to me that these showed discolouration similar to that which was evident on Dr Nicola’s samples. The Defendants submitted that her evidence as to what her samples showed was not satisfactory, because she had not produced them or, apparently, looked back at them after Dr Nicola had raised his discolouration argument. If the samples had been critical to Dr Bennett’s evidence, I could see the force in this submission – although I also note that the Defendants could have asked to see her samples, particularly after she had made the point (as she did in paragraph 4.6 of the joint report) that her samples did not show discolouration.

However, I was able to see – and Dr Bennett was cross-examined upon – whether discolouration was visible in the places on the statue from which her samples were taken. Dr Bennett was cross-examined by Ms Renton on that topic whilst the statue was examined in court, with the lights turned down and with the benefit of torchlight. She was insistent that it was not visible, and that her sample areas looked very different to those of Dr Nicola. I was helpfully provided, during the Claimants’ closing, with the relevant transcript accompanied by photographs of the areas being discussed in the cross-examination. I consider that those photographs do bear out the point that Dr Bennett was making, as to the difference between her sample areas and those of Dr Nicola.

Dr Bennett also explained, persuasively in my view, the likely reason why the discolouration could be seen on Dr Nicola’s samples (and sample sites), but not on hers, was the way in which samples had been taken. Dr Nicola’s had been taken with a chisel. This would remove a piece which, because of micro-cracking, would fracture along the line of least resistance. In contrast, her samples had been taken differently, using a cutting-wheel, which would not cut along a line of least resistance. If the discoloration had been the result of weathering, with iron moving towards the surface of the stone from its interior, one would expect that phenomenon to be visible on all the samples. She explained, again convincingly in my view, that the micro-cracking explains why the discolouration was seen on Dr Nicola’s samples or sample sites, but not on her own. The discolouration or staining would have to come from somewhere: it came in via the micro-cracks, otherwise it would not have got into the stone. Those areas were the weakest areas of the stone and hence permitted the ingress of the liquid that resulted in the staining.

The significance of this point is that the discolouration relied upon by Dr Nicola cannot be seen generally at the surface of the stone. I accept the logical point made by Dr Bennett that if the cause of the discolouration was the weathering described by Dr Nicola, then one would expect to see it generally at the surface of the stone, including in Dr Bennett’s samples, and at all the sample sites on the Hari Hara.

Accordingly, I do not accept Dr Nicola’s theory that discolouration demonstrates that the statue is ancient. I consider it more likely than not that the discolouration is the result of the introduction of an external liquid, albeit that Dr Bennett could not identify precisely what had been introduced. However, it is clearly possible, and in my view probable, that the statue had been subject to liquid artificial treatment, bearing in mind all the evidence which I have discussed, including the evidence of significant modern working on the statue and the improbability that this was restoration work. It is in my view no answer to this argument that Dr Bennett had not identified the liquid, or conducted tests to identify it. In a case where there has been a modern forgery, with obviously a careful and determined effort by the forger to pass off the item as genuine, it is unlikely that the full story of exactly what has been done will be known or capable of being revealed. Ultimately, I must look at all the evidence. In my view, when looking at art history, provenance, or materials science – both separately and in the round – I am satisfied that the Hari Hara is not authentic.

5 of the 7 pieces at issue in the present proceedings (the Head of Dionysus, the Head of a Goddess, the Head of a Bodhisattva, the Krodha, and the Frieze) are referred to in JEL’s invoices as associated with “Bactria” or “Gandhara”. These pieces are said to date from various times between the 2^nd – 6^th centuries CE.

Before dealing with the detail of those objects, this section introduces the geography and history of those periods during the time period relevant to the disputes between the parties. This section is based on the evidence of the Claimants’ experts, whose evidence in this regard was in my view reliable, and was to a very large extent not the subject of substantial challenge.

The ancient place names “Bactria” and Gandhara” indicate two distinct regions (see the map below). Bactria corresponds to northern Afghanistan and the southern parts of Uzbekistan and Tajikistan. Gandhara has its nucleus in the Peshawar Valley (North-Pakistan), but includes (in the modern archaeological and art-historical understanding) the neighbouring areas in today’s Afghanistan (south of the mountain range of the Hindu Kush) and Pakistani Panjab, to form what scholars currently name “Greater Gandhara”. Bactria belonged to Central Asia, whereas Gandhara represented the North Western periphery of the Indian Subcontinent.

There was a debate in the written expert evidence as to whether the term “Greater Gandhara” should include a more extensive area. Dr Jager used the expression “Greater Gandhara” to describe a region that stretched from the borders of Iran into the Ganges valley in India, during the period of Kushan rule described below. Professor Lo Muzio (whose evidence generally I prefer, for reasons explained elsewhere) considered that the term should be limited to the territory from the southern foothills of the Hindu Kush, in Afghanistan, to the Pakistani Punjab, and also that it excludes Bactria. This approach was also consistent with the evidence of Dr Luczanits. He described the core area of Greater Gandhara or cultural Gandhara as being the Peshawar basin in Northern Pakistan, and Bactria as being the region north of the Hindu Kush mountains.

The map below is based upon Professor Lo Muzio’s evidence. I acknowledge, however, that there is room for argument as to what territory was covered by “Greater Gandhara”. Ultimately, I did not think that this was a significant issue that needed to be resolved. In addition, as Dr Luczanits said in his evidence, different people use the expression Gandhara to refer to different periods. Usually, the phrase is associated with the Kushan period, from the 1^st to 3^rd centuries. However, some people extend that to the 5^th century.

In the late 4th century BCE Bactria became part of the Greek (Macedonian) empire created by Alexander the Great. In the mid-3rd century BCE, Bactria became an independent Greek polity, named by the scholars “Greco-Bactrian kingdom”.

Apart from a rich collection of numismatic finds (coinage of the Seleucid period, 300-245 BCE circa, and the subsequent Greco-Bactrian period, 245-145 BCE circa), the archaeological record on the Greek period in Bactria is represented by a major urban site, Ai Khanum (north-eastern Afghanistan, circa 280-145 BCE). Ai Khanum has revealed an impressive picture of the topography, defensive system, architecture (public, domestic, religious and funerary) of a Greek settlement in the heart of Central Asia, and on the extent to which the Macedonian elite came to terms with the local traditions. Beside structures which are patently Greek (the gymnasium, the urban mausoleums, the theatre), Ai Khanum bears witness to an Oriental architectural tradition (the temples, the large complex known as “palace” as well as the treasury included in it).

A few minor sites bearing evidence of the Greek presence, but much less informative than Ai Khanum, have been excavated in Uzbekistan. In the 2000s, sporadic Greek architectural findings have been brought to light in Balkh (the ancient Greek capital, Bactra), in northern Afghanistan.

In the early 2nd century BCE, the Bactrian Greeks extended their power to Gandhara and the neighbouring areas. These territories became a separate Greek polity (or several Greek polities), known as “Indo-Greek Kingdom(s)”, documented by numismatic finds throughout the 1st century BCE and, to a much more limited extent, as late as first decades of the 1st century CE.

The archaeological evidence of the presence of the Greeks in the Gandharan area (other than numismatic), datable to the period they ruled the area, is, in comparison with Greek Bactria, still elusive or, in any case, not as straightforward as at Ai Khanum. A Greek foundation has been proposed for Sirkap, one of the sites in the Taxila area (Pakistani Panjab), but the material evidence supporting this hypothesis is now debated. Mentions of Greek settlements (or garrisons) in this region, scattered in Western sources, still wait to be corroborated by firm archaeological proof.

The Greek legacy left more substantial and recognisable traces in the artistic production flourishing in this region. This, however, dates from the very end of the Greek period or well after the fall of the Greek kingdom. Apart from Gandharan art, so called “toilet-trays” have been found. These small stone artifacts possibly date from the 1st century BCE to the 1st century CE. Their function is not yet clear, but they provide evidence of a repertoire of objects which are partly influenced by the Greek imagery (mythological subjects and cult scenes). The number of specimens revealing a “cleaner” Greek inspiration (subject, style and workmanship) is relatively limited, the largest part of the repertoire denoting a more or less advanced degree of hybridisation.

The Greco-Bactrian kingdom came to an end in the mid-2nd century BCE, partly as a consequence of successive attacks launched by nomadic tribes who came from the north, among which the Sakas and, slightly later, the Yuezhi (as they are named in the Chinese historical sources). One of the Yuezhi tribes, the Guishuang, imposed its hegemony on all the others, and, during the 1st century BCE, laid the foundation of a kingdom. Starting from Bactria, this kingdom included, by the 1st century CE, the Gandharan area and a large part of northern India. This was the Kushan empire (1st-3rd centuries CE), in which Bactria, Gandhara and Northern India were ruled by one and the same dynasty.

The Kushan rule represents an important chapter in the political, religious and cultural history of Central Asia and of ancient India. It sparked a flowering of Sanskrit literature and encouraged the dissemination of Buddhism and Buddhist figural art. Canonical depictions of the Buddha in anthropomorphic form were developed for the first time in the Kushan period, in the main artistic centres of the time – Mathurā and Gandhara – and thence disseminated into the remainder of the subcontinent as well as Central Asia, along with a narrative repertoire connected with stories from the life of the Buddha.

As far as Bactria is concerned, the period between the fall of the Greek kingdom (mid-2nd century BCE) and the rise of the Kushan empire (1st century CE) is, from an archaeological viewpoint, poorly documented. The only major exception is a Temple on the Oxus river (today Amudarya), in Tajikistan, dedicated to a river god (or goddess).

For the Kushan period (1st to 3rd centuries CE), archaeological diggings have brought to light urban settlements, free-standing monumental buildings (Khalchayan, in Uzbekistan) and shrines (Surkh Kotal, in Afghanistan) of dynastic patronage.

Of the greatest importance, for architecture, epigraphy and pictorial art, are the Buddhist monasteries and single urban Buddhist shrines excavated in Uzbekistan, in the Termez oasis (Kara Tepe, Fayaz Tepe), and at Dalverzin Tepe. In addition to the materials unearthed in these sites, there are sporadic minor findings associated with the Buddhist cult. Throughout the Buddhist Bactrian sites of the Kushan period, the Gandharan component (in architectural forms and figurative arts) is predominant.

In Kushan Bactria, there is no evidence of sculptures or mural paintings or other artifacts of reliable provenance that can be associated with the practice of Greek cults in loco. There is also no evidence of the persistence of a Greek or Hellenised elite, to which could be attributed the patronage of religious buildings that may have hosted large size marble sculptures of Greek gods or goddesses. Furthermore, in the Kushan period, there is no evidence of marble sculptures altogether. These are matters which are particularly relevant to the issues in this case concerning the two marble sculptures, the Head of Dionysus and the Head of Goddess.

In Bactria, fragments of Hellenistic stone sculpture are known at Ai Khanum only (North-Eastern Afghanistan, early 3rd to mid-2nd centuries BCE). Almost all the few unearthed specimens are worked in limestone, not in marble. A well-known exception is a marble foot wearing a sandal, once belonging to a large size sculpture, most probably an acrolith, representing Zeus. No further acroliths are recorded in Bactria, either in the Greek or in the subsequent periods.

In the following centuries (including the 2nd to 4th centuries), only rare Bactrian examples of stone (limestone) sculptures are recorded, and they mostly belong to Buddhist contexts. Fragments of limestone sculpture were also found in the Kushan dynastic shrine at Surkh Kotal (Afghanistan). These stone examples apart, the bulk of Bactrian sculpture was modelled in clay and/or gypsum.

During the Kushan period, Gandhara was a thriving and influential artistic province and a prestigious centre of Buddhism. The Hellenistic component is a major factor in the genesis of Gandharan art. There is a rich corpus of sculptures, mainly reliefs, which once decorated religious monuments in the Buddhist monastic settlements of the region. Gandharan sculpture is a Buddhist art: the subjects are largely determined by the Buddhist narrative and doctrinal tradition. Themes with no apparent relationship with Buddhism are illustrated in marginal spaces. The Greek legacy mostly impacts on style and workmanship (naturalistic rendering of physiognomic traits, postures, clothes), in the relationship of figures with space, and in the architectural frame of the reliefs, which usually employs forms and ornaments borrowed from the Hellenistic repertoire.

Gandharan art has been popular with Western academics and collectors from the time of its discovery in the 19^th century. This is substantially due to the aesthetic links to the Greek and Roman worlds. The western elements found in Gandharan art can be interpreted as the long-term consequence of the 4^th century BCE conquests of Alexander the Great, the subsequent establishment of a dominion of Greek culture in Bactria, and the subsequent spread of ruling elites into the Gandharan area, all as described above.

The flourishing of Gandharan sculpture in stone (schist and phyllite, which is a stone closely related to schist) covers the period of the Kushan rule, 1st to 3rd centuries CE. Accordingly, its inception postdates the Indo-Greek kingdom described above. In the following centuries sculptures are largely worked in clay and/or stucco. Sculptures (friezes, cornices, reliefs, stelae) are worked in several varieties of schist and, in certain areas, phyllite. Sculptures worked in limestone are sporadically recorded. In Gandhara or Greater Gandhara, no marble sculptures are recorded from archaeological excavations or, in general, from documented or reliable contexts.

Along with stone sculptures, the excavations of Gandharan Buddhist monasteries have yielded a large number of clay and/or stucco sculptures, modelled on a core made of clay and straw applied on a wooden “skeleton”. Large size images were modelled on a core containing stone rubble or even bricks. The clay and/or stucco production is much better documented in the period following the end of the Kushan dynasty, i.e. from the fourth to the sixth-seventh centuries CE; in certain areas south of the Hindu Kush (Afghanistan), well into the eighth century.

The Head of Dionysus was the larger of two marble sculptures in excellent condition, the other being referred to as “Head of Goddess”. It is a work of substantial size. According to the invoice to QIPCO, its dimensions were 48.5 cm (height), 36 cm (width) and 39 cm (depth). There were a number of gemstone inlays in a diadem (or band) above Dionysus’ forehead. The gemstones were garnets.

Professor Lo Muzio said that he suspected that both of these sculptures were carved in one and the same workshop. This was suggested by elements of workmanship and style. For example, both heads showed ornaments enriched with garnet inlays: the diadem on the Dionysus, and various features on the Head of Goddess. Both sculptures also contained gemstone inlays in the pupils of the eyes as well. According to Professor Lo Muzio, all of the various ornaments show rough finishing. In addition, both heads appeared to have essentially the same workmanship in the treatment of wavy tufts of hair, in the female head, and in the beard of Dionysus. Dr Jager thought that the similarities in the two pieces made it likely that they did indeed come from the same workshop. I consider that it is indeed likely that they did so. It follows that if one of the pieces is proved, on the balance of probabilities, to be inauthentic, the same conclusion follows in relation to the other.

Dr Jager’s view was that the Head of Dionysus formed part of an acrolithic sculpture. Similarly, he considered that the Head of Goddess was conceived as part of an acrolithic sculpture, as indicated by a rectangular recess at the back of her head. An acrolith is a large-scale full length sculpture where certain features, such as the head and feet, are made from an expensive stone such as marble, but other features are made from much cheaper materials such as wood.

Neither the Head of Dionysus, nor the Head of Goddess, had any clear provenance. It was common ground that they were not referred to as having been unearthed in any archaeological excavation of which records have been maintained. Apart from the transfer of ownership from a company called FEAL to JEL, and Mr Eskenazi’s evidence that they had been in the private collection of Mr Hubbard for some time and that FEAL was a company owned by an Italian dealer Mr Vigorelli, nothing is known of the origins of the two pieces; for example, their ownership during the substantial period of time from around the 2^nd century CE (the date referred to in JEL’s invoice) until their acquisition by the individual who sold them to JEL.

There was no significant dispute that the sculpture referred to as the Head of Dionysus did purport to show Dionysus. One reason for this was the presence of grape bunches on the temples of the head. Dionysus, the god of nature, of wine and drunkenness, is one of the Greek gods, possibly one of the earliest. In the Roman Empire, he was known as Bacchus. The grape bunches on the temples were of a small size, as Professor Lo Muzio pointed out. He made clear in cross-examination, however, that this was not a significant point. I shall therefore proceed on the basis that the sculpture did indeed purport to be Dionysus.

In broad summary, Professor Lo Muzio made the following key points in his written evidence.

Bactrian archaeological and art historical record. First, he considered the archaeological and art historical record in the Kushan period relating to Bactria, which was alleged to be the geographical origin of the piece. The piece was said to date from around the 2^nd century CE, which falls squarely within the Kushan period. The Greek kingdom had fallen in the 2^nd century BCE, and the Kushan empire had risen in the 1^st century CE. The Kushan period was the 1^st to 3^rd centuries CE. However, in Kushan Bactria, there is no evidence of sculptures or mural paintings or other artifacts of reliable provenance which could be associated with the practice of Greek cults. More specifically, there was no evidence of a Greek or Hellenised elite to which could be attributed the patronage of religious buildings that may have hosted large size marble sculptures of Greek gods or goddesses. Furthermore, in the Kushan period in relation to Bactria, there is no evidence of marble sculptures altogether.

Bactrian materials and technique. Secondly, he considered the materials and technique which were used in Bactria. In Bactria, fragments of Hellenistic stone sculpture are known at the site at Ai Khanum. This is a major urban site providing an important archaeological record on the Greek period in Bactria, dating from around 280 – 145 BCE (i.e. some centuries before the Head of Dionysus). Almost all of the unearthed specimens from that period are worked in limestone, not in marble. There is a well-known exception: a marble foot wearing a sandal, once belonging to a large size sculpture, most probably an acrolith, representing Zeus. However, no further acroliths are recorded in Bactria, either in the Greek or in the subsequent periods.

In the following centuries, including the 2^nd – 4^th centuries, only rare Bactrian examples of stone (limestone) sculptures are recorded and they belong mostly to Buddhist contexts. Fragments of limestone sculpture have been found in a Kushan dynastic shrine, but these are clearly unrelated to the marble heads in issue in the present case. Apart from these stone sculptures, the bulk of Bactrian sculpture was modelled in clay and/or gypsum.

Accordingly, as far as the archaeological record is the period around the 2^nd century in Bactria is concerned, there is simply nothing in marble, let alone anything comparable to the sculptures in issue.

Gandharan archaeological and art historical record. Thirdly, Professor Lo Muzio considered the archaeological and art historical record, and the materials and technique, in Gandhara. Gandhara is, possibly, more relevant to the Head of a Goddess, which was referred to in JEL’s invoice as: “Greater Gandhara region; Late Bactrian period, 2^nd/3^rd century”, whereas Bactria is referred to in the invoice for the Head of Dionysus. Nevertheless, because of the reliance placed by JEL and Dr Jager on Gandharan art in relation to the Head of Dionysus – and also because the important issue in the case is whether the Head of Dionysus was ancient, rather than whether it came from Bactria or Gandhara – it is appropriate to summarise Professor Lo Muzio’s evidence on Gandhara here.

During the Kushan period, Gandhara was a thriving and influential artistic province, and a prestigious centre of Buddhist “doctrinal elaboration and irradiation”. The Hellenistic component is a major factor in the genesis of Gandharan art, as can be seen in a rich corpus of sculptures, mainly reliefs, once decorating religious monuments in the Buddhist monastic settlements of the region. Gandharan sculpture is a Buddhist art, with the subjects largely determined by the Buddhist narrative and doctrinal tradition. Themes with no apparent relationship with Buddhism are illustrated in marginal spaces. The Greek legacy mostly impacts on style and workmanship (such as the naturalistic rendering of physiognomic traits), in the relationship of figures with space, and in the architectural frame of reliefs. The flourishing of Gandharan sculpture in stone (schist and phyllite) covers the period of the Kushan rule, 1^st to 3^rd centuries CE. In the following centuries, sculptures are largely worked in clay and/or stucco.

Gandharan materials and technique. Professor Lo Muzio’s evidence was that sculptures (friezes, cornices, reliefs, stelae) are worked in several varieties of schist and in certain areas, phyllite, a stone closely related to schist. Sculptures worked in limestone are sporadically recorded. In Gandhara or Greater Gandhara, no marble sculptures are recorded from archaeological excavations or, in general, from documented or reliable contexts.

Along with stone sculptures, excavations of Buddhist monasteries have yielded a large number of clay and/or stucco sculptures. Production of these sculptures is much better documented in the period following the end of the Kushan dynasty, i.e. from the 4^th to 7^th centuries CE. Terracotta sculptures yielded by certain archaeological excavations in Greater Gandhara proved to be clay sculptures accidentally burned by a fire. Professor Lo Muzio considered that the authenticity of alleged Gandharan terracottas of unknown provenance had to be considered with the greatest caution.

Iconography. Fourth, Professor Lo Muzio referred to the iconography of the Head of Dionysus. As previously described, Bactria was identified in the JEL invoice as being the geographical region from which the sculpture was said to emanate. Bactria was not, however, within or the same as Gandhara or Greater Gandhara, as illustrated by the map in Section F above. Professor Lo Muzio therefore questioned the validity of an analysis of the authenticity of a Bactrian piece by reference to the art of Gandhara.

Even assuming, however, that reference could be made to Gandharan art, he said that there was nothing more than a generic association between any cult of worshipping Dionysus (or Dionysism) and Gandharan art. In Gandharan art, drinking scenes often accompanied by musicians and dancers are not uncommon. However, they occur on the fringes of the major narratives dictated by Buddhism. They can include figures dressed in Hellenistic or Iranian clothes, playing on western musical instruments or figures in Indian dress, playing on local instruments. This was what he described as a “generic” association. The key point, however, was that representations of true Dionysiac processions, and/or figures identifiable as Dionysus, either in reliefs or large size sculptures, or fragments of sculptures, are unrecorded in both Bactrian and Gandharan materials of ascertained provenance.

Facial and other features. Professor Lo Muzio said that the physiognomic traits, in particular the long hooked nose and the frowning countenance of the sculpture, were inconsistent with reliable Bactrian, Gandharan or Greater Gandharan works, both in stone and in clay/ stucco sculpture. Although accepting that he was not a classicist, he also said that in classical art Dionysus was not expected to have a frowning countenance.

The diadem, embellished with a surprising number of gemstones, shows an elaborate design for which he knew no parallel in Bactrian or Gandharan art works of reliable provenance. In cross-examination, he said that he would not describe the diadem, with its gemstones, as a rare phenomenon: “I think I never saw anything comparable to this”. Despite its rich ornamentation, the diadem denotes rough workmanship; the frames and bezels were inaccurately executed and the surfaces looked as if they were left unpolished, which is at variance with the general treatment of the face.

There was a strange feature on the back of the head: allegedly a peg which was used as part of a prong designed to secure the head of the acrolithic sculpture to the wall behind. In a later section of his report, Professor Lo Muzio specifically discussed the theory that the Head of Dionysus (and the Head of Goddess) were part of acroliths. In Bactria, the only piece of evidence of a probable acrolithic marble sculpture was the sandalled foot of a large statute depicting Zeus from the Greek town of Ai Khanum. This dates from the 3^rd century BCE. No acroliths are recorded in the following centuries. As far as Gandhara was concerned, no marble sculptures of small or large dimensions are known from documented excavations in Gandharan sites or through objects of ascertained provenance. The same is true of acroliths.

Finally on this topic, Professor Lo Muzio said that the garnet inlays in the eyes of the Dionysus were “no less surprising” than the peg at the back of the head.

In summary, Professor Lo Muzio considered that both the Head of Dionysus, and the Head of Goddess, were inauthentic and were modern artifacts. He reached this conclusion by reason of the lack of reliable comparators and a substantial inconsistency with all that is known of Bactrian and Gandharan sculptures of ascertained authenticity. The authenticity of the Dionysus was undermined by iconography and various aspects of material and workmanship: the use of marble, the “acrolithic technique”, and extensive use of garnet inlays, and the “astonishing preservations of almost all of them”. He also referred to the inconsistency with the historical and cultural context to which the artifacts are supposed to belong. He said that a rare iconography in an art work of unascertained provenance is the first alarm signal of a disputable authenticity. Overall, the two heads raised serious doubts in so many respects as to be hardly comparable with anything known in the credited set of evidence of Gandharan and Bactrian art.

Professor Lo Muzio was cross-examined on the views expressed in his written report. I refer to aspects of his cross-examination in my conclusions below.

Contrary to a suggestion in cross-examination, I did not think that Professor Lo Muzio, in his oral evidence, sought to make points which had not in substance been made in his written reports. For example, in his oral evidence, he discussed the suggestion that the Head of Dionysus was part of an enormous sculpture. He said that it would require a major economic financial outlay to create it, and only a wealthy social group could have afforded it. This was unlikely because, from what is known of the 2^nd – 4^th century, “there were no social groups that were wealthy enough and that still practised the cult to Dionysus to actually commission a piece of work like this. We no longer have a Greek elite after the 1^st century CE”. This was essentially the same point that Professor Lo Muzio had made in paragraph 18 of his report.

There were two particular, allegedly comparable, heads on which questions were asked in cross-examination.

One was the terracotta head of Dionysus, ascribed to Gandhara 4^th/5^th century, in the Metropolitan Museum of Art in New York. This head was, as I saw it, potentially relevant in showing that, as late as the 4^th/5^th century, Dionysus was part of a continuing cult in Gandhara and Ghandaran art, and that this would be reflected in a large sculpted head of Dionysus himself. Professor Lo Muzio questioned the authenticity of this head, which was of unknown provenance. He referred in particular to the absence of terracotta objects in Gandhara, other than those which had been accidentally fired.

The other was the so-called “Ortiz” head, which was a substantial marble head in a private collection. This head had been the subject of some scholarly analysis by a reputable scholar, Osmund Bopearachchi. The significance of this piece, if authentic, was that it was in marble, thereby giving rise to the possibility that there would be other marble figures. The authenticity of this piece was, however, not accepted by Professor Lo Muzio. Given the absence of evidence in the archaeological record of the use of marble, as summarised above, he considered the Ortiz head as no more authentic than the Head of Dionysus in this case.

In his written report, Dr Jager’s conclusions, expressed towards the start of his report, were as follows.

It was likely that the Head of Dionysus originates from the Bactria region (modern Afghanistan) which formed part of Greater Gandhara under the Kushan Empire in the early part of the first millennium CE and which until around 1^st century BCE had been a Greek kingdom ruled by successors of the armies of Alexander the Great. In this formerly Greek region of Bactria, Hellenistic culture and religious and artistic practices survived over the centuries.

The presentation of Dionysus as an older man with a formidable gaze is consistent with evolving representations of Dionysus within the Greater Gandhara region where he was presented in a more mature, forbidding guise reflecting his expanded role as a more universal god of truth, judgment and awesome power. Stylistically, it is consistent with origin from the Greater Gandhara region. It stands in a long tradition of Hellenistic sculpture and reflects the enduring Hellenistic influences within the region and fresh Roman/Hellenistic influences which travelled along the trade routes between the Eastern Mediterranean and the Kushan Empire in the first centuries of the first millennium CE as well as other syncretic influences within the region.

The Head is of exceptional quality. It is likely to have formed part of an acrolithic sculpture which itself would have been the central figure in one of the remote hillside sanctuaries in which Dionysus was worshipped in Greater Gandhara in the first half of the first millennium CE. It reflects the traditions and practices which we know existed at this time and is a rare and important addition to our knowledge in this area. So far, the finds have been smaller and we have never yet had the chance to see what an acrolith of Dionysus would have looked like. For him, this outstanding and rare sculpture fills the gap which he had always looked for and which he had hoped would be filled.

The fact that this Head is made of marble (which was known to be used, but less commonly in Gandhara than schist, due to its cost and rarity) and so outstandingly carved was to him an indication of its authenticity. He did not consider that any of the elements in the Head of Dionysus could be said to be anachronistic or indicate a lack of authenticity.

He would date the Head of Dionysus to around 2^nd or 3^rd century CE although he could not rule out that it may be of slightly later origin (c. 100 or 200 years later). It was likely that it is from the Bactrian region in Greater Gandhara. It was also likely that it came from the same workshop as the Goddess Head.

He therefore considered that the Head of Dionysus was authentic.

These conclusions were then developed and explained in the subsequent pages of his report. A large number of illustrations of other objects were used for comparative purposes. However, Dr Jager acknowledged that the Head of Dionysus was not directly comparable with anything that had previously been found or whose existence was previously known. He described the head as “extraordinary”, and that it filled the gap that:

“we know existed: we know that Dionysian cults existed in Bactria and the depictions of Dionysus in this region were that of an older and more forbidding figure, reflecting the expansion of Dionysus’ rule. We also know that hillside sanctuaries existed in which huge statutes of Dionysus were worshipped. This acrolithic head, which to me clearly formed the central figure in a large sanctuary, is the missing piece in our knowledge as to how Dionysus was portrayed in such settings”.

The head therefore exceeded anything found in a known excavation.

In cross-examination in relation to this piece (as well as the Head of Goddess), Dr Jager accepted that, but for the sculptures themselves, he would not have considered that there was any possible archaeological or art historical base for the objects. He said, however, that this would be the case for a great number of pieces in international museums. They were extraordinary, but he was not surprised to see them. So if there were no comparable items, then “that’s just the way it is”.

His evidence as to the absence of an archaeological or art historical base reflected his views in a report which he had written for JEL in 2016: he had described them as opening up “a totally new, yet unknown field of knowledge! Without these sculptures, we would never really thought that they could exist …”. He accepted, in evidence, that he and Mr Eskenazi had agreed that there were no comparable examples.

I consider that the evidence of Professor Lo Muzio, as summarised above, was persuasive. In summary, the weight of archaeological and art historical evidence leads to the conclusion that the Head of Dionysus is an inauthentic recent piece of work.

Professor Lo Muzio is a distinguished scholar and expert in the field of Bactrian and Gandharan art history and archaeology. He is an Associate Professor in Indian and Central Asian Archaeology and Art History at the Sapienza University of Rome. He is the author of many scholarly articles, as well as a 2017 book on Pre-Islamic Central Asian Archaeology. The focus of his research has been religious iconography, in particular Buddhist sculpture and mural painting in Central Asia (including, specifically, Bactria) and in Gandhara. Between 1990 and 1997, he took active part in the creation of a catalogue of the Gandharan reliefs housed in the National Museum of Oriental Art, Rome. This involved contributing with descriptive files of 300 reliefs from excavations of Buddhist monasteries in North Pakistan. There was no suggestion by the Defendants that Professor Lo Muzio was not qualified to give the evidence which he gave.

By contrast, Dr Jager has no equivalent academic achievements or appointments. He has never held a full-time academic position. He has not published widely on the subject of Bactrian or Gandharan art or art history. Indeed, he described himself as an archaeologist, and accepted that he was not an expert art historian at all. Thus, in relation to iconography, Dr Jager had no relevant expertise at all, as he accepted in cross-examination. Iconography is, however, a crucial element of art history.

Dr Jager is a conscientious and indeed imaginative researcher, as his lengthy reports in these proceedings attest. But he really was no match for Professor Lo Muzio, or indeed Dr Luczanits. In so far as Dr Jager’s views may differ from those of Professor Lo Muzio (and also Dr Luczanits) I have no hesitation in preferring the latter. Indeed, as previously described, the Defendants in their closing submissions accepted that Dr Jager’s evidence was highly unsatisfactory, and they did not place any reliance upon it in those submissions.

The Defendants nevertheless submitted that I should not place any reliance on the evidence of Professor Lo Muzio, and that therefore that the Claimants had not provided satisfactory art historical evidence in support of their argument that the marble heads were inauthentic. To a large extent, this was based upon the argument that Professor Lo Muzio’s methodology was flawed, because he was only willing to compare the Dionysus (and the other marble heads) with indisputably authentic pieces from the archaeological record. I have already addressed, in Section D above, the issue of whether his methodology was flawed. I do not think that it was.

In their closing submissions, the Defendants made a number of further criticisms of Professor Lo Muzio’s evidence. However, it did not seem to me that these criticisms (concerning, for example, an academic debate as to whether there may have been a Dionysiac cult in Gandhara) addressed the central points that Professor Lo Muzio made in relation to authenticity. In any event, I thought that he dealt well with the questions asked. The professor was in my view a most impressive witness. His written evidence was concise and to the point. In cross-examination, he dealt courteously and clearly with all questions asked, and it is fair to say that none of them caused him to alter the views expressed in his report.

Professor Lo Muzio emphasised in cross-examination the importance of looking at comparable pieces. For example, it was not helpful, when considering Gandharan art, to look at a piece which was from the classical period in Greece: “it is useless to search for comparators in Greece for a Dionysus head in Bactria”. This was in my view an important point since many of the illustrations in Dr Jager’s report were unrelated to Bactria or Gandhara.

Professor Lo Muzio also said that it was not appropriate to jump from Bactria to Gandhara as if they were one and the same: they should be kept distinct. I agree. The origin of the Head of Dionysus, as set out in the JEL invoice, was Bactria. I accept Professor Lo Muzio’s evidence that this was a distinct region which should not be conflated with Gandhara and Gandharan art.

In his evidence, Professor Lo Muzio also emphasised, as he said in his report, the importance of making comparisons with pieces of known authenticity: “it is not good practice to prove the authenticity of a disputed piece using other disputed piece”. As will be apparent from my general approach to the evidence (see Section D above), I agree with this point. In the present case, as Dr Jager’s evidence acknowledged, there is nothing in the archaeological record which would support the authenticity of the Head of Dionysus. This is in my view a very weighty piece of evidence against authenticity.

Another impressive aspect of Professor Lo Muzio’s evidence is that he did not seek to overstate the importance of every point that he had made. For example, he made it clear that his opinion was not based on small points such as the size of the grapes. He also acknowledged that his point about the garnets, and the rough workmanship that surrounded them, was a point against authenticity, but was not the main point.

His main points, consistent with his written report, were summarised in the following answer:

“This is not to deny the existence of a Dionysiac trend in Gandharan art, but it is important to identify it, identify its weight, the space it takes, because so-called Dionysiac scenes always appear in marginal spaces. We don’t have representations of Dionysus because, let’s not forget, this sculpture is identified with the God Dionysus. In Gandhara, we don’t have representations of God Dionysus, first; second, we don’t have sculptures of large size having anything to do with Dionysism in Gandhara. We don’t have marble sculpture altogether and we don’t have marble acroliths. That’s why I said that the small grapes is just a [light] remark. We have many problems with this sculpture”.

This broad summary encapsulates two points which I consider to be very important.

First, Professor Lo Muzio explained the iconography that did exist in Gandharan art. This did not provide any support for the possible existence of a circa 2^nd century CE large statue of Dionysus, even assuming that Bactria (the geographical area referred to in the JEL invoice) could be equated with Gandhara. It was common ground that there was no evidence either in Bactria or in Gandhara of any shrines, whether in caves or structural, dedicated to Dionysus.

Secondly, the absence in the archaeological record of representations of Dionysus, of large sculptures of Dionysus, of marble sculpture generally, and marble acroliths, all point convincingly to inauthenticity. Against this background, it is not surprising that Dr Jager said, in his 2016 report: “Without these sculptures, we would never really thought that they could exist”. In the joint report, Dr Jager fairly accepted that there was no material evidence of marble sculptures from archaeological excavations in both areas. In my view, given the fact that such excavations have taken place since the 19th century, the absence of any material evidence of archaeological excavations is powerful, indeed decisive, evidence against authenticity in this case.

In that regard, Dr Jager’s evidence in the joint report was that he believed that “we cannot rule out that marble sculptures were produced in Gandhara and in Bactria”. Similarly, Dr Jager postulated the possibility that the original location of the statues may have been a cave shrine dedicated to Dionysus. Professor Lo Muzio’s response was that there was no evidence whatsoever of cave shrines dedicated to the worship of Dionysus, not only in Bactria and Greater Gandhara, but in the whole of the Hellenised East altogether. Dr Jager’s response was that there was indeed no evidence of cave shrines, but that “we cannot rule out that such caves existed, and that the two heads may represent the evidence we lacked”.

It seems to me that the debate on this issue, outlined above, illustrates one of the difficulties with the evidence of Dr Jager and the case advanced by the Defendants on this issue. There can of course be no certainty as to what the position was some 2000 years ago. The court must, however, form a view on the basis of such evidence as does actually exist. If a particular matter cannot be “ruled out” on the basis of the existing evidence, that does not mean that the evidence supports the existence of that matter. Thus, even if it could be said that there remains a possibility, as yet unproven, that marble sculptures existed in that area at the material time, or that there is a possibility that there were cave shrines, and that neither of these possibilities can be “ruled out”, that is not a significant factor when it comes to weighing the evidence which does actually exist. I am concerned with what is likely, on the balance of probabilities; not with the question of what is possible despite the absence of evidence. As Professor Lo Muzio neatly expressed it: one cannot support the hypothesis that something existed by simply stating that we cannot rule it out.

All of these matters must also, in my view, be seen in the context not only of the flooding of the market with fakes, but also the inherent improbability of the Head of Dionysus (and indeed the Head of Goddess) having survived for nearly 2 millennia in the condition in which they now appear. I return to this issue in the context of the materials science evidence.

In addition to the points summarised in Professor Lo Muzio’s answer set out above, he also referred to various specific features of the Head of Dionysus, such as the physiognomy and the gemstones. I did not think that these were the most important points, but in my view they all form part of an overall picture which leads to my conclusion that the substantial weight of evidence is that the Head of Dionysus is indeed inauthentic.

In reaching this conclusion, I have not disregarded the evidence of two particular heads on which a number of questions were asked in cross-examination. These two heads – one in the Metropolitan Museum of Art in New York (“the Met”), and the other in a private collection – have no documented provenance in the archaeological record. One is in private ownership, and the one in the Met was formerly in private ownership. Even if these two items are to be weighed in the balance when considering the evidence as a whole, neither of them causes the balance to tip in favour of authenticity, given that the pieces with which I am concerned are not supported by or comparable with anything in the undisputed historical record.

The first piece is the terracotta head of Dionysus, ascribed to Gandhara 4^th/5^th century, in the Metropolitan Museum of Art in New York. Professor Lo Muzio questioned the authenticity of this head, which was of unknown provenance. He referred in particular to the absence of terracotta objects in Gandhara, other than those which had been accidentally fired. I can well understand why this fact would, for an art historian, raise question marks as to the authenticity of this head, as do the points made by Professor Lo Muzio concerning the absence of comparable representations of Dionysus in the historical record. But even leaving those powerful points aside, I do not consider that the Metropolitan Museum terracotta Dionysus Head, alleged to date from the 4^th/ 5^th centuries in Gandhara, and of unknown origin, is of any assistance or weight in the context of objects sculpted in marble alleged to date from 2^nd century Bactria.

The second piece was the so-called “Ortiz” head, which was a substantial marble head in a private collection. This head had been the subject of some analysis by Bopearachchi. The significance of this piece, if authentic, was that it was in marble, thereby giving rise to the possibility that there would be other marble figures. The authenticity of this piece was, however, not accepted by Professor Lo Muzio. Given the absence of evidence in the archaeological record of the use of marble, he considered the Ortiz head as no more authentic than the Head of Dionysus in this case. The only actual evidence of a marble sculpture, in the archaeological record, was the marble fragment from Ai Khanum, in Bactria, of the foot of Zeus, possibly as part of an acrolith. This dated to the 2^nd – 3^rd century BCE. That apart, the only marble artifacts are those in private collections, such as the Ortiz head, which have no documented provenance. And although Bopearachchi has provided a description of that head, he did not identify any comparators. So Bopearachchi did not apply any criteria which could test and possibly prove the authenticity of the artifact.

I consider that there are obvious and legitimate question marks against the authenticity of the Ortiz head, in view of the archaeological evidence to which I have referred. Furthermore, the Ortiz head – unlike the Met terracotta head discussed earlier – is not even found in a public collection of a major museum. It is in a private collection, and the expert evidence is clear that many objects in private collections are fakes. It has no verifiable provenance. I do not see how such evidence as is provided by this object (whether or not taken in conjunction with the Metropolitan Museum head) could outweigh the weight of the evidence that there is no object, comparable to the Dionysus head, in the archaeological record, and the other matters referred to by Professor Lo Muzio in his evidence.

The reality is, in my view, that a search for comparables or any support for authenticity is elusive. There was correspondence in 2016, after Dr Jager had produced a lengthy paper on Dionysus and his cult in Bactria, in which he considered images of the Head of Dionysus and the Head of Goddess which Mr Eskenazi had sent him. In his e-mail dated 10 October 2016, Mr Eskenazi said:

Your work is very impressive indeed, especially when one considers that there are no comparable examples and very little literature to draw on

I therefore conclude, on the basis of the art historical evidence, that the Head of Dionysus is inauthentic.

Prior to its acquisition by JEL, there is no record of the Head of Dionysus having previously been sold. Mr Eskenazi’s evidence was that the Dionysus, as well as the Goddess, were originally part of a collection of marble heads owned by Mr Mark Hubbard. Mr Eskenazi said that he had acquired 6 or 7 marble Bactrian heads over a period of years. All of these, except for a head with bronze inlaid eyes which was in JEL’s research paper, had come from Mr Hubbard. Mr Hubbard’s father had himself acquired these heads over a number of years: he had been a dealer in tea and coffee and was frequently in India and elsewhere, and had acquired the heads from dealers. All of the heads were in immaculate condition. Mr Eskenazi had seen the collection in Paris a long time ago, and after his death “Mark slowly sold them to me”. There were, however, no photographs of the collection of heads, or indeed any individual head, in situ.

In my view, as the Claimants submitted, this explanation of the provenance of the pieces lends considerable weight to their inauthenticity. There are, as the art historical evidence shows, no marble heads from the relevant place and period in the archaeological record. Indeed, as Mr Eskenazi said in evidence, the only documented marble head is the Ortiz head discussed above. In my view, it is inherently improbable, against this background, that one collector could amass a large collection of genuine marble heads in immaculate condition, and indeed for there to be no photographic or other convincing evidence concerning the existence of the collection or when it was acquired. If these pieces were genuine, they would have been objects of major importance. This makes it even more surprising that there is no evidence at all concerning their provenance, or indeed their existence, prior to the sale by FEAL to Mr Eskenazi, the sale to QIPCO, and the two letters written by Mark Hubbard.

There is nothing else about the provenance of these pieces that suggests authenticity. For example, there is no evidence as to the identity of the dealers from whom Mr Hubbard’s father is said to have acquired his collection.

The Claimants also referred to some other unsatisfactory features of the evidence as to the acquisition of these pieces. Although the owner of the Dionysus was said to be Mark Hubbard, the sales were made by Far Eastern Antiquities Limited. This was the $ 1 Hong Kong company incorporated in October 2012, although Mr Eskenazi had never looked up the company details and did not know who the directors were. However, he understood FEAL to be a company owned by Mr Vigorelli, the Italian dealer whom he had known for many years. The invoice for the piece, dated 23 October 2014, described it as “Head of a Bacchus”, with its origin as Tajikistan, Afghanistan and its date “3^rd b.C. cent”. When it was later sold to the QIPCO, the invoice gave its origin as Bactria and it was dated to “circa 2^nd century CE”. A provenance letter was provided by Mr Mark Hubbard. This is dated 15 September 2014. It is clear, however, that this was backdated: since the description in the letter matched the JEL invoice to QIPCO which was only created in November 2014. Although these points were valid, I did not consider that they materially added to the more important points described in the previous paragraphs.

Accordingly, although the art historical evidence is in itself sufficient to establish inauthenticity, the evidence concerning the provenance of the piece adds to and reinforces that conclusion.

In view of my conclusions in relation to the art historical evidence (reinforced by the provenance evidence), it is not in fact necessary to consider the materials science evidence, or decide the many disputes between Dr Bennett and Dr Nicola. This is because even if I were to accept Dr Nicola’s evidence in full (which I am certainly not inclined to do), it would not have the effect of negating any aspect of the art historical or provenance evidence that I have considered. My conclusions as to lack of authenticity, based on that evidence, would remain.

This is because the effect of Dr Nicola’s evidence on the Head of Dionysus (and indeed all of the other pieces except for the Hari Hara) is that the materials science evidence does not assist, one way or the other, in establishing authenticity. Towards the end of his cross-examination on the two marble heads, he was asked whether, as a fair-minded expert, he was “going to tell the court that it is likely that these are ancient objects”. His response was that it was “conjecture” and that his opinion was conjecture. He went on to say that it was difficult to reconstruct exactly what had happened. He went on to say that it: “Could be a fake maker that make all this reasoning. I feel it a little too much constructed, but it could be”.

These answers reflected the substance of his report on both marble objects. That report recognised that there was some evidence which might point towards modern work in relation to the objects. For example, there was no dispute between the experts that there were two layers on the marble objects which did not date from antiquity. One layer was a distinct fluorite (calcium fluoride) layer close to the surface of the marble. This layer was the result of either the application of a hydrofluoric acid based product (as Dr Bennett’s evidence indicated, and Dr Nicola accepted as a possibility), or the application of a fluorosilicate in order to bind, strengthen and protect the surface of the item (a possibility which Dr Nicola advanced, but Dr Bennett rejected). Either way, this layer had not been applied in antiquity. There was also a second layer, which Dr Nicola described as an “applied deposit”. Again, there was no dispute that this had not been applied in antiquity. In addition to this positive evidence of work on the objects in modern times, it was common ground between the experts that there was no evidence, or at least (looking at Dr Nicola’s evidence as a whole) no substantial evidence, of the objects having been subject to natural weathering over a period of approximately 2,000 years.

Dr Nicola was therefore unable to point to any positive materials science evidence which indicated that they were ancient objects. However, in order to respond to the points made by Dr Bennett, he also needed to provide an explanation for the evidence of modern work and the absence of evidence of weathering. His explanation for the evidence of modern work was that this may have been the result of past conservation methods, and therefore that it was not possible to draw the conclusion that modern work meant modern original manufacture. For example, in relation to the layer of calcium fluoride on both objects, he said that this “could be the result of common past interventions”. The same was “also true for the surface deposit which appears to sit atop of this layer”. It was, he said, “not possible to ascertain whether these were applied as restoration and/or conservation, or because of an intent to deceive”. Accordingly, the existence of these layers did not exclude the possibility that the items were of ancient production. His conclusion in relation to both objects was therefore:

“In my opinion, it is not possible to state that the item is uniquely of recent production as the observed phenomena or treatments relied upon by the Claimants could be due to common restoration practices very widespread in the near past”.

Accordingly, Dr Nicola expressed the view in the joint report (on both marble heads) that all data found were compatible with those expected for an ancient object that had undergone restoration interventions common in the past, such as aesthetic restoration after an overcleaning conducted even with aggressive materials. In summary, the essential points that he made were that the presence of the chemically induced transformation of the surface and of a deposit containing pigments applied as an aesthetic patina over the surface was typical of conservation treatments which were common in the past. Any assumption about authenticity based on the deposit layer should be avoided, because it was agreed that this was of modern application. In Dr Nicola’s view, the chemical treatment had impacted upon all the elements that Dr Bennett had relied upon, including the presence or absence of visible weathering and/or biological material and/or a natural patina to the marble. The absence of two features which might be expected with ancient marble – namely dedolomitization and dissolution of calcite – did not lead to a conclusion of inauthenticity. These matters depended upon an object’s conservation and restoration history, and no conclusion could be drawn here, because it was agreed between the experts that strong chemical treatment had been applied that would in all likelihood have changed their characteristics. He therefore reiterated his opinion that it was not possible to ascertain whether the fluoride layer or the deposit layer had been applied as restoration and/or conservation or because of an intent to deceive. However, he thought that the use of a fluorinated layer as a forgery technique would not be expected even by an unskilled forger; for example, because it was easier to remove toolmarks by hydrochloric acid which is much easier to use and far less hazardous than hydrofluoric acid. He ventured the opinion that the presence of this layer pointed to a restoration intervention and not a falsification.

Dr Bennett took a completely different view. She summarised her conclusions in the joint report on the Head of Dionysus (which was materially the same as her conclusion in relation to the Head of Goddess) as follows. The work was of modern manufacture based on: the carving techniques; the presence of modern tool marks; the shape of the sockets used to accommodate the gemstone inlays; the absence of weathering; the absence of a natural patina; the absence of any natural mineral deposits; the absence of any biological activity; the absence of any dissolution of the calcite; the absence of any de-dolomitization; the presence of a chemically induced fluoride layer over the entire surface of the work; the presence of a decorative paint layer in a modern binder and of deposits which had been applied over the induced fluoride layer to simulate a burial sediment. In her opinion, the objects were, after carving, treated with hydrofluoric acid and air abrasion, followed by an application of applied deposits. She did not consider that a cleaning or conservation treatment could account for the fluoride surface: in her opinion, this could only have been created by the application of hydrofluoric acid.

Since, the Claimants’ case of inauthenticity is established on the basis of the art historical evidence, I will state my conclusions in relation to the materials science evidence briefly. In short, I accept the evidence of Dr Bennett, as summarised above, for the following reasons.

First, since a case of inauthenticity is established on the basis of the art historical evidence, it is inherently probable that the matters relied upon by Dr Bennett do indeed reflect modern manufacture, rather than (as Dr Nicola postulated) an ancient object which was then subject to aggressive restoration. Modern manufacture provides a clear and satisfactory explanation for all of the features identified by Dr Bennett.

Secondly, there is no evidence at all of the conservation and restoration history of the pieces in question. If there had been some evidence as to these matters, then it might have provided support for the evidence of Dr Nicola. I am, however, considering the evidence on the balance of probabilities. On the one hand, there is considerable evidence of what might be called modern interventions (for example, the fluoride layer, the deposit on top of that layer, the modern tool marks), and a lack of evidence that would ordinarily be expected on 2,000 year old marble statues (for example, lack of weathering, or evidence of biological activity). In my view, in the absence of any evidence as to the conservation and restoration history, it is inherently probable that this evidence indicates modern manufacture.

Thirdly, I consider it most improbable that, if the two objects were of ancient origin, they would now be in – as Dr Bennett aptly described them – almost pristine condition. Dr Bennett’s evidence, which I accept (and which I did not understand to have been substantially disputed by Dr Nicola) is that both objects have an unweathered surface and there is no evidence of a re-precipitated layer of calcium carbonate or dissolution of calcite which would be expected on a marble which had aged naturally for almost 2,000 years. Dr Bennett said (and Dr Nicola acknowledged in cross examination) that over a prolonged period, marble surfaces become porous and prone to staining. Natural patinas derived from the weathering of the underlying marble, and the incorporation of elements from the deposition environment into the re-precipitated layer of calcium carbonate which has resulted from the dissolution of the marble surface, produce a boundary or interface with the underlying material of an object. These signs of weathering are not present here. Dr Bennett’s view was that restoration and cleaning would not be capable of removing all signs of weathering, and would certainly not be capable of doing so whilst also leaving other carved details, as here, sharp and distinct. In her opinion, even if an object had been sealed in a cave for almost 2,000 years, a natural patina would be evident. Here, the objects had surface cracks, but they are free of debris and completely devoid of soil or weathering products.

This evidence made considerable sense to me, and I found it persuasive. Moreover, Dr Nicola’s response did not in my view provide any effective answer, when it comes to assessing the position on the balance of probabilities. Dr Nicola’s point (as expressed in the joint report) was to draw a distinction between items stored at archaeological sites in contrast to the present item which

“… has presumably been kept in controlled indoor environment and has been subject to multiple conservation treatments also specifically aimed at removing the signs of time on it”.

He made the point that an object’s weathering will depend on its conservation and restoration history, including whether signs of weathering have been removed by restoration treatment:

“For example, an object’s weathering is likely to be slower in an arid or stable climate, or can be impacted by the fact that it was buried or sealed in a cave. It is possible that the Work may have been deposited in a grotto. If that is the case, the item would potentially have been protected for significant periods such that it will appear less weathered than if it had been buried or preserved in a more active environment. However, in this case, we have no proven information regarding the environment in which the Work has been conserved and the conservation and restoration history of the Work is unknown. … In [Dr Nicola’s] opinion, the presence of an evident weathering deposit will depend on the conservation history and the environment in which the object was kept and since we do not know what these were it is not possible to draw conclusions”.

In my view, however, the evidence as a whole indicates that it is far more likely than not that marble objects which are 2,000 years old will have some evidence, almost certainly considerable evidence, of weathering. It is improbable, that an object will have benefited from storage over that extended period of time in conditions that resulted in the absence of any material evidence of weathering. Indeed, Dr Nicola accepted in cross-examination that with marble, even in the most protected place in a remote zone, he would expect to see some signs of weathering over 2,000 years. On the balance of probabilities, the absence of weathering therefore evidences inauthenticity, as Dr Bennett said. I do not accept Dr Nicola’s point that it is not possible to draw conclusions in the absence of evidence as to the object’s conservation history and storage environment. In my view, the absence of evidence as to an object’s conservation history and storage environment means that there is nothing to counter the ordinary conclusion that would be drawn, on the balance of probabilities, from the absence of evidence of weathering.

Fourth, I consider it far more likely than not that the fluoride layer resulted from the use of hydrofluoric acid, even though that should not be used in conservation or restoration of marble. I cannot see why there would have been any need to use fluorosilicates. Those are generally used, as Dr Bennett said, on sandstone and it acts as a consolidant of the sand grains. Fluorosilicates can be, or at least were, used on a sandstone surface to strengthen it. I cannot see, however, how or why the marble heads would have required strengthening or the use of fluorosilicates.

Fifth, I have already (for reasons discussed in section D above) expressed my general preference for the evidence of Dr Bennett where she differs from Dr Nicola. That general preference is now reinforced by the conclusions which I have reached on the Hari Hara. My preference for Dr Bennett’s evidence remains after having re-read the reports on these two marble objects, and the cross-examination of both witnesses in relation to those objects. As it seemed to me, the thrust of Dr Nicola’s evidence was to identify difficulties in the way of drawing firm conclusions from the materials science evidence, and to highlight uncertainties which were said to prevent definitive conclusions from being drawn. In my view, however, when the evidence is considered as a whole, the case for inauthenticity advanced by Dr Bennett is compelling, and in any event is made good on the balance of probabilities.

This sculpture was fairly described by Dr Jager as follows. It is the head of a woman with a serene expression. She has a soft full face, large almond shaped eyes inset with garnets and full lips. Her waved hair is swept back from her face. She wears a headdress with a garland of flowers: each flower consisting of six external petals and six smaller internal petals, studded with garnets. A semi-circular garnet pendant falls on her forehead which appears to be attached to a chain studded with garnets which then loops through her hair and in front of her ears, which are not seen, covered by her hair.

There is a rectangular recess at the back of her head. Dr Jager suggested that this indicated that the head was originally conceived as part of an acrolithic sculpture.

It was not possible for either expert to identify any particular goddess that was represented by this sculpture. The woman may therefore have been an ordinary mortal: as Professor Lo Muzio said, there were no specific features allowing an identification as a goddess rather than a well-to-do-lady or worshipper. This is, however, immaterial to the issues requiring resolution. The designation “Head of Goddess” is convenient although it may not be accurate.

JEL’s invoice referred to this piece as “Greater Gandhara region - Late Bactrian period, 2^nd/3^rd century”. Professor Lo Muzio, correctly in my view, described this as a confusing description. Bactria is not properly to be regarded as part of Greater Gandhara. Furthermore, the expression “Late Bactrian period” does not make any real sense: Bactria was an area, not a period.

Professor Lo Muzio and Dr Jager were more or less agreed that the Head of a Goddess was made in the same workshop as the Head of Dionysus. I consider that this is probable. Since I have already concluded that the Head of Dionysus is an inauthentic modern piece, it follows that the same conclusion follows in relation to the Head of a Goddess.

Even if the same workshop were not responsible, most of the important reasons for considering the Head of Dionysus to be inauthentic are equally applicable to the Head of Goddess. Specifically, as discussed above in the context of the Head of Dionysus, there is no reliable evidence of any marble sculpture either in Greater Gandhara or Bactria during the 2^nd/ 3^rd centuries CE. As Dr Jager said in his 2016 report: without these two sculptures “we would never really thought that they could exist”. Or as Mr Eskenazi said in his 10 October 2016 e-mail: “there are no comparable examples and very little literature to draw on”. I regard these as very significant points against authenticity.

Against this background, I can deal with the further points discussed in the expert art historical evidence briefly.

Professor Lo Muzio dealt with various aspects of the iconography of the piece. He made a number of points, including that the large wide-opened eyes of the lady had nothing to do with Gandharan art altogether. As with the Dionysus, he drew attention to the large number of gemstones, and the fact that these were of poor workmanship. He said that the shape of the earrings, and above all the way that they were worn (they seemed to encapsulate the ears) were unrecorded in Gandharan iconography. The same applied to the eye inlays, for which no reliable comparators are known. The central gem in the forehead’s ornament was also grossly off-centre.

He also said that another concerning feature of the work was the hair tufts apparently growing out of the ears or out of the narrow space between the ear ornament and the ears. He described this as a gross inaccuracy in a head worked in broadly naturalistic style, and at any rate unthinkable in a Gandharan sculpture.

Just as in the Head of Dionysus, all ornaments showed rough execution and poor finishing. Despite this, one of the striking features of the head was the preservation of almost all garnet and quartz inlays.

It seemed to me that these various points involved both iconography and style, perhaps with a degree of overlap between them. These are both areas of expertise of an art historian, such as Professor Lo Muzio, and I have already indicated that I considered him to be an impressive witness upon whose evidence I can rely. Accordingly, I accept that Professor Lo Muzio’s additional points, concerning iconography and style, add to the weight of evidence of inauthenticity.

As far as provenance is concerned: the key points are those which I have addressed in relation to the Head of Dionysus, and which in my view add to and reinforce the conclusion (based on the art historical evidence) that the Head of Goddess is inauthentic.

As far as materials science is concerned: whilst there were some differences in the expert evidence concerning the Head of Goddess and the Head of Dionysus, it can fairly be said that the central issues were the same. Those issues and my conclusions are set out in Section G3 above. Accordingly, my conclusion, based on the materials science evidence, is that the Head of Goddess is inauthentic.

The Frieze is a frieze depicting a number of figures. In contrast to the previous marble objects associated with classical Greece, the Frieze is in schist (a stone) and is associated with Buddha. It depicts the emaciated Bodhisattva seated frontally on a simple throne. He is flanked by two monks behind his shoulder, and two pairs of women are represented in dialogue with each other. Two more women standing on vases each occupy a pilaster which flanks the scene. The entire scene is placed on lotus flowers.

The moment depicted in the Frieze is part of the Buddha’s life and progress. The story is narrated in Chapter 17 of one of the latest biographies of the Buddha, the Lalitavistara. Prior to being reborn on earth, Prince Siddhartha was the Bodhisattva Siddhartha. He had renounced his princely life of comfort. He had refused all food, hoping that this would assist him to reach enlightenment. A group of folk deities become concerned and travelled to the highest of heavens that remain physically connected with the world. This is where the Bodhisattva’s mother, Mayadevi, resides. She had died shortly after his birth. She descends with a retinue of goddesses to visit her son, and entreats him to break his fast, reminding him of the prediction that he would be a Buddha in this world. The Bodhisattva reassures her that this will come to pass, and she returns. The fast then continued for six years, until he finally realised that his starvation would not help him reach enlightenment. Following the breaking of the fast, he walked to the bank of a river. There, seated beneath the Bodhi tree, he ultimately attained enlightenment.

There was no dispute that the Frieze bore some considerable similarity to a work which is in the Linden Museum, Stuttgart. Dr Luczanits considered that the Linden Museum piece had served as a direct model for the Stuttgart Frieze, but also that there were similar doubts about the authenticity of that Stuttgart piece for many of the same reasons that he considered that the Frieze was inauthentic. He also considered that aspects of both the Frieze, and indeed the Stuttgart piece, were derived from an important and authentic sculpture of the emaciated Bodhisattva (4^th or 5^th century) in the Lahore Museum. Dr Jager, by contrast, considered that there was no reason to regard the Stuttgart piece as inauthentic, and he referred to one academic paper which had considered that piece without any suggestion of inauthenticity. The effect of Dr Jager’s evidence was that the similarities with the Stuttgart and Lahore Museum pieces demonstrated that there was a sound basis in existing Gandharan art for the Frieze, and therefore that this supported its authenticity.

Dr Luczanits’ assessment, in his written report, was made on the basis of two main criteria, style and content, in the context of the stylistic and thematic development of Gandharan art as well as its religious meaning. Using these criteria, he relied upon a very large number of matters in support of inauthenticity. These were, in summary, as follows. There were doubts as to where a frieze of this kind would have been located. The carving of the figures, and the perspective shown, was not consistent. The overall narrative content of the relief did not make sense. When each separate element of the relief was considered, the Frieze was simply a modern pastiche of Gandharan art. It combined a number of matters, but the combination made no sense. The Frieze had most likely been derived from, and perhaps produced in the same workshop as, the Stuttgart piece. It was also in remarkably good condition, with only minor damage which would not impact upon the aesthetic appearance of the piece.

His overall conclusion was that the relief in question could not be a Gandharan object. It bore all the hallmarks of modern productions, which catered to western taste: often symmetrical, perfectly preserved (all chips are at marginal places only), and made up by combining elements from different, not necessarily contemporaneous sources in one. Authentic Gandharan art, by contrast, is never this stereotypical, and in narratives the figures relate to each other in ways one can make sense of. Further, one can usually reconstruct how the piece was used, and if perspective is used in carving the relief is consistent throughout. There was no doubt in his mind that it was the product of the late 20^th century, a period when publications on Gandharan art became widely available. Indeed, he went so far as to say that there was no way that somebody claiming expertise in Gandharan art could mistake the relief as authentic.

I will not describe in detail every point which Dr Luczanits made in his carefully considered but nevertheless comprehensive attack on virtually every aspect of the Frieze – albeit that he acknowledged that the carving was well done and was oriented on some of the best Gandharan sculpture. However, the following gives a sufficient flavour of the points which he made.

Potential location. Dr Luczanits considered in detail the place where a composition of this kind might be placed. Whilst it would fit a front panel of a throne, its subject matter would be unusual in that position. The Frieze had holes for iron clamps at the top of the relief, indicating that it would need to be attached to other objects. This pointed towards an original mounting in a narrative frieze, comprising a band of successive panels around the dome or base of a monument. In Gandhara, the life of the Buddha is most frequently depicted in such a narrative, with panels arranged clockwise in chronological succession. However, there are no strictly symmetrical compositions of scenes in narrative friezes, whereas this Frieze was symmetrical: a feature designed to appeal to western taste.

Dr Luczanits also considered the possibility that the Frieze would be a panel forming part of a frieze on the base of a square stupa (a Buddhist shrine containing relics). However, again the figural composition of such panels would not be symmetrical. He also considered, but rejected, the idea that the Frieze would be part of a false window gable. Overall, his conclusion was that consideration of the potential original location of the object puts it in doubt, as it was unclear what the original context of the relief could have been.

Carving and perspective. Whilst the carving was well done, the artist did not understand the models fully. Gandharan artists used perspective, and allowed for the fact that certain pieces were designed to be viewed from below. In this Frieze, the lotus bases of the figures are cut at an angle to be seen from slightly below, indicating that the relief would have originally been placed above head level. However, this approach was not fully followed through into certain aspects of the figures in the Frieze. Accordingly, there was no uniform concept for the carving of the figures and the perspective in which they were presented. This indicated that this was a composition built of disparate elements and without a future context in mind.

Narrative content. The background to some of Dr Luczanits’ points in relation to both narrative content and “modern pastiche” is one of the problems which confronts forgers. If they simply copy an existing work, the piece may be easily recognised as a modern copy. Accordingly, forgers may seek to introduce some elements of originality into the work. This is done by picking and choosing elements from a number of sources. However, forgers lack the necessary background in what Dr Luczanits called the “language” of Gandharan art. What can happen is that an element is used in the wrong place or for the wrong subject.

One example of this was the usage of a lotus base on the relief. The lotus base had two layers of petals, with whirls between them indicating the water from which the lotuses grow. However, there was no reason at all to place this narrative scene on such a base. There are authentic depictions of Gandharan art where there is a lotus base. Dr Luczanits exhibited a picture of a sculpture of the Buddha in the National Museum of Pakistan, Karachi. However, this shows a Buddha in his own world. A miraculous Bodhi tree and lotuses at the base make sense in this context, as it indicates a world in which beings are born as lotuses. But there is no justification to depict water and lotuses growing from it underneath the emaciated Bodhisattva and his company.

Whilst this might be regarded as a point of some subtlety, Dr Luczanits had some more basic points about the overall narrative. The story of the Mayadevi descending from the heavens to see her emaciated son is dramatic. However, the Frieze does not directly refer to this divine intervention. It does not make clear that the mother has by now become a goddess. Indeed, when Dr Jager was cross-examined on the Frieze, he (understandably) could not say where the Mayadevi was actually depicted on the Frieze. Furthermore, there is nothing to suggest the drama of the episode, such as the worry of the mother that her son may not survive.

In addition, the scene contains elements which make no sense in the context that they are employed. Both the flanking monks and the leaves of the tree – which appeared to Dr Luczanits to be a bodhi tree – make no sense in this part of the story. As described above, it was only some years after the Mayadevi’s visit that the Bodhisattva achieved enlightenment, becoming a Buddha. It is at that stage that the leaves of a bodhi tree and monks might make sense. But those elements only took place after the event supposedly depicted in the relief.

In addition, the depiction of the two pairs of women holding offerings in their hands towards each other, rather than towards the Bodhisattva, make no sense in the context of the story. In traditional representations, the position to the immediate left of the central image is the most prominent position in terms of the hierarchy, and thus should represent the most important of the women and thus the mother of the Buddha in the story. In that context, it makes no sense for her to be looking away from the Bodhisattva and offering her blossoms to the lady next to her. More broadly, each figure in an authentic Gandharan narrative of this kind would have a meaning, even if that may be difficult to deduce. However, there is no (indisputably) authentic Gandharan relief in which women simply offer flowers to each other. The two pairs of women, turning away from the Bodhisattva, with offerings in their hands are not consistent with the story depicted.

Modern Pastiche. Dr Luczanits also considered each element of the Frieze in detail. His essential point was that this showed that it had a “composite nature”: in other words, that elements were borrowed from different sources, and made no sense in the context of the story, or as a whole.

For example, the body of the Bodhisattva in the Frieze, and the way in which the cloth falls around the body, are direct quotations from the famous emaciated Bodhisattva in the Lahore Museum. However, in that piece, the Bodhisattva is bearded and sits on a grass mat, which makes much more sense for an ascetic than sitting on a bejewelled throne supported by a lotus.

Another example is the tree under which the Bodhisattva is sitting. This appears to reflect the miraculous tree that can be seen in pieces such as that in the Karachi museum showing the Buddha. However, it has the heart-shaped leaves of the bodhi tree, which is usually depicted in a much more natural way. The leaves of the tree in this case are unlike any other representation of the bodhi tree that Dr Luczanits could remember. In any event, the Bodhisattva should not be sitting under either a miraculous or a bodhi tree: there is no tree in the Lahore museum piece, or any other (indisputably) authentic relief with the emaciated Bodhisattva.

Comparison with the Linden Museum piece in Stuttgart. Dr Luczanits considered that the relief in the Linden Museum in Stuttgart served as a direct model for the Frieze. Both pieces shared some of the contradictions described above. There were also a number of features which appeared to be a direct copy: the pair of females on the right of the Frieze, and the leaf-shaped crotch on the lady on the left side. There was also an awkward right arm of the first lady on the right of the Bodhisattva. The arm in the Stuttgart piece had broken off, and the artist has invented something for his copy.

Dr Luczanits was cross-examined for the best part of a day in relation to his expert report. There was no material change in his evidence. I refer briefly to that evidence in my discussion below.

Dr Jager’s view was that the piece was authentic. He said that the narrative which it depicts was known to be in circulation in Gandhara in this period. There were comparable friezes in the British Museum and the Linden Museum, Stuttgart and in publications which depict the same narrative in a similar manner. Although Dr Luczanits considered the Linden Museum frieze to be doubtful, that was a personal view which is not in accord with accepted academic opinion.

The quality of the carving is outstanding and reflects the sculptural style of Swabi, in Gandhara. Stylistically and iconographically, the mixture of imagery, combining Hellenistic and Indian/Hindu influences in a Buddhist context, was referable to and consistent with origin from the Greater Gandhara region and reflected the rich variety of influences that are found in Gandharan art.

Dr Jager did not consider that any of the elements in the Frieze can be said to be anachronistic or indicate a lack of authenticity. His written report addressed in some detail each of the points which Dr Luczanits had identified in an earlier report. That report made many of the points which Dr Luczanits made in his expert report served in the proceedings. In summary, Dr Jager did not consider that any of the points raised by Dr Luczanits created any difficulties in relation to the authenticity of the piece.

Dr Jager therefore placed the date of the Frieze as either late 2nd or early 3rd century CE, and he considered that it was from the Greater Gandharan region. He considered it to be an outstanding example of a frieze in the narrative tradition.

I have no hesitation in preferring the evidence of Dr Luczanits to that given by Dr Jager. Dr Luczanits could, in my view, speak with real authority in relation to style and content of the frieze. He was an art historian, with very considerable knowledge of Gandharan art.

In my view, Dr Luczanits’ evidence was, throughout, thoughtful and careful, and indeed convincing. He was cross-examined at length. I did not consider that any material damage was inflicted on the views which Dr Luczanits had expressed in his written report, and which I have summarised above. Indeed, by the end of the cross-examination, which I have re-read in full, I took the view that there was a considerable amount of ground that Dr Jager would need to make up if I were to be persuaded that his evidence was to be preferred.

I was also, even at that stage, wondering (on the basis of Dr Jager’s description of his background in his written report) whether Dr Jager had anything approaching the expertise in this area of Dr Luczanits. It clearly, and fairly rapidly, transpired in cross-examination that he did not. He agreed that he was not an expert in iconography. Indeed, he had indicated in his written report that he did not consider himself to specialise in iconography. Ultimately, the Defendants did not place any reliance on Dr Jager’s evidence.

It also seemed to me that, Dr Jager’s cross examination reinforced the persuasiveness of many of the points made by Dr Luczanits.

Thus, Dr Jager was asked whether it was possible to tell, from the Frieze, who the Bodhisattva’s mother was, given that her descent was an important feature of the story. He said that it was a “question of interpretation”. He accepted, however, that to anyone with a knowledge of Buddhist iconography, it was unlikely that there would be a lack of clarity as to who the mother was. He could not think of any other representation of the story in which there was a lack of clarity. He indicated that there were always new ways of telling a story, and new ideas, and therefore he did not have a problem with it. In my view, however, the absence of clarity as to the mother’s identity, in a story where her presence is central, is indeed very problematic.

I consider that this lack of clarity as to the mother is the result of the points made by Dr Luczanits’ concerning the forger’s lack of a true understanding of the nature and context of the story that is being depicted. Forgers are, clearly, extremely skilled. They can look at various pieces of Gandharan art, since they are now widely available in publications such as Kurita. They can carry out high quality carvings of different representations of Gandharan art, but at the same time need to avoid simply copying a single piece. They also do not necessarily know which works, which form the raw material for their work, are themselves of questionable authenticity. The result is that a forger’s sculpture, attractive though it might be, may be a composite of features which can be found in different works (not always authentic) but which do not make sense as a whole.

Similarly, Dr Jager was asked about the tree. It was put to him that the leaves were (as Dr Luczanits had said) the leaves of a bodhi tree. He was reluctant to accept that they were. However, I see no reason to doubt the evidence of Dr Luczanits that this is what they purport to be. If so, then there would be an iconographic problem with the story, since the emaciated Bodhisattva would not be sitting under a bodhi tree at that stage in his life. This was implicitly acknowledged by Dr Jager, who suggested that the tree might be a different type of tree (as suggested in his report).

Dr Jager was then asked about the monks: the point being that monks would not make sense at that stage in the story, because the Bodhisattva had not yet become a Buddha. His response was that they could be other followers. However, I again see no reason to doubt the evidence of Dr Luczanits that these purport to be monks. This is how they are described in the JEL research paper. Indeed, Dr Jager’s report (paragraph 14) initially refers to them as monks, prior to offering a number of other possible explanations as to who they are. These explanations did not seem convincing. For example, he referred to the fact that some translations of the relevant story refer to 5 ascetics. But if these are being portrayed, why should there only be two of them?

Finally, it was put to Dr Jager that the women looking away from the Bodhisattva was a sign of rudeness in Buddhist art. His response was that they may have been horrified by the state of the Bodhisattva. In my view, however, the far more likely explanation is, as Dr Luczanits said, that the artist was simply copying the Stuttgart piece, and did not appreciate that it made no real sense for the women to be looking away from the Bodhisattva.

Accordingly, I accept the Claimants’ case, based upon the art history, that this Frieze is inauthentic.

That conclusion is to my mind reinforced by the point, upon which Dr Luczanits also relied, as to the improbability of the Frieze surviving unscathed in the condition in which it now appears, coupled with the absence of any indication that the holes at the top of the Frieze were ever used to secure it. This leads into the materials science evidence concerning the Frieze.

The shape of the debate between Dr Bennett and Dr Nicola concerning the Frieze was similar in many ways to the debate concerning the two marble pieces. Dr Bennett relied upon various matters which, in her opinion, demonstrated that the Frieze was of modern origin. For his part, Dr Nicola could again not positively assert that the evidence demonstrated, on the balance of probabilities, that the piece dated from antiquity. In his view, the evidence was inconclusive.

For the same reasons as discussed in the context of the marble objects, the evidence of Dr Nicola does not ultimately assist JEL. Even if I were to accept that evidence in full (which I am not inclined to do), the Claimants’ case on inauthenticity would succeed on the basis of the art historical evidence. Furthermore, as with the marble objects, my conclusion of inauthenticity based on the art historical evidence makes it more likely than not that the matters relied upon by Dr Bennett are to be explained by modern manufacture, rather than (for example) restoration and conservatory work or the remarkable survival of an ancient object without any material evidence of weathering or biological activity.

I can therefore address the expert evidence relatively briefly.

There was considerable debate between the experts concerning a deposit which had been applied to the stone: Dr Nicola accepted that the presence of organic material suggested that a deposit had been applied, rather than that it had somehow formed naturally. However, Dr Nicola put forward a number of possible hypotheses concerning the origin or purpose of the layer. One hypothesis was that there had been a preparatory or pictorial layer, followed by a conservatory layer. But he considered that it was not possible to establish when these layers were applied; i.e. whether in modern or ancient times. Another possibility was patina reintegration, which in his view was commonly applied for restoration purposes until very recently. These were possibilities, as was the Claimants’ hypothesis that the deposit had been applied by a forger with an intent to deceive. However, it was in his view not possible to reach any conclusion in relation to authenticity on the basis of the deposit. He had not seen anything in the deposit that was reliable evidence of modern production.

By contrast, Dr Bennett considered that the deposit was the result of modern work. She considered that the deposit was a modern fresh applied deposit which was intended to simulate a burial deposit; i.e. the sort of deposit that might be formed if the object had been buried. Her report identified, with photographs, what she (visually) considered to be brush hairs which were incorporated within the deposit and which were visible to the naked eye, under a hand lens and under the microscope. She drew the conclusion, which was in my view reasonable, that the brush hairs had been used to apply the deposit, and it was therefore an indication of modernity.

However, even if this conclusion were incorrect – Dr Nicola in his supplementary report said that he could not see brush hairs embedded in the deposit – Dr Bennett had another powerful point indicating the modernity of the deposit. She identified (and photographed) what she described as the profiles of a modern pendant drill which were visible to the naked eye and under a hand lens throughout the decorative carving. A pendant drill produces engraved grooves of a totally different shape to those produced (in antiquity) by a chisel. These profiles were beneath the deposit, and thus necessarily indicated the modernity of the deposit.

Dr Nicola accepted (as discussed above in the context of the Hari Hara) that the interpretation of tool marks was beyond his expertise as a chemist and conservation scientist. He could not therefore dispute what Dr Bennett said about tool marks, which is an area in which Dr Bennett had considerable expertise. However, he suggested that these tool marks could derive from previous restorations or conservation treatments, and that they were therefore inconclusive as to modern manufacture. In my view, however, when viewed in the context of the art historical evidence of modern manufacture, it is far more likely than not that the tool marks are simply further evidence of modern manufacture.

Whilst the existence of the deposit, the brush hairs embedded within it, and the modern tool marks were all part of the overall picture on which Dr Bennett relied, perhaps the most significant point was the absence of weathering.

In their joint report, the experts agreed that there was no evident weathering of the stone. They also agreed that schists could undergo physical effects of exfoliation (which I describe below), and had a tendency to split along the planes of their metamorphic deformation (something known as schistosity). They were agreed that there were no accumulations of soil or debris in the cracks in the surface of the work. They also agreed that there was no visible evidence of root marks or the mineral deposits that accumulate along the paths of plant root surfaces during natural deposition. They were agreed that there was no iron staining in the rectangular sockets at the top of the Frieze.

In cross-examination by Ms Renton, Dr Bennett summarised the position as follows. The Frieze had a very uniform surface. There was absolutely no difference in the weathering – or rather the absence of weathering – of any of the figures. If the object had been buried, it would have been buried either face up or face down. If it had been exposed on an architectural structure, there would have been rain or sand or wind coming from one direction far more than the other. There were no fossilised root marks, which are often (but, as she accepted, not always) found on the surface of Gandharan schists. There was no evidence of any weathering deposit, either on the surface or within the fissures or folia of the schist stone. Those fissures were very clean. There was no “exfoliation”, which would normally be expected. This is because water penetrates through the foliae of the stone and dissolves the clay minerals, which then expand. There was also no evidence of any rising damp or mineral transport from the surrounding environment. There were also clamps or holes for fixing the Frieze, had it been used. The fixings would have been iron, but there was no evidence of any rust staining.

In my view, this evidence was persuasive, even if taken on its own as a standalone point. In that regard, for reasons similar to those which I have expressed in the context of the marble objects, I consider it inherently probable that the features described in the previous paragraph reflect modern manufacture. That is a far more probable explanation than (as Dr Nicola’s evidence postulated) the possibility that the Frieze was kept for most of all of the past 2,000 years in arid conditions or a “controlled indoor environment”, and that it has been subject to multiple conservation treatments aimed at removing all the signs of time on it.

However, Dr Bennett’s evidence in relation to weathering does not standalone. When considered in conjunction with the other matters relied upon by Dr Bennett (as summarised above), and the art historical evidence, it is in my view compelling evidence in support of inauthenticity and modern manufacture.

Bodhisattvas are spiritual beings who have progressed through numerous lives during which they have accrued sacred knowledge to the point that they are able to move away from the cycle of death and rebirth and attain Buddhahood. However, they generally choose not to do so, remaining within contact of the human sphere in order to guide and encourage others on the same path to enlightenment. Bodhisattvas seem to have been invariably male: there are in general not many female figures from Gandhara that survive.

The object with which I am concerned was described in the invoice from JEL as “Head of a Bodhisattva” and was attributed to 4^th century Gandhara. It is 42 cm in height, and is made of schist. There is no suggestion that it depicts Prince Siddhartha himself: i.e. the Bodhisattva depicted in The Frieze. It has huge loose hair, an elaborately decorated wreath wrapped around his hair-knot, and facial hair above the lips, at the chin and to the side of the cheeks.

The wreath contained a number of elements commented upon in the expert evidence. Towards the top there was a central ornament of a lion’s head, with strings of pearls coming from its mouth. There were 9 medallions or “rosettes” set within the headdress. These contained various figures, including two seated preaching Buddhas on either side of the front of the lion’s head. The strings of pearls finish adjacent to these two Buddhas. Other figures depicted in the medallions are Vajrapani, the Buddha’s spiritual guide, depicted by a thunderbolt (or vajra) held in the hand

The object is made of a fine-grained greenish stone that is found in some excavated Gandharan objects. It is common ground that it was well-carved, with damages only on minor features. It was in exceptional condition, with no trace of wear or corrosion that indicated a long-term exposure to the elements.

Dr Luczanits’ evidence was that the Head of a Bodhisattva was not authentic. His evidence covered: the potential location of such a head; a comparison with excavated objects; a comparison with a terracotta head in the Asian Civilisation Museum in Singapore; and iconological considerations.

Potential location: Dr Luczanits accepted that a few heads of the size of this object, and indeed even larger, have been recovered in Gandharan excavations. They belonged to massive figures occupying dedicated niches, which were added to the Gandharan monastic complexes in a period of time usually dated to the 3^rd century CE. In this phase, the mass production of images in stucco largely replaced the production in stone, albeit that there was an overlap between the two materials and the switch in materials may have happened in different areas at different times. The fact that the present head was in stone, instead of stucco, therefore tended to contradict the overall trend observable in Gandharan art.

Most of the large images were Buddhas, for which the massive size is justifiable. There were also large images of the Bodhisattva Maitreya, the future Buddha in our world: the largest Bodhisattva image in stone of the Lahore Museum is a Maitreya. That piece is clearly made for being placed in a niche, and is flat from the side. A few Buddha heads of comparable size from Taxila (in Pakistan) are preserved in stucco, but there is no Bodhisattva head of comparable size.

Dr Luczanits accepted that the fact that massive images were in niches made it theoretically possible that a head of the size and state of preservation of the Head of Bodhisattva may have survived unscathed. However, the three-dimensional quality of this head contradicted its usage in a niche, and pointed towards a free-standing image. There is scant evidence for such a free-standing image in Gandhara, and none in the time period into which the head was dated.

Comparisons with excavated objects. Dr Luczanits accepted that the Head of Bodhisattva did have some features which were comparable or bore some resemblance to those found in authentic objects. For example, the loose hair was characteristic of a peculiar Bodhisattva type for which there were a few (albeit rare) examples with a secure provenance. However, he identified various features which, in his view, indicated inauthenticity.

Thus, an authentic head in the Peshawar museum, dating most likely to the end of Gandharan production in stone in around the 3^rd century CE, had rather rough squarish features. By contrast, the Head of Bodhisattva showed much more naturalistic features, including the enormous hair with varied locks worked out in deep relief. These, as well as the awkward facial hair at the sides and on the chin and the elegant moustache, were catering to Western taste rather than representing Gandharan art. The hair at the chin and the cheek in front of the ear had no comparison in (indisputably) authentic Gandharan art.

Dr Luczanits thought that the Peshawar museum head, with a large mass of hair, was likely to have been one of the models for the head in question: for example, there were similarities in the circular formation of the hair locks at the root of the hair-knot. However, in contrast to the Peshawar museum head, the earrings on the Head of Bodhisattva had a bodhi-leaf pattern (rather than simply being metal earrings). There is no precedent in excavated Gandharan art for the usage of the bodhi leaf in such a way. Furthermore, the bodhi leaf was an awkward motif to use: it is associated with the awakening of the Buddha, rather than a Bodhisattva.

The ornate wreath wrapped around the hair-knot, decorated with a lion rosette and medallions with figures on them, was described by Dr Luczanits as unique. A garland headdress was found, albeit rarely, in Gandharan art: for example, a Bodhisattva Maitreya in the Asian Art Museum in Berlin. But the garland there was very different. In the present case, the garland wrapped around the Head of a Bodhisattva did not compact the hair, but sat on top of it as if it would have no weight. This made little sense, particularly given the rosette and medallions in the garland. It also contributed to the rather strange proportions of the head, not found elsewhere in authentic Gandharan art. It also enabled the head to be viewed in a front-on perspective. But if the head had been once mounted on a large free-standing image, it would never have been seen from this angle. There was no evidence that heads were made as stand-alone objects in Gandhara.

Comparison with Singapore terracotta head. A head with very similar characteristics was acquired by the Asian Civilisation Museum in Singapore in 2011. This head was itself sold to the museum by JEL. Dr Luczanits said that he had been most surprised to encounter the terracotta head when he had visited the Singapore museum. He also said that his discussions with the curator had indicated that it may be removed from display, because of concerns as to its possible lack of authenticity. Dr Luczanits made the point in his report that there is not a single case among excavated Gandharan art in which two objects in different materials (here terracotta and stone) relate to each other in such a similar way. Moreover, terracotta was itself not a common material for Gandharan art at all: it was not used as such, and the pieces which do survive were the result of accidental firing.

Iconography. Dr Luczanits relied upon a number of iconographic anomalies. I have given bracketed numbers to his individual points

(1) In excavated Gandharan objects, there are two principal types of Bodhisattva depictions. One type is Brahmanic, which has some form of bejewelled headdress. Loose hair is an early Brahmanic subtype. However, it was discontinued and gave way to a jewel-net-type headdress. This shift reflected the increasing divinisation of Bodhisattvas in early Mahayana Buddhism that is reflected in the art of Gandhara. Thus, the loose hair type had actually fallen out of use when Buddhas came to be depicted in headdresses. However, this head had loose hair with two Buddhas in the headdress flanking the central lion rosette.

(2) A laurel wreath would generally be preferred as a sign of victory, as can be seen in some authentic objects. The wreath in the present case had no clear meaning as a headdress.

(3) The figured medallions decorating the headdress were not possible in this combination in authentic Gandharan art. Figured medallions are not common at all on headdresses but Buddhas and other figures do occur in the central rosette. Where there is a Buddha figure in the central rosette, it denotes the temporal relation between this Buddha and the Bodhisattva wearing it. Even if there is more than one Buddha represented, a central one is indispensable. In the Head of Bodhisattva, however, the two images of Buddha are only a secondary decorative feature in relation to other medallions spread across the sides of the headdress.

(4) One of the medallions has a Bodhisattva with a raised hand. This appears to have been taken from a figure in a stele depicting a main Buddha, and numerous other figures, in the Lahore museum. However, a turbaned Bodhisattva should not hold a flask, as it does in the Head of a Bodhisattva: this is an element which belongs to different Bodhisattva types. Also, authentic Gandharan Bodhisattvas should not sit on lions.

(5) There are two figures, known as Vajrapani, each holding a “vajra” or thunderbolt. This doubling of the vajra holding figures, is meaningless, as is their symmetrical composition.

(6) The wrathful face on one of the medallions at the back is unusual. It is reminiscent of some faces on Hindu sculpture in Kashmir. This is an esoteric or tantric element which is not documented in Buddhist sculpture elsewhere in the period that the head is supposed to relate to.

(7) The central position on the headdress is taken up by a decorative figure, the lion head spitting pearls. With this feature, the entire composition of figured medallions around the head appears absurd. Gandharan headdresses often feature a variety of figures and even fable animals. They are, however, composed in relation to a central feature and often are also directed towards it. Here, however, placing Buddhas at the sides and relating the other medallions (with Bodhisattvas and other figures) to them has no precedence, but is also a caricature of Buddhism. That these figures are composed of elements that do not belong to the same phase of the development of Buddhism further attested, in Dr Luczanits’ view, to the “farcical nature of this composition”.

Accordingly, Dr Luczanits concluded that it was not possible to give satisfactory answers to questions as to the overall meaning of the Head of a Bodhisattva and how it fits into what was known about Gandharan art and the development of Buddhism in the region. There was no overall meaning this sculpture would allow for within a Gandharan context or even beyond it. In particular, the doubling of the Buddhas on the medallions, and their secondary position in the headdress, would force one to interpret this Bodhisattva head in cosmic terms, and even then, the figuration of the head could not be explained. The figured medallions themselves combine early types in Gandharan art, such as the seated Vajrapaṇi, with later ones, such as the Bodhisattvas. The animal vehicles for the Bodhisattvas and the wrathful head added a third historical layer that dated after Gandharan art altogether.

However, to make a cosmic or esoteric Buddhist interpretation possible at all, the Bodhisattva head would then need to come at the very end of the development of stone sculpture in Gandharan art, which, as is evident from the sculptures of Gandharan Buddhist sites such as Taxila, Shari Bahlol or Butkara I, goes hand in hand with a diminishing of carving quality. This head, however, would then go against this development documented in excavated art.

Overall, this was extremely far-fetched, and there was no doubt that this head, with a fancy headdress, is a modern invention, in part fed by knowledge that is not restricted to Gandharan art alone.

Dr Jager summarised his conclusions as follows.

He described the Head of Bodhisattva as having a serene face framed by beautifully carved loose curls and an elaborate headdress. It was possible from the lion headdress that the Bodhisattva depicted is Shakyamuni, which means “Lion of the Shakyas”, who was identified by some Gandharan Buddhist sects as the Buddha prior to his final rebirth as Prince Siddhartha. The fact that it was not possible conclusively to identify the bodhisattva depicted is not unusual. The central images on headdresses of bodhisattvas could depict a variety of subjects, and sometimes have no known divine or spiritual association at all; or indeed no textual explanation known to us. Indeed that can be said of their iconography as a whole. There were many elements, the precise meaning of which remain a mystery.

The Head of a Bodhisattva sits in the context of clearly comparable Greater Gandharan examples, such as the famous Ortiz Head which Dr Jager considered to be close in both style, appearance and the extreme skilfulness of the workmanship. The diverse imagery found in the headdress was iconographically sound and reflected the diversity and richness of the influences at play in Greater Gandharan art. It was a unique piece and rare in its quality with diverse elements that reflect the richness of Greater Gandharan art. Dr Jager did not consider that any of the elements in the Bodhisattva Head could be said to be anachronistic or indicate a lack of authenticity.

He would date the Head of a Bodhisattva to around 4th or 5th century CE (although it could possibly be earlier) and considered that it comes from the Greater Gandharan region. Most of the large-scale figures of Buddhas and bodhisattvas from Greater Gandhara are now believed to come from the 3rd to 5th century CE period.

He therefore considered that the Bodhisattva Head was authentic.

In his written evidence, Dr Jager compared the Head of a Bodhisattva with other heads, in particular the terracotta head in the Singapore museum and the white marble Ortiz Head (discussed in relation to the Head of Dionysus above). He drew attention to various similarities; for example, both the Head of a Bodhisattva and the Ortiz Head had intricate, beautifully worked locks of hair, carved out in deep relief. These two heads had a very similar style in the form of their workmanship.

Dr Jager also drew a comparison with a large Bodhisattva in the Lahore Museum. This was made of dark grey schist. It was, as Dr Jager recognised, a full length image, and “not an acrolithic one”. But it was possible that the Head of a Bodhisattva and the Lahore Museum object came from the same workshop. He also referred to a terracotta head (not alleged to be of a Bodhisattva) in the Metropolitan Museum of Art, as well as a Bodhisattva head in the same museum.

In relation to Dr Luczanits’ points concerning the iconography of the Head of Bodhisattva, Dr Jager recognised that this was not his area of specialism. However, he nevertheless considered that the medallions (which Dr Luczanits had addressed at some length) were iconographically sound. One should not expect to be able to give a precise and reasoned explanation for every iconographic element of a Gandharan bodhisattva. It was well recognised that many elements of their portrayal remain a mystery to us. We cannot now say in respect of each piece of imagery when precisely it was used, or ceased to be used.

I again considered that Dr Luczanits’ evidence was persuasive and authoritative (far more so than that of Dr Jager, upon which the Defendants ultimately placed no reliance) and that the Head of a Bodhisattva is not an authentic Gandharan piece. Whilst I accept, generally, Dr Luczanits’ evidence in relation to this piece, I regarded the following matters as particularly important.

First, it is improbable that the Head of Bodhisattva would have survived, in such exceptional condition, for around 2,000 years, with no significant damage affecting the overall aesthetic of the piece. I accept Dr Jager’s point that there have indeed been some notable and remarkable objects found, in good condition, as shown by some of the photographs (for example of an object in the British Museum) which Dr Jager produced. However, survival in good condition is very much the exception to the norm, and this is a relevant factor to take into account when assessing authenticity. Of course, the improbability of survival in good condition, with the odds heavily stacked against it, cannot be a decisive consideration when assessing the authenticity of antiquities. Otherwise, one might reach the perverse conclusion that, on the balance of probabilities, no antiquity is authentic. Nevertheless, it remains an important factor to be weighed in the balance.

Secondly, I think that Dr Luczanits made powerful points in relation to the potential location of this object. The Head of a Bodhisattva has the appearance of being a standalone object: a very well carved head which appears designed to be displayed as a head, in other words as a free-standing object. The piece can be walked round, with much to admire on this circular tour, including detailed medallions at the rear. However, as Dr Luczanits said, this three-dimensional quality contradicts the usage that would be expected of such a piece; namely as part of a large piece that would be placed in a niche. There is no evidence that free-standing objects of this kind were made in Gandhara in the relevant period. In cross-examination, Dr Jager accepted that the stucco heads which have been excavated, and which were in niches, were essentially forward-facing representations, although he thought that there might be some stucco heads which had an all-round aspect.

Thirdly, the date of the piece, as stated in the invoice, was 4^th century CE. Dr Jager dated the piece to around the 4^th to 5^th century CE, although he said that it could be earlier. However, as Dr Luczanits said, the overall trend in Gandharan art by around the 3^rd century CE was for mass production of pieces in stucco, rather than stone. Dr Jager agreed in cross-examination that this was so, on the basis of what was known.

Fourth, I considered that Dr Luczanits made convincing points, as summarised above, in relation to the iconography. Dr Jager acknowledged that he had no real expertise in this area. Although Dr Luczanits was cross-examined on some aspects of his evidence on iconography, I did not consider that any damage was done to his evidence on that topic, or indeed generally. On the contrary, I thought that the points which he made in relation to the anomalies in the iconography, and related points which he made when comparing the Head of a Bodhisattva with other authentic objects, were both well-explained and well-made. I see no basis on which I can, or should, reject his evidence that (as he said in cross-examination) the distribution of the medallions and the iconography were clear indications of forgery.

Fifth, the expertise of Dr Luczanits as an art historian with particular knowledge of Gandharan art does not simply concern his conclusions on iconography. Just as an expert in Old Masters might have an ‘eye’ for what is genuine, I consider that Dr Luczanits’ expertise and experience enables him to identify features of pieces which are aimed at being attractive to Western purchasers, and which are not features of authentic pieces. Dr Luczanits identified a number of such features on this piece. In a telling part of his cross-examination, Dr Luczanits explained that forgers catered for western taste, because it was “the west who buys these objects. Nobody in Gandhara buys them”. Mr Green then put to Dr Luczanits a passage in the evidence of Dr Jager, where the latter described the realism and naturalism of Gandharan art, including the beautifully rendered hair of the piece in question. Dr Luczanits was able to say, with an authority which Dr Jager lacked: “Essentially, this doesn’t describe Gandharan art at all. This describes the Roman taste, or the western taste, of what they like on Gandharan art”. True Gandharan art has much less sensuality and emotion.

Sixth, I thought that Dr Luczanits also made a very powerful point in relation to improbable similarity between the Head of a Bodhisattva and the terracotta piece now in the Singapore museum. It was, as he said in cross-examination, a “unique” case in Gandharan art where there are two heads “completely unusual in different materials of perfect quality, preserved, 2000 years old”. There are, on the evidence before me in this case, significant doubts as to the authenticity of the Singapore piece; a doubt now apparently shared by the curator in Singapore. However, even if it were to be accepted that the Singapore object is genuine, it would be highly improbable that a very similar piece was made, and indeed survived, in a completely different medium. This is reinforced by the fact that the skills required to sculpt a terracotta object (which would involve building up the sculpture in clay) would be very different to the skills required to sculp in schist (which would require chiselling away at stone). As Dr Jager acknowledged in cross-examination, the skills involved are “opposing processes”. Furthermore, not only is there an improbable similarity but, as Dr Jager accepted in cross-examination, the Head of a Bodhisattva and the Singapore museum piece are unlike anything else to be found in the archaeological record.

Seventh, I did not consider that any of Dr Jager’s other comparative pieces, on which he drew as support for authenticity, carried the case any further. They certainly did not outweigh the force of the points already described. For example, the Ortiz Head is in a private collection, and the evidence (already discussed in the context of the Head of Dionysus) indicates that there must be doubt as to its authenticity. Dr Jager also referred to other pieces which are in private collections, and (again as previously discussed) it is necessary to be wary about the authenticity of such pieces. Reference was also made to two pieces in the Metropolitan Museum of Art in New York. One of these was a Bodhisattva head, allegedly dating to the 3^rd/ 4^th centuries. Whilst some features of this head could be said to be comparable to the Head of Bodhisattva (for example the moustache), it could not be suggested that all of the features on which Dr Luczanits commented (adversely) are present in the Met piece. For example, the anomalous medallions, which formed much of Dr Luczanits’ evidence on iconography, are not present in the piece. Thus, neither the Ortiz Head, nor the two heads in the Metropolitan Museum of Art, had the elaborate wreath containing the various medallions.

In that regard, it should also be borne in mind that Dr Luczanits accepted that, in relation to various features of the Head of a Bodhisattva, a comparison could be drawn with features found in indisputably genuine objects. However, it does not follow that the existence of such comparators establishes the authenticity of the piece that I am considering. The comparators are, at least, equally or even more consistent with such comparators forming the inspiration for the head that I am considering. For example, one of the medallions shows a Bodhisattva seated on a lion raising his hand above his head. This appears to have been inspired by one of the figures in the stele in the Lahore Museum.

Overall, in my view there is a considerable weight of evidence which supports the conclusion that the piece is not authentic.

The nature of the debate between the experts covered much of the same ground as was traversed in relation to the Frieze. Again, Dr Nicola did not give positive evidence that the materials science demonstrated that it was more likely than not that the Head of Bodhisattva was ancient. Ultimately, the effect of his evidence was that the materials science did not enable a conclusion to be drawn one way or the other.

As with the Frieze (and indeed the marble objects) there was again evidence of additional layers of material above the stone itself. In his principal report, Dr Nicola identified these layers as being a white substrate layer with a thin red layer on top. The white substrate, which he described as a deposit, was made principally of calcium carbonate. The thin red layer was probably composed, in his view, of iron oxide and aluminium silicates. He said that he had found nothing in either layer that contraindicated an ancient production. However, he also said that it was not possible to ascertain from the tests available whether these layers were ancient or modern. He could not exclude the possibility that they were the result of the application of a modern patina, whether as restoration for aesthetic purposes or as an attempt to deceive. If they were of modern application, it was in his view not possible to discern the intent of the individual who applied them.

There was a substantial issue between the experts as to whether Dr Nicola was correct in his conclusion that there was nothing in the thin red layer which contraindicated ancient production. Dr Bennett’s work led her to the conclusion that the contents of that layer evidenced the use of Portland cement. There was a substantial debate between the experts as to whether this was so. Indeed, Mr Green’s cross-examination of Dr Bennett on the Head of Bodhisattva was principally directed at that issue.

I do not consider it necessary to resolve that particular dispute between the experts. It is sufficient in my view to reach the conclusion that the two layers are of modern origin (which Dr Nicola accepts as a possibility) simply on the basis that this is the conclusion that I have reached on the basis of the art history.

Furthermore, I do not consider that the evidence as to the layers is the most significant point in any event. Towards the beginning of her cross-examination, Dr Bennett was asked whether her point on Portland cement was a significant part of her reasoning. Her answer was that it was “part of the story of the object”, but she explained (in a passage quoted in Section D above) that one had to be very careful when considering a deposit because deposits could occur when a genuine object was overcleaned or had some other process applied to it:

“So, you know, an applied deposit is – you note it, but it may not be the most paramount for the authenticity of an object because the actual weathering of the object itself, that’s the most important in terms of authenticity”.

I considered that this was a sensible and persuasive answer, and it will be apparent from this judgment that I have been persuaded that (in relation to all of the objects) the absence of evidence of weathering is a significant point which confirms and reinforces the art historical evidence as to the inauthenticity of the various objects.

In relation to weathering, Dr Bennett’s evidence on the Head of Bodhisattva was very similar to that which she gave in relation to the Frieze.

She said that some of the most informative evidence of an ancient date of a Gandharan sculpture comes from the physical consequences of weathering that are due to effects such as rising damp and the transport of minerals from the environment and the groundwater onto the sculpture itself. Weathering is highly dependent on the prevailing environmental conditions. The different mineral phases will be susceptible to different weathering processes dependent on the ambient conditions of temperature and moisture. She made what seemed to me to be the important point, in the context of both the Frieze (as already discussed) and the Head of a Bodhisattva, that the differing weathering processes will not therefore be distributed over an object’s surface. In the case of the two schist objects, it seemed to me that a very powerful point against authenticity was the pristine condition of the whole object, without any discernible difference between areas which might have been more susceptible to weathering than others.

Whilst accepting that the extent of weathering will be dependent on the ambient conditions of temperature and relative humidity, she said that over prolonged time, even in dry conditions, there would be a degree of surface alteration. Deposits caused by root growth were commonly observed on genuine ancient Gandharan sculpture, particularly if an object had been buried. Ancient carved schist sculptures and friezes often exhibit fractures and exfoliation. Because schists are relatively soft stones, they are easily abraded and scratched (referred to as “physical weathering”). The effect of weathering will include the reprecipitation of iron oxides and hydroxides created by the dissolution of iron-bearing mineral phases within the schist, creating “iron staining” on the surface.

In the case of the Head of a Bodhisattva, Dr Bennett described the absence of weathering. There was no evidence of any natural weathering of the stone surface. The uniform unweathered appearance of the surface of the work with no signs of any degradation of the stone is “exceptional and unrealistic”. There were no mineral encrustations which would have been acquired under conditions of natural exposure or of long term burial. There were no deposits caused by root growth of the sort frequently encountered on ancient Gandharan stone. Apart from a few small chips, there was no physical weathering. There were no natural abrasions such as would be expected on ancient surfaces, no evidence of any natural wear and no patters of superimposed scratches – even though the stone was so soft it could be scratched with a fingernail. There was no dissolution of the iron-bearing mineral phases within the chlorite schist. There was no chemical variation in the composition of the stone nearer the outer surface.

Dr Nicola’s answer to this point, as summarised in the joint report, was that under normal conditions of weather exposure, schists would not undergo mineralogical changes or dissolution over historical time spans. This type of rock is not normally subject to very evident weathering. The limited weathering was not surprising, because the stone had been coated by a preparatory layer which would have been quite protected from physical weathering. The presence of root marks was dependent on the conservation history. In his view “given the presumed conservation environment is likely to be very arid, significant biological attack was unlikely”.

On this issue, as will be apparent from my earlier discussion of similar points, I consider that Dr Bennett’s view is far more persuasive than that of Dr Nicola. It is a view which is supported not only by common sense, but also by the evidence of Dr Luczanits who has great experience of seeing ancient objects. On the balance of probabilities, the absence of any material evidence of weathering is in itself sufficient to establish inauthenticity. However, as previously indicated, the absence of material evidence of weathering does not stand alone. When viewed in conjunction with my conclusions on the art historical evidence, the most probable explanation for the absence of weathering is recent manufacture.

This object is made (at least principally) from unfired clay: the materials scientists were agreed (in their joint report) that it is made of clay, that the material did not have the structure of a fired object, and that it could be described as being made of a clay matrix. Unfired clay is the least durable of the materials from which any of the objects in issue was made. It is also, obviously, less durable than clay which has gone through a firing process, for example terracotta.

The object is the wrathful head of a demon like figure. In its gaping mouth there is another face, which can fairly be described (as Dr Luczanits described it) as a rather sad-looking middle-aged man. Dr Jager did not consider the middle-aged man to be sad, but rather as having a worried or almost haunted look.

The head was said by Dr Luczanits to have, overall, a pale blue colour, although it is fair to say that this colour is not particularly apparent. Rather clearer is the reddish colour of the main head (i.e. not the head in the gaping mouth). There are two short horn-like appendages, on the top of the main head, and these project its wavy hair.

The piece was referred to in the JEL invoice as the “Head of a Krodha Vighnakarta”. The invoice further stated its origin and date as Greater Gandhara; and 5th/ 6th century, and also that it was “Unfired clay with traces of coloured pigments” and that its height was 34 cm.

During the trial, this was usually referred to as the Head of a Krodha. The invoice reference to “Krohda Vighnakarta” should have been Krodha Vighnantaka in a Buddhist context. This is a wrathful remover of obstacles in early esoteric Buddhism. Unless otherwise appropriate, I shall simply refer to this object as the “Krodha”.

Dr Luczanits’ evidence was that, iconographically, he had not seen an object like this previously. In his written evidence, he described it as a unique piece. In cross-examination, Dr Jager agreed that, as far as Gandhara was concerned, he was not aware of any comparator in the accepted record to the wrathful face containing a face.

Dr Luczanits’ evidence that this was an inauthentic object was reached after consideration of its potential location, condition, iconography, and style and colouration.

Potential location. Dr Luczanits said that there was no direct comparison for this object that would reveal its potential location. He said that the frontality of the object (in other words, the fact that it is clearly facing forward, and had been so designed) would make one assume that, if it were original, the head would have crowned a sculpture. However, there was no deity in the relevant region that would have a face in its mouth.

There were some sculptures which did include this element. But these are not main sculptures, and they would be expected to be modelled in a way that they are to be seen at an angle. Thus, Dr Luczanits described how a face inside the mouth might fit one of the demons attacking the Buddha at his awakening. However, this Krodha would not have belonged to such a sculpture for a large number of reasons. Gandharan art rarely used the dramatic features for which this scene is known in post-Gandharan art. The only clay depiction of this scene dates from the 11th century at the earliest. If this Krodha had belonged to such a scene, then it would have to have been part of a very large depiction. If this well-preserved head had survived, one would expect other parts of the composition to have come to light as well. In addition, the scene would not have involved the head being represented frontally, and the sad middle-aged face in the mouth would make no sense in the context of that scene.

Condition. It would be nothing short of a miracle if an object of this type had been preserved for 1,400 years in the condition in which the Krodha was in. Although unfired clay was the preferred material of sculpture in the second half of the first millennium CE, it is extremely rare that objects of this size were recovered either archaeologically or in situ. Even then, they are not found in this pristine condition, including the preservation of a thin paint layer. Even in regions of relatively dry climate, recovered clay objects were predominantly fragments, with a few rare pieces that could be restored into a comparable condition. Clay objects are not usually preserved in good condition at all, even less so if they are more or less solid clay, as this object appeared to be. Here, all elements of the crowns and headdress are perfectly preserved, when even in situ such elements are most vulnerable.

Iconography. Dr Luczanits had never seen an object like this previously. The identification of the head as Krodha Vignakarta made little sense and had no support. The head showed none of the characteristics that would identify the head as belonging to an early esoteric image, and such a deity would not have a peaceful, sad looking face in his mouth. In fact, the earliest depictions of this type of deity post-dated the date proposed for this piece. In addition, wrathful deities would wear jewellery, and the horns have no comparison in the region. For reasons summarised above, Dr Luczanits considered and rejected the idea that the head might have formed part of a scene of demons attacking the Buddha at his awakening.

Style and colouration. Stylistically and aesthetically, nothing about the head associated it with the region and time period it was attributed to. Comparable heads originally modelled in clay are known from a number of sites, but their semi-wrathful form is based on entirely different artistic conceptions. Wrath is shown by frowns and bulging eyes, and if the mouth is open it likely has fangs and indicates the emitted cry or roar. Heads from Central Asia (for example a 10th/ 11th century piece from Khocho), were closer to the Krodha. But even then, they were not directly comparable. The Khocho piece is from a different region, a different time period, and it has fangs in its mouth, not another head. Whilst the wide gaping mouth of the Krodha could be compared to Hellenistic fountainheads – such as the one from Ai Khanum – there were again significant differences: the Ai Khanum head has no proper lips, and does not have an object in its mouth. The Hellenistic period was many centuries earlier than the 5th/6th century CE to which the Krodha was dated: the Ai Khanum site has been dated to circa 280-145 BCE.

The colouration did not conform to traditional Gandharan polychromy, but compared better to Central Asian examples. Gandharan tradition often had dark areas of the face, such as corners along the nose and ears, accentuated with a red line. This feature is missing in the head in question.

Overall, Dr Luczanits considered that everything about the object conformed to western taste and a western cultural background. The Krodha appeared to be a pastiche, combining elements of clay images from the entire region, including Central Asia, with elements not known there at all, or deriving from western motives, such as the horns deriving from the Satyr.

Dr Jager described the main head as resembling Hellenistic depictions of a satyr with a wide-open mouth. He said that technically, the Krodha corresponded with the body of painted, unfired clay sculptures which survive from the Greater Gandhara region (many in excellent condition), where this was a known and frequently used sculptural medium.

Stylistically and iconographically the elements of the Krodha are all referable to and consistent with origin from the Greater Gandhara region. The larger satyr’s head reflects the strongly Hellenistic influences, including the enduring use of Dionysiac images in Gandharan art (satyrs being the companions of the God Dionysus). The concept of one head enclosed within another also derives from Hellenistic influences and is found within Gandharan art. So is the use of grotesque imagery that is found here. The head as a whole reflects the syncretism which was such a prominent feature of Gandharan art.

It was possible that the object depicted a Krodha Vighnakarta (a Buddhist concept meaning the wrathful destroyer of obstacles, which emerged in the 5th and 6th centuries CE). Its precise iconographic meaning could not certainly be known. He did not, however, consider this to be problematic. The precise iconographical meaning of many objects of Gandharan sculpture remains unknown to us. The fact that an ancient sculpture’s ultimate meaning cannot be fully understood by a modern audience does not detract from its authenticity. That is particularly so in a case where the image is as unique as this object. The “book” of iconography should not be considered as closed. Technically, the head made proper sense in the extensive group of Gandharan unfired sculptures known from excavations.

Dr Jager considered that the uniqueness of the piece was in his view an indicator of authenticity. It was unlikely that a forger would venture to create something so unusual. He did not consider that any of the elements in Krodha could be regarded as anachronistic or indicating a lack of authenticity.

He therefore considered this to be an outstanding and rare piece of sculpture from the Greater Gandhara region. He would place the date of the object as between the 1st and 6th centuries CE. He considered that it was authentic.

Again, I considered Dr Luczanits’ evidence and opinions to be far more convincing than the evidence and opinions of Dr Jager, upon whose evidence the Defendants ultimately placed no reliance. The most significant points in my judgment are as follows.

It is inherently improbable that a piece such as the Krodha, made from unfired clay, would have survived over 1500 years so as to be in the condition in which it now exists. This is in my view a weighty point against authenticity, albeit that I accept that it is not in itself conclusive; because it is theoretically possible that there might have been very extensive restoration. This possibility does not seem to have featured, in Dr Jager’s written report, as the reason for the excellent condition of the piece. Rather, the thrust of his evidence is that pieces do, surprising as it may seem, survive from antiquity in good condition.

I consider, for the reasons given by Dr Luczanits, that the possibility of an unfired clay piece surviving in its present condition, from antiquity, is so remote as to be fanciful. The choice is therefore between the possibility that this is a modern forgery, or that it is a heavily restored antiquity. It seemed to me that there was very little if anything which indicated that it was a Gandharan antiquity at all, and that all of the evidence pointed to a contrary conclusion.

Thus, it was common ground that the piece is unique, and that there is nothing to which it can be compared in authentic Gandharan art. Indeed, there was also nothing to which it could be compared in any Gandharan art, whether authentic or allegedly authentic: in other words, Dr Jager could not point to any comparable piece at all, even a piece in a private collection, as to which there might be an argument about authenticity. I consider that the uniqueness of the piece, and the absence of any comparator in authentic Gandharan art, is a very telling point against authenticity; at least in the absence of strong countervailing evidence, which in my view does not exist here.

I do not accept Dr Jager’s argument that uniqueness is a sign of authenticity, because a forger would have to be very imaginative. The evidence in this case, taken as a whole, indicates to me that there are forgers who are both skilled and imaginative, and who are adept at producing objects that will appeal to western tastes.

Ultimately, Dr Jager’s evidence consisted of a reasoned argument as to how, based upon the continuing impact of Hellenism, a piece such as the Krodha might have been produced. I regard that as no more than a theory. It depended, in significant part, on an argument that iconography did not really matter; because each piece had its own special iconography, and much was unknown about Gandharan art. This argument was advanced in order to counter the fairly obvious point that the absence of any comparable object is an indication of inauthenticity. Dr Jager’s argument, that each sculpture “will have its own iconography”, was put to Dr Luczanits. His response was neatly expressed and convincing: “If each sculpture has its own iconography, we don’t need iconography as a discipline”. Indeed, it is clear to me that iconography is of immense importance in assessing the genuineness of pieces, and that it cannot be swept aside on the basis that each piece is unique.

I also consider that Dr Luczanits, as is apparent from his report and as he said in cross-examination, considered the issue of authenticity from a number of different perspectives. The condition of the object was, as he said, “very important”; but the miraculous condition of the piece was not the only criterion. This was a frontal sculpture, which must have been life-size at least. But there was no “religious context” that existed for such a piece. It was, he said, interesting that Dr Jager had drawn an analogy with a satyr. There was, however, absolutely no evidence that in the 6th century satyrs were depicted in this area. The difficulty of dating the piece was also telling: Dr Jager had dated it as between the 1st and 6th century, which is 500 years, and in “Greater Gandhara” which was a huge region. I considered that all of these points, which were made in response to questions in cross-examination, were powerful and convincing.

Accordingly, I conclude that the weight of art historical evidence is that this piece was not authentic. This is sufficient to establish the inauthenticity of the object, since there is nothing in the evidence as to provenance, or materials science, which supports a conclusion or provides evidence of authenticity. Indeed, as discussed below, I consider that both aspects serve to reinforce the conclusion of inauthenticity.

The Defendants did not provide any evidence as to the provenance of this piece, beyond the fact that it was purchased by JEL from FEAL approximately a year prior to the sale to the Sheikh. The invoice for that sale gave a very different description of the object: it was purchased as “Head of Sathirus” from “Afghanistan”, and was dated to the 3rd/4th century CE. There were no documents evidencing any prior sale, the length of time that it had been owned by its previous owner, or how the object had come onto the art market in the first place. These facts provide no support for the authenticity of the piece. On the contrary, they add to the weight of evidence of inauthenticity.

There were numerous disagreements between the materials scientists which were debated at length in a joint report running to nearly 17 pages of single-spaced text. As with all of the other objects (apart from the Hari Hara), however, Dr Nicola did not suggest that the materials science enabled a positive conclusion of authenticity to be reached. His argument was the familiar one, namely that the evidence of modern materials and work was consistent with conservation and restoration. In their closing, the Defendants submitted that the key issue was the extent to which this object had been restored.

There was no dispute, and in any event I accept Dr Bennett’s evidence, that there was very substantial evidence of modern materials and work in relation to the Krodha. For example, Dr Bennett drew attention to, and photographed, a visible piece of plastic protruding from the cheek: Dr Nicola accepted that this was visible on a visual inspection. Dr Bennett’s evidence showed, that there were fragments of plastic sheeting in the object. She also identified, in her samples, various modern particles such as modern brass, nickel chrome steel with a vanadium hardener, gold dust and pure molybdenum. Dr Nicola accepted that this was what Dr Bennett had found (albeit that his own samples may perhaps not have shown the same thing), but identified in his cross-examination the possibility that particles in a workshop could be absorbed and trapped into a deep part of the object. Dr Bennett also identified modern fibres protruding from the surface of the work, and deeply embedded within the matrix of the samples. She also identified modern tool marks.

It was, no doubt, the evidence of modern work and materials that led Dr Nicola to say, in his oral evidence, that he was sure that the Krodha was not 100% an original object. He said that he did not know the exact percentage that was original. It could be that it was 80% original, 50% original, 30% original. He did not know. In his second report, he had said that there were quite widespread areas of apparent restoration across the Krodha.

I have no doubt, however, that the evidence of modern work and materials cannot be explained by the theory of restoration and conservation of an ancient object. The reason that there is this evidence of modern work and materials is because the object is modern. In order to reach that conclusion, it is sufficient to refer to three fairly simple points.

First, the art historical evidence proves modernity and therefore inauthenticity. Accordingly, modern manufacture is the likely, indeed obvious, explanation of modernity in the work and materials discussed in the materials science evidence.

Secondly, it is inherently improbable that an ancient unfired clay object would have survived for 1400 years at all, let alone that it would have survived so as to be in the condition that it now appears. As Dr Bennett said, the very friable nature of the work would make it highly improbable that such a large piece of fragile clay could have survived from the 5th/6th centuries to the present day. Dr Nicola said that, in evidence, that it was “very strange” for an object of this age, in unfired clay, to survive into modern times in this condition. His suggestion, as with some of the other objects, was that the object may have benefited from extremely favourable storage conditions. As he said in the joint report, the item had presumably been kept in more controlled environments than, for example, objects stored on archaeological sites, and that the Krodha was likely to have been preserved in an enclosed grotto or cave. There was a “presumed very arid conservation environment”.

However, there is no evidence at all to support the thesis that this is where the object was in fact kept in such favourable conditions. In my view, the likelihood is that the object would (if ancient) have been exposed to a range of ordinary weather conditions which would impact upon it. The overwhelming likelihood, in my view, is that the reason that the object now exists in its current condition is that it was manufactured recently.

Thirdly, Dr Bennett made (as with other objects) a number of powerful and convincing points in relation to the absence of evidence of weathering. The experts were agreed that under normal conditions of weather exposure, an unfired clay object would show extensive deterioration. They agreed that the head crumbled easily and was very friable. They also agreed that there were no evident accumulations of soil or debris in the cracks of the surface of the work. They also agreed that there was no visible evidence of root marks.

Dr Bennett said that the appearance of the entire surface of the work is very clean and devoid of dirt or dust embedded in it. This was not consistent with the likely appearance of an ancient clay object. She said that it would not have been feasible to effectively remove all dust and debris from the surface of an unfired object such as this, without causing obvious damage to the surface. There was, therefore, an absence of natural weathering over the entire surface of the work, and absence of any deposits within the cracks in the work, and the presence of artificial distressing over the surface. All of these matters indicate that the entire work is of modern manufacture. I consider that this evidence was far more convincing than the view of Dr Nicola that these matters could be explained by extensive restoration, with the piece benefitting from “a lot of care and attention in the years since it was found”.

Accordingly, the materials science evidence, including the views of Dr Bennett, reinforces the conclusion that I have reached on the basis of the art historical evidence.

Before leaving the Krodha, I return to the point discussed in Section D above concerning the Defendants’ criticism of Dr Bennett, and the suggestion that she knowingly made false statements in a number of documents. Those statements concerned the presence of ground-up ceramic particles in the Krodha. In order to explain this point, it is necessary to describe the history of the reports that Dr Bennett made.

When Dr Bennett wrote her first report for the Claimants, prior to the commencement of the proceedings, she came to the conclusion that the Krodha was made of unfired clay: in other words, clay which had not gone through a firing process, unlike for example terracotta or other ceramics. Ultimately, this was common ground between the experts, who (as set out in the joint report) agreed that the head could be described as having a clay matrix, and agreed that it did not have the structure of a fired object. Indeed, at the start of her cross-examination, the following passage from Dr Nicola’s report was put to Dr Bennett:

“The analysis carried out on samples from the item suggest that the object is overall made of unfired clay or at least of clay that never reached temperatures above 500 degrees and this makes it not possible to conduct thermoluminescence testing on all samples”.

Dr Bennett said, without hesitation, that Dr Nicola was right.

The above question asked referred to “thermoluminescence testing”, often abbreviated to “TL testing”. TL testing can be used to date ceramic objects. It dates the last time that a ceramic was fired. It is not necessary to explain in detail how this process works: it is based on the physical principle that the time at which objects are pyrotechnologically manufactured at temperatures above 500 degrees C is dependent on an accumulation of electrons and holes with the lattice of certain minerals and other solid state phases.

Subsequent to her first report, samples of the Krodha were sent for TL testing. Such testing is carried out by specialists, and it was not something that Dr Bennett did or could do herself. They were sent to a company called CIRAM. This report concluded that the TL results indicated “an ancient firing and may be consistent with the presumed period of manufacturing”.

It is clear that Dr Bennett accepted, at the time, that this TL analysis was correct. She was then concerned to examine whether the results meant that the Krodha was indeed of ancient origin.

She then wrote a second pre-action report, which referred to the CIRAM results. The aim of her investigation, as described in that report, was to “establish whether the head was a ceramic object which was fired 1650 years ago, or whether it was made of ground fired ceramic particles which were of the reported age and which were held together with a synthetic binder”. The conclusion of her further technical examination was that “while the ceramic material may date to 1650 [plus or minus 150] years old, the object itself was fabricated using ground up ceramic particles held together with a modern binder”.

Contrary to the Defendants’ submissions, I did not consider that this report, read in context, was intending to suggest that Dr Bennett had done her own TL or other analysis in order to identify the presence of ceramic materials. Nor was she seeking to mislead anyone. She was quite clearly using the TL analysis of CIRAM, the specialists. The purpose of her investigation, as she explained both in the report and in evidence, was to see whether or not this was a fired ceramic object, or whether it was made up of particles of something. She was seeking to determine, and did determine that it was not a ceramic structure. As she said in evidence: “it was made up of a chaotic mix of particles fired or not fired”. In the joint expert report, the experts were (as I have said) agreed that the material did not have the structure of a fired object. This was essentially what Dr Bennett was seeking to establish in her second report.

Subsequently, however, some further TL testing was carried out by Oxford Authentication, who are a leading specialist in this area. They came to the conclusion that, contrary to the CIRAM work, all the samples which they analysed were made of unfired material so they could not be tested.

Dr Nicola’s view, as expressed in the joint report, was that the results of CIRAM’s testing was not reliable. Dr Bennett did not, as I read her court report and the joint report, express a definitive view as to whether CIRAM’s results were unreliable. It is clear from those reports that it did not matter to her analysis of inauthenticity whether CIRAM or Oxford Authentication were correct. Indeed, she made this clear in paragraph 7.3 of her court report:

“7.3

My conclusions

My conclusion that the Work is not ancient is not based on whether it was made of unfired clay as indicated by Oxford Authentication or fired clay as indicated by CIRAM. My examination clearly indicates that the material used to sculpt the Work was modern. My conclusions are based on my examinations and analyses which have shown that the work is made of materials contaminated with modern metals and plastics. The Work could not have been modelled nearly two thousand years ago. The material identification also indicted that the Work did not have the structure of a fired ceramic object of 1650 years +_ 150 years as determined by the CIRAM laboratory (report 1117-OA-353B).”

In 2020 and 2021, prior to service of Dr Bennett’s court report, including in the lead up to service of that report, there was correspondence to which Dr Bennett and/or Pinsent Masons were party. Questions were asked of Oxford Authentication which were aimed essentially at the question of whether the two different results of CIRAM and Oxford Authentication could be reconciled. In May 2020, Greg Adamiec of Oxford Authentication expressed the view that it was possible that the “binder was an unfired material whose signal could swamp the signal of the genuinely old constituent, ground-up fired material”. In other words, Oxford Authentication was recognising the possibility that ground-up fired material may have been part of the Krodha.

On 19 March 2021, Dr Bennett asked Oxford Authentication to confirm that this conclusion could not be ruled out. Oxford Authentication then did so, in a response dated 23 March 2021. I accept that Dr Bennett’s letter of 19 March 2021 could be read as suggesting that she herself had performed an analysis from which, as she said, “it is clear that the entire Krodha is composed of previously fired, ground up ceramic”. However, I am quite sure that she had no intention to mislead Oxford Authentication, and I cannot see that she had any motive to do so. She was seeking to understand whether, and suggest a possible way in which, the two sets of results could stand together. Oxford Authentication indicated that there were indeed various possibilities. She was clearly, and in my view understandably, baffled by the suggestion in cross examination that she was seeking to mislead anyone. This correspondence with Oxford Authentication only came to light because Dr Bennett, in her court report, gave a full account of the background concerning TL testing (see paragraphs 7.2.12 – 7.2.14), and decided to exhibit the correspondence. She then went on to make it clear (paragraph 7.3, quoted above) that her conclusions were not based on the conclusion of either CIRAM or Oxford Authentication.

I do not consider that any of this causes me to regard Dr Bennett as an unreliable witness. Indeed, as with many of the criticisms of Dr Bennett which were advanced by the Defendants, it seemed to me that it was not directed at the really important points that she was making. In relation to the Krodha, the ceramic particles point was the only point on which she was cross-examined. Yet, as will be apparent, this formed no part of her reasoning as to why the piece was inauthentic.

Since I have rejected QIPCO’s case that it is sufficient for them to establish liability by proving inauthenticity, I now turn to consider the question of whether the opinions expressed by Mr Eskenazi were based on reasonable grounds. I will also consider QIPCO’s case, in relation to the Hari Hara, that JEL and Mr Eskenazi personally acted fraudulently, because the latter knew that the piece was a modern forgery or at least made statements as to antiquity not caring whether they were true and false. I start by giving a summary of the cases advanced by the parties.

The Claimants submitted that Mr Eskenazi should have undertaken appropriate inspection, research and other work necessary to establish reasonable grounds for its opinions. That process may well be arduous. Where the client is paying considerable sums for world-class museum quality pieces, proper scholarly investigations can reasonably be expected. It was not enough for a dealer, however, renowned, just to assert their credentials and trust their gut. A proper process would involve buying from a reputable source, doing all necessary scientific and art historical research, and looking at the history of the piece. It would also involve taking steps to ensure that the work had not been stolen or illegally exported. The Claimants referred in that context to: JEL’s accounts, which referred to the company’s “strict verification process” for procuring genuine objects; Mr Eskenazi’s oral evidence as to what was required; the terms of a legal complaint served on behalf of JEL in New York legal proceedings; and to the Code of Practice of the British Antique Dealers Association (“BADA”).

They submitted that despite claiming to adhere to these high standards, JEL did not carry out any proper investigations. There was no proper basis for the attributions. There was a failure properly to investigate provenance. There were no, or no substantial, attempts to obtain scientific analysis. There were many features of the pieces which should have alerted Mr Eskenazi to the fact that the objects were modern forgeries, or at least that there were real doubts as to their authenticity. JEL should have qualified its attribution if necessary. If there is a real as opposed to a fanciful doubt, the doubt must be expressed. Here, however, Mr Eskenazi in all cases gave emphatic unqualified statements as to the antiquity of the pieces.

The Claimants’ submissions identified specific features of each piece which led to the conclusion that appropriate steps were not taken and that there was no reasonable basis for the emphatic statements which were made. The Claimants submitted that reasonable grounds required each piece to be considered separately. For example, there was a relatively simple test, concerning cadmium, that could have been carried out on the Serpent Bracelet. That test would have provided a simple answer. In other cases, more work might have been required. Although JEL had produced various research papers with large numbers of illustrations, these only revealed the paucity of the research carried out and lack of proper analysis. In relation to the Hari Hara, the Claimants placed particular reliance on the trail of false or backdated documents concerning the shipment of the object and the figures stated in JEL’s accounts.

In his oral submissions, Mr Stewart referred to the fact that 4 of the pieces (the Hari Hara, marble heads and Head of a Bodhisattva) were, if genuine, very major items for which large sums of money were being taken. The fact that they had materialised, without any prior record, was very surprising. It was also very important, given the backdrop of numerous fakes. Any analysis of those pieces had to start with the fact that these objects, and indeed all the objects, were in undocumented, unverified private collections of uncertain provenance.

Mr Eskenazi did not take any steps to verify the authenticity of the pieces. For example, all that he had in relation to the marble heads was his “own ipse dixit”. He did not go to a materials scientist, or to an acknowledged expert art historian. Had he done so, he would never have been in a position to express an unqualified opinion, as evidenced by the expert evidence called in the present case. At best, Dr Nicola, if consulted, would have drawn attention to the modern features on all of the pieces, and would have said that (at least on 6 of them) he could not say whether they were authentic or not. A positive opinion would not have been expressed by an art historian: here, the Defendants had spent very significant sums of money on experts, and had not been able to call an expert art historian to express a positive opinion that 6 of the pieces were authentic, and had only been able to adduce written evidence from Dr Roveda on the 7^th.

The Defendants identified a number of principles which, they submitted, appeared from the authorities.

Attribution is inherently subjective, and it was to be expected that one expert (for example, a dealer or an auction house) may reasonably reach a different view to another: any potential attribution is likely to be arguable. All that is required is a logical basis and that the opinion is one that a reasonable expert could have reached. It may often be impossible to reach an objectively “correct” attribution. The same approach should be taken in the context of antiquities, where the question is authenticity rather than attribution to a particular artist.

An attribution can depend on various matters including an expert’s experienced eye, provenance and historical research. It is possible that neither provenance nor history gives the answer. In reaching a view, connoisseurship and expertise is of great importance. A judge cannot presume to have an expert’s eye.

Where auction houses are concerned, at least some due diligence (likely to include checking provenance and historical research) is likely to be expected, but the amount of due diligence that is reasonable will depend on the facts of the specific case. That includes the nature of the party carrying out the due diligence: for example, the standard expected of a provincial auction house is lower than that expected of an international house. Likewise a sole trader dealer should not be expected to have the same resources as an auction house.

Compliance with the required standard is to be judged by reference to the actual circumstances confronting the practitioner at the material time, rather than with the benefit of hindsight.

Against this background, the Defendants submitted that the court’s assessment of the conduct of an antiques dealer would be assisted by expert evidence: to explain what an antiquities dealer in Mr Eskenazi’s position would normally have been expected to do, what inquiries would have been expected in terms of provenance, and what scientific testing would have been expected. Here, no such expert evidence has been called by the Claimants, and the absence of such evidence puts the court in a near-impossible situation.

In this context, the evidence showed that many museums had bought pieces from Mr Eskenazi, including the Bodhisattva head in the Singapore museum. Those museums were “plainly unconcerned about provenance”, and certainly did not consider that Mr Eskenazi had taken insufficient steps to consider and authenticate provenance. The Claimants’ case therefore requires Mr Eskenazi to do something which multiple museums had not considered necessary. If reputable museums had historically been prepared to buy pieces and regularly display objects without any apparent extensive provenance, including buying works from Mr Eskenazi, it is far from obvious what standard Mr Eskenazi should be held to in the context of sales to Sheikh Hamad.

Overall, the court was therefore not in a position to draw any conclusions as to what, in 2014/2015, an antiquities dealer in Mr Eskenazi’s position should have done in terms of provenance and/or scientific/ technical testing. It was not in a position to conclude that the Defendants fell below the standard of reasonableness.

In relation to the Hari Hara, Mr Green submitted that Mr Eskenazi was both honest and reasonable. Mr Green relied upon: the absence of evidence from dealers with a similar standing to suggest that the steps taken were unreasonable; that Mr Eskenazi was purchasing an object via Mr Jewell, whom he trusted and from whom he had previously purchased; the historical research conducted by JEL prior to the sale to Sheikh Hamad; the positive views expressed by an expert, Dr Czuma, who gave his views in an e-mail, again prior to sale. In his oral submissions, Mr Green also emphasised the fact that Mr Eskenazi had the piece in his own home, and therefore knew it very well indeed by the time of the sale.

There was also here, and generally in relation to the objects, the further crucial intangible element to assessing such antique artworks, which does not lend itself well to legal scrutiny: “an eye”. The importance of having “an eye” informs the assessment of Mr Eskenazi’s reasonableness. The fact that museums have regularly bought from him is validation of his practices and his “eye”. There was no basis for concluding that Mr Eskenazi acted unreasonably in forming a view, based on an eye trained over 50 years, that these objects were authentic.

In relation to the other objects, the Defendants relied upon broadly similar points. They had been purchased from sellers that Mr Eskenazi had known for many years and trusted. He had some, albeit limited, provenance information. But this was sufficient information, when sent to Dr Jaffer, to satisfy him. He had some scientific tests: the Adamantio report on the Head of a Bodhisattva. He produced detailed research papers, and had been living with the objects for a period of time. There could therefore be no suggestion that Mr Eskenazi had not devoted enough time to considering each object. Based on his eye, and acknowledged expertise, he considered these to be genuine objects. There was no need for Mr Eskenazi to consult an expert: he was an expert himself.

In assessing reasonableness, it was important that Professor Ahuja later expressed positive views on 5 of the objects. Mr Eskenazi was not aware of his views at the time. But the fact that those were the views of a world-renowned expert in such antiquities reflects both “the eye” of Mr Eskenazi and the reasonableness of his conclusion that the objects were authentic. Professor Ahuja’s views therefore validates Mr Eskenazi’s opinion.

In his oral closing, Mr Green identified two “red flags”. First, he submitted that the Claimants’ art historical evidence, based on an analysis of the antiquities in 2021/2022, was not addressing the duties of an antiquities dealer in 2014/2015. Those “contemporaneous” views on authenticity should not therefore influence the court’s conclusions as to what a reasonable antiquities dealer ought to have done in 2014/2015. Secondly, anything that took place after the sales, particularly long after the sales, is likely to be irrelevant.

In the present case, Mr Eskenazi gave unqualified or emphatic opinions as to the antiquity of each of the pieces. I have discussed the relevance of an opinion being unqualified in Section C3 above. Leaving aside the issue of fraud in relation to the Hari Hara, the critical question is whether Mr Eskenazi had reasonable grounds for those unqualified or emphatic statements. If there is a real (as opposed to a fanciful) doubt, an opinion will need to be qualified by appropriate words. By contrast, if a certain or definite opinion is reasonably held, then a person is not required to express doubts about it: see Thomson v Christie Manson & Woods [2005] EWCA Civ 555 at paras [155] – [157] and the judgment of the trial judge (Jack J) at paras [188] – [189] quoted at paragraph [159] above.

I do not accept the Defendants’ submission that the present case cannot fairly or properly be decided without the benefit of expert evidence from antiquities dealers as to practices in the market in 2014/2015. Neither side in the present case sought permission to adduce such evidence, and both were therefore content to proceed on the basis that such evidence was not required or not appropriate. I therefore treat with scepticism, and reject, the suggestion that this case cannot be decided in the absence of the expert evidence which the Defendants now say is essential.

Had an application been made for such evidence to be adduced, issues would likely have arisen as to whether it should be admitted at all. It is clear from the discussion in Jackson & Powell on Professional Liability 9^th edn paras 6-008 – 6-011 that there are many cases where expert evidence is not required. One category is where experts are only giving their views as to what they would have done in the particular circumstances of the case, rather than evidence as to a recognised practice within the profession. Another category is where it is not necessary to apply any particular professional expertise in order to decide whether the defendant has failed to exercise the skill and care expected of an ordinary member of his profession. In this category are the cases where the trial judge does not need to “get under the skin of a different profession” in order to assess whether or not the defendants had failed to use reasonable skill and care. If an application to adduce expert evidence from an antiquities dealer had been made in the present case, it is uncertain as to whether it would have succeeded. There might have been a healthy argument as to whether, for example, such an expert would only have been giving evidence as to what he or she would have done, rather than as to any practice within the profession. Indeed, there might have been a debate as to whether antiquities dealers were properly to be regarded, for the purposes of considering expert evidence, as a profession. As it is, no such application required determination.

In the present case, there is in my view sufficient evidence as to the matters which an antiquities dealer should have considered in the context of considering whether a piece is authentic and making statements to that effect, as well as other matters which are relevant to the conduct of business by antique dealers. That evidence was not controversial, and indeed seems both obvious and in line with the approach taken in the auctioneer cases where the courts have identified the matters relevant to the auctioneer’s work.

Thus, Mr Eskenazi was asked about the “strict verification process for all stock purchased” referred to in successive statutory accounts of JEL. He said that this was not written down anywhere, but was an automatic process. He said that it involved buying from a reputable source, doing all the necessary scientific and especially art historical research, and then looking into the history of the piece, if any. In my view, this answer neatly identified the key matters which one would expect a dealer of antiquities to focus on. Some additional colour or detail, concerning those key matters, was provided by two other sources on which Mr Eskenazi was asked questions in cross-examination.

First, he was asked about passages in a Complaint filed by JEL in New York proceedings. He confirmed that these fairly set out the nature of national ownership laws and market practice in 2013 when the item (which was the subject of the New York proceedings) was purchased. I do not accept that Defendants’ submission that this answer, or the Complaint, can be disregarded because Mr Eskenazi’s evidence about it was given at the end of a long day. It seems to me that the relevant parts of the Complaint, set out below, are unsurprising:

“[9] The international antiquities trade is increasingly subject to national and international regulation. Many “source nations” rich in antiquities, artifacts and other cultural materials (collectively “antiquities”) have enacted national ownership laws, export controls and/or other regulations governing the domestic and international markets in their antiquities.

[10] There is no registry, database or catalogue of Indian antiquities in situ in India or those traded on the international art market. Indian law requires domestic Indian dealers to register their inventory but this registry is neither comprehensive nor reliable. To avoid trading in stolen or illegally-exported antiquities, dealers such as Eskenazi, who desire to maintain a reputation in the trade for dealing only in objects that are lawfully on the market, are careful to (i) determine an object’s “provenience” (i.e., country of origin or place of origin, discovery or find-spot) and “provenance” (i.e., history of ownership and chain of title), and (ii) obtain contractual representations from the seller as to lawful provenience and provenance when they purchase an object.”

Secondly, he was also asked about the Byelaws of BADA. Membership of BADA is determined by its Council based on various criteria, including that the applicant is a fit and proper person. Members are entitled to describe themselves as members of this Association, and Mr Eskenazi did so. The Byelaws set out a number of “Principles of Conduct of the UK Art Market”. Thus, Members undertake not to purchase, sell or offer any item of property that they know has been stolen, illegally exported or illegally excavated. They agree to ask sellers to confirm in writing that they own the property concerned, and to record sellers’ details such as name and address.

In addition, each member is required to comply with the “Code of Practice for the Control of International Trading in Works of Art” to which BADA had subscribed. This Code of Practice provides that members of the UK antiques trade undertake, to the best of their ability, not to import, export or transfer the ownership of objects where they have reasonable cause to be believe that an imported object has been exported in violation of that country’s laws.

In the light of this evidence and these sources, as well as ordinary common sense, I do not consider that there is any difficulty in identifying the matters with which Mr Eskenazi should have concerned himself when giving his unqualified opinions as to authenticity.

One of the matters identified by Mr Eskenazi was doing the necessary historical research. In that regard, I do not accept Mr Green’s argument that the court should somehow disregard, or attach little weight to, the “contemporaneous” views on authenticity expressed by the Claimants’ art historians. This argument was advanced on the basis that their views post-date the dates of the sales and therefore provide no assistance on the question of what a reasonable antiquities dealer ought to have done in 2014/2015.

It seems to me, however, that the views of the Claimants’ art historians are relevant to the question of what the necessary historical research would have revealed as to the authenticity of the pieces in question. Mr Stewart accepted, and I agree, that one would not necessarily expect an antiquities dealer to do the same level of very detailed research which might be carried out by an art historian who is writing a scholarly article, nor indeed as the Claimants’ experts had carried out for the purposes of writing their reports in the present proceedings. In the passage from Thomson quoted above, the judge made the point that an auctioneer’s evaluation cannot and need not be as exhaustive as the examinations carried out for the purposes of trial. However, there can in my view be no doubt that some art historical research was required, as Mr Eskenazi accepted and indeed as reflected in the fact that JEL did produce their research papers. In the light of my conclusions in Section D above, and in the context of the giving of an unqualified opinion as to authenticity, this research required consideration of and focus on the authentic historical record.

Furthermore, there is nothing in the evidence which suggests that the evidence of the Claimants’ art historians owes anything at all to developments after the 2014/2015 period. On the contrary, their evidence was itself focused on the authentic historical record based upon excavations and discoveries during more than 100 years prior to that time.

I do not consider that the fact that Mr Eskenazi had successfully sold various pieces to museums, over many years, is of any real assistance in deciding whether there were reasonable grounds for his unqualified opinion on the pieces with which I am concerned. I am concerned with statements as to authenticity made in relation to these 7 pieces, and it is upon those pieces and their circumstances that I must focus.

I accept that there were, on the evidence, many successful sales by Mr Eskenazi to museums. But none of those sales were of the pieces with which I am concerned. Furthermore, whilst there were successful sales, it is likely that – for one reason or another – there were other proposed sales which were unsuccessful. Even in relation to successful sales, I do not know what work was carried out, either by Mr Eskenazi or the acquiring museum, into the piece that was being acquired. For example, Mr Green referred to the acquisition of the terracotta Bodhisattva head by the Singapore museum. Mr Eskenazi’s evidence was that considerable work had been carried out by that museum, and that this included TL (thermoluminescence) testing. Such testing would have been possible, because the piece was a ceramic. I have no evidence, however, as to precisely what work was done or what documents were provided. I know nothing about the expertise of those responsible for the acquisition, or how they carried out their evaluation of the potential acquisition. Overall, I do not see how other sales of different objects, as to which there is scant evidence, assists materially on the question of the reasonableness of the unqualified opinion expressed on the pieces with which I am concerned.

In view of the reliance placed by the Defendants on the reports of Professor Ahuja, it is appropriate here to start by summarising the facts concerning his involvement.

By early 2016, Sheikh Hamad was (according to his witness statement) having doubts about the objects, having already been told that the Serpent Bracelet was inauthentic. He asked Dr Jaffer to identify a scholar who could write a series of reports about them. Dr Jaffer’s evidence (according to his witness statement) was that, at this time, he had no reason to doubt the authenticity of the objects, and that at this time discussions were ongoing with two leading museums (the Met in New York and the Art Institute of Chicago) that were interested in having certain of the pieces in QIPCO’s collection on long-term loan.

Professor Ahuja was introduced by Dr Jaffer to Sheikh Hamad in March 2016, and matters were taken forward by Dr Jaffer, who asked Professor Ahuja to look at the Gandharan/ Bactrian pieces and to write reports. Dr Jaffer’s evidence (on which he was not cross-examined) was that he expected the reports to provide important context about the objects and the culture that produced them, as well as providing stylistic comparisons, but that Professor Ahuja was not specifically asked to opine on the authenticity or the condition of the objects.

Professor Ahuja then looked at the objects at their London storage facility in August 2016. Mr John Baker was told by Laura Stuart (Head of Collections Management at The Fine Art Group, which oversaw the management of QIPCO’s collection) that Professor Ahuja was going to write a report and that she believed “he will include stylistic comparisons as well as recommendations for X-raying the heads”. Mr Baker told Ms Stuart that they should indeed do x-rays as per his advice. This was not, however, a matter that was pursued at that time.

Professor Ahuja’s reports were sent by e-mail to Mr Jaffer on 19 December 2016. They contain a discussion of the historical and cultural background, with stylistic comparisons. In his evidence, Sheikh Hamad described them as polite reports, but that they were inconclusive on the question of authenticity. I think that this is fair. They do not express an opinion, one way or the other, on authenticity. There may be a number of reasons for this, including that Professor Ahuja had not been asked specifically to opine on that issue.

Sheikh Hamad referred in his evidence to statements in Professor Ahuja’s reports that some of the objects were unique and also that gems will usually have been lost or stolen over the course of time. He said that this gave him the impression that the sculptures, given their excellent condition and that all of the gems remained intact on the Heads of Dionysus and Goddess, could be “too good to be true”. Sheikh Hamad thought that, reading between the lines, Professor Ahuja was politely and diplomatically saying that something was unusual about the objects.

It seems to me that the reports are something of a mixed bag. In relation to some of the objects, particularly the Head of Dionysus and the Krodha, Professor Ahuja was struck by and commented upon the uniqueness of the objects. He also described one feature of the Head of a Bodhisattva as making it unique. This could be viewed as raising question marks as to their authenticity. In relation to other objects, for example the Frieze, the reports do not seem to me to raise the same question marks based on uniqueness.

There is also evidence as to what Professor Ahuja was saying at around the time when the reports were produced. This too presents something of a mixed bag. In August 2016, he had recommended x-raying the objects. In his covering e-mail dated 19 December 2016 to Dr Jaffer, he referred to the “six wonderful pieces” in the collection of Sheikh Hamad. He said that two of the pieces (the Head of Dionysus and the Krodha) had “tremendous potential for further study”. Professor Ahuja said that he would love to find out what adhesive had been used to affix the jewelled eyes and gemstones in the jewellery, and if there had been any conservation to secure them. He asked if “the pieces” had ever been x-rayed. It is not clear whether he was referring to all of the pieces, or just the two which had the tremendous potential for further study.

At some stage, Mr Jaffer spoke to Professor Ahuja after receiving the reports. Professor Ahuja said that it was incredibly remarkable and unlikely that after some 1,800 years or so, the garnets remained so completely in the Heads of Dionysus and Goddess, given that over such a long period one would expect most of the garnets to have fallen out. This suggested that the pieces might have been significantly restored or enhanced in recent times, as was somewhat common in Gandharan works. He said that a scientist should therefore look at them.

In a subsequent WhatsApp exchange in November 2017, when the question arose as to which reports should be given to Mr Eskenazi, Dr Jaffer told Sheikh Hamad that Professor Ahuja had been impressed with the pieces, but that he had raised questions as to the excellent condition of the pieces for example the survival of the garnets intact on the Head of Goddess. This survival was remarkable.

In early 2017, after some prompting from Sheikh Hamad, Dr Jaffer followed up with Professor Ahuja regarding the suggestion of engaging a scientist. It was Professor Ahuja’s recommendation that led to the engagement of Dr Bennett. She suggested, in May 2017, examination in the first instance under a microscope and with surface-based analytical equipment, rather than x-rays. The proposed cost of the work was relatively modest: £ 4,500 for examination under microscope and digital recording of the surface.

I will refer briefly to some of the detail of Professor Ahuja’s reports when considering the individual objects. However, I do not think that they provide any support for the Defendants’ case that their unqualified opinions as to authenticity were reasonably based. Professor Ahuja did not express an unqualified opinion on authenticity: he was not asked to do so. There are aspects of his reports, as well as his statements that the pieces were “wonderful” and that he was impressed, which indicate that he may have thought that the pieces were authentic. However, there are other aspects which point in a different direction, including his comments as to the uniqueness of some of the pieces as well as his advice – given on various occasions – that a scientist should look at the pieces. He was also clearly surprised at the condition of the pieces.

If the facts had been that Professor Ahuja had expressed an unqualified opinion on authenticity, and Mr Eskenazi had seen this at the time and taken it into account in forming his own opinion, it would be easy to see the potential relevance of Professor Ahuja’s views on the question of whether there were reasonable grounds for Mr Eskenazi’s opinion. However, those are not the present facts. Mr Eskenazi did not see Professor Ahuja’s reports until disclosure. Those reports did not express unqualified opinions, and in fact the Professor was suggesting further scientific analysis. I cannot see how, in these circumstances, Mr Eskenazi’s unqualified opinion could be reasonably based by reference to reports which (i) he did not see and (ii) did not express unqualified opinions, and where the author was recommending further analysis (which Mr Eskenazi did not carry out).

I also consider that reliance on Professor Ahuja’s reports, in the present context, is objectionable, or at least cannot be afforded any significant weight, for other reasons. Professor Ahuja has not been called as an expert witness, and he has not produced any Part 35 compliant evidence. However, the statements in his reports are in substance being relied upon as expert evidence on the question of both whether the objects were authentic and whether there were reasonable grounds for believing in their authenticity. In so doing, the Defendants thereby invite the court to reject, in material parts, the evidence of the Claimants’ experts, notwithstanding the inadequacies (as previously discussed) in the expert evidence called by the Defendants at trial.

The basis of the argument in the present “reasonable grounds” context, as I see it, is that the court should have regard to the views of Professor Ahuja, because they demonstrate that a well-qualified expert could take the view that the 5 pieces, on which he wrote reports, were authentic or at least not obviously inauthentic. Accordingly, this supports the reasonableness of Mr Eskenazi’s views.

I do not think that this can be a sound approach. Even if it were correct to say that Professor Ahuja had expressed a positive view on the issue of authenticity, that cannot in itself be a sufficient basis for saying that there were reasonable grounds for Mr Eskenazi’s opinion, or that this is the right way to approach the evidence. A very similar argument was advanced, and rejected, by Rose J in Thwaytes at [98]. The nature of the argument, and Rose J’s rejection of it, appears in the following passage:

“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 Sothebys 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 v Messenger May Baverstock [1990] 1 WLR 1009. A similar point was rejected by the House of Lords in Bolitho v City and Hackney Health Authority [1998] AC 232.

…

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 the 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.”

Here too, I have to consider the evidence and come to a conclusion as to whether there were reasonable grounds for the unqualified opinions which Mr Eskenazi expressed. This means considering whether it can be said that no reasonable leading specialist antique dealer would have concluded that these objects were ancient or expressed an unqualified opinion to that effect: see Section C4 above. In a case where I have heard from impressive art historical witnesses called by the Claimants, where the expert evidence called by the Defendants has been no match for that evidence, and where Professor Ahuja has not given any evidence or produced a compliant Part 35 report, I can attach no weight to Professor Ahuja’s views.

Two other features of Professor Ahuja’s reports should also be noted in the context of reasonableness. First, Mr Eskenazi knew much more than Professor Ahuja about the provenance of the pieces, and in relation to some of the objects this is an important matter in the context of the reasonableness of the opinions expressed. Secondly, Professor Ahuja did not opine either on the Serpent Bracelet or the Hari Hara.

Against this background, I turn to consider the question of reasonableness in relation to the various pieces.

Having admitted the inauthenticity of the Serpent Bracelet, the Defendants in their closing submissions did not advance any real argument as to why Mr Eskenazi’s unqualified opinion as to authenticity was based on reasonable grounds. In my view, there was nothing in the evidence upon which such an argument could be based.

Mr Eskenazi’s evidence was that he had no relevant expertise in jewellery such as the bracelet. It is therefore surprising that he felt able to change the dating of the bracelet from 3^rd century BC (which was the date in the invoice from FEAL) to 1^st century BCE to 1^st century CE. In his witness statement, Mr Eskenazi suggested that Mr Vigorelli, whose company sold the bracelet, was himself an expert. However, it became clear in cross-examination that, as Mr Eskenazi accepted, Mr Vigorelli was not himself a scholarly expert in Gandharan jewellery, or indeed jewellery more generally. He had handled jewellery over the years, and had a good eye. Accordingly, there was nothing in Mr Eskenazi’s expertise, or that of Mr Vigorelli, which enabled an unqualified statement to be made that this bracelet was of ancient origin.

Furthermore, as previously discussed, there is a relatively simple test for cadmium which can be carried out as a means of identifying whether gold is of modern origin. Mr Eskenazi’s evidence indicated that he had in the past had his “metalwork” tested, by Pieter Meyers in California. Given the availability of a simple test, and also in the light of Mr Eskenazi’s lack of expertise, it was in my view necessary for Mr Eskenazi to have that simple test carried out before he gave his unqualified opinion. Indeed, the thrust of Mr Eskenazi’s evidence was that he had intended to have the bracelet tested, but that there were other priorities and “the bracelet was forgotten”. I do not accept that this was a good reason for his failure to have the bracelet tested. At one stage in his evidence, Mr Eskenazi sought to ascribe some blame to Mr Colin Bowles, because he had taken a long time to mount the bracelet. But this explanation had no substance: Mr Bowles had returned it a month after JEL had bought it.

I also do not accept that the bracelet, which was (on my findings) displayed at the gallery, had been forgotten by Mr Eskenazi, and that this was the reason why it was not tested. When one considers the above matters in the round, I consider that they demonstrate a casual attitude on the part of Mr Eskenazi to the making of an unqualified representation as to the dating of this piece. I consider that Mr Eskenazi must, in the context of the present proceedings, have later recognised this. This explains his evidence (which I have rejected) as to a conversation or agreement with the Sheikh that he would take the piece back if shown to be inauthentic. Mr Eskenazi was there seeking to show that he was nevertheless acting responsibly at the time of sale, given the uncertainty that existed as to whether this piece was ancient. Even if I had accepted that evidence, it would have served to reinforce the real doubts as to authenticity that existed, and therefore the absence of reasonable grounds for the unqualified ancient attribution.

Accordingly, I do not consider that there was a reasonable basis for the unqualified opinion expressed. No reasonable leading specialist antique dealer would have expressed an unqualified opinion that this object was ancient.

In reaching this conclusion, I do not attach any significance in the present context to the points made by Mr Stewart, by reference to the BADA byelaws, that there was a failure by JEL to ask the seller to confirm in writing that it owned the property concerned, and to record sellers’ details such as name and address.

As far as the latter requirement is concerned, there was an invoice from FEAL which identified the name and address of the seller. It is true that there is nothing in the documentation which gives the name and address of Mr Vigorelli, the apparent owner of FEAL. However, I do not read the byelaws as requiring the name and address of the owners of companies with whom the buyer is dealing. In any event, I see no reason to doubt Mr Eskenazi’s evidence that he had dealt with Mr Vigorelli for many years and knew him well.

As for the former requirement, there was indeed no confirmation from FEAL that it owned the bracelet, or indeed any of the other 4 objects that were sold by FEAL and with which I am concerned. This is primarily relevant to title, but I accept that it has some bearing on authenticity; because if there is uncertainty as who is the owner of the item, then that gives rise to questions relevant to provenance which is an important matter in the context of authenticity. The point is perhaps most significant in relation to the marble heads, where the evidence is unclear as to whether they were in fact owned by Mr Hubbard (with FEAL receiving a commission on the sale) or by FEAL itself. And if they were owned by FEAL itself, then there is a question as to when and where FEAL acquired them. Whilst these points can be made, it seemed to me that in relation to all the items sold by FEAL, including the Serpent Bracelet which I am considering in this section, there are far more significant points.

I will deal with these two objects together. In my view, there was nothing about these objects which enabled a person in Mr Eskenazi’s position to give, reasonably, an unqualified opinion that they were ancient. On the contrary, the available evidence indicated that it was most unlikely that they were ancient, or at best that it was very doubtful indeed (such doubts being very real, not fanciful) that they were.

First, there is the art historical evidence which I have addressed in detail in Sections G and H above. The most important evidence as to inauthenticity is not concerned with points of detail such as the size of the grapes in the Head of Dionysus. There were instead fundamental points as summarised by Professor Lo Muzio in his answer (already quoted) in cross-examination:

“This is not to deny the existence of a Dionysiac trend in Gandharan art, but it is important to identify it, identify its weight, the space it takes, because so-called Dionysiac scenes always appear in marginal spaces. We don’t have representations of Dionysus because, let’s not forget, this sculpture is identified with the God Dionysus. In Gandhara, we don’t have representations of God Dionysus, first; second, we don’t have sculptures of large size having anything to do with Dionysism in Gandhara. We don’t have marble sculpture altogether and we don’t have marble acroliths. That’s why I said that the small grapes is just a [light] remark. We have many problems with this sculpture”.

That answer was primarily directed at the Head of Dionysus, but the absence of marble sculpture and marble acroliths is significant in relation to the Head of Goddess as well. Dr Jager said in cross-examination that but for the sculptures themselves, he would not have considered that there was any possible archaeological or art historical base for the objects. This was something that Mr Eskenazi is likely to have understood, and certainly should have understood. Indeed, in the correspondence with Dr Jager prior to the proceedings, Mr Eskenazi said that there were no comparables and very little literature to draw on. A search for comparables, or any support for authenticity in the historical record, is elusive.

Accordingly, there was nothing in art history, which Mr Eskenazi accepted was a matter that needed to be researched, which enabled an unqualified statement as to ancient origin to be given. Indeed, the reverse was true: art history indicated that the pieces were not authentic, or at least that their ancient origin was very doubtful.

Secondly, and again as discussed in the Section G above, the provenance of the pieces again firmly pointed towards inauthenticity. There were, as it seemed to me, unsatisfactory features of the evidence that both pieces, together with other marble heads, were in the collection of Mr Mark Hubbard’s father and then had passed to his son. It is surprising that such a collection had never been exhibited, or even photographed as a collection. The provenance letter for the Head of Dionysus was backdated. The provenance letter for the Head of Goddess was created in March 2016, some years after the acquisition of the piece by JEL and the sale to QIPCO, in the fortuitous circumstances (according to Mr Eskenazi’s e-mail) of Mr Hubbard happening to drop into the gallery.

The Claimants submitted, however, that Mr Eskenazi’s account (if accepted) only adds weight to the absence of reasonable grounds for an unqualified opinion as to ancient origin. I agree. That account needs to be considered in the context of the evidence as to the proliferation of fakes in the market, and the absence of comparators for either of the pieces in the authentic historical record. Mr Eskenazi’s account was that these two pieces were part of a collection of 6 or 7 pieces acquired by Mr Hubbard’s father. All of these pieces were in immaculate condition. They would, if genuine, have been a quite remarkable collection, given the absence of such pieces in the historical record. It is surprising, if not astonishing, that this collection had never been exhibited, or even photographed. It is in my view even more astonishing to think that 6 or 7 genuine pieces of this kind could survive from antiquity in immaculate condition, and all find themselves in the same collection, when there is nothing in the historical record to support the survival of even a single comparable piece. (Even the Ortiz head – which does not purport to be a Gandharan “Greek” object – was alleged to have been found in many pieces, and its damaged-looking condition is very different to the pristine pieces with which I am concerned). All of these matters detracted from any reasonable basis for the unqualified opinion that Mr Eskenazi expressed.

Thirdly, and related to the previous point, there is the condition of the two pieces. Their immaculate condition is quite contrary to the way in which one would expect genuine ancient marble objects to be, after approximately 2,000 years.

Accordingly, there was in my view nothing on which Mr Eskenazi could base, reasonably, an unqualified opinion as to authenticity.

I do not accept that, against this background, any significant weight is to be attached to Mr Eskenazi’s experience in the business and “eye”. I am doubtful, on the evidence presented in this case, that “eye” is a particularly significant factor in assessing the authenticity of an ancient object. This is not a case where the question is whether a painting is to be attributed to a particular artist, in which context it has been held that “eye” is important. This case is concerned with the authenticity of ancient objects, and the evidence has persuaded me that the most important factor – in the art historical context – is knowledge of and comparison with the known and undisputed historical record. This is not to say that “eye” may not have some relevance: for example, a person with expertise in this area may well look at an object (for example the Frieze and the Head of a Bodhisattva) and be able to recognise a modern pastiche or an object catering for western taste. However, the matters which I have discussed above – art history, provenance, and appearance – are far more significant in the present context than Mr Eskenazi’s “eye”.

I also do not accept that Mr Eskenazi’s research papers into these two objects demonstrate that there was a reasonable basis for his opinion as to ancient origin. The reality is, as Mr Eskenazi recognised in his comment to Dr Jager, that there were no comparables. The comparators which were used in JEL’s research papers were based on items of unverified provenance or in private collections. They included items which JEL had itself owned or had sold. Some of the pictures were taken from Kurita, which is well-known as including many fake objects. Other “comparables”, such as objects emanating from ancient Greece or Turkey or the Vatican, were thrown in despite their obvious irrelevance. The paper on the Head of Goddess referred to acroliths, and the paper on the Head of Dionysus said that the scale of the head suggested that it belonged to an acrolithic image. However, there is a complete absence of evidence of acroliths – let alone marble acroliths – at the material time. Another theory in the research paper was that there had been the worship of Dionysus in caves in Bactria. However, Mr Eskenazi agreed in cross-examination that there was no evidence of the worship of Dionysus in caves in Bactria. The JEL research paper had given some examples to support the theory. But the first example (Tapa i Shotor) were not caves at all. The second example (Dunhuang Cave 16) had not produced a single item with anything to do with Dionysus, as Mr Eskenazi accepted in cross-examination. Ultimately, the nicely produced research papers, with lavish illustrations, cannot disguise the absence of comparables. This was why the papers did not, and could not, refer to any known, genuine documented head from the region.

Apart from his research paper, and his eye, Mr Eskenazi did not and could not rely on anything else to support his unqualified opinion. He did have an e-mail exchange with an expert, Mr Stan Czuma. Some pictures had been sent to Mr Czuma, but it is not clear what the pictures were: they were not produced as part of JEL’s disclosure. Mr Czuma commented briefly on what he described as the “Bactrian head”, and this may perhaps have been one of the objects with which I am concerned. He described it as “impressive”, but did not offer an analysis let alone a positive opinion on authenticity. Indeed, he said that it was “very difficult” to comment upon them, without seeing them in the flesh. Apart from this correspondence with Mr Czuma, there was some limited consultation with a Mr Spink in relation to the inlays, and an instruction to a Dr Neunteufel to test the marble. However, the result of the latter test was not received by JEL until after QIPCO had bought the work, and it could not therefore have affected JEL’s attribution. In any event, an analysis of the marble does not greatly assist in relation to an issue of authenticity, and it is of no weight in the light of the significant matters which I have already described.

Furthermore, in the context of a market with a proliferation of fakes, there was nothing in the research paper – or indeed in any internal or external document produced at the time – which evidences Mr Eskenazi giving consideration to the question of whether these objects were or were not genuine, and why these objects were to be regarded as being on the genuine rather than the fake side of the line.

Mr Green also relied upon the opinions expressed by Professor Ahuja. For reasons already given, I did not consider that these assisted JEL’s case in relation to any of the objects. In addition, Professor Ahuja did not know what Mr Eskenazi knew about their provenance; in particular the extraordinary circumstance that these heads formed part of a collection of 6 or 7 marble heads. His paper on the Head of Dionysus also said when a sculpture was so unique, “questions are also raised on its authenticity and it becomes necessary, therefore, to also undertake a study of comparable material for that purpose”. His description of the object as unique reflected an earlier statement that neither museum collections nor deeply researched books on art had “brought to light a classical Greek god of this importance previously”.

Accordingly, I consider that there was no reasonable basis for the unqualified opinion, as to ancient origin, which Mr Eskenazi expressed. No reasonable leading specialist antique dealer would have expressed an unqualified opinion that these objects were ancient.

I shall first address the question of whether there was a reasonable basis for the unqualified opinion, as to ancient origin, which Mr Eskenazi expressed. The answer to that question has some bearing on the further question of whether Mr Eskenazi was fraudulent. If there was a reasonable basis for the opinion, then a case of fraud would obviously face considerable difficulties. If there was no reasonable basis, then that might provide some evidence on the question of fraud. But it must be borne in mind that fraud is very different to negligence, since it would require Mr Eskenazi to have known that the Hari Hara was a modern forgery, or at least to have made statements as to its ancient origin not caring whether those statements were true.

In relation to the Hari Hara, there was no dispute that its iconography was consistent with an ancient object. However, that alone does not provide a reasonable basis for an unqualified statement that the object was ancient, bearing in mind what was unknown about the object and indeed what was known.

When this piece came to be sold, Mr Eskenazi had no real idea where it had come from. As discussed in Section E above, there was no history to the piece, and no identification of where it was found, despite its considerable importance if genuine. There were no photographs showing it standing in a Hindu temple, no record of its being excavated, nor even a record of its being in someone else’s collection 30 years before. It appeared “by magic” (Mr Stewart’s expression) from an unknown dealer in Vietnam, with nothing behind it. There is also the oddity of a Hindu statue being owned by a Buddhist monastery, if indeed that was what Mr Eskenazi was told. I am doubtful as to whether Mr Eskenazi was in fact told this, since his evidence to that effect was first given in cross-examination.

At the time of its acquisition in early 2013, and indeed at the time of the sale to QIPCO in 2015, JEL had no genuine document of provenance. On 7 February 2013, Mr Jewell sent an e-mail (to Manuella at Lapo N.A. Fine Arts Ltd which was a company owned and controlled by Mr Eskenazi and his wife) saying that he had “sold something else to Johnny, and he [i.e. Mr Eskenazi] asked for invoice to Lapo which is attached”. The invoice purports to record a sale from Mr Jewell to Lapo for a price of US$ 155,000. The invoice goes on to state:

“I purchased the piece in Singapore about 22 years ago, and since 1998 it has been in my house in London”.

This statement was clearly untrue, to Mr Eskenazi’s knowledge. Since he was the person who had asked for the invoice, I consider it probable that he would have seen it at the time. The position was therefore that Mr Eskenazi had a false document of provenance, or at best had not seen any genuine document of provenance.

Furthermore, I do not consider that there was in fact any genuine sale from Mr Jewell to Lapo at that time or at all. Mr Eskenazi’s evidence in cross-examination was inconsistent. At the start of his evidence, he said that Mr Jewell was the seller, acting as principal on his own behalf. Subsequently, however, he said that Mr Jewell was acting as his agent. This seems to me to reflect the true position, as reflected in the e-mails where Mr Jewell was asking for Mr Eskenazi’s instructions relating to the possible purchase of the Hari Hara. Whether or not Mr Jewell was principal or agent is not really an important question in the context of whether there were reasonable grounds for the unqualified opinion expressed. More important is the fact that this piece had come from a dealer in Vietnam about whom Mr Eskenazi knew nothing, and whose name he did not even know. Mr Stewart submitted that his attitude to the dealer was cavalier: he was untroubled by the fact that he did not know their identity and was uninterested by it. I agree with this submission.

All of these matters, including the complete absence of pedigree, raised serious question-marks as to the authenticity of the piece. They negated the appropriateness of an unqualified statement that the object was ancient.

So too, in my view, did some of the important features of the state of the piece which are discussed in Section E above. I accept that there may be room for argument as to whether Mr Eskenazi should have appreciated some of the points of detail made by Mr Beguin; for example the design of the sampot. However, there were other features which firmly suggested inauthenticity. The piece showed no sign of the weathering and damage that might be expected if this was indeed an ancient piece. There was also the highly polished appearance of the statue. The polish meant that the appearance of the piece was quite unlike authentic pieces which had survived from antiquity, including those which had some signs of polish. Moreover, it gave rise to questions as to why, if this was a genuine piece, it had been polished into this state at all, in circumstances in which polish should not be applied to an ancient stone statue of this kind? If the suggested answer is that there had been an inappropriate restoration, the further question arises as to why the statue was in pieces and had not been fully restored? Mr Eskenazi, however, knew nothing at all about the restoration or conservation history of the piece.

I also consider that the similarity to the Vishnu in the Ho Chi Minh City museum is a further significant point which raised serious question marks as to authenticity. There is here both the similarity of the sampot, and the surprising coincidence that two statues would both have similar parts missing.

Some time after the Hari Hara had been acquired and restored, Mr Eskenazi did send photographs of the piece to Mr Czuma. His brief response was sent by e-mail on 4 July 2013. I do not, however, accept the Defendants’ submission that this was a highly enthusiastic affirmation of the genuineness of the Hari Hara. Rather, it contains a clear warning that it was an object whose authenticity may be questioned by some people and that Mr Czuma himself “did not respond to it without any hesitation as being “above board” authentic”. He went on to say that he looked and studied it very carefully, and that the more he did so the better it looked. The e-mail therefore had a positive aspect. However, Mr Czuma appears to have been told nothing about the provenance of the piece, and he said that it was “very difficult to comment” on the pieces whose photographs he was sent, “without seeing them in the flesh”. It does not seem to me that this provided a basis for an unqualified opinion of ancient origin.

JEL did produce a research note on the piece, and this was sent to Sheikh Hamad in 2015 at the time of purchase. The research note is lavishly illustrated with photographs of the Hari Hara itself, and other Hari Haras and Hindu gods. It contains a discussion of matters such as iconographical history, regional history and religious practice. However, the question-marks against authenticity do not arise from iconography or these matters. As with other papers, there is no discussion here – or in any other JEL document – as to why this piece was considered to be authentic in a world where forgeries are rife and where Mr Eskenazi had been told that its authenticity may be questioned.

In his closing submissions, Mr Green emphasised the importance of the “crucial intangible element to assessing such antique artworks, which does not lend itself well to legal scrutiny”, namely “an eye”. He relied upon Mr Eskenazi’s “eye”, developed over many years, and here reinforced by the fact that Mr Eskenazi had the Hari Hara in his own home for some time: he was living with the piece, and therefore came to know it very well. For reasons already given, I am not persuaded that, in the context of assessing the authenticity of a piece such as the Hari Hara, an “eye” is the most important factor; particularly bearing in mind the proliferation of well-made fake objects. In my view, no amount of “eye” can provide an answer to the serious question marks raised by the matters discussed above, namely the lack of provenance, the unusual polish on the statue, its overall very good condition, and its similarity to another object. In so far as “eye” is relevant, I bear in mind that I have independent evidence from a good witness, Mr Beguin, who had no doubt that the statue was inauthentic.

Accordingly, I do not consider that there was a reasonable basis for the unqualified opinions that were expressed. No reasonable leading specialist antique dealer would have expressed an unqualified opinion that this object was ancient.

This leaves the question of whether Mr Eskenazi was fraudulent in his representations as to the ancient origin of the statue. I do not consider that he was. There is contemporaneous evidence that a substantial sum was paid for the piece. The exchanges between Mr Eskenazi and Mr Jewell in late 2012 indicate that a substantial sum was being asked for the piece: in October. Mr Jewell indicated that the price was creeping down from “200” (which I infer was USD 200,000). There is nothing in the contemporary exchanges which contains any suggestion from Mr Jewell, or indeed Mr Eskenazi, that they are not dealing with a genuine piece. In early January 2013, Mr Jewell asked whether Mr Eskenazi wanted to resurrect the negotiations: he said that he had “made a few stories to try to put them to sleep for a few months”. These exchanges seem to me to have the flavour of a genuine negotiation for a desirable piece for a reasonably substantial sum of money: all of which is consistent with Mr Eskenazi believing that he would, if successful, acquire a genuine antiquity. This is reinforced by Mr Jewell’s e-mail of 4 February 2013, where he expresses concern as to the need to “act quickly to stop the chance of it being shown more generally”. If this was understood to be a fake, then it is difficult to see why there would have been this concern that others might acquire the piece.

It is not possible for me to make a fact finding as to whether Mr Eskenazi paid more than USD 85,000 (described below) for the piece. There is very limited contemporaneous documentation which substantiates the amounts paid, and most of the documentation that does exist is unreliable. Thus, although there is an invoice from Mr Jewell to Lapo dated 6 February 2013, in the sum of USD 155,000, that invoice is in material respects untrue: it contained the false statement that Mr Jewell had purchased the piece in Singapore about 22 years ago and that it had been in his house in London since 1998. I see no reason to attach any credence to the figure of USD 155,000 stated in the invoice, particularly when it is not supported by documentary evidence of actual payment.

Similarly, the invoice and packing list provided by Mr Jewell, and sent to (amongst others) the e-mail addresses of Mr Eskenazi and Lapo in March 2013, was false. It purports to record a sale of “Garden Stone Decoration” for a price of USD 575, with the seller being the Hien Minh Gallery, Hanoi to an unidentified buyer. However, it is apparent from the prior e-mails relating to the negotiation for the purchase that the price paid for the piece was substantially in excess of USD 575. There is no doubt in my mind that this invoice, as well as the bill of lading which describes a container said to contain Garden Stone Decoration, was a subterfuge designed to avoid questions being asked by Vietnamese authorities and to allay any concern on their part that an ancient object was being illegally exported. No credence can therefore be attached to that invoice, or the description given. I have no real doubt, however, that Mr Eskenazi – despite his evidence expressing shock at the false documentation when it emerged in the disclosure process – would have known full well that some such subterfuge would have been required in order for the object to avoid the attention of the Vietnamese authorities. Furthermore, the shipping documents were copied to the same e-mail address of Mr Eskenazi as had been used for the acquisition negotiations. I consider it likely that Mr Eskenazi saw these documents at the time.

There is some contemporaneous evidence which suggests that a substantial sum was indeed paid for the piece. In his e-mail enclosing the shipping documents on 11 March 2013, Mr Jewell asked Manuella (at Lapo) to “process the balance of 75k” when the shipment arrives. On 15 March, Winnie Lam of Michelle International Transport (who was dealing with the shipment) advised that the shipment had arrived (in Hong Kong) and had been unpacked. The shipment arrived in Hong Kong on around 15 March 2013, Mr Jewell asked Manuella, since all had been delivered safe and sound, to transfer “the balance of 75k to my normal Vietnam HSBC account”.

This e-mail indicates, therefore, that a sum had previously been paid, and that a further USD 75,000 was to be paid. Mr Eskenazi’s evidence was that the sum previously paid was USD 10,000, making a total of USD 85,000. I think that it is likely that this USD 85,000 was the amount paid by Mr Eskenazi (or at least by one of the companies that he owned or controlled). That was Mr Eskenazi’s evidence – although he said that further sums were paid to Mr Jewell at a later stage, when Mr Eskenazi had seen how good the statue was. I make no findings as to the amount of further sums paid at a later stage, or indeed whether any further amounts were actually paid. I do, however, accept Mr Eskenazi’s evidence that USD 85,000 was the amount paid to Mr Jewell in around March/ April 2013. That is the amount stated in a proforma invoice dated 2 April 2013 from Lapo as consignor to Mr Neil Perry Smith (the restorer) as consignee. That invoice too had a false statement: it sought to avoid VAT by pretending that this was a temporary import for restoration purposes. However, it is likely that USD 85,000 did represent the amount which had been paid to Mr Jewell. It is unclear as to exactly how much of this money found its way to the Vietnamese dealer. But in view of the fact that he had originally been looking for USD 200,000, I think it likely that the majority of it did so.

The upshot of all of this is that Mr Eskenazi, or one of his companies, did pay out a substantial amount of money, not less than USD 85,000, for the Hari Hara. I do not think that he would have done this unless he believed that he was buying a genuine antiquity. His evidence was that this was what he thought, and on this point I accept it. It seems improbable that USD 85,000 would be the going rate for a forgery, even a good one, and there is nothing in the evidence to suggest that it was.

Much of the Claimants’ argument on fraud was based around two features of the acquisition and import story: the illegality of the export from Vietnam, and the creation of false documents concerning the chain of sales. Whilst both points raise serious concerns as to Mr Eskenazi’s integrity, I did not think that either of them assisted on the question of whether there were fraudulent statements as to the antiquity of this piece, particularly in the light of the other matters to which I have referred.

As far as illegal export is concerned, it does seem to me that – if the piece was genuine – Mr Eskenazi was involved in taking an antiquity out of Vietnam which in all likelihood it was illegal to export, and also that he knew of false documents which disguised what was happening. On my findings, however, the piece was inauthentic and therefore there was no illegal export. However, I did not see that the argument about illegal export said anything about whether or not Mr Eskenazi believed in the genuineness of the object. If, as he maintained, it was genuine, some method would have to be found for getting the object out of the country without alerting the authorities. Mr Eskenazi’s attitude was that it was more important to obtain and save ancient treasures, even if this meant disregarding export laws and indeed the principles on which members of BADA are required to operate. This attitude can certainly be heavily criticised, but it does not lead to a finding of fraud in relation to the Hari Hara.

As far as the creation of false documents concerning the chain of sales are concerned, I do not propose to discuss all the documents in detail. Mr Eskenazi was asked about them, and he accepted that at least some documents were backdated and reflected a sale that had not taken place: for example, an invoice from Mrs Eskenazi to SCL dated 13 February 2015 was not created on that date, and did not record a genuine sale at that price between those two parties. I accept that there are strong grounds for saying that this document was part of a scheme to create the false impression that JEL had acquired the Hari Hara for USD 1.8 million, thereby significantly reducing JEL’s tax liability. However, I am not concerned in the present case with questions of tax fraud, but rather with the question of whether Mr Eskenazi was fraudulent in relation to his statements to the Claimants as to the antiquity of the Hari Hara. In that context, I consider that the contemporaneous evidence discussed above, including the significant amount which was paid out (albeit far less than the USD 1.8 million shown in JEL’s accounts) is far more important than the creation of documents which suggest false accounting to HMRC.

Mr Green made some other powerful points in support of the argument that Mr Eskenazi was not fraudulent, whether the case was put on the basis of knowingly making false statements or not caring whether the statements made were false.

He pointed out that, at the time when Mr Eskenazi paid out USD 85,000, he did not have a buyer in mind, let alone one who had agreed to buy the piece. Accordingly, he would have been taking a substantial risk if he had paid out that sum knowing, or not caring, whether he was buying a forged item. I thought that there was force in that point.

He also submitted that Mr Eskenazi would not have risked the relationship with Sheikh Hamad, a dream client, by selling a known forgery to him. I thought that there was also much force in that point. Mr Eskenazi was clearly an intelligent man and he had cultivated a good relationship with Sheikh Hamad. I accept that he would not have wanted to risk his reputation or the relationship by selling a known forgery to the Sheikh. I consider that Mr Eskenazi – who had kept this object in his own home – had persuaded himself, at the outset, that he had purchased an outstanding treasure. When he sold the piece to the Sheikh, he did genuinely believe – wrongly and without reasonable grounds, for the reasons which I have given – that it was a treasure.

Accordingly, I dismiss the case of fraud against both Mr Eskenazi personally and JEL.

In Section I above, I have set out what can fairly be described as a plethora of matters relied upon by Dr Luczanits in support of his view that the Frieze was inauthentic, and was what he described as a “modern pastiche” of Gandharan art. One of the many striking points was that it was not possible to say (as Dr Jager agreed) where the Mayadevi – Prince Siddhartha’s mother, and a central part of the story – was actually depicted on the Frieze. He also referred to the improbability of a genuinely authentic schist frieze, such as this, surviving from antiquity so as to be in its current condition.

In Section I above, I have accepted the points made by Dr Luczanits in support of inauthenticity.

Dr Luczanits also said that no-one claiming expertise in Gandharan art could mistake the Frieze as authentic. Dr Luczanits was, as I have already said, an impressive witness, and the Defendants have not been able to call a reliable expert to give any contrary opinion to that expressed by Dr Luczanits. The best that they can do is to refer to Professor Ahuja’s report, and to make the point that he has expertise but did not immediately recognise the Frieze as inauthentic. However, for reasons already given, I do not attach weight to the views expressed by Professor Ahuja. I accept the evidence of Dr Luczanits on this issue.

The Defendants also relied upon the views or “eye” of Mr Eskenazi himself, including the fact that a detailed research paper was produced. In my view, however, the research paper and Mr Eskenazi’s views or “eye” do not address the plethora of reasons – including its remarkable condition – as to why this was not an authentic piece, or at least that there were very real question marks as to its authenticity. Much of the paper is addressed to an explanation of the story itself, and again it does not address the question of why, in a world of fakes, this is a genuine antiquity.

I therefore conclude that there were no reasonable grounds for the unqualified statements as to the antiquity of this piece. No reasonable leading specialist antique dealer would have expressed an unqualified opinion that this object was ancient.

In reaching that conclusion, I do not attach any significance to the Claimants’ argument that Mr Eskenazi failed to make proper enquiries as to the provenance of the piece. Mr Eskenazi gave contradictory evidence as to whether, when he bought the piece from Mr Vigorelli, he knew of Corinna Mazza’s ownership since the 1970’s; a fact to which she attested in a declaration signed November 2012. Despite that contradictory evidence, I am willing to accept that he did so: since the research paper describes the provenance as “Italian Collection, early 1970’s”.

In relation to this piece, I have accepted Dr Luczanits’ evidence that there were a large number of matters which indicated inauthenticity. These include: the exceptional condition of the piece; the difficulty in understanding its potential location; the dating of the piece to a time when pieces were generally being produced in stucco not stone; anomalies in the iconography, leading to the conclusion that the piece as a whole was extremely far-fetched; and that, to an experienced eye, this piece was aimed at Western purchasers.

In my view, these matters, taken individually and as a whole, meant that there was no reasonable basis for an unqualified opinion. However, it also seems to me that one other matter is decisive on this question. This is the improbable similarity between this piece and the terracotta Bodhisattva head in the Singapore museum. I have accepted Dr Luczanits’ evidence that the similarity should have rung an alarm bell with Mr Eskenazi, because of the improbability that a very similar piece would have been made, and indeed survived, in a completely different medium. He knew about the Singapore piece, because he had himself sold it to the museum.

Mr Eskenazi appears to have given no consideration to this matter at all. Indeed, the Singapore piece was not even referred to in the JEL research paper. Mr Eskenazi said in evidence that this was a mistake on the part of the “researcher who puts it together”. It is clear, however (as Mr Eskenazi accepted at one stage in his evidence) that there is indeed a remarkable resemblance between the two pieces. I consider, contrary to Mr Eskenazi’s evidence, that this must be an indication of copying, and hence a matter which should have raised, to an experienced dealer, a real doubt as to the authenticity of the piece.

In the present case, the alarm bell rings even stronger because of the fact that both pieces came from the same source, Mr Pilati. Accordingly, as the Claimants submitted, JEL had fortuitously alighted on a common source for two highly unusual, but incredibly similar objects – in one case (the terracotta piece) in a highly unusual material.

For reasons already given, Professor Ahuja’s report does not assist the Defendants. Furthermore, the Singapore museum piece is not referred to in his report as a comparator, and it would appear that Professor Ahuja did not take it into account when writing his paper. Mr Eskenazi did, however, know about it and I have accepted the evidence that an alarm bell should have rung.

I again conclude that there were no reasonable grounds for the unqualified statements as to the antiquity of this piece. No reasonable leading specialist antique dealer would have expressed an unqualified opinion that this object was ancient.

There were many indications of inauthenticity, as discussed in Section K above. This piece was unique, without any comparator in the accepted record. (Professor Ahuja described its iconography as “utterly unique”, and this was a piece that he said specifically should be further studied). The survival of an unfired clay object of this kind, for 1500 years, would be astonishing: it is so remote as to be fanciful. There was no religious context for this piece. Nor did Mr Eskenazi have any evidence as to its provenance. In summary, everything pointed towards inauthenticity. On any view, there were obvious question marks against the piece which precluded the giving of an unqualified opinion that this was of ancient origin.

I therefore again conclude that there were no reasonable grounds for the unqualified statements as to the antiquity of this piece. No reasonable leading specialist antique dealer would have expressed an unqualified opinion that this object was ancient.

In relation to all of the objects, the Claimants have proved their inauthenticity, and the absence of reasonable grounds for the unqualified opinion as to their ancient origin, which JEL gave. However, the Claimants have failed to establish fraud in relation to the one object (the Hari Hara) where this was alleged.

QIPCO is therefore entitled to rescind the contracts for misrepresentation, and to recover the price paid. QIPCO is also entitled to damages for breach of contract based upon breach of the admitted implied term that JEL (honestly and) reasonably held the opinion that the objects were of ancient origin. QIPCO is also entitled to damages for negligence. Each of these claims leads to an award of restitution or damages equal to the amount which QIPCO paid for the objects. I will hear submissions as to the precise form of order at a “consequentials” hearing to be held in due course.