[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]
	England and Wales Court of Appeal (Civil Division) Decisions
You are here: BAILII >> Databases >> England and Wales Court of Appeal (Civil Division) Decisions >> Volpi & Anor v Volpi [2022] EWCA Civ 464 (05 April 2022) URL: http://www.bailii.org/ew/cases/EWCA/Civ/2022/464.html Cite as: [2022] 4 WLR 48, [2022] WLR(D) 173, [2022] EWCA Civ 464

[New search] [Printable PDF version] [View ICLR summary: [2022] WLR(D) 173] [Buy ICLR report: [2022] 4 WLR 48] [Help]

IN THE COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE HIGH COURT OF JUSTICE
BUSINESS AND PROPERTY COURTS OF ENGLAND AND WALES
His Honour Judge Jarman QC
BL-2019-001956

		Royal Courts of Justice Strand, London, WC2A 2LL
		5 April 2022

This judgment was handed down remotely by circulation to the parties' representatives by email, and release to BAILII. The date and time for hand down is deemed to be 10am on 5 April 2022.

Lord Justice Lewison:

1. The issue on this appeal is whether HHJ Jarman QC was wrong to find that a sum of CHF 4 million paid by Delta Ltd to Matteo Volpi for the purchase of an apartment in Lugano; and a further CHF 2 million paid for works to that property was an interest-free loan rather than a gift. Delta is a company under the control of Gabriele Volpi, Matteo Volpi's father.

Appeals on fact

2. The appeal is therefore an appeal on a pure question of fact. The approach of an appeal court to that kind of appeal is a well-trodden path. It is unnecessary to refer in detail to the many cases that have discussed it; but the following principles are well-settled:

i) An appeal court should not interfere with the trial judge's conclusions on primary facts unless it is satisfied that he was plainly wrong.

ii) The adverb "plainly" does not refer to the degree of confidence felt by the appeal court that it would not have reached the same conclusion as the trial judge. It does not matter, with whatever degree of certainty, that the appeal court considers that it would have reached a different conclusion. What matters is whether the decision under appeal is one that no reasonable judge could have reached.

iii) An appeal court is bound, unless there is compelling reason to the contrary, to assume that the trial judge has taken the whole of the evidence into his consideration. The mere fact that a judge does not mention a specific piece of evidence does not mean that he overlooked it.

iv) The validity of the findings of fact made by a trial judge is not aptly tested by considering whether the judgment presents a balanced account of the evidence. The trial judge must of course consider all the material evidence (although it need not all be discussed in his judgment). The weight which he gives to it is however pre-eminently a matter for him.

v) An appeal court can therefore set aside a judgment on the basis that the judge failed to give the evidence a balanced consideration only if the judge's conclusion was rationally insupportable.

vi) Reasons for judgment will always be capable of having been better expressed. An appeal court should not subject a judgment to narrow textual analysis. Nor should it be picked over or construed as though it was a piece of legislation or a contract.

3. If authority for all these propositions is needed, it may be found in Piglowska v Piglowski [1999] 1 WLR 1360; McGraddie v McGraddie [2013] UKSC 58, [2013] 1 WLR 2477; Fage UK Ltd v Chobani UK Ltd [2014] EWCA Civ 5, [2014] FSR 29; Henderson v Foxworth Investments Ltd [2014] UKSC 41, [2014] 1 WLR 2600; Elliston v Glencore Services (UK) Ltd [2016] EWCA Civ 407; JSC BTA Bank v Ablyazov [2018] EWCA Civ 1176, [2019] BCC 96; Staechelin v ACLBDD Holdings Ltd [2019] EWCA Civ 817, [2019] 3 All ER 429 and Perry v Raleys Solicitors [2019] UKSC 5, [2020] AC 352.
4. Similar caution applies to appeals against a trial judge's evaluation of expert evidence: Byers v Saudi National Bank [2022] EWCA Civ 43, [2022] 4 WLR 22. It is also pertinent to recall that where facts are disputed it is for the judge, not the expert, to decide those facts. Even where expert evidence is uncontroverted, a trial judge is not bound to accept it: see, most recently, Griffiths v TUI (UK) Ltd [2021] EWCA Civ 1442, [2022] 1 WLR 973 (although the court was divided over whether it was necessary to cross-examine an expert before challenging their evidence). In a handwriting case, for example, where the issue is whether a party signed a document a judge may prefer the evidence of a witness to the opinion of a handwriting expert based on stylistic comparisons: Kingley Developments Ltd v Brudenell [2016] EWCA Civ 980.
5. Tribunals are free to draw, or to decline to draw, inferences from the facts of the case before them using their common sense. Whether any positive significance should be attached to the fact that a person has not given evidence, or to the lack of contemporaneous documentation, depends entirely on the context and particular circumstances: Royal Mail Group Ltd v Efobi [2021] UKSC 33, [2021] 1 WLR 3863.

Trial

6. The judge conducted a four-day trial, in the course of which he heard live evidence from witnesses who were cross-examined; and he also had expert evidence from document examiners on the question whether certain signatures, purportedly made by Matteo Volpi, were genuine. There was, however, little contemporaneous documentary evidence which, the judge said, had made his fact-finding task more difficult. Any trial judge will have been faced with the task of trying to do a jigsaw puzzle when some of the pieces are missing; and many of the others do not precisely fit together.
7. In evaluating the evidence adduced at trial a judge will consider not only the live evidence called before him and contemporaneous documents but also the inherent probabilities of each side's case. The judge observed at [52] that the contemporaneous documents (such as they were) and inherent likelihoods were "particularly important".

Main facts

8. I begin by setting out the main uncontested facts.
9. Gabriele Volpi has two adult sons by his ex-wife Rosaria Rota, Matteo and Simone. He acquired his considerable wealth in the oil business in Nigeria. Both sons had, at times, worked in that business. Gabriele Volpi was, as the judge found, obsessed to protect his assets for himself and his blood line and against any claim, for example from his daughters-in-law upon any subsequent breakdown of their marriage to his sons. Matteo Volpi described it as "paranoia".
10. It was common ground at trial that in late 2011, in connection with an impending move from Nigeria to Switzerland, Matteo Volpi asked Mr Cuzzocrea (a long-time business associate of and adviser to Gabriele Volpi) to help him obtain a loan in order to buy the apartment in Lugano. Mr Cuzzocrea ran the family office in Lugano at that time. At about the same time, Simone Volpi also moved from Nigeria to Switzerland.
11. That approach led to a conversation between Matteo Volpi and his father. Matteo Volpi gave evidence about that conversation. His evidence was that as a result of the conversation he understood that the property would be bought by a trust of which he was a beneficiary. He accepted in cross-examination that his father did not use the word "gift"; but maintained that he thought the property would be bought by family trusts which his father had set up. Gabriele Volpi was due to give evidence, but in the event he was unable to do so because he became ill. The judge refused to admit his statement as hearsay evidence and did not refer to it in his judgment.
12. Dr Baggi (an attorney and notary who acted for the family) told the judge that he spoke to both Gabriele Volpi and Matteo Volpi in 2011 about the proposed loan from father to son, but the judge was not satisfied that his recollection was accurate in that respect.
13. In late 2011 Dr Baggi was instructed by Mr Cuzzocrea on behalf of Mr Matteo Volpi to prepare all necessary documents for the purchase of the apartment; and this he delegated to Mrs Catta (his secretary and personal assistant), under his supervision. As Matteo Volpi was then resident in Nigeria, permission had to be sought from the First Instance Authority of the District of Lugano to make the purchase. The permission was granted, subject to appeal, on 8 February 2012, and the permission noted that Matteo Volpi was represented by Dr Baggi.
14. The next day Matteo Volpi executed a power of attorney in favour of Mr Cuzzocrea to sign on his behalf the sale and purchase deed in respect of the apartment. Those documents were drafted by Mrs Catta under the supervision of Dr Baggi. The same day Mr Cuzzocrea executed the deed, which recorded the purchase price of CHF 4 million, to which the power of attorney was then attached. An entry was made in the land register that Dr Baggi was irrevocably instructed by the parties to register the deed as soon as the time for appealing the permission had passed.
15. The deposit of CHF 400,000 was sent by Delta to Dr Baggi's firm in two tranches by 13 February 2012. On 20 February Dr Baggi emailed Mr Cuzzocrea saying that he was sending the draft of a loan agreement between Gabriele Volpi and Matteo Volpi and attached an unsigned draft in the sum of CHF 4 million. By 28 February 2012 Gabriele Volpi had signed that agreement. The balance of the purchase price was transferred to Dr Baggi's firm by Delta on 2 March.
16. On 4 April 2012 Dr Baggi emailed Mr Cuzzocrea saying he was sending two drafts of the loan agreements "between father and sons" and asked him to check if they were "ok". The reference to "sons" (plural) was a reference to both Matteo and Simone. He attached a draft of an unsigned loan agreement between Gabriele Volpi and Matteo Volpi, this time in the sum of CHF 5 million, and another unsigned loan agreement between Gabriele Volpi and his other son Simone in the sum of CHF 3 million. The former was sometime later signed by Gabriele Volpi.
17. At the time of purchase, the apartment was incomplete. In the summer of 2012, Matteo Volpi and his family went on holiday to Italy. In late August 2012 they moved into rented accommodation in Switzerland whilst the works to the apartment were completed. He was at this time still working for his father's business in Nigeria, and travelling back and forth between that country and Switzerland.
18. On 21 August Mrs Catta emailed Mr Cuzzocrea saying that she had been instructed by Dr Baggi to issue "mortgage notes on the properties." He emailed her shortly beforehand, copying in two members of his staff including Ivana Sala, saying that they would inform her shortly what amounts were to be stated in the mortgages and that the power of attorney (or "procura") would be granted "to Ivana."
19. The mortgages which were in the event executed are governed by the Swiss Civil Code. Section 3 of Chapter 3 deals with "mortgage certificates on paper", by which the bearer or a specific person may be named as the creditor under the paper mortgage certificate. It must be registered with the land register to be valid. Where the debt secured is discharged the debtor may request the creditor to hand over the paper mortgage certificate without cancelling it; and it may be used again to secure a new debt. It can be passed by the creditor to another creditor to secure any borrowing of the owner of the charged property. Such a mortgage was referred to in some of the contemporaneous documentation as a bearer mortgage; and that is how the judge described it in his judgment.
20. The next day, 22 August 2012, Mr Cuzzocrea emailed Mrs Catta again, copying in Dr Baggi, saying that the bearer mortgages needed to be "Matteo 6,000,000 Simone 4,000,000." He attached to that email two loan agreements, bearing what appeared to be the signature of Gabriele Volpi, and that of Matteo and Simone Volpi respectively. In the case of Matteo Volpi the loan agreement recited his confirmation that he had received several loans from his father amounting to CHF 6 million. The document purporting to evidence that loan consisted of three pages, each page bearing signatures of the parties. The unchallenged metadata demonstrated that it was a contemporaneous document. The authenticity of Matteo Volpi's signatures was in issue. The authenticity of Gabriele Volpi's signatures was not. Nor was there any issue about the authenticity of Simone Volpi's signature on the loan agreement applicable to him.
21. The loan agreement recited (among other things) Matteo Volpi's confirmation that he had received several loans from his father for a total of CHF 6 million; and the body of the document contained his confirmation that that sum had been made available to him.
22. On the same day Mrs Catta signed bearer mortgages on behalf of Matteo and Simone Volpi over their respective properties. She purported to do so pursuant to two powers of attorney (procuras), each bearing the date 14 August 2012, in her favour and purportedly signed by each of them respectively. The authenticity of Matteo Volpi's signature on that document was also in issue.
23. On 22 August, copies of the two bearer mortgages and the corresponding powers of attorney were stamped with the notary stamp of Dr Baggi and signed by him as authentic copies.
24. The next day Mrs Catta emailed Mr Cuzzocrea, copying in Dr Baggi, attaching scanned copies of the two bearer mortgages. Dr Baggi confirmed in an email on 27 August to Mr Cuzzocrea, copying in Mrs Catta, that he would send the authenticated copies of the bearer mortgages the next day by messenger. Next day by letter he confirmed that that is what he had done.
25. After the works to the apartment were completed, Matteo Volpi and his family continued to live in the property until the summer of 2017 when he decided to sell and move to the UK. In the meantime, in late 2013 or early 2014 Mrs Catta handed the bearer mortgage to him. There was no conversation about it at the time. He was subsequently told that he could use the mortgage as security for a loan and it seems that he did so. The certificate constituting the bearer mortgage was not in evidence before the judge; but it was accepted that it would have stated that it secured borrowing of a sum up to CHF 6 million.
26. The notary public instructed to prepare the deed of sale in 2017 emailed Matteo Volpi on 14 September referring to the bearer mortgage in the sum of CHF 6 million and asking whether the apartment was free of debt or whether a bank had financed him. He forwarded this to Mrs Catta. She emailed him on 19 September forwarding the draft deed of sale, which provided that the bearer mortgage would be delivered to the purchaser without claim, and said that he could talk to the notary public about a possible reimbursement of the costs of the bearer mortgage. On the 25 September, she emailed him again referring to a reimbursement of CHF 10,000 for such costs. On 2 November she confirmed to him by email that the bearer mortgage had been delivered to the notary public on behalf of the purchaser.
27. By 2018 relations between Mr Matteo Volpi and his father had broken down. On 10 July 2019 Gabriele Volpi and Delta served a statutory demand on his son for repayment of a CHF 4 million loan which it was said was recorded in a loan agreement. The current claim was begun on 13 November 2019. The pleaded claim eventually claimed repayment of a loan of CHF 6 million.

The evidence

28. Because of the dispute about the authenticity of Mr Matteo Volpi's signatures both sides instructed experts: Mr Stockton for Delta and Gabriele Volpi and Ms Radley for Matteo Volpi. One of the grounds of appeal is that the judge misunderstood or mischaracterised that evidence.
29. Mr Stockton's report set out his methodology. He also set out a glossary of the terms that he used. He used a scale from 1 to 7; the first being very strong evidence that the questioned signature was genuine and 7 being very strong evidence that it was not. He explained that his opinion was based on a visual comparison between signatures accepted as genuine and those which were disputed. He explained that there were a number of tests that he was unable to carry out. In particular, the questioned documents were themselves copies, so that he was unable to examine a "wet ink" signature. In addition, Matteo Volpi's signature was short and simple which made it easier to simulate. In considering the signatures on the loan agreement, Mr Stockton noted that the signature on page 3 was similar in overall style and appearance to the reference signatures, with no significant differences; and that it appeared to have been fluently executed with no signs of hesitation. Because the signature was relatively simple the evidence that it was genuine was "not strong" but "moderate". "Moderate" is on the positive side of the scale in favour of authenticity. In relation to the other two signatures on that agreement, he found that the evidence was "inconclusive". Inconclusive means that he was unable to express an opinion.
30. Turning to the signature on the power of attorney, he concluded that there was "moderate" evidence that the signature was not genuine.
31. Ms Radley used the same methods as Mr Stockton, and described similar shortcomings in the material upon which she was able to work. But her report was more guarded. She considered that the signature on page 3 of the loan agreement corresponded well with the known signatures, and that there were no significant differences between them. There were some unusual features in the signature on page 2 of the loan agreement. She was unable to say whether the observed differences were accidental variations in genuine signatures or signs of forgery. Overall she regarded the evidence about those signatures as inconclusive. The reason why she formed that opinion was, she said, "due to the restrictions encountered in my examination". She went on to explain that "inconclusive" meant that the signatures straddled the 50-50 point. It is important to note, however, that neither expert expressed the positive opinion that the other two signatures on the loan agreement were not genuine.
32. So far as the power of attorney or procura was concerned, Ms Radley considered that there was "limited positive evidence" that it was a forgery. On Ms Radley's scale "limited positive evidence" is between "moderate" and "inconclusive" on Mr Stockton's scale.
33. In their joint statement, made after the exchange of reports, the experts agreed that it was "more probable than not" that the signature on the power of attorney was a simulation.
34. It is important to note that the experts' opinions were not based on scientific analysis (such as testing the paper for indentations or pencil guidelines or examining inks to see whether the same or different inks had been used). Moreover their opinions were based on copy documents only. Expert evidence of that kind can be no more than inferential.
35. The judge summarised the expert evidence at [27] and [28] and in my view his summary was accurate.
36. Both Dr Baggi and Ms Catta were called as witnesses and cross-examined. The judge found that Mrs Catta was an honest witness; although he was more cautious about the evidence of Dr Baggi. Mrs Catta gave evidence that she had been told by Dr Baggi and by Mr Cuzzocrea (who contacted her directly) in 2011 that Gabriele Volpi would lend Mr Matteo Volpi money to purchase the property. That evidence was not challenged.
37. Matteo Volpi also gave evidence before the judge and was cross-examined. The judge found parts of his evidence to be unsatisfactory in a number of respects; and he was critical about Matteo Volpi's lack of disclosure of contemporaneous emails. One matter that particularly troubled the judge was Matteo Volpi's attempt to explain why his father would have wanted to make a substantial gift to him; as well as the form that the transactions took, especially the bearer mortgage. It is, I think, worth quoting what the judge said about that:

"[49] Although Matteo Volpi said he thought originally that the apartment would be purchased by the family trusts of which he was a beneficiary, he accepted that he saw the purchase deed at the time which showed the purchase in his own name. He says when he saw the bearer mortgage on the apartment, which he did much later, he assumed this was his father's way of seeking to protect family assets. As to whether he believed in the meantime that his father was seeking to protect family assets, he gave these explanations:

"I told Cuzzocrea we were married under separation of assets in our marriage certificate to cut him off, because the truth is that I didn't want to protect my own assets against my wife, I don't think it's fair, and considering what happened to my mother, I think I was right…since I told Cuzzocrea that our matrimonial law was in separation of assets as far as my father was concerned, it would be satisfied."

[50] His reference to what happened to his mother relates to her divorce from his father. In my judgment these explanations were unsatisfactory. If his father was very concerned about, or obsessed with, asset protection, it is unlikely that he would make a gift to him of the purchase price of the apartment on the basis of an oral assurance given to Mr Cuzzocrea that this would be achieved by a separation agreement which he claimed he had with his wife, but which he did not in fact have."

38. It is important to note that what the judge was dealing with in that passage was Matteo Volpi's state of mind not only after he saw the bearer mortgage but also in the intervening period between the time when he saw that the purchase had been made in his own name and the time when he saw the bearer mortgage.
39. In addition, Matteo Volpi did not assert in evidence that his father had actually told him that the monies were to be a gift. He said that he had assumed that the property would be bought by a family trust of which he was "a" (not "the") beneficiary. In fact his pleaded defence asserted that the family trusts were discretionary trusts under which he was one of the class of beneficiaries. A purchase by trustees of a discretionary trust could not have amounted to a gift to Matteo Volpi personally. He subsequently found out that the property had been put into his name. Nor in his witness statement did he positively say that he had not signed the power of attorney or procura which led to the creation of the bearer mortgage. His evidence was that he could not say categorically whether or not he signed it because he had previously signed powers of attorney: he simply could not recall.

The judge's consideration of the evidence

40. The judge set out the issue he had to decide in paragraph [1] of his judgment; namely whether the funding of the purchase of the property was by way of an interest free loan or a gift. The second issue he described was whether a further CHF 2 million was advanced to Matteo Volpi by way of gift.
41. Having set out the background, and his evaluation of the witnesses, the judge then turned to consider a number of strands in the evidence (which he called "factors") on which Gabriele Volpi and Delta relied in order to establish that the money was a loan rather than a gift. He weighed each of them. I should particularly refer to what he said about the signatures on the loan agreement:

"[57] Fifth, the expert evidence suggests that the signature on the final page of the loan agreement is likely to be genuine. Although Ms Radley took a more cautious approach because of the ease of simulation, she accepted that in respect of that signature, there were no significant differences to known signatures. In my judgment the preponderance of the expert evidence gives a fairly strong indication that this signature is genuine. This conclusion is not detracted from significantly by the unchallenged evidence of Matteo Volpi that he would regularly sign blank pieces of paper at the request of Mr Cuzzocrea. The signatures on each page of the loan agreement, including in particular the third signature which appropriately follows a few lines of text, fit neatly with the text on each page."

42. He then considered the factors on which Matteo Volpi relied to show that the money was a gift, and weighed those. One of them was that the claim had originally been for CHF 4 million, rather than CHF 6 million. But the judge said that the increase in amount was "not surprising" given that works to the property were carried out; and he noted that no documentation was put before him showing the source of that extra finding. His overall conclusion was expressed thus:

"[64] Weighing up these factors for and against, in my judgment the balance of probability tips firmly on the side of a loan. The two most weighty factors are, first, that all arrangements were put in place for a loan agreement to be signed by each of his sons in the context that Gabriele Volpi was at the time obsessed with protecting assets against possible claims by his daughters-in-law. The second is that the preponderance of the expert evidence suggests that it is likely that the loan agreement was signed by Matteo Volpi."

The appeal

43. Mr Beltrami QC, for Matteo Volpi, submits that there was no oral or contemporaneous documentary evidence that the parties ever agreed a loan or that Matteo Volpi signed the purported loan agreement. That left only the expert evidence; but this was inconclusive and could never have satisfied the burden of proof.
44. We were not addressed in any detail on the burden of proof or, indeed, on what was required to be proved and by whom. Nor, it seems, was the judge, although there was a brief mention in Matteo Volpi's opening submissions at trial that "the presumption of advancement is clearly engaged".
45. But although the case was not argued in this way, I think it is appropriate for me to set out my understanding of the correct legal framework. The relevant events took place in Switzerland. But no evidence of Swiss law was called; so it was assumed that in all material respects Swiss law was the same as English law. Under normal circumstances in English law, where A receives money from B the money is prima facie repayable unless B can establish that the money was a gift: Seldon v Davidson [1968] 1 WLR 1083. But in a case like this, the position is complicated by what remains of the presumption of advancement as between father and child. That is the presumption to which Matteo Volpi's opening submissions referred. The modern law "is increasingly unenthusiastic about the presumption, even in relationships where it does apply": Stack v Dowden [2007] UKHL 17, [2007] 2 AC 432 at [117]. Even where the presumption applies, it is a weak presumption as between a father and his adult child, in a case in which the adult child is not financially dependent on the father: Laskar v Laskar [2008] EWCA Civ 347, [2008] 1 WLR 2695 at [20]. The position is yet further complicated in this case because the money came from Delta rather than from Gabriele Volpi personally. Although the judge found that Delta was controlled by Gabriele Volpi, the fact that the money came from a company rather than directly from a parent seems to me to weaken any presumption of advancement.
46. This species of presumption is a factual presumption; not a legal one: Pettitt v Pettitt [1970] AC 777 at 823 (Lord Diplock). It is "no more than a consensus of judicial opinion disclosed by reported cases as to the most likely inference of fact to be drawn in the absence of any evidence to the contrary." It would not often happen that when evidence had been given, the presumption would today have any decisive effect: ibid at 811 (Lord Hodson).
47. Where the presumption applies, it is rebuttable. The important point is that in deciding whether the presumption has been rebutted, what the court looks for is the subjective intention of the putative donor: Lavelle v Lavelle [2004] EWCA Civ 223, [2004] 2 FCR 418 at [18]. The intention of the putative donor is highly relevant, and in most cases will be determinative: Meisels v Lichtman [2008] EWHC 661 (QB) at [71]; Scott v Bridge [2020] EWHC 3116 (Ch) at [121]. Thus the question is not whether putative donor and donee have agreed that the money was a loan, but whether the putative donor had an intention to make a gift. There are no rigid rules about what evidence is admissible to decide that question: Lavelle at [18].

Was the judge's evaluation wrong?

48. The starting point is that what began the whole sequence of events was Matteo Volpi's approach to Mr Cuzzocrea to help him obtain a loan. Then there is the evidence that Dr Baggi was instructed by Mr Cuzzocrea on behalf of Matteo Volpi to prepare all necessary documents for the purchase of the apartment. The judge appears to have accepted that evidence. Mrs Catta gave evidence that she had been told by Dr Baggi and by Mr Cuzzocrea in 2011 that Gabriele Volpi would lend Matteo Volpi money to purchase the property. Although hearsay, that was admissible evidence, and even though the judge did not specifically mention it, it cannot be assumed that he overlooked it. Next there is the draft loan agreement which the judge found had been signed by Gabriele Volpi by 28 February 2012. That is in itself contemporaneous evidence of Gabriele Volpi's intention. It is his intention that was important. The loan agreement in issue was sent by email (now bearing or purporting to bear both men's signatures) on 22 August 2012. That document, too, is evidence. The fact that it was signed by Gabriele Volpi was not challenged. It, too, is direct contemporaneous evidence of Gabriele Volpi's intention, whether Matteo Volpi signed the agreement or not. It is also of some significance (although again not specifically mentioned by the judge in his discussion) that he had found that the provenance of the signed loan agreement was as an attachment to an email to Mrs Catta from Mr Cuzzocrea (who had originally been tasked by Matteo Volpi with arranging a loan). Matteo Volpi confirmed in his evidence that Mr Cuzzocrea did "everything" for him in respect of the purchase. In addition to that, the family trusts which, according to Matteo Volpi, were to buy the properties were discretionary trusts; so that a purchase by any such trust would not have resulted in Matteo Volpi himself having become entitled to the purchased asset. Thus Matteo Volpi's contemporaneous understanding that the property was to be purchased by the trustees of a discretionary trust is itself inconsistent with the making of an immediate outright gift to him personally. According to Matteo Volpi, the reason why the property was put into his name was because of a complication in Swiss tax law. That, too, is a pointer away from an outright gift.
49. Next there is the bearer mortgage. The ostensible purpose of a mortgage is to secure a loan. Matteo Volpi knew about the existence of that mortgage (at the latest) in 2013 or 2014 and indeed relied on it himself in order to raise money. He made no suggestion at the time that it was not genuine. When he decided to move to the UK in the summer of 2017, relations between father and son were still good. He did not question the authenticity of the bearer mortgage at that time either.
50. The judge was also entitled to draw inferences from the lack of any disclosure of emails passing between Matteo Volpi and Mr Cuzzocrea. At the time of the relevant events Matteo Volpi was in Nigeria and Mr Cuzzocrea was in Switzerland. Matteo Volpi's evidence about that was that there were many communications between them. He said that he could not remember how many emails there were and then said that there were "not many emails". The judge observed:

"Even if this answer is to be taken to refer to his communications generally with Mr Cuzzocrea, it would be surprising, given that he was then based in Nigeria although travelling to Switzerland from time to time, and Mr Cuzzocrea's office was based in Lugano, that this did not also apply to the purchase of the apartment. He did not, in my judgment, deal satisfactorily with the reason why not one such email has been disclosed."

51. The judge noted the fact that neither side had called Mr Cuzzocrea. He considered that that fact was more telling against the claimants than against Mr Matteo Volpi. Nevertheless, it is a fact that he took into account in weighing the evidence; so it cannot be said that he overlooked it.
52. In addition to these specific pieces of evidence, there were the inherent probabilities. All the documents were created at a time when relations between father and son were good, and Matteo Volpi was still working in the family oil business. Why, one asks, would Gabriele Volpi have created a false paper trail at a time when there was no apparent reason to do so?
53. In the light of Gabriele Volpi's obsession with protecting assets, the judge was entitled to find that an outright gift was improbable. In addition, at more or less the same time, Gabriele Volpi was making arrangements for the accommodation of his other son, Simone. The judge had said at [64] that documentation was put in place for each of the two sons. The judge inferred (as he was entitled to do) that Gabriele Volpi would have treated his sons equally, and that there was no suggestion that monies used in the purchase of a home for Simone were a gift.
54. As far as the expert evidence was concerned, where a document is produced from reputable custody (as the loan agreement was) then it seems to me that if it is alleged that a signature on it is forged, the burden of proof in that respect lies on the person who asserts that it is a forgery. Indeed, CPR Part 32.20 provides expressly that a notarial act or instrument may be received as duly authenticated unless the contrary is proved.
55. The judge was careful to distinguish between the different signatures on the loan agreement. He referred to the signature on page 3 of the loan agreement. Mr Stockton considered that there was moderate evidence (i.e. more probable than not) that it was genuine; and Ms Radley considered that it corresponded well with the known signatures, and that there were no significant differences between them. The judge was entitled to find that, as regards the signature on the third page of the loan agreement, the probability was that it was genuine; and that in the light of that, since neither expert had said that the other two signatures were not genuine, it was more probable than not that the other signatures on that document were also genuine.
56. The expert evidence, taken at its highest in Matteo Volpi's favour, did not positively assert that any of the signatures on the loan agreement had been forged.
57. But in any event, whether Matteo Volpi signed the loan agreement is of secondary importance. It would have mattered more if it had been alleged that he had agreed to pay interest on the loan. But the case against him was always that the loan was interest-free. So the real question is whether Gabriele Volpi and/or Delta intended to make a gift. There was ample evidence to support the judge's conclusion that they did not.
58. The judge also distinguished between the signatures on the loan agreement and the signature on the procura. He was not satisfied that the latter was genuine; but said that it did not affect his conclusion about the loan. It is important to note that the judge did not find that the signature had been forged; merely that he was unable to conclude that it was genuine. It is also important that Matteo Volpi did not positively assert that he had not signed the procura. He neither pleaded nor asserted in evidence that the signature was a forgery. His counsel did, however, cross-examine Mrs Catta and Dr Baggi to the effect that they had dishonestly forged the procura. The judge dealt with that in some detail and rejected the allegations made against them.
59. Mr Beltrami also submitted that the judge's approach of setting out the "factors" which he said might indicate a loan rather than a gift and vice versa; and ascribing weight to them without making any actual factual findings, was tantamount to exercising a discretion. He submitted that this was an erroneous approach and it led to an unjustified conclusion. In my judgment that mischaracterises what the judge did. On his way to reaching a conclusion on the issue in the case a judge is not required to make individual findings of fact on each piece of evidence. Each of the factors that the judge considered was either based on a finding of fact; or it was an argument by one or other party based on the lack of evidence on a particular point; or it was the judge's assessment of the inherent probabilities. Any judge, called upon to make findings of fact in a contested case will have evidence pointing in each direction and it is a commonplace of the exercise of the judicial function that he will weigh the competing considerations against each other.
60. Mr Beltrami also complained that the judge ignored evidence. In substance all that that submission amounted to was that the judge should have accepted Matteo Volpi's evidence. Whether to accept the evidence of a witness is a matter for the trial judge; and is particularly difficult to upset on appeal. It is even more difficult where, as here, the judge found Matteo Volpi to have been an unreliable witness.
61. There were three specific points that Mr Beltrami concentrated on that I need to deal with. First, he criticised the judge for saying at [55] that the contemporaneous documentation showed that "all necessary arrangements were put in place for Matteo Volpi to sign a loan agreement". He argued that, on the contrary, all arrangements were not in place. This submission, in my judgment, concentrates too much on the judge's use of the word "arrangements." It is the sort of narrow textual analysis which the cases deprecate. If the judge's finding is reworded as all necessary documents were put in place, then it is amply borne out by the contemporaneous documents. The very fact that all necessary documents were put in place was a fact to which the judge was entitled to give weight. What would have been the point of creating the documents if they were not to be used for their ostensible purpose?
62. Second, Mr Beltrami criticised the judge for saying at [57] and [64] that the "preponderance of expert evidence" suggested that the signature on the final page of the loan agreement was genuine. Once again, I consider that this concentrates too much on the judge's precise choice of words. The fact is that Mr Stockton's view was that it was more likely than not that that signature was genuine; and Ms Radley's opinion that the evidence was "inconclusive" straddled the 50:50 point. So a finding that the balance of evidence pointed in favour of the genuineness of the signature was a finding open to the judge. It did not, as Mr Beltrami tried to suggest, require him to reject the evidence of Ms Radley. His finding was consistent with her view.
63. Third, Mr Beltrami attacked the judge's discussion of the increase in the loan from CHF 4 million to CHF 6 million. First, he said that there was no evidence that the works had cost CHF 2 million. But in fact that was the evidence given by Matteo Volpi himself in the course of his cross-examination. Next, he said that there was no evidence to show where the CHF 2 million had come from. But at this stage in his consideration of the evidence, the judge was dealing with the question whether the loan agreement was genuine or not. That loan agreement itself referred to a loan of CHF 6 million. If genuine, the loan agreement would have been an acknowledgement of a debt of CHF 6 million. Although Matteo Volpi claimed to have paid for the work to the property himself, he disclosed no documentary evidence (whether bank statements or receipts or, indeed, anything) to corroborate his evidence. It would have been an odd coincidence if the amount of the loan had increased by the same amount that had been spent on the property if the expenditure had been made by Matteo Volpi out of his own funds. In my judgment the judge was entitled to draw inferences from the lack of any documentary evidence to back Matteo Volpi's assertion. In addition, this was a factor that the judge took into account as one of the list of factors that pointed against his ultimate conclusion that Matteo Volpi signed the loan agreement. The weight of that factor was a matter for him.
64. Finally, Mr Beltrami attacked the way that the judge dealt with the signature on the procura. The experts agreed that it was more likely than not that the signature was a simulation. But balanced against that was the judge's consideration of other factors which pointed towards the genuineness of the document. In short, the judge's view was that the same factors which had led him to conclude that the loan agreement was genuine also applied to the procura, with the exception of the expert evidence. The experts were looking only at one piece of evidence, whereas the judge had to evaluate all the evidence, both direct and circumstantial. In addition, as I have said, the experts were hampered in their opinions because of the limited material they had to work on; and their opinions were based on stylistic inferences rather than scientific tests. Ultimately the judge was unable to reach a conclusion one way or the other. He was fully alive to the relevance of Matteo Volpi's argument that if the procura had been forged it cast doubt on the genuineness of the loan agreement. But he rejected the specific allegations of forgery that had been raised; and no alternative case had been raised before him. It must also not be forgotten that the purpose of the procura was to pave the way for the bearer mortgage in the sum of CHF 6 million, which Matteo Volpi knew about and indeed exploited for his own benefit. In the judge's view the evidence about the procura was not strong enough to displace the conclusion he had reached about the loan agreement. Another judge might have taken a different view, but that does not mean that the judge's conclusion was not rationally supportable.

Conclusion and result

65. This appeal demonstrates many features of appeals against findings of fact:

i) It seeks to retry the case afresh.

ii) It rests on a selection of evidence rather than the whole of the evidence that the judge heard (what I have elsewhere called "island hopping").

iii) It seeks to persuade an appeal court to form its own evaluation of the reliability of witness evidence when that is the quintessential function of the trial judge who has seen and heard the witnesses.

iv) It seeks to persuade the appeal court to reattribute weight to the different strands of evidence.

v) It concentrates on particular verbal expressions that the judge used rather than engaging with the substance of his findings.

66. I re-emphasise the point that it is not for an appeal court to come to an independent conclusion as a result of its own consideration of the evidence. Whether we would have reached the same conclusion as the judge is not the point; although I am far from saying that I would not have done. The question for us is whether the judge's finding that the money was a loan rather than a gift was rationally insupportable. In my judgment it was not. In my judgment the judge was entitled to reach the conclusion that he did. I would dismiss the appeal.

Lord Justice Males:

67. I agree.

Lord Justice Snowden:

68. I also agree.