THE key issue for determination by the Supreme Court in Ivey v Genting Casinos (UK) Ltd. [2017] UKSC 67, [2017] 3 W.L.R. 1212 was whether the crime of cheating at gambling, contrary to s. 42 of the Gambling Act 2005, requires proof of dishonesty. Even though the Court unanimously held that it did not, the Court went on to consider the appropriate test of dishonesty in both criminal and civil law and, in doing so, adopted a unified test which is essentially objective.
Ivey concerned a claim brought by a professional gambler against the defendant casino for payment of £7.7 million following a game of Punto Banco Baccarat, during which the claimant had employed a technique known as “edge-sorting”. This technique involved the claimant, having identified subtle differences in the pattern on the back of playing cards, persuading the croupier to rotate particular cards to enable him to identify whether a particular card was of high or low value and so improve his chance of winning. At no point did the claimant touch the cards. Neither was the croupier compelled to sort the cards as the claimant requested; she was willing to do so because many players adopt superstitious practices. When challenged by the casino, the claimant admitted to edge-sorting. The casino refused to pay the claimant his winnings on the ground that he had cheated in breach of an implied term in the contract between them. The claimant denied this since he considered his conduct to involve deployment of a legitimate advantage.
The Supreme Court held that the claimant had cheated and so could not recover his winnings since he had breached the implied term in the contract. It was held not to be necessary to show that the claimant had been dishonest to establish cheating, because cheating served the same function as dishonesty in establishing the illegitimacy and wrongfulness which made the conduct criminal. Determining whether the claimant had cheated simply involved making a value judgment on the facts. But how is that judgment to be made? Whilst the court was reluctant to provide a definition of cheating, the essentials were held to involve a deliberate act which was designed to gain an objectively improper advantage, given the nature, parameters and formal or informal rules of the relevant game. The claimant had cheated because this variation of Baccarat was characterised as a game of pure chance with cards delivered at random and unknowable. The claimant's conduct, albeit exhibiting skill, had increased his chances of winning by means of what was described as a carefully planned and executed sting, since he had not simply observed differences in the cards but had taken positive steps to fix the deck.
Even though it was not necessary to consider whether the claimant was also dishonest, the Supreme Court examined the meaning of dishonesty in both criminal and civil law. Lord Hughes acknowledged that dishonesty is not a defined concept but, like an elephant, will be recognised when encountered; it is not a matter of law, but a question of fact about standards of conduct. In the criminal law, if a simple judicial direction about objective dishonesty is not sufficient, dishonesty is determined by the test with two limbs identified in Ghosh [1982] Q.B. 1053: whether the defendant's conduct was objectively dishonest by the standards of reasonable honest people, and, if so, whether the defendant realised that such people would consider the conduct to be dishonest. The Supreme Court recognised that, if Ghosh applied to cheating, the claimant would not have been dishonest because, although his conduct was objectively dishonest, he would not have satisfied the second limb.
In civil law, particularly for the equitable claim for dishonestly assisting a breach of trust or fiduciary duty, the determination of dishonesty has proved to be controversial. Whilst the House of Lords in Twinsectra Ltd. v Yardley [2002] UKHL 12, [2002] 2 A.C. 164 adopted Ghosh, in two Privy Council cases starting with Royal Brunei Airlines v Tan [1995] A.C. 378, followed in various decisions of the Court of Appeal, dishonesty was said to be determined by applying the standard of the ordinary honest person, but in the light of the facts known or believed by the defendant.
In Ivey, the Supreme Court considered that directions based on Ghosh should no longer be given in criminal trials and that a unified approach to dishonesty should be adopted in the criminal and civil law, one which accords with the civil law test. Although this analysis of dishonesty was clearly obiter the Supreme Court considered it was binding. In the civil law it will follow that the conflict in the cases, which was not acknowledged by the Supreme Court, has been resolved in favour of the Privy Council's objective test. In the criminal law, strictly the Court of Appeal will still need to determine whether Ghosh has been overruled, although the clear intention of the Supreme Court that this has occurred means that trial judges can legitimately adopt the Ivey test of dishonesty on the assumption that the Court of Appeal would prefer Ivey over Ghosh, as was acknowledged in DPP v Patterson [2017] EWHC 2820 (Admin).
But can the rejection of the second limb of the Ghosh test be justified? That limb was rejected for five reasons. First, because it meant that the more warped the defendant's standard of honesty the less likely they would be convicted. Secondly, because Ghosh was premised on the assumption that criminal responsibility depends on the actual state of the defendant's mind. Thirdly, because the test was considered to be difficult for juries to apply. Fourthly, because it involved an unprincipled divergence between the tests of dishonesty in criminal and civil law. Fifthly, because its recognition departed from the law before the enactment of the Theft Act 1968 and was not compelled by later authority.
Although certain of these reasons are correct, especially the last one, others are less convincing. For example, as regards the first reason, the second limb of the Ghosh test was specifically formulated with reference to the defendant's perception of the reasonable person's standard of honesty rather their own, although it is certainly true that a defendant who considered their own conduct to be honest would be more likely to believe that reasonable people would share that belief. Further, there is no evidence from the cases that juries found the Ghosh direction difficult to apply, although it did require the jury to consider what the defendant thought they (as reasonable people) thought about defendant's conduct, which is not conceptually and evidentially straightforward. The fourth reason relating to divergence between criminal and civil law is not convincing, at least in the broad terms identified by Lord Hughes. There are many examples of situations where the criminal and civil law use the same concepts which are defined differently, and such divergence may be justified by virtue of the different functions of the criminal and civil law. Different tests of dishonesty could be justified because in the civil law dishonesty determines unacceptable conduct in order to impose liability, whereas dishonesty in the criminal law is concerned with identifying culpability, which requires consideration of the defendant's mental state. The effect of Ivey is to treat dishonesty in the criminal law as a mechanism for assessing conduct rather than culpability, albeit that the defendant's knowledge or belief about the facts is relevant to this objective assessment.
Whether the rejection of Ghosh is defensible turns on whether it is appropriate to convict a defendant of a property or economic offence without needing to prove that the defendant was aware that their conduct was objectively dishonest. In most cases this will not make any difference to the result, either because the jury would conclude that no defendant would believe that a reasonable person would characterise their conduct as dishonest, or because, in the light of the defendant's belief as to the facts, the jury would characterise the conduct as honest. So, for example, a defendant who travels on a train without paying for a ticket would be considered dishonest, save, as the Supreme Court acknowledged, if the defendant was a tourist from a country where public transport was free and who believed that the same was true in England. But there will be some cases where a defendant genuinely believes their conduct to be honest when this view would not be shared by reasonable people. Ivey is such a case. If, despite what the Supreme Court held, dishonesty was a component of the crime of cheating, did Ivey exhibit such wrongfulness in his inducement of edge-sorting that, had he been charged with the crime of cheating, he deserved to be punished? The Supreme Court considered that he did, especially because dishonesty and cheating were considered to fulfil the same function of characterising illegitimate conduct.
Ivey will impact other offences, notably theft for which it is dishonesty alone which renders conduct criminal, since, following Hinks [2001] 2 A.C. 241, appropriation does not require adverse interference with property rights contrary to the owner's wishes. When the Ghosh test of dishonesty applied, the defendant's belief that reasonable people would consider their conduct to be honest provided some protection against criminalising an otherwise legitimate interference with property rights. That safeguard has now gone, meaning that conviction for interference with property rights even though the owner of the property consents depends only on whether reasonable people would consider the conduct to be dishonest. Theft is now a crime which requires neither proof of harm nor subjective fault. Together with Hinks, Ivey has resulted in an unacceptable expansion of the criminal jurisdiction, one which is inconsistent with the civil jurisdiction and so constitutes an unprincipled divergence between criminal and civil law.
It is the Supreme Court's rejection of the assumption that criminal liability depends on the state of the defendant's mind which is likely to be the most significant implication of the decision. This is at odds with the rejection of an objective test of recklessness in favour of a subjective test in G [2003] UKHL 50, [2004] 1 A.C. 1034. The consequent dissonance between the objective test of dishonesty and the subjective test of recklessness was not acknowledged by the Supreme Court. The Supreme Court was concerned with avoiding unprincipled divergence between criminal and civil law. Perhaps the next step will be to remove unprincipled divergence between subjective and objective fault within the criminal law by moving towards the latter.
